jump to navigation

Is auto-plagiarism acceptable ? September 3, 2007

Posted by dorigo in Blogroll, books, personal, physics, science.

Here is an instance of a question that is best asked in a blog. I know this site is read by a few academics who frequently publish their work and attend to conferences, and I thus expect to hear from them what they think of the following practice.

Imagine you give a talk at a conference, on some subject, and a few months later you give a very similar talk at another conference. You have changed your slides only to accommodate new results, some new considerations, added some material and subtracted some. It is legal to do that: in fact, you are sometimes even expected – or asked – to give the same seminar in different places.

However, when you later have to write the proceedings for the second conference, you are assailed by a doubt: should you start writing your proceedings article anew, or make heavy use of the one you published just a few months back ?

Proceedings are printed matter, and as the old adage teaches – verba volant, scripta manent – you know somebody could one day confront you with the question: “What advancement to the field did you grant with this second paper?”. None at all, of course: a writeup of a conference talk is not meant for that purpose. Still, if you cut and paste large portions of stuff that is already published elsewhere, you feel guilty.That is, I do.

I am less interested of the legal side of the question, but there is one. I signed a transfer of copyright agreement with AIP (the American Institute of Physics) when I submitted my manuscript last September. Now, that very document explains that I have “The nonexclusive right, after publication by AIP, to give permission to third parties to republish print versions of the Article or a translation thereof, or excerpts therefrom, without obtaining permission from AIP, provided the AIP-prepared version is not used for this purpose, the Article is not published in another conference proceedings or journal, and the third party does not charge a fee. If the AIP version is used, or the third party republishes in a publication or product charging a fee for use, permission from AIP must be obtained.” So, since the PASCOS conference will also use AIP as a publisher of their proceedings,  it looks as if I will have to ask AIP to use excerpts of the first proceedings for the second one, of which they will still be the publishers!

Hmmm… I am more interested in the moral issue. Any ideas ? What do you do in similar cases ? Try to reinvent the wheel ? 



1. Jon Lester - September 3, 2007

Dear Tommaso,

I would simply avoid to publish further about. I mean, having already published the same material in another conference proceedings, I would not give another contribution.

I have seen largely applied this practice as I have partecipated to several conference where, in the proceedings, often the contributions by the most important speakers were missing. These are the ones mostly expected to give a lot of conference presentations.


2. dorigo - September 3, 2007

Hi Jon,

funny! It never occurred to me that not writing a paper was an option, but it is… I tend to consider it impolite to deny one’s contribution, mainly because I always hate it when I pick up a proceedings book and find empty pages throughout. However, it is indeed a sensible thing to do in cases such as this one.

Mumble mumble. I have always given high priority to publishing papers with my only name on them, since I belong to 500+ collaborations and by some metric my publishing record is ridiculous if you take away the proceedings (I refer to weighting the number of papers by the inverse of the number of authors, as some advocate). On the other hand, I do not need one more paper so much…

Other opinions on this issue ?


3. Andrea Giammanco - September 3, 2007

I’m in favour of re-publishing.
As you said, a conference proceeding is not supposed to say something original, in any case (at most it will contain an original way of showing and contestualizing results published elsewhere).
Of course, not publishing at all is an option, but every time I see missing proceedings by some famous scientist I’m quite disappointed. I don’t think they do because they feel a moral issue in recycling a text, I just feel it as a lack of respect towards the reader.
How may I (the reader) know where their original proceeding has been published ?
And a collection of proceedings is, in a sense, a thematical book, which is a great place where to find content. If I don’t find the expected content where it belongs, I’m disappointed more than by discovering that the same text has been already published elsewhere.
In some cases, I even re-read the same text that I’ve read elsewhere. In the new context, my second reading will give me something anyway.

By the way, I almost never wrote two very similar proceedings: I’m a perfectionist, so I never like something that I wrote in the past (whether it is scientific or non scientific), so I recycle very little text. How little is my “very little”, well, it depends on time constraints…

4. dorigo - September 3, 2007

Lol Andrea, I am a perfectionist too, but a more successful one 🙂
In fact, my embarassment in having to rewrite something is that I will be unable to surpass my previous effort!


