Why Do Scholars Only Care about Plagiarism when it's Students?
Post 3 of a series on lessons learned since October on campus
How will we explain to future generations that one of the consequences of the war unleashed by the Hamas invasion of 10/7 was…. an attack on academia for not policing the rampant plagiarism in its ranks? How indeed will we explain that such allegations were the ‘straw’ that brought down the president of Harvard?
There are many ways to unpack this episode. What I’d like to do here is zero in what I consider what I consider its most extraordinary aspect, which I’d formulate as follows:
Plagiarism is ultimately an instance of intellectual property theft. And like any case of theft, the people who should be most likely to complain about it are the victims of the theft. But the plagiarism-accusers in this case were third-party/non-academics! And they were right that plagiarism— which indeed is policed when the miscreants are students— is rarely if ever alleged by academics. Don’t academics care when they are theft victims? Why would observers to a crime be more exercised about it than the victims?
This is indeed a mystery, right?
You might be tempted to resolve it by just saying that we’re a bunch of hypocrites— we police intellectual IP theft when it’s students, but we don’t actually care about it.
But that explanation doesn’t work for two reasons.
First, it begs the question of why, if we don’t care when our colleagues do it, we do care when students do it.
Second, we do very much care about intellectual IP theft! After all, we’re highly competitive, perhaps overly much so. As anyone who has even passing knowledge of academia knows, the name of the game is getting there first with a credible claim to novelty. This is an essential question underlying the presentation of any new piece of scholarship (just as it is with a patent claim): We need to make the case that we’re saying something that’s not only coherent and well-substantiated; we must also make the case that our contribution is novel.
Accordingly, to make our claim to novelty stick, we need to reference all relevant prior work and show that it doesn’t fully anticipate what we’re offering. But this is hard! Most papers are rejected, and lack of novelty is a common reason. Indeed, even those papers that are published aren’t as novel as the authors claim, and this is one reason they may not get much traction in their fields (as reflected in citations).
Of course, there’s much more to the game than this. Highly novel work is often not recognized as such.1 Some of that is innocent (life is unfair!) and some of it is corruption of various kinds (the game is rigged!).
For now though, the implication of our discussion so far is to reinforce our puzzle:
This highly competitive arena is structured in such a way that participants can be expected to be zealous about protecting claims to novelty. And so it’s really strange that they apparently are indifferent to their failure to protect such claims, yet they get worked up about student plagiarism.
Why Do We Care about Student Plagiarism?
To build towards a resolution to our puzzle, let’s first compound it. In particular, let’s zero in on the archetypical case of the student who’s caught red-handed engaging in plagiarism. We have strong intuitions about why they should be punished. But what really is the crime? It does seem like a kind of theft, but it can’t be of the type that we imagine faculty should get worked up over. After all, students aren’t really expected to say something new.
Sure, a great undergraduate thesis can sometimes become recognized as a valuable piece of original scholarship. But that’s very rare. And this shouldn’t surprise us. Students simply don’t have the training or the time necessary to make such a contribution— or at least to be perceived as being capable of such a contribution. That’s what grad school is for.
What’s the problem with plagiarism among students then?
The issue is not novelty but fraud.
It is extremely unlikely that a student in a Chaucer class will say something that Chaucer scholars have never considered; and if they do, they might not even get credit for it (because the Chaucer scholar is their teacher and may not like it!). But since they don’t know much about the relevant scholarship, it’s an accomplishment for the student just to anticipate what others have said, or some variation thereof. The key is that they have to get their on their own. For that to work though, their work has to seem naive to the existing literature.
Students are thus expected to produce independent work. They are meant to engage with material provided to them by their professors, sometimes augmented by their independent efforts at research, and to process it independently of others. When students plagiarize, the crime is that the work is not in fact theirs, as they claim. But their work is not expected to be novel.
To be sure, when scholars appropriate the words of others without attribution, they too are feigning independence, when in fact their work is actually derivative. But whereas the heart of scholarly plagiarism involves faking the claim to novelty, in student plagiarism the issue is faking the claim to naivete. The former is a case of theft as well as a case of fraud. The latter is merely a case of fraud.
In the first instance, this just reinforces our original question. If plagiarism by scholars is both fraud and theft, whereas student plagiarism is just fraud, it should be a bigger deal, not a smaller deal.
Ah, but there’s another important aspect of this ‘novelty vs. naivete’ difference.
What Makes a Scholar’s Work Novel?
In particular, we have yet to say what it means for scholarly work to be novel. And in most fields (certainly in the sciences and social sciences), what matters is not the specific words that are used but the ideas and methods that are claimed as the central contributions of the scholarly work.
There are exceptions to this rule. For instance, once in a while snickering arises on the internet over the phrase “the Strength of Weak Ties”— which is the title of perhaps the most influential sociology paper of all time (Granovetter 1973). That’s because this paper was preceded in 1972 by Liu & Duff’s “The Strength in Weak Ties.” Apparently, the direction of influence ran in the other direction: some time before either paper was published, Granovetter presented his paper (under an earlier title, “Alienation Reconsidered,” while mentioning the phrase “strength of weak ties”) in a seminar that Liu & Duff attended. For our purposes, the key takeaway is that originality in wording can sometimes matter.
But here’s the main thing: If you actually read the papers, you can see why the Granovetter paper ended up being much more influential, and this has little to do with the wording. Rather, it’s that he framed his ideas in terms of the emerging framework of social network analysis, a framework that had an enormous impact on the sociology of the 1970s-1990s and then on many other fields subsequently.
This is emblematic. The novelty in scholarly work (certainly in the social sciences) lies in ideas and methods, and sometimes in specific empirical findings; but not in wording.
Why Steal and How to Get Away with It?
The matter can be put in terms of the incentives to steal others’ work. Is there anything to be gained if you can pass off others’ novel ideas, methods, and/or findings as your own? Definitely yes! It is wrong to do so, of course. And it is risky to try: if you’re caught, you could incur major reputational damage. But if you have loose morals and you can mitigate the risk of getting caught, the lure of being recognized as an innovator is very strong.
By contrast, is there anything to be gained by appropriating others’ words? Nope. Because scholars don’t care about the words that are used. We care about ideas, methods, and findings.
In fact, if you want to steal other scholars’ ideas and/or methods (it’s tougher to steal findings because they’re so specific), it would be foolish to steal their words because this makes it easier to catch you. That is, if you want to pass off others’ innovative ideas or methods as your own, you’d be smart to use different language in doing so!
This brings me to a related point. Scholars steal each other’s ideas and methods all the time. I myself have experienced at least half a dozen ugly cases, and I’ve witnessed even more cases as experienced by my colleagues and students. In pretty much every case though, the theft was subtle. And one of the main reasons is that the thief did not phrase their contribution in the same language as the original contribution.
More generally, scholarly thieves have a wide array of tools in their arsenal. One strategy I’ve become attuned to over the years is what I’ve come to refer to as the “gaslighting citation.” This is when a scholar (or journalist, etc.) cites prior work, but deliberately mischaracterizes it. I’ve had this happen to me at least twice in recent years, in a major organizations journal. In each case, the authors diminished the ideas in a paper of mine so that they could effectively claim my ideas as theirs. You might think that they couldn’t get away with this. But I wasn’t a referee on their submissions to that journal and most people don’t read unless they have to; so the referees on those papers presumably just took it on faith that the authors were characterizing my work accurately.
Gaslighting citations are just one strategy. I’ve seen others. I believe that scholars regularly steal each other’s ideas and methods And this should not surprise us, for two reasons I discussed above: a) the incentives for staking a claim to novelty are very high; and b) the risks of being caught red-handed are often (perceived as being) quite low, precisely because one doesn’t have to appropriate others’ language to pull it off.
Here’s a painfully ironic takeaway from the discussion so far:
The plagiarism witch-hunt of fall 2023 is symptomatic of a naïve understanding of how scholarship works, one that effectively imagines that scholars are like undergraduates.
To be clear, this doesn’t mean that the hunt for plagiarists didn’t uncover questionable behavior. It is at least odd that Claudine Gay recycled some language from prior work. But some of that recycling involved reporting on standard methods, which some have argued— quite reasonably, in my opinion— should be immune from plagiarism norms. What troubled me more were accusations that her papers were weak methodologically2 and that she had appropriated ideas from prior scholars. So the smoke of the (minor) plagiarism may lead to some actual fires. On the other hand, I bet that if many scholars were subject to such scrutiny, similar fires would be found.
OK, so if It’s a Big Deal, Why Don’t We Police It?
This brings us back to our original question, which we can now phrase as follows:
If scholars are in fact quite sensitive to their ideas and methods being stolen, and if it actually happens all the time, why don’t we call it out?!
There are two reasons, in my opinion, and they interact with one another.
The first is one that I already touched on above: the theft is usually very subtle. In pretty much every case I know of, it’s at least arguable that the action was innocent rather than deliberate. We often don’t remember where our ideas come from, and it’s easy to for us to convince ourselves that we are more original than others might think.3
The second issue is the problem that accusers are generally suspected of ulterior motive.
As Minjae Kim and I argue and show in our research, norm enforcement is tricky, especially when the would-be enforcer doesn’t have a formal mandate to enforce (like, say, an investigating magistrate) but instead, must take initiative to accuse others without any authority to do so. Under these conditions, the “norm entrepreneur” is typically suspected as having ulterior motives of some kind. In particular, the suspicion may be that they’re trying to deflect attention away from their own sins. Whoever smelt it, dealt it, you know.
This problem certainly came up with regard to the attacks against Claudine Gay and others. Indeed, there were many who rejected the claims of plagiarism out of hand due to the ulterior motives they suspected.
And if third-parties are suspected in this fashion, so too are primary parties. Remember when I claimed above that some scholars had deployed gaslighting citations to steal from me? Well, did you actually believe me? Didn’t you think that I can’t be trusted because I am biased? After all, I have a lot to gain from taking an expansive view of the novelty of my work. Who would be surprised if a scholar exaggerated the novelty of their ideas or methods more than a neutral third-party would? To the contrary, such accusations are often taken as evidence for what we all believe about academics: that we have an inflated sense of our value, that we are petty.
This is what I regarded as the most painful irony of the plagiarism witch-hunt. In short:
Not only did the accusations reflect a naïve, undergraduate’s understanding of scholarship, but the charge that academics don’t police IP theft enough was tantamount to alleging that academics aren’t self-regarding and petty enough! Which of course is an insanely laughable charge.
This is another indicator of the accusers’ naivete. What they don’t seem to get is that academics would love to police theft of their (and that of their colleagues’ and students’) work. And in fact, we often do— say, in rejecting submissions that don’t reference prior work in the way we think they should. But we generally refrain from doing this openly because we don’t want to look like schmucks.
OK, So What do We do?
Finally, you might wonder what kind of institutional reform might limit IP theft among scholars. To that questions, I say: I have no idea. Given how much of this behavior (for reasons discussed above) occurs in the dark, and given how unenthusiastic most academics would be to engage in an honest assessment of their own (and their colleagues’ and students’) contribution to the problem, it should be no surprise that there is little discussion of the problem and what to do about it. By contrast, in recent years the social sciences have had a very robust discussion, with some institutional reforms, around research integrity and fraud. But if successful reform efforts focused on intellectual theft ever do emerge, I’d bet dollars to doughnuts that this will not come from third parties who have obvious ulterior motives and no understanding of scholarship.
There’s some good news though:
Non-academics really shouldn’t care much about this. After all, if a scholar produces novel work but the work isn’t regarded as novel by other scholars, that doesn’t affect the quality of the scholarship. All that is affected is the ego of the scholar. Sure, indirectly, one could argue that if scholars anticipate that their contributions won’t be properly regarded then they will produce less of it. But this is a pretty indirect effect, and there’s no evidence for it.
And so ultimately, perhaps the most bizarre aspect of the plagiarism with-hunt of 2023-24 is not that the third-party accusers were so naïve about scholarship or that they effectively were accusing academics of being insufficiently schmucky, but that anyone outside academia actually cared.
Of course, this would not be so bizarre if in fact they had other grievances about academia (perhaps with some justification; see my earlier posts) and this was an effective vehicle for pursuing them.
Here’s a paper of mine, With Pierre Azoulay and Michael Wahlen, that demonstrates how inefficient the market for science is. Tragically, it might take the death of a scientist for some papers to attract attention, as the death (especially when the scientist is young) mobilizes colleagues and students to promote the scientist’s work.
Can’t recall now where I saw this, on Twitter.
An interesting case is that of the relationship between a 1955 paper by the sociologist and demographer Evelyn Kitagawa and the highly related work by economists Ronald Oaxaca and (independently) Alan Blinder. Whereas the work of the economists’ work became widely cited and celebrated as the Oaxaca-Blinder Decomposition (OBD), Kitagawa’s work was less widely known, and not at all by economists. After Kitagawa’s paper came to economists’ attention in recent years, it led to a thoughtful treatment by Oaxaca and a colleague. The paper essentially argues that OBD is superior because Kitagawa’s method is a “special case” of OBD. Even if that is correct, it is notable that they don’t address the elephant in the room, which is whether Oaxaca or Blinder were aware of Kitagawa’s work (or others who had built on her, like Duncan). Nor do they acknowledge that it was unfortunate that they hadn’t known of work that preceded theirs by 20 years. But given the stakes, this stinginess in acknowledging the novelty of prior work should not surprise us.