Retraction Watch

Tracking retractions as a window into the scientific process

Ask Retraction Watch: Can I publish identical data in two different languages?

with 32 comments

questionTime for another installment of Ask Retraction Watch. Today’s query comes from a postdoc:

I have a question regarding publications in different languages: Can I publish the same identical data in two different languages? Would this be considered duplication?

Take our poll, and comment below.

Written by Ivan Oransky

September 13th, 2013 at 9:30 am

  • DT September 13, 2013 at 9:33 am

    It’s clearly duplication, and there’s papers already retracted because of that. Quite common by brazilian authors, by the way.

    • jesse emspak September 13, 2013 at 9:44 am

      But what if the original journal isn’t published in English, or you want to e-publish something for people who don’t speak it? Seems to me ok if you call the original journal up and say, “Hey, I want to get this into a place where few people can read your journal, is it ok if I re-publish and say where it originally posted?”

  • Irene September 13, 2013 at 9:36 am

    It could be published as a translation, with reference to the original paper and language.

    • pjie2 (@pjie2) September 13, 2013 at 2:49 pm

      Exactly, and that would require informed consent from both editors.

    • Booker September 18, 2013 at 12:11 am

      Completely agree. I was actually at a Cochrane Collaboration Workshop on systematic reviews and got into an argument with the speaker about this issue (who believed it was duplication and anathema).

      I’ve always found a very Anglo-centric view in these debates about publication in more than one language (and I say this as a someone from an Anglophone country who’s native language is English). For reasons which probably come back to the course of history and ‘Western’ dominance, we’ve ended up with a science system today which is very much English dominated, and my impression of this issue is definitely that it is usually English speakers that have a somewhat disdainful view of the idea of publishing in two languages. This works well for them of course because they never have to entertain the idea of publishing in a language other than English (since it’s the dominant science language and all the ‘high impact’ journals are English). But for people from other countries or cultures there can be very legitimate reasons for doing so. Unfortunately, a lot of people (as can also be seen from this survey) believe this is to be illegitimate scientific practice, and frankly I find it hard to believe that there’s not some degree of racism involved.

      I’d like to see a move away from a scientific publishing system based around simple rules like this and towards one where transparency is the name of the game – if someone publishes something in a second language, as long as it’s noted that it’s previously published elsewhere in a different language, referenced, and is clear to anyone reading that article, then I find it hard to believe that this is somehow ‘damaging to science’. Having a system which bans this practice would also place science publishing at direct odds with other forms of publication (literature, novels, etc.) where translations are commonly done in order to disseminate the work more widely. Science is a global endeavor after all, and we have to acknowledge that there is more to the world than English-only mechanisms of dissemination.

      • Miguel Roig September 18, 2013 at 6:42 am

        As a non-native English speaker who is now dominant in English, I very much appreciate your post, except perhaps for the racism part. I just don’t believe it to be the operating construct underlying the view against publication in more than one language (is linguistic anglocentrism perhaps the correct construct?). Anyway, I would only add that as we move toward a more transparent system, a mechanism must also be developed to prevent those who publish the same material in two or more languages to use those translated publications as different publications for purposes of vita padding.

  • giulio September 13, 2013 at 9:40 am

    In social sciences and humanities may be a common practice because national and international audiences exist.

    • foobar September 14, 2013 at 9:47 am


      First, the lab-science bias is again evident.

      Second, I think there might a small cultural barrier between particularly the American scientists and the ones that come from smaller countries. The usually small native-language scientific community is important and often considered as a prerequisite for international publication, although to a lesser extent nowadays. Depends on the field of course.

      Third, in my opinion it is somewhat ridiculous to even think that one could write a “duplication” in two different languages. You are welcome to prove me wrong, but then you probably have to be a translator by trade. Usually the publication in native language may moreover have a different angle and relate more to the domestic discussion. In other words, the audiences are different.

      Fourth, a publication in a native language journal hardly matters anything when it comes to tenures and alike. Didn’t we have these impact factors and whatnot?

      Last, while (broken) English might be the lingua franca, publication in the native language is valuable on its own right, to preserve the historical and cultural academic traditions. One thinks in one’s mother tongue.

  • Andrew Brown September 13, 2013 at 9:47 am

    Duplication is a copyright issue (though covert duplication is also an integrity issue), so if the copyright has been cleared then it should not be a problem GIVEN that it is cleared by both journals AND that it is made clear that it is being published again in another language (very important for meta-analysis, for instance). The purpose of publishing scientific results is to communicate them to a wide audience. Unless we force a universal scientific publishing language, or journals start offering translated versions, this is a reasonable way to widely disseminate results. This, