12

So say you are writing a review which includes paper A, which claims to introduce a new idea to your field. In the course of reading surrounding literature, you find earlier paper B, which puts forward a very similar idea. Paper B is not cited by paper A. Normally the thing to do would be to add a note saying the paper B offered a similar idea some years earlier, then synthesise the discussion in both.

The catch here is that paper B is pretty idiosyncratic. It makes various claims that contradict the established wisdom, and does not (in my opinion) sufficiently back them up. It is a long paper, but contains no numerical analysis, which on this topic is a bit unusual. It contains sentences such as;

What may appear as fancy mathematical formulations is primarily intended as an invitation to mathematical physicists to fill in the remaining gaps.

(Which, if you are minded to, is probably sufficient to find the paper I'm referring to.) It has been cited (by reputable papers) as a a publication, multiple times.

Is it reasonable to just mention this papers existence in passing, without really commenting on its odd nature?

Edit; the advice of a commentator helped me find the publication in question. It's not in Russian, it's in English, I just failed to find it before. It's not a lot changed from the preprint, and I still don't feel comfortable making a strong statement about the accuracy of its claims. The discussion of errors in particular seems to miss key aspects, but clearly at least one referee did not agree with that assessment. So it's quite possible that I'm missing the point.

So far I'm inclined to go with @Captain Emacs' suggestion, and possibly also email the author (politely) to see if they are willing to clear up my confusion.

Clumsy cat
  • 6,926
  • 2
  • 27
  • 37
  • 8
    "An early attempt/exploration of XYZ [reference] ..." Makes clear it is not a completed work and it is not really mature. YMMV. – Captain Emacs Sep 05 '21 at 12:27
  • 1
    Was the paper actually published or was it only ever a "preprint"? – Kimball Sep 05 '21 at 14:08
  • 1
    Are the "idiosyncratic claims" of the original actually confirmed/proven in the later paper? – Buffy Sep 05 '21 at 14:20
  • @Buffy No. The two papers do have some similarity in their central premise, which is a reconstruction algorithm. Paper A does demonstrate that this premise leads to good results, by offering numeric analysis. But paper B makes a lot of assertions, many more than paper A. – Clumsy cat Sep 05 '21 at 15:01
  • @Kimball it was published, in "International Journal of Modern Physics A". I can't seem to find the published copy, but I trust the citations would be accurate. – Clumsy cat Sep 05 '21 at 15:04
  • 2
    @Clumsycat But the journal version might contain the "central premise" without so many idiosyncrasies/eccentricities; if so, it would render this question moot. I've managed to find it on the World Scientific website, but I guess you'd rather I didn't link to it in case it identifies/offends the author. Put the first sentence of the abstract in double quotes, copy-paste that into a plain Google search, and it's the third hit. – Daniel Hatton Sep 05 '21 at 15:49
  • @DanielHatton many thanks! I will go check out the published version. Yes, I'm being coy with it. – Clumsy cat Sep 05 '21 at 16:00
  • @Clumsycat Okay, I was confused because at one point you said publication but then you say it's was cited as a preprint. – Kimball Sep 05 '21 at 16:29
  • @Kimball yes your right. The citation contained a link to a preprint, but that isn't the same as citing a preprint. – Clumsy cat Sep 05 '21 at 16:53
  • @CaptainEmacs likely going to go with your suggestion. If you'd like to make it an answer I would accept it. – Clumsy cat Sep 05 '21 at 17:23
  • @Clumsycat Glad you like it. I slightly expanded it to an answer. – Captain Emacs Sep 05 '21 at 17:30

1 Answers1

24

Problems like that can often be solved by suitable formulations.

In your case, something along the line of

"An early attempt/exploration of XYZ has been undertaken in [reference] ..."

This makes clear it is not a completed work and it is not really mature. It assumes that this is at least somehow scientific and not crank work, of course.

Captain Emacs
  • 47,698
  • 12
  • 112
  • 165