Hacker News new | comments | ask | show | jobs | submit login
The “phantom reference”: How a made-up article got almost 400 citations (2017) (retractionwatch.com)
59 points by DanielleMolloy 10 days ago | hide | past | web | favorite | 25 comments

From Rota's Ten Lessons I wish I had been Taught:

"8. Give lavish acknowledgments

I have always felt miffed after reading a paper in which I felt I was not being given proper credit, and it is safe to conjecture that the same happens to everyone else. One day, I tried an experiment. After writing a rather long paper, I began to draft a thorough bibliography. On the spur of the moment, I decided to cite a few papers which had nothing whatsoever to do with the content of my paper, to see what might happen.

Somewhat to my surprise, I received letters from two of the authors whose papers I believed were irrelevant to my article. Both letters were written in an emotionally charged tone. Each of the authors warmly congratulated me for being the first to acknowledge their contribution to the field."


There's an old saying that the only people who ever read your acknowledgements are those you've forgotten to include. We'll have to expand that to those whose work you've cited unnecessarily.

The flip side of this - helping with some stuff that the authors weren’t well versed in and then getting acknowledged for some really trivial work. I’ve liked it but it didn’t deserve any credit.

> It was a “phantom reference” that had been created merely to illustrate Elsevier’s desired reference format.

I think that explains a lot. Some authors cut and paste the reference into their document to have an example, and then forget to remove it before they submit the paper. Then the reviewers don't notice or do not remark it in their review report, and it get published.

I did a very similar error myself when I was including the price of a book I cited as part of its title, and this passed through review without notice. This paper was part of my doctoral dissertation, and while none of the opponents made a remark, the institute leader that otherwise was in a completely different field and probably read only the citations did notice. During the "questions from the audience" session, he asked why I did that, which caused a bit of amusement.

There should be a distinct difference then between whether the paper is only referenced in the bibliography, or whether it is also cited.

I wonder how many of the papers actually cite the article? An article that is only referenced is pretty meaningless. Like a function that is implemented but never called. Or should be. Perhaps the problem is the way indexers assume references equal citations.

To continue yesterday's interesting discussion of Latex, references should not appear unless they are cited. I used to use a huge master bibliography for everything, and assumed that Bib/Latex would sort out the things I actually cited.

“Because, ma’am, I wanted to know who had read with sufficient attention that they would catch something so minor in the footnotes. Then I would know who had invested their time and attention in my paper.”

That was probably more of a good-natured dig at the reviewers (his peers) for failing to read the paper as closely as he had!

Ok, so I write "The art of writing a scientific article." and change my name to Van der Geer. Publishing it eight years ago may take some faking but 400 citations out of the box.

Now: _that's_ karma whoring ...

I once proposed a conference paper and had to withdraw it. Nonetheless, a citation shows up in Google Scholar, and researchgate keeps asking me to confirm authorship. I can't figure out how to get them to stop.

Edit: I just checked, and my nonexistant paper now has two citations, according to Google Scholar!

Assuming someone got ahold of a preprint, it could be appropriate to cite it even if the paper was withdrawn, right?

There was no preprint -- it didn't get that far! I'd only submitted an abstract. The citations are based on that alone.

Google Scholar should also be able to show you which papers are citing yours. Can you give a link?

Title is a bit clickbaity... From the article:

> “most citations to the phantom reference occurred in fairly low-quality conference papers,” and were written by authors with poor English.

Something similar happens quite a bit with news reporting. Someone references a non existent story online, then a whole bunch of journalists and media outlets assume it's real based on the 'source' link/text.

Saw it with a story about a game Niantic Labs was supposedly working on, all based on a non existent IGN story that linked to an error page.

But it's depressing to see something like this happen with journals and what not. Would have thought these academics would be a bit better at checking their sources than this.

I swear I've gotten stuck in reference loops online before. A cites B who cites C who cites A again. Not sure how it could happen.

Yeah, that's definitely a thing. It's especially common with wikis like Wikipedia, where the wiki article ends up being quoted in the press, which then end up as references for the wiki article etc:


Edit: Wikipedia also calls it 'citogenesis':


Ooooh nice find! Thanks!

There was a fascinating Defcon 26 talk on "predatory publishing" which may explain how this kind of thing happens:



From the article, "It was a “phantom reference” that had been created merely to illustrate Elsevier’s desired reference format." and "nearly 90% of the citations were for conference proceedings papers, and nearly two-thirds of these appeared in Procedia conference volumes, which are published by Elsevier."

Thus, it is unlikely to be due to predatory publishing.

at 35c3 another good talk Inside the Fake Science Factories https://media.ccc.de/v/35c3-9744-inside_the_fake_science_fac... English audio track available.

Was taking forever to load for me. In case others are having trouble, here's a cached version: https://webcache.googleusercontent.com/search?q=cache:fZ-W5g...

Journals don't use tools that check citation validity?

That's not a nice word for reviewers!

Fake publications get cited in fake journals. Nothing to see here. Stop the needless science bashing. Every scientist here knows how much fake journal spam we get in our emails every day. It is an industry. That industry is not science. It is spam. And frankly, people read this kinf od headline, think science is fake, corrupt, and stupid and get in their SUVs and let the world turn into an oven. OK I am ranting.

Procedia isn't a fake journal. There is no science bashing in the linked-to article. This isn't about fake journal spam. It's about people who accidentally leave an example citation in their citation list.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact