6

I recently completed my PhD in Computer Science in Australia and am now a postdoc at a U.S. university, where I am assisting a PhD student with his thesis. In our field, it is common to publish several papers before completing the thesis, which often becomes a compilation of these publications. In my thesis, I began each technical chapter with a note saying, 'This chapter is based on my publication [A],' but I combined all Related Work sections from these publications into a single literature review and introduction, paraphrasing the content without specific citations or acknowledgments. The student I'm helping questioned whether this approach might be considered self-plagiarism. I can't advise on that, but it suddenly makes me realize this might not be appropriate from a point of view.

Then, I scanned my thesis with an online plagiarism checker, which showed that some parts of my thesis involved plagiarism. However, what is odd is that I've observed similar structures in many other theses, including those from U.S. universities. Some PhDs even directly copy and paste the Related Work sections without any paraphrase and citations in the Introduction section, although they explicitly mentioned reusing their own publications in the technical chapters. I guess this might be acceptable, as many people have done the same thing, but I can't find explicit guidelines on any university website about this practice. They all only state that recycling your own text without attribution is considered self-plagiarism.

My question is: Does integrating and paraphrasing content from one's own published works into different thesis sections without specific citations or acknowledgments constitute self-plagiarism? Why did so many PhDs not follow the rule if recycling your own text without attribution is considered self-plagiarism? I'm concerned about potential repercussions, similar to what happened to the Harvard President, even years after graduation. Maybe I am just overthinking. Should I apply to revise my thesis at the university where I graduated? I guess this might be extremely difficult.

1
  • Comments have been moved to chat; please do not continue the discussion here. Before posting a comment below this one, please review the purposes of comments. Comments that do not request clarification or suggest improvements usually belong as an answer, on Academia Meta, or in Academia Chat. Comments continuing discussion or containing answers may be removed.
    – cag51
    Feb 4 at 6:14

5 Answers 5

16

As @GeorgeSavva said in the comments, self-plagiarism is the practice of presenting your old work in a way that misleads people into thinking that it's new work. The main contexts in which the distinction between old work and new work matters is important is when you have something to gain by misleading other people into thinking that the work you are presenting is entirely new. Some examples of situations when people tend to commit self-plagiarism are:

  1. A student submits work for course A, and then resubmits the same work (or a repurposed version of the original work that's substantially similar) for an assignment in course B without disclosing that this work was done for another course.

  2. A researcher publishes a paper containing substantially similar work to a previous paper they published without citing the earlier work.

In the first example above, the student's deception undermines the learning process, since the instructors for both credit are assuming that the student will learn something from doing their assignment, but this is not true for the second duplicate assignment. The student will unfairly get double the academic credit for the same assignment, and that's not how college is designed to work.

In the second example, the researcher misleads the journal into thinking the work is novel when it isn't (thereby gaining the benefit of publishing another paper), and misleads their employer into thinking the researcher is scientifically productive and continuing to publish novel work when they aren't.

Now, the type of situation you're describing, where a PhD student reuses material from publications in their PhD thesis, is not a situation where the student tries to mislead anyone or to unfairly get double the credit for a given work. Rather, they are trying to get two different (and roughly orthogonal) types of credit: first, they publish the work in a journal, and second, they include it in their PhD thesis. This is completely acceptable and appropriate: publishing research is the standard way to disseminate it in the literature, and including it in the thesis is the standard way to get a PhD degree, which is the academic recognition of your ability to do research.

Indeed, in my field it is common for PhD theses to contain entire chapters that are either identical to, or are mildly paraphrased versions of, papers the PhD student published during their PhD. As I said, this is entirely acceptable and appropriate.

Since it is acceptable and appropriate, there is also nothing for a student to gain by hiding the fact that they are reusing material. As you say, it is standard for the thesis to include citations of the relevant papers where the work originally appeared, along with a brief explanatory statement of the sort you referred to ("This chapter is based on my publication [A]"). This can be done in the introduction, or at the head of each chapter, or at some other appropriate place.

Is it self-plagiarism if you don't include citations? Well, it's poor scholarship, but mainly it's just a strange way for anyone to behave. You won't gain anything by it, and you would undermine yourself, since if you're lucky enough to have someone reading through your PhD thesis, you should want them to know about the papers you've published instead of wanting to hide those papers' existence. So to the extent anyone would consider this to be a kind of misconduct, it's a stupid form of misconduct that doesn't provide any benefits. For this reason it seems misleading to me to describe such behavior as self-plagiarism. This is like asking if it's misconduct to write with bad grammar or punctuation, or to be a sloppy writer or researcher. No, it's not misconduct, it's just something that has only disadvantages and no advantages.

15
  • 3
    Why creating work suitable for two courses to be a problem? Student was adept enough to create something that suited both needs. Learned whatever it was to be learned in both subjects. I see no issue with that. Sounds like blaming an engineer to reuse previous invention into a new invention. Jan 16 at 11:32
  • 3
    @akostadinov Is absolutely correct. The 1st example is clearly a failure on the institution that is then blaming the student. If the classes, or instructors, fail to differentiate themselves from other classes, then it is their fault if homework for one class is passable for another. If the point was for the student to explore more, or learn more, then the onus is on the instructor to provide an assignment that would fulfill that goal. If the classes are so similar, then it is an institutional failure for poorly designing a program with too much overlap. But its easier to blame the student.
    – David S
    Jan 16 at 16:17
  • 2
    @DavidS While I agree with the sentiment that the student is not at fault, if it's just one assignment with overlap I don't think it's an "institutional failure". Ensuring that all classes in a subject have disjoint syllabi, especially when there are basic and advanced versions of the classes, seems hard. So an early assignment in a class might duplicate something in the more advanced version of the prerequisite.
    – Barmar
    Jan 16 at 17:43
  • 2
    @DavidS a student who behaves dishonestly cannot blame other people for their dishonesty. The right thing for a student in that situation to do would be to go to their professor, explain that they have already studied the topic of the assignment in another class, and ask whether it is permitted to reuse the work from the other class. The professor is then free to allow it if they agree with your logic, or they can give an alternative assignment, thereby addressing your criticism about duplication of material and avoiding the “institutional failure”.
    – Dan Romik
    Jan 16 at 17:48
  • 1
    @DanRomik In the general sense of this discussion, the student is not automatically dishonest here. There are certain conditions that are not universal that would have to be present for the student to be dishonest here. If those conditions are not met, then it is an undue burden to put the onus on the student to bring it up to the instructor. That's just blaming the student because the instructor or institution's failure to recognize or address the duplication. Now, if the conditions are met where the student is being dishonest, then you have two problems.
    – David S
    Jan 16 at 19:08
5

Paraphrasing is not a guard against plagiarism. Reusing work without citation is plagiarism. If it is the work of others then it is classical plagiarism. If it is your own work then it is self plagiarism.

The issues are slightly different, however. In classic plagiarism the infraction is one of attributing the work of others to yourself. In self plagiarism, however, the infraction is breaking the chain of development of ideas, so that the entire context of the idea can't be properly followed to the source. The other/earlier papers provide important context, including the citations made there as well as associated ideas. Self plagiarism breaks that chain.

You can "self plagiarize" in things that won't be published, as you are the only consumer. But for publications, and likely theses, the assumption is that they will be made public (the literal definition) and so things should be cited. Likewise, you can self plagiarize things of your own that haven't been (and won't be) published. That isn't actually any form of plagiarism, actually, but I think you get the meaning. Your unpublished ideas are still just your own ideas, perhaps awaiting publication.

A "stapled thesis" that includes the earlier work can take some liberties in the summative material as the link to the earlier work is unneeded when the earlier work is literally included.

Your literature review should have pointed explicitly to the literature. Otherwise it is difficult to evaluate and even more difficult for a future researcher to make use of what you did to go forward.

The fact that you find "transgressions" isn't surprising. At the moment, for political reasons, the rules seem to be getting stricter. This is a good thing, IMO, though it might make some writing more pedantic. But in academic writing, links to past ideas are critical to the forward movement of the art. Avoid "islands" of thought, which appear when there is insufficient citation.


Note, importantly, that you can plagiarize a work using, literally, none of the key words of the original. It is still plagiarism. Plagiarism is about the underlying ideas, not the specific words used to express those ideas.

5
  • Yes, the literature review pointed to the source literature. Here is the common approach I used: in the publication, I mentioned that the works are classified into categories A, B, and C, each followed by respective citations. Then, in the thesis, I expanded this to 'The works are classified into A, B, and C. A is defined as (detailed description of A), with (citations). B is (description of B), with (citations), and C is (description of C), with (citations). Most of these are described in descriptive language without any creative content. The categorization way might be somewhat novel.
    – oldbee
    Jan 15 at 13:24
  • I'm worried about "without any creative content". If you mean "new" creative content, but describe old creative content without citation, then it is probably some form of plagiarism. My advice is to say more. Popular literature emphasizes readability. Scientific/academic literature emphasizes precision. Make it clear.
    – Buffy
    Jan 15 at 13:29
  • And you description of "sections" isn't very clear. If you are clearly pointing to the earlier work and that is obvious to a reader then you are fine. Otherwise, not so fine.
    – Buffy
    Jan 15 at 13:31
  • Yeah, I meant "I clearly pointed to the earlier work with citations". It is just that the way I organize the earlier works is from my publication. That is without any pointing to the original source of my previous publication. Like the example I mentioned in the comment.
    – oldbee
    Jan 15 at 13:41
  • Plagiarism is about both: words AND ideas. Feb 4 at 22:30
2

Why did so many PhDs not follow the rule if recycling your own text without attribution is considered self-plagiarism?

An honest answer is that in some subjects, people don't care so much about "self-plagiarism", to the point that some will argue that it is an incoherent concept. You can see a sample of opinions here.

In the coursework context, we never know whether the student has actually done any of the work, we can only judge by what they submitted. Hence, the tendency to inflate the penalty if caught and act whenever there's a clear proof. If a student lifted a sentence verbatim from a source without citation, we punish them so harshly not because it is such a heinous crime, but for deterrence purposes, and because it puts the independence of the whole work in question.

None of this applies to the thesis situation; you are judged on substantial contribution and people have little appetite to examine how far from the paraphrased text you have quoted the source.

I'm concerned about potential repercussions, similar to what happened to the Harvard President, even years after graduation.

There are three thing here to calibrate: 1) this was as high profile political case as it gets, normally nobody would care; 2) what Claudine Gay was accused of is far more serious than what you describe; 3) once we get to a public scandal, there's an important formal aspect of it: using someone else's text without citation or quotes is undeniable, and clear to non-experts.

1
  • If you're bringing in non-experts, or laymen in to the picture, the average layman will look at you like you're the crazy one if you try to compare any form of "self-plagiarism" to actual plagiarism.
    – David S
    Jan 16 at 17:45
0

I largely agree with this answer, which cuts right to the heart of the ethics of it all, but I have one caveat important enough to include in my own answer:

Sometimes you can get into copyright issues with the first publishers of your papers. I've seen this fall under the umbrella of self-plagiarism, too. A great many (most? probably varies by field) of those venues have carve-outs allowing for republication in a dissertation, but not all of them.

My US-based institution had a formal process for squaring away that issue, along with a bunch of other plagiarism issues. I thought it was very tedious at the time, but in the wake of certain scandals, I find it retroactively comforting.

1
  • Yeah. At that moment, our university asked us to sign a declaration to ensure that we tried every way to secure the copyright. But actually, I signed the declaration without knowing what that declaration meant. So, this time, I went back to check the copyright of the publications. Luckily, they all allow redistribution... otherwise, it would have been another story.
    – oldbee
    Jan 17 at 4:25
0

I have the same problem with my doctoral dissertation: three chapters are based on three articles I wrote previously, and they include some of the text and images that also constitute the articles. I did not cite the articles in the doctoral thesis or write the sentence "this chapter is based on the article ..." and later I became concerned about this. I did not do it because after having asked to my supervisor, I did not get any clear explanation about how do it properly (I was basecally the first PhD student who had published articles before writing the PhD thesis).

I did some research and found that the editor allows articles to be reproduced in a doctoral dissertation. However, the editor asks you to quote the article in the doctoral thesis, not as a citation, but to write the sentence "this chapter is based on the article ...". At the same time, if you publish the doctoral thesis first and then the article, you must write that the article is based on the doctoral thesis. I came to this conclusion: the problem is not plagiarism, self-plagiarism or copyright. The editor can just claim that I did not mention it correctly, that's all.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .