« The Price Is Too Damn High, Canadian Edition | Main | Employment growth in Texas »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

great post. as someone entering college, i appreciate this :)

I've been wondering about self-plagiarism. I've been reading review articles, and paragraphs keep re-appearing...once you've figured out how to phrase a thought, it is difficult to alter the wording, but how much duplication is too much?

http://edmonton.ctv.ca/servlet/an/local/CTVNews/20110617/dean-resigns-plagiarism-accusations-110617/20110617/?hub=EdmontonHome

Out damned spot! Or Ubermenschen of the world unite?

Academics are people too. Ambitious people, and ambitious often people cut corners to get what they want. Mostly, it works pretty well.

Apart from Schadenfreude, I admit I find it satisfying when ambitious people are busted for cutting corners. But then I wonder if I'm celebrating a victory for justice, or am I snickering as overman is ground beneath the boot heels of my slave morality? Would it be as satisfying if the difference in wages and status between them and me wasn't so damn big? Probably not.

Linda: "I've been wondering about self-plagiarism" - yup, when I think about how much I've written over the past 20 years. Would every word stand up to scrutiny?

I can see why the Frey case would definitely raise a question of adherence to contract (with the journal), and thus, by extension, of personal integrity, but why does the Frey case also raise questions of academic integrity? Presumably, beyond the contract issue, copying in large part one's own previously published articles is dumb because it reduces their value to potential readers and thus to publishers. But I don't see anything morally wrong in copying sections of one's own previous work, for efficiency, where, for example, one might have been particularly happy with the previous characterization of a complex problem, etc. Is the expectation that one must not only use one's own words, but always a different set of one's own words?

I've often been skeptical about the publish-or-die horse race for academia. I can't characterize it as anything else. You spend a good deal of time getting a PhD and then are put in the race to publish. Given the stakes I'm not surprised some people cheat.

Though I think it raises the question that given all this publishing and all the effort put into educating professors, does all this education and publishing actually produce results that the rest of academia and the general public considers useful and worthwhile? Or are faculty just publishing for the sake of publishing and their work is little-read?

Seems like academia has become a prisoner of the rules of it's own game.

I gotta agree with david NH on this. Apart from the contractual issue (which is a big caveat) it's not clear what's morally wrong with recycling your own work. To be sure, if you publish essentially the same paper three different ways, that doesn't necessarily say much about your work, but that's a qualitative issue (i.e., what kind of goon are you that you can't come up with more than one semi-interesting thought), not a moral one.

Granted, if you're doing it to game the system in order to secure tenure based on the quantity (rather than the quality) of your work (on the theory that university tenure committees will look at the CV, not the underlying publications), that might be objectionable. Even then, though, what's objectionable is not that the professors are submitting the same work to multiple publications in order to pad their resume, but that the universities really are awarding tenure on the basis of quantity, not quality (since if they looked at underlying quality, the strategy doesn't work). It's a systemic problem, not a moral one.

Then again, no sooner did I write the last post, and it dawned on me that, if a student were to do the same thing (i.e., submit the same paper in two different courses), there'd be hell to pay. I suppose the distinction is that implicit in a student's evaluation is the understanding that the student will do some sort of semi-original work for the course.

That reminds me of one of my father's story. He had given one of his students a solid "D" for a history paper she written for him. She showed up a few days later at his office quite distraught and demanding to know why she'd done so poorly. "I don't understand" she said "I submitted the same essay last year to Professor Smidlap [the name has been changed to protect the innocent] and he gave me an "A"". My recollection is that my father gently advised her that, for obvious reasons, she might not want to pursue her appeal, because while he would gladly give her paper the grade it merited, she might not be too pleased with the outcome. Then he went down the hall to give Professor Smidlap the gears.

david nh "But I don't see anything morally wrong in copying sections of one's own previous work, for efficiency, where, for example, one might have been particularly happy with the previous characterization of a complex problem"

Agreed. There's nothing wrong with adding a footnote saying "this discussion draws from Greatmind (2010)" and then quoting yourself. What is problematic is pretending that something is novel when in fact it's been published before. Because in a lot of cases an academic journal is only interested in publishing novel results - they wouldn't publish it if they knew it had been published before.

Determinant - these are serious and difficult questions.

Bob - actually, for tenure and promotion re-publishing yourself is not a great strategy, because an outside referee will usually read the complete file and detect any self-plagiarism. So for tenure, the temptation is to just fake the results - do dubious statistical manipulations and in other ways torture the data until it confesses. That's more what Frey is talking about in the Withering Academia article.