5. delio - September 3, 2007

do physical proceedings really include blank pages if a participant hasn’t submitted anything? funny.

well, in maths it’s quite ok to deny a contribution – in fact it’s quite common; and it’s even more common to publish a contribution about another (minor) topic which has NOT been discussed during own’s talk at the conference, and is not worth a a journal publication.

that said, i must admit that whenever i (as co-editor of a proceedings volume) happen to find clear cases of self-plagiarism – typically, blowsy cut-and-paste – i am quite pissed off.

> a conference proceeding is not supposed to say something original

well, 30 years ago it used to be different. sadly, only few libraries bother to buy proceedings because only few scientists bother to publish interesting, new (or at least newly explained) results there.

6. Moshe - September 3, 2007

What I find most sensible is one writeup sent to the arxiv as “talk given in X,Y,Z”, possibly slightly extended or modified, instead of 3 variations on the theme. The proceedings themselves (unless they are refereed) are a minor issue, but for any kind of paper I’d refuse to publish in a place that demands exclusive copyrights.

7. Moshe - September 3, 2007

Just to be clear, if the proceedings do not demand copyrights, I’d publish exactly identical version, which I find better than slight variations.

In any event given much cheaper and more universal ways of distribution, I really see no reason libraries should spend hundreds of dollars on conference proceedings, but that’s a different story.

8. dorigo - September 4, 2007

Hi Delio,

yes, our proceedings usually have a blank page with “The contribution was not received in time for publication” or a similar apologetic remark. As for writing something different, I would be killed! I am writing on behalf of the CDF and D0 collaborations… Anyway, I will take in account your feeling about plagiarism!

Dear Moshe,

what you propose is sensible, but I am talking about different contributions, with some parts cut and paste, rather than the same content. What I mean is, the presentation of 2006 was based on results that have since been updated: all the results I quote in the 2007 proceedings are different. Only, I have been considering some cut and paste of description of the experimental apparatus (which has not changed) and some considerations about global SM fits which still hold. As for copyrights, I agree with you, but what can we do – if the organizers of a conference decide to publish with a publisher that demands copyrights, you either comply or leave a blank page….
And of course I agree on the uselessness of conference proceedings, but that’s another story.


9. Proofread my PASCOS 2007 proceedings « A Quantum Diaries Survivor - September 5, 2007

[…] September 5, 2007 Posted by dorigo in language, physics, books, personal. trackback After polling my twentyfive readers (a wink to A.Manzoni’s “I promessi sposi”, cap.I, par.31) […]

10. Count Iblis - September 8, 2007

I often use an older article as a template when writing a new article. Sometimes I keep the introduction, figures and their captions, etc. (almost) exactly the same. Why make changes if it is not necessary?

11. dorigo - September 8, 2007

Hello count iblis,

thank you, in fact I mostly do the same… It is a bit awkward to admit, but in fact when one is writing on more or less the same topic, well… It just seems the reasonable thing to do.


12. Thomas D - September 12, 2007

Unfortunately, even if you publish, almost no-one will read it. Compare the space you have available on your slides versus in the proceedings. Unless they are talking about a single, simple, result, no-one can say anything coherent and useful in the tiny space you are allowed. If you want to understand the physics you have to read the paper anyway, or even better, attend the talk.

It is much better simply to maintain an online electronic record of the slides presented. Almost all proceedings are a scam of publishers to make more money.

Being a theorist though, I have the luxury of having independent papers (ie not within a huge collaboration) of my own. However, does a proceedings talk, whose results come precisely from the efforts of the huge collaboration, really count as your ‘own’ publication?

And if the results you present in two different conferences are essentially the same, what point could there possibly be in reformatting them?

It would be nice if conferences and proceedings were used to present some slightly eccentric ideas which are perhaps not ready for publication yet. But almost everyone (quite understandably) is trying to sell their latest paper.

Sorry comments are closed for this entry

%d bloggers like this: