11

This paper was published in a high impact journal for physics, Physical Review Letters, in 2010, and basically duplicates work done in 2000 by another author in the field, here.

What is strange to me is that, in our subfield of optical polarization theory and measurement, the author of the second, older paper, is well known. There is no way that an adequate reviewer couldn't have known about the second author's work in 2010. The 2010 paper doesn't cite this former work.

What is the protocol for this situation?

The first, newer paper, has a more elegant and easily understandable derivation, but it still duplicates the older paper, and isn't original, just more concise. However, the 2010 paper, disingenuously states that "non-quantum entanglement" is used to resolve an issue (the issue solved in 2000). This needs to be called out, because mathematically all we have is a 2 element vector with elements that are not linearly dependent, and this is what they are denoting the "non-quantum entanglement."

In my opinion, this lack of citation of the original work by the authors of the first paper above (newer), likely where they got the idea in the first place, is potentially unethical.

Edit:

Now that I've been in academia for awhile, I find myself often disappointed and disheartened by the following problems, most of which are primarily political :

  • The unfortunate view that retractions are so unpleasant that they almost never occur. Everyone is wrong occasionally, the literature shouldn't be polluted with incorrect results.

  • Everyone always being afraid to be wrong, and subsequently afraid to fail. Failure is essential to developing new frontiers.

  • The lack of risk taking in mainstream science.

  • The focus on positive results, especially in experimental sciences. I'm sure we would all love to know when not to spend years going down a fruitless path.

  • Ethical issues being addressed as just ignore it.

Retractions (for ethically sound reasons, like mistakes, etc.) really shouldn't be so unpalatable that we end up with incorrect results proliferating in the literature.

Ethical problems like the above, failing to cite likely known prior work shouldn't be swept under the rug and just ignored for political reasons.

  • The usual course of action would be to contact an editor of the journal with your concerns. They would normally not identify you to the authors or anyone else. But I wouldn't be surprised if someone reading this is an editor, or knows someone who is, so there's a fair chance your concern will reach the right quarters simply by you posting it here :-) – Nate Eldredge Oct 10 '14 at 22:11
  • 7
    Note that Gil (the author of the first paper) cites both papers in this review article. It's worth taking a look at how he handles it. – Anonymous Mathematician Oct 10 '14 at 22:54
  • Thanks @AnonymousMathematician, I haven't read that paper yet... – daaxix Oct 10 '14 at 23:36
3

This is a very opinion-based answer, but ...

What is the protocol for this situation?

Letting it go.

You say that the paper got published in a high-impact journal 4 years ago. It reproduces results from another well-known paper. It is pretty much guaranteed that you are not the first person to notice that. It seems unlikely to me that the paper is indeed perceived as "suspect and unethical" by your peers.

Maybe your peers value the "more concise" write-up? Maybe there is a subtle difference that you have not noticed? Maybe, for some types of (empirical) papers, having a reproduction paper that basically just confirms that the observations from the first paper are indeed general is a really valuable contribution (although not citing the first paper is of course very bad in this case).

If you feel the case is indeed bad enough that something needs to be done, you have two options:

  • Mail the authors of the first paper.
  • Mail the EIC of the journal.

However, realistically, the chance of retracting the paper based on a years-delayed decision that the contribution was not novel seems very small, as long as there was no clear case of unethical behavior or a significant technical problem with the results.

  • 3
    Another outcome could be not retracting the second paper (assuming it is still valid but done independently) but adding a brief note acknowledging the first paper's priority. I don't know enough about this area to know whether that would be appropriate here. – Anonymous Mathematician Oct 10 '14 at 22:56
  • 1
    “It is pretty much guaranteed that you are not the first person to notice that.” – I would be very careful with such a statement. I have seen several reviewers and authors not noticing that a paper proposes an overcomplicated method to calculate 1. – Wrzlprmft Oct 10 '14 at 23:08
  • 1
    I added something to the question, I wouldn't care as much if the 2010 paper had cited the older paper, which it doesn't... – daaxix Oct 10 '14 at 23:34
  • @Wrzlprmft Sure, but after a paper has been in print for multiple years in a well-known journal, it is very unlikely that nobody is aware that the paper, as you say, "proposes an overcomplicated method to calculate 1". – xLeitix Oct 10 '14 at 23:41
  • 2
    As a graduate student, I suppose it would be foolhardy for me to pursue this further, but I must say that the "let it go" attitude is a huge part of the problem with the explosion in erroneous and fraudulent results in the literature, and the slaps on the wrists professors receive for unethical behavior. The "let it go attitude" allows more gaming of the system, and for the cheaters and unethical people to win (the game)... – daaxix Mar 29 '15 at 16:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.