54

There have been a couple of occasions in my research in which I've come across a preprint that is several years old and is very relevant to the work that I'm doing.

Often these preprints have very promising initial results. However, when looking at the CVs or Google Scholar pages of the authors on the preprint, I can't seem to find a version that ended up getting published in a peer-reviewed journal, even if the preprint is several years old already. Why would would a researcher abandon a manuscript that they obviously put a lot of time into?

Do researchers sometimes just abandon lines of inquiry because they get too busy? Or, is this an indication that their promising initial results were not robust enough for peer-review, and I should be wary of attempting a similar study?

Peter Mortensen
  • 836
  • 7
  • 7
Amadou Kone
  • 1,033
  • 9
  • 12
  • In my field, important credibility indicators for preprints without a formal publication would be: who the authors are (yes, of course, this is a bit unfair); and how the preprint is cited (can be unfair too, because it encourages a rich-get-richer mentality). If there is subsequent work with formal publications who cite the preprint in their related work study, you would expect them to say if the preprint is known to contain errors (or, sometimes, if there is a reason why it wasn't published). But if subsequent work ignores the preprint, it can be a bad sign (... or they may have missed it). – a3nm Nov 26 '23 at 19:31

3 Answers3

50

There might be any number of reasons. You might try to contact the author(s) to get more information. But... (not all with the same likelihood)

They might have left academia for various reasons and not bothered. Is the CV also old?

They might have incorporated the key ideas into another paper with a very different title. You search is then fruitless.

They might have discovered errors.

Reviewers might have considered the results trivial.

Their attempts to publish might have been rejected by journals for other reasons.

They might have changed sub-fields. (This one less likely, I think.)

But you should be wary, at least, of following up on unpublished work and, at least, be sure that you can verify the claims independently.

Buffy
  • 363,966
  • 84
  • 956
  • 1,406
  • 82
    One other important possibility (if it's math), everything with the preprint is basically fine but they submitted to a top journal and the refereeing process took 2 years but the paper was rejected, they then spent a year revising based on those reports and other feedback they'd gotten, spent half a year deciding where to resubmit, then it took another year and a half to get accepted at the second top journal, but their backlog is such that it takes another year and a half for it to be published. So now 6 years have passed and the preprint isn't published anywhere. – Noah Snyder Jul 18 '19 at 18:00
  • 30
    @NoahSnyder that sounds oddly specific... – Mark Omo Jul 18 '19 at 23:20
  • 18
    That’s not actually intended to be the exact story of a particular paper (mine or others), but more a realistic amalgam of different stories of mine and others. – Noah Snyder Jul 18 '19 at 23:36
  • 5
    I second the suggestion to consider contacting the author. You should have a low threshold for doing that. – Mark Foskey Jul 19 '19 at 02:14
  • 1
    To abbreviate @NoahSnyder: Maybe the authors lost interest. – JeffE Jul 19 '19 at 03:27
  • 25
    @NoahSnyder this hits home. I think this happens more than people think. – nimcap Jul 19 '19 at 07:35
  • 1
  • They might not be able or willing to afford the publication costs.
  • They might not have thought highly of the result at that time, and when it proved popular, it was not new enough to interest journals. (What do you get from republishing something that's already known by everyone in the field and put to test for validity "in the wild"?)
  • – The Vee Jul 19 '19 at 11:06
  • 3
  • They might be in the lucky situation to not feel pressured to publish by their employer, so why bother?
  • – Sylvain Ribault Jul 19 '19 at 14:03
  • 1
    Just to give a concrete example: this paper solved a famous open problem in the field. As you can see it was essentially written in 2009. It was published in 2016. I'm not privy to the editorial misadventures it has encountered, but no doubts about its correctness or its relevance were present in the community in the intervening years, as far as I can remember – Denis Nardin Jul 20 '19 at 09:24
  • 1
    This kind of thing happens a lot, publishing papers can be tough at times and it can take several years before something gets published. Someone told me they knew someone who wrote a paper and couldn't get it published for many years, but now the technique in that paper is used all the time and it is referenced a lot. – Tom Jul 21 '19 at 15:37
  • 2
  • They work in an environment where preprints are accepted as publications already, so no need to re-publish it, espectially if the preprint got a lot of citations already. (I believe this can be the case for senior researchers in theoretical physics and computer science, and I expect it's also dependent on which country you're in, whether you're in industry or academia, etc.)
  • – N. Virgo Jul 21 '19 at 17:17