Where there is a huge incentive to re-publish is when there's exercises like the UK Research Assessment Exercise, and universities are ranked on the basis of the number of publications. Or when pay raises/merit increments are based on the number of publications, and there isn't a process whereby the publications are actually read and evaluated.

But, yes, any kind of research assessment that uses a quantity metric will encounter this kind of problem.

I wonder how fine the line is between 'churn' - where a new approach or model is applied to several different problems or data sets - and self-plagiarism.

I've never had to struggle with this issue, because I can never seem to stay on the same research project long enough to publish more than 2 or 3 papers.

@Frances:

" What is problematic is pretending that something is novel when in fact it's been published before. Because in a lot of cases an academic journal is only interested in publishing novel results - they wouldn't publish it if they knew it had been published before."

Again, though, that's a contractual or disclosure issue where the academic is misleading the journal, not some sort crime against the academic pursuit of Truth and Knowledge.

Like any other field, there are those who will attempt get their free lunch. Cheating is just a matter that the rules state which free lunches are not to be consumed. If the point of academic publication is to add to scholarly debate and further research then it follows that the articles published therein should be original.

If articles were not required to be original then a field with three journals would see the same articles in each of those journals. It would do nothing to stimulate debate and further the body of knowledge. The rules prevent a stream of recycling that would lead to a loss of relevance for the journals concerned.

More deeply, the problem seems to be epistemological. How do we determine whether research is valuable and how do we measure the value of a researcher? What metric do we use? Potential alternatives vary by field. In Engineering you can go with patents filed or technologies transferred, but that leads to a strong bias to applied knowledge to the detriment of theory and at worst becomes corporate research at public expense. In the Humanities publication can be a worthwhile metric but then how do you make room for the popular commentators vs. the iconoclastic theorists?

Academia seems to have settled on "articles published" as a general metric across fields and faculties but this favours bulk over creativity and republication is merely one strategy to increase bulk.

Or given the role of government research grants to universities, what exactly is the value of these grants if all they produce is an article count in publications that few read and with results never seem to go beyond the narrow audience of the journal in question?

Given the metric what have, journal publications, and the inputs, research grants, what exactly are we trying to achieve as an agreed outcome for society? Simply increasing the body of knowledge? Or are we waiting for that one proverbial "lightning strike" idea that changes everything after thousands of ideas that didn't?

"Linda: "I've been wondering about self-plagiarism" - yup, when I think about how much I've written over the past 20 years. Would every word stand up to scrutiny?"

Posted by: Frances Woolley

(a) How much is is written in peer-reviewed journals?
(b) To what standard of scrutiny, and to what consequences? Being wrong is not being dishonest.

Funding agencies often have policies on what constitutes academic misconduct, with violations often leading to cut off of funding and a ban from seeking future funding.

"If articles were not required to be original then a field with three journals would see the same articles in each of those journals. It would do nothing to stimulate debate and further the body of knowledge. The rules prevent a stream of recycling that would lead to a loss of relevance for the journals concerned."

I'm not sure that's right. Requiring original work helps stimulate debate and further the body of knowledge. But the diffusion of knowledge is also a component of stimulating debate and futhering the body of knowledge. Publishing the same work in three separate journals may well be helpful if those journals have different audiences or appeal to different disciplines. For example, if someone publishes a brilliant insight in the Scottish Journal of Political Economy (which, I'd suggest, is not as well read as the AER or the JEP), do we really want to prevent them from republishing the same point in a more widely read publication. Granted, this is less of an issue in the era of google and electronic databases, but still it probably isn't entirely irrelevant.

Bob: "do we really want to prevent them from republishing the same point in a more widely read publication."

I'm not sure who the 'we' is here.

Academics generally have a 'submission tree' - that is, they submit to AER first and then to SJPE only after their work has been rejected by AER. So in practice the kind of situation that you're imagining almost never arises.

Also, any journal will require that an author assign some kind of copyright or exclusive licensing arrangement, giving that journal rights to any commercial use of that work. So republication is almost invariably a violation of copyright.

If AER wants to republish something that's already been published in the SJPE, that's fine - but the author needs to tell the AER about the previous publication and the AER will need to pay royalties to SJPE. Most journals will have a line in their submission policy saying something along the lines of 'by submitting this article you're confirming that the article has not been published or is not under consideration for publication elsewhere.'

Sure, disseminating research ideas widely is a great idea - that's what books, magazines, blogs, etc are for. But academic journals are generally intended to be places for the publication of original research.

Or perhaps your point is that copyright protection is just a bad idea?

Frances,

Not at all. I don't disagree with any of your points, my comment was only that diffusion of new ideas is as important as the development of such ideas, so republishing original work isn't inherently undesirable, where it aids the diffusion of that work. Obviously, it has to be done in compliance with relevant laws and policies. I think "we" all agree that if the journal only accepts original work (or even if it republishes previously published works), it is unethical to submit previously published work without putting the journal on notice that it was previously published. It's the misreprensentation that gets Frey in trouble, not the resubmitting of his prior work per se. If the journal is put on notice, copyright is something that can be dealt with, in appropriate cases.

With respect to the decision tree, I get that. But that assumes the person with the brilliant original insight is an economist. If they were, say, a tax lawyer (or, more realistically, a mathmatician, physicist, statistician, political scientist or moral philosopher - i.e., groups from which no shortage of clever insights into economics have emerged in the past) the AER would probably not be their first choice for publication, in fact, it would probably be a fair ways down the list. It also assumes that journals and referees don't make mistakes about what is significant and what isn't. That might be a bold assumption.

Indeed, how do we know that the situation I suggested never (or rarely) arises? If journals say they won't republish articles, I suspect most people won't bother to submit previously published articles to those journals. But is the absence of such submissions, how do we know whether ideas ever arise in obscure journals that might be of significance for economists if they were republished in a more accessible journal. Isn't that the problem? Indeed, query whether journals shouldn't modify their submission policies to provide that they might consider republishing previously published work in exceptional circumstances (i.e., where the work is important and is unlikely to have received, or to receive, adequate diffusion in the field by virtue of its prior publication). In fairness, I can understand the practical reasons for not having such a policy (i.e., having to deal with every half-wit sending along their piece from the Mongolian Journal of Agricultural Economics as an important work qualifying for that exception).

Unfortunately, not every article in the Mongolian Journal of Agricultural Economics is by a half-wit. In the '60's , american physicists had been stuck for five or six years in the design of some magnet in a particle accelerator. Till someone discovered that the problem had been solved in the 50's by a greek electrical elevator engineer who had published the results in a small greek journal. AER (or Science or Physical Review) does not publish top notch brilliant papers. They publish top-notch brilliant papers by members of the club. Had the greek guy submitted his paper to Phys Rev, there is a good chance the enveloppe would not have been opened...

Jacques,

Agreed. I think the history of the sciences is replete with examples of seemingly intractable problems which turn out to have been resolved a half century earlier by some obscure academic publishing in some obscure journals. And the history of the sciences is also filled with example of brilliant ideas by "nobodies" being squashed by the scientific establishment (by some accounts Lord Kelvin held back the development of Geology as a science for half a century and who knows how many lives and treasure were lost due to the stone-walling of John Harrison's chronometer by the British scientific establishment). I wouldn't expect that economics (or any other discipline) would neccesarily be different.

But I think that ties back to the point that prior publication shouldn't, in all circumstances, be a bar to republication if the idea is sufficiently important.

The points raised about the "audiences" of different journals are almost completely irrelevant in this century. A quick search on Google Scholar (or wherever) for a certain keyword will return papers published in the AER, the Mongolian Journal of Agricultural Economics (to quote Jacques), the Icelandic Journal of Small Rodents (my own creation), etc. One could argue that some economists would overlook results published outside of the journals they feel are "highly-ranked" economics journals, but that is to their detriment. Ignorantia juris non excusat.

People are really concerned about some guy's paper being too close to his own paper? Really?

The point is gradual intellectual advancement, not to necessarily create something completely new every time. Maybe the author should have been more forthcoming. Maybe not. Maybe he assumed the journal already knew of his past work. Who cares? That's not plagiarism. What matters is if the work is good.

If the journal feels that the work isn't fresh enough for their liking, fine, but let's not label it as plagiarism.

On a side note, Martin Luther King plagiarized parts of his PhD dissertation -- and I'm okay with that.

Sina - the questions still remain: what kinds of behaviour are unacceptable? And when people behave in these ways, what happens?

If a student had submitted two papers that were as similar as these two papers in two different courses, that student would be hauled before the associate dean and asked to explain him or herself. Why shouldn't faculty be held to the same standards?

Because faculty are part of the club and have put in their dues. Who is going to discipline them? A committee of their fellow profs who have every potential to do the same thing or the Associate Dean (Research) who was regular prof until her promotion a year ago?

More generally and in more economic terms how do you get a circle of insiders to effectively police themselves?

Jacques: You must be referring to Nicholas Christofolos

Determinant makes a good point, but I don't think it's a hopeless situation. Look at the case of Dr. Baker at U of A medicine. He plagiarized a chunk of a commencement speech - it wasn't even close to an academic paper - and he was effectively canned (well, he 'voluntarily' resigned as Dean and took a few months gardening leave - no doubt to find a 7 figure salary in private industry given that his academic career is toast). I think if the University puts it's mind to it they CAN police faculty. Tenure is, at least in theory, a shield against gratuitous dismissal. It isn't carte blanche to do whatever one wants. If the University perceives that it's reputation or funding is at risk, it will act.


Patrick - and, you know, I have some sympathy for that Dean.

If there was an academic integrity policy for faculty it might be possible to find some middle ground between turning a blind eye and forcing people to resign in a blaze of public shame.

Frances,

I think there is a distinction (or there may be a distinction) between a student who submits the same paper for different courses and the professor who republishes an original paper. There is nothing inherently wrong with reusing your work, it's the context that matters.

The expectation atuniversities (and perhaps this isn't always made explicit enough to students) is that they will do original work for their courses (for example, UofT's code of academic behaviour makes it an explicit offense to resubmit work for which you have already received credit, whether at UofT or elsewher - I expect that most universities have similar rules). A student who resubmmits previously submitted work is both violating the rules of his or her university and, inherently, misleading his or her professors.

Similarly, a professor who submits previously published work under the guise of being new and original work is both in violation of his contract with the journal and is misleading the journal. It's unethical conduct, even if it isn't explicitly prohibited by a university code of conduct. On the other hand, a professor who submits a paper for republication and puts the would-be publisher on notice that it was previously published isn't doing anything wrong. Similarly, a student who takes previously submitted work and uses it for some other purposes, maybe gets it published, maybe sells it, whatever, isn't doing anything unethical.

Now, I suppose one might argue that it's an expectation that Professors will also do original work, but it isn't clear that you need a prohibition on republishing papers to enforce that expectation. Presumably, a university can review a professor's publications - which are inherently readily available - if it turns its mind to it and assess them based on his or her aggregate work. In contrast, students are evaluated on a work-by-work basis, so you need a rule prohibiting the resubmission of work and, because detecting such resubmissions can be difficult, hefty punishments to deter it.

Bob "Similarly, a professor who submits previously published work under the guise of being new and original work is both in violation of his contract with the journal and is misleading the journal. It's unethical conduct, even if it isn't explicitly prohibited by a university code of conduct."

And when a professor does this, she is tarnishing the reputation of her institution. That's why institutions have an interest in deterring such behaviour.

Plus an employer is in a better position than a journal to enforce ethical conduct. As a practical matter, it's very difficult for journals, especially small journals, to take action against people who infringe copyright. When an author submits to two journals, and the two journals publish substantively identical pieces simultaneously, who has infringed copyright? Is it the submitting author? What are the costs for a US journal of taking legal action against an author who lives in Europe?

So, instead, people end up getting tried in the court of public opinion.

The most amazing thing about this incident is EJMR was apparently instrumental in uncovering the plagiarism. Who would have thought something good could ever come from that cesspool?


Considering that the public pays the bills for most Universities in Canada, and that public R&D funds go to universities, it isn't surprising that the public take an interest.

In the Baker case, I'm not sure a 'code of conduct' would have helped. The code would, no doubt, have said "thou shalt not plagiarize". The public would have just said "Look! He broke the rules! Off with his head! And, oh by the by, we just become much less willing to have our tax money go to the UofA since you hire liars and cheats". They had no choice but to throw him under the bus. The Faculty of Medicine at the UofA is a gigantic money pump. They weren't going to risk damaging it. I suspect any publicly prominent academic whose conduct puts funding at risk would likely receive the same treatment just about anywhere.

Frances, perhaps we're talking past one another here. I agree that the conduct in that case in unethical (although the extent to which it tarnishes the reputation of the university is perhaps not as clear), but do you need a rule against resubmitting articles (which might catch scenarios which aren't unethical) or do you have a general morals clause and, in appropirate cases say: "Gee Bob, we think your behaviour is making us look bad, and by the way, you haven't written anything new in a decade, you're out". And if we're only worried about catching cases which are in breach of agreements with journals, we already have a set of rules dealing with that (contract and copyright law).

Moreover, while I agree that an employer is better situated to enforce ethical conduct, that's only by virtue of having position of relative power over the employee that an outside journal doesn't enjoy (although I suspect the punishment imposed by the JEP in the Frey case is more damning than anything that the University of Zurich could do). But we wouldn't accept that as a reason for having employers enforce other ethical or legal violations (I mean, if you've got a professor who's not paying his child support, that's unethical, and possibily illegal, behaviour on a far more depraved scale than republishing papers, but do universities want to be enforcing that).

Furthermore, is an employer in a position to adjudicate between a journal and an author as to who is in the wrong? That's questionable (and does a university want that responsibility?). Does an employer know that resubmitting an arcticle resulted in a copyright violation or mislead the publisher? Is it better situated than a court, or other parties, to determine whether the deception was intentional, or a genuine error. Perhaps the agreement between the original journal and the author permits republication or is ambiguous on the point (I'm a lawyer, and for the life of me, I can't figure out if the non-exclusive license agreement that I enter into with the Canadian Tax Journal every time I make a submission allows me to republish the work or not - being a non-exclusive license agreement which provides that I retain copyright, you think it would, but it isn't obvious. Not, I hasten to add, that anyone is rushing to republish my works).

In fact, in my mind, the JEP handled the Frey case just right. They (presumably after consulting their lawyers - don't want to be defaming anyone) called Frey out. To his credit, Frey acknowledged his error and has to live with the stain on his reputation (which, if it truly was a one-off mistake, may wash away with time).

And to come back to the comparison with students, the different contexts requires different arrangements. Just because you have a code of conduct for some behaviour for students doesn't mean that it's neccesary or appropriate for professors.

Patrick: Yep. Off by a decade and a few details. Couldn't verify as I was in the middle of the "Characteristics of a good model" lesson. Friends don't let friends comment and teach....

Bob: Frey only acknowledged his error after Autor slammed in the JEP. Immediately prior when the issue was only being kicked around on blogs, he dismissed his error as a simple oversight and blamed the scandal on---get ready for it---German jealousy of Swiss success:

Storbeck: "Do you think you’re treated unfairly?"

Frey: "I ask myself if the German are envious. Maybe some people in our neighbouring country don’t like the fact that a Swiss like me is often No. 1 in the rankings."

Yesterday someone on EJMR pointed out the following. Possibly what's going on is Frey habitually sends new papers to multiple journals in the hope it'll hit at at least one of them, and when it hits at several of them he winds up self-plagiarizing.

Frey (2003) "Publishing as Prostitution? – Choosing Between One's Own Ideas and Academic Success," Public Choice.

Abstract. Survival in academia depends on publications in refereed journals. Authors only get their papers accepted if they intellectually prostitute themselves by slavishly following the demands made by anonymous referees who have no property rights to the journals they advise. Intellectual prostitution is neither beneficial to suppliers nor consumers. But it isavoidable. The editor (with property rights to the journal)should make the basic decision of whether a paper is worthpublishing or not. The referees should only offer suggestions for improvement. The author may disregard this advice. This reduces intellectual prostitution and produces more original publications.

Frey (2005) "Problems with Publishing: Existing State and Solutions." European Journal of Law and Politics.

Abstract. Survival in academia depends on publications in refereed journals. Authors only get their papers accepted if they intellectually prostitute themselves by slavishly following the demands made by anonymous referees without property rights to the journals they advise. Intellectual prostitution is neither beneficial to suppliers nor consumers. But it is avoidable. The editor (with property rights to the journal) should make the basic decision of whether a paper is worth publishing or not. The referees should only offer suggestions on how to improve the paper. The author may disregard this advice. This reduces intellectual prostitution and produces more original publications.


Am I the only one who thinks it both funny and outrageous when an economics paper has the thesis that academic publishing is prostitution. It seems that rhetoric and hyperbole has invaded the research domain.

Perhaps "Commit an act of Intellectual Dishonesty" instead?

Chris - that is outrageous, at least the various Titanic papers all had different abstracts!

Though it shows you the importance of getting the right journal and the right title - "Publishing as Prostitution" is cited almost three times as often as the more mundanely titled "Problems with publishing" (I couldn't believe this, I had to look up the papers myself on google scholar).

Bob Smith - " But we wouldn't accept that as a reason for having employers enforce other ethical or legal violations "

The university is paying for professors' research activities, and the university's research reputation depends upon the quality of professors' research.

Universities as employers could certainly specify in the employment contract that professors are not allowed to, say, falsify data, or pass of students' ideas as their own, and specify procedures to follow in the event that a professor was suspected of these activities - this is just setting out part of the job requirements. And the point of this post is that this idea is something worth considering.

This is completely different from universities requiring that professors, say, pay child support on time. It's hard to see paying child support as a bona fide job requirement, in the way that being an ethical researcher is.

The exclusive licence agreement doesn't allow you to republish the work in a commercial venue. What you can do is post the version of the paper that you submitted to the journal - the one before it was typeset and copyedited etc - on your website. At least that's the usual contract.

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad