Skip to content

“Translation Plagiarism”

Michael Dougherty writes:

Disguised plagiarism often goes undetected. An especially subtle type of disguised plagiarism is translation plagiarism, which occurs when the work of one author is republished in a different language with authorship credit taken by someone else.

I’ve seen this done, where the original language is statistics and the translated language is political science.

Translating ideas into another field can be useful, and I think there can be value in the acts of translation and rediscovery. For example, many, probably most, of the ideas in our path sampling paper already existed in the physics literature, but the connection to statistical problems was not always clear. We didn’t plagiarize anything in our paper—we worked it all out ourselves and cited everything relevant that we knew about—but there was still some translation going on, if only indirectly.

The problem is when the translator doesn’t acknowledge the original source, and then garbles the translation. Garbling the translation happens a lot: ideas that are worth translating, and that aren’t already known in the secondary field, can be complicated. And the bestest scholars aren’t the ones who plagiarize: plagiarism is, among other things, a shortcut to scholarly renown.

Here’s how Dougherty puts it:

Through no fault of their own, the authors of articles that engage [the translated article, without knowing about the original] fundamentally misidentify the author they are addressing, creating a fundamental corruption of scholarly communication. This point is often overlooked in discussions about the harm of plagiarism.

Basbøll and I discuss this in our article, “To throw away data: Plagiarism as a statistical crime”:

Much has been written on the ethics of plagiarism. One aspect that has received less notice is plagiarism’s role in corrupting our ability to learn from data: We propose that plagiarism is a statistical crime. It involves the hiding of important information regarding the source and context of the copied work in its original form. Such information can dramatically alter the statistical inferences made about the work. . . .

A statistical perspective on plagiarism might seem relevant only to cases in which raw data are unceremoniously and secretively transferred from one urn to another. But statistical consequences also result from plagiarism of a very different kind of material: stories. To underestimate the importance of contextual information, even when it does not concern numbers, is dangerous.

This point is key. Copying with appropriate citation is fine—it’s a way to get ideas out there to new audiences. Copying without appropriate citation—plagiarism—does get the ideas out there, but often in scrambled, contextless, and misleading form. See here and here for a couple of notorious examples done by a much-decorated statistician.


  1. Radford Neal says:

    The first link in your last sentence is NOT an example of plagiarism. It’s just an example of a not-very-good article of which Wegman was an author.

    You can tell from the comments on that post of yours that despite the post explicitly saying that it’s not an example of plagiarism, many readers took it to be such. And now it seems that you yourself have forgotten that it’s not an example of plagiarism.

    I think this validates the comment I made on that post that going from pointing out instances of plagiarism by an author to criticizing all papers by that author that aren’t very good is getting carried away…

    • Andrew says:


      Good point. This was an example in which Wegman would’ve been better off copying (and citing his source) rather than trying to make stuff up!

    • Joshua says:


      Quoting Radford…

      > The first link in your last sentence is NOT an example of plagiarism.

      Which brings us back to you saying in the article that it wasn’t plagiarism.

      > The above passage does not appear to be plagiarized; instead it looks like Wegman read some material and rephrased it in his own words, adding error in the process

      I”m trying to understand that. You seem to be saying that paraphrasing someone else’s work (rather then generating your own ideas) without citing a reference isn’t plagiarism?

      Of course, there’s alway going to be some level of subjectivity…and it’s hard to trace back exactly what someone else has done (despite some claims made here recently, motive-impugning seems to me to be highly problematic). But I don’t understand on what basis you are making your distinction.

      • Radford Neal says:

        The passage in question is describing what the simulated annealing algorithm does. Everyone who describes this is in some sense “paraphrasing” what they’ve read. Without reading something else, how could they possibly know what the phrase “simulated annealing” refers to? This isn’t plagiarism (unless the description is unnaturally close to someone else’s words).

        • Joshua says:

          > Without reading something else, how could they possibly know what the phrase “simulated annealing” refers to?

          So it seems to me that’s where the “common knowledge” aspect comes into play. So you’re saying that within a particular progressional context, he wasn’t saying anything about what the simulated annealing algorithm does that wasn’t effectively “common knowledge?”

          There wasn’t anything that he was saying there that was rather specific to an analysis that a particular individual had done?

          • Andrew says:


            The Wegman article is terrible. It never should’ve been published. Had they just copied straight from wikipedia (with citation), that would’ve been an improvement. Perhaps what happened is that the publisher wanted to publish this volume and make some money, Wegman volunteered to write this article, but then he didn’t feel like doing the work to write something new but didn’t want to back out, so he put together this bad article. Nobody noticed because nobody cares—publishers used to be able to make money by selling this sort of thing to university libraries—but then the plagiarism scandal came out (as Radford notes, these were other articles of Wegman) so people looked into the quality of other articles he’d published.

  2. Rahul says:

    What we need is a fundamental reform in the process of how we reward and assess academics for grants, promotions etc.

    In the absence of this perverse incentive system which gives excessive importance to publications, impact factors and citations I doubt people would spend so much time worrying about plagiarism.

    • Delfina Kahn-Sue, Ph.D. says:

      Damn straight, everything should be better than it is. Harder to come up with an implemantable, acceptable alternative. Sometimes better alternatives when other things are equal turn out not to be better, when other things aren’t equal or haven’t been adequately pre-considered.

      Not arguing against fundamental reform at all, far from it (I demand fundamental reform!), just waiting to “hear the plan,” to quote a famous Beatle.

      • Rahul says:

        Here’s a suggestion: Much more reliance on just in person interviews for promotions, grants etc.

        It’s not untested. Outside of academia most hiring relies very heavily on interviews anyways!

        Let’s just reduce this formulaic reliance on publication metrics as a surrogate for quality. The goal should not be to publish but to produce something useful.

  3. gec says:

    Obligatory Tom Lehrer reference:

    On a different note, I suspect many cases of “translation plagiarism” in the scientific literature arise from a desire to claim priority. They understand they didn’t invent the original idea, but they feel their contribution will only “count” if, for some reason, they are the first to introduce it to a new field, so they gesture to the original idea but dance around it enough to make it seem like it was theirs to begin with.

    I never really got this. Making the connection to the other field and acknowledging what is retained and what is changed—that seems like a more valuable contribution to me, and one that would garner a lot more respect than “I did it first [but not really].” Besides, any credit for priority rapidly fades as the ideas diffuse into the field and people develop and refine them further.

    Pretending priority just seems like playing the wrong game. Or, to put it differently, playing a game where every move causes you to lose, just maybe at different times.

    Now I’ve gone from Tom Lehrer to War Games!

    • Andrew says:


      Indeed, and let me clarify that there are many excellent political science papers that openly take an idea from statistics or some other field and apply it to political science. This can be useful, and interested readers can follow the references to the original work. When the ideas are stolen, the problem is that readers aren’t given the opportunity to get the background and context of the methods that are being applied. Also, as discussed in the context of those Wegman links, the kind of people who copy without attribution are also the kind of people who copy errors or who simply garble the original source.

  4. Spiros says:

    > I’ve seen this done, where the original language is statistics and the translated language is political science.

    This feels like quite the damning claim Andrew! Do you have any examples?

Leave a Reply