Saturday, May 01, 2010

Publication Cut-off

The German Research Foundation (DFG) has taken an important and overdue step. To limit their applicant's attempts to blind the reviewer with publications, from July 1st 2010 on a maximum of 5 publications can be listed in the CV. In addition to this, only papers that are already published can be listed. Previously, it was possible to also list papers that are submitted, but not yet published. The change in this policy is apparently a reaction to an instance last year in which applicants (in the area of biodiversity) invented publications. (More details on the new regulation here.) It remains unclear to me whether a paper on the arxiv counts as published or unpublished.

With this decision, the DFG is clearly signaling that it's quality that matters, and not quantity. Or at least that's what should matter for their referees. Another reason for the change is that other countries have similar restrictions. The NSF for example also has a limit of 5 publications relevant for the project, and the NIH 15.

Matthias Kleiner, President of the DFG said
“With this we want to show: For us it is the content that matters for the judgement and the support of science.”

And he bemoans that today
“The first question is often not anymore what somebody's research is but where and how much he has published.”

(As quoted in Physik Journal, April 2010, my translation).

The DFG is the funding source for scientific research in Germany. Not the only one, but without doubt the most important one. This decision will therefore have a large impact. The impact however is limited in that the other major reason publication numbers are ever increasing is that hiring committees pay attention to these numbers - or at least are believed to pay attention, which is sufficient already to create the effect. The President of the German Higher Education Association (DHV*), Bernhard Kempen, comments
“To assess a candidate's qualification in a hiring process it should also be solely the content of provided publications, not their number, that is decisive for an appointment.”
(as quoted here, my translation.)

Since I have written many times that it hinders scientific progress when selection criteria set incentives for researchers to strive for secondary goals (many publications) instead of primary goals (good research), it should be clear that I welcome this decision by the DFG.

* DHV stands for Deutscher Hochschulverband. The literal translation of the German word "Hochschule" is "high school" but the meaning is different. "Hochschule" in Germany is basically all sorts of higher education, past finishing what's "high school" in America. The American "high school" is in German instead called "Oberstufe," lit. "upper step." See also Wikipedia.


  1. Hello Bee,

    "Multum, non multa"


  2. Hi Bee,

    This definitely sounds like progress to me, which is long overdue, with Germany being the first to take the high ground. Although I wouldn`t find it as unexpected since I`ve always conceived such thinking as to be part of the collective national conscious. Perhaps now as a next step they should require all graduate students to read Robert Pirsig and summit an essay on it just to be sure that its understood :-)

    “The difference between a good mechanic and a bad one, like the difference between a good mathematician and a bad one, is precisely this ability to select the good facts from the bad ones on the basis of quality. He has to care! This is an ability about which normal traditional scientific method has nothing to say. It's long pasttime to take a closer look at this qualitative preselection of facts which has seemed so scrupulously ignored by those who make so much of these facts after they are "observed." I think that it will be found that a formal acknowledgment of the role of Quality in the scientific process doesn't destroy the empirical vision at all. It expands it, strengthens it and brings it far closer to actual scientific practice.”

    -Robert M. Pirsig- Zen and the Art of Motorcycle Maintenance - page 253



  3. Hi Bee,

    I like this decision. Quality is better than quantity. On the other hand, are arXiv preprints publications or not ? I think they should be regarded as publications, since they usually get published in journals.

    Best, Kay

  4. What is the fraction of arxiv papers that end up published in peer review journals within, say, one year?

    I think arxiv preprints should be tagged - refereed, non-refereed.

  5. Hi Kay,

    Yes, I too think arxiv publications should be accepted. They are clearly public, so for what I am concerned they are published. I didn't really figure out what the story about the biologists in Göttingen was though. Did you hear of that? There was only a short note in the article I read. It seemed to me they simply made up publications that didn't exist. If a paper is on the arxiv, it might not have a journal reference but it clearly does exist, so the problem isn't there in this case. And if it's the content that matters, then it should be okay to have a published paper though it's not, or not yet, published in a journal. It will of course make a difference anyway through the magic stamp of "peer review" that comes with a journal reference that the reviewers will certainly pay attention to.

    In any case, it occurs to me there's a linguistic degeneracy here. Is there a good way to distinguish between a paper that's publicly available but not published in a journal and one that is published in a journal (though possibley access restricted, ie not even publicly available?). Best,


  6. Hi Christine,

    Well, the way the situation is right now, a journal reference is basically the tag "refereed." If you leave aside the referee reports that you'd rather forget about ;-) Such a tag would make sense though if one had the "peer review agencies" that I suggested here to have a possibility for peer review that's not coupled to the journals.

    But yes, it's an interesting question how many arxiv papers get published within a year. (Or maybe two. A standard review process can take a year if you're unlucky.) I guess it depends on the arxiv category, the type of articles that dominate there and the standards of the community. Best,


  7. Hi Bee,

    no I didn't heard of the biologists. This is really a mad story.

    Best, Kay

  8. Yes I would agree too, as to the quality, and not the quantity.

    But again, what does this mean and of course the suggestion as too how to go about perceiving this quality of the work.

    Link on name.

    Bee:I vividly recall that in my first semesters at the university the most exciting moments were when a professor or a tutor (usually a postdoc) mentioned an unsolved problem, an open questions. There it was, the frontier of knowledge, and I wanted to go and poke around in the dark.

    While it may appear "dark asto choice of wording," the expansion is with the idea "that quality abounds here in unknown territory."

    You already know the facts. This glimmer of light then and is part of the idea a quality ensues after given those facts. The perception has been allowed beyond it's circumference, to create a new pathway for the thinking minds.

    What does it take then to give it this quality? It is a perspective that once sensing the containment and constraints applied to the discussion, "the mold," allows for many more insights as participators with "points of view" then ever before.

    In no way have can I defined for any one individual, yet, it is within the individual that they will recognize that moment.


  9. This comment has been removed by the author.

  10. When product not process is defining, more product appears. When science works, the person is expendable but the job is not. Anything worth having lies on the far side of risk.

    Mediocrity is a vice of the doomed - with no child left behind.

  11. a journal reference is basically the tag "refereed."

    Yes, you're right.

  12. So at the end an arxiv reference should always be implicitly seen as either "non-refereed" or "under review" at best, because it is clear that if a given arxiv paper was accepted in a refereed journal, one will obviously give the journal reference instead of the arxiv number...

  13. ... So one would wonder whether an arxiv paper has the same status as a refereed journal paper.

    One could argue that an arxiv paper can be eventually scrutinized by the community, through instant responses around the blogosphere, or by comment papers submitted also to the arxiv, etc.

    While that may happen, which is positive, I think that it is still missing some systematic feedback system that could be attached to a given arxiv paper in order to raise its status. But as you said, such a perception depends on several issues.


  14. I think it is obvious that this won't work. Who is going to say that his best papers were published between 15 and 10 years ago, say? *Of course* all of my best papers are my recent ones, because I am getting better and better, right? And of course I want to stress that I am "research-active". So of course my five most recent papers appeared in the last 18 months. To ensure all this, I will have to generate a vast number of papers. The publish or perish syndrome will continue undisturbed.

  15. Hi Rastus,

    I don't think that's much of an issue. With the 5 paper rule, it will be abundantly clear to the referee that the list is not *all* publications, but a few selected ones that are most relevant to the proposal or otherwise impressive. As a matter of fact, it is usually the older papers that have collected more citations and thus score higher in that regard, thus I would expect that in fact more often older publications will be listed. Best,



COMMENTS ON THIS BLOG ARE PERMANENTLY CLOSED. You can join the discussion on Patreon.

Note: Only a member of this blog may post a comment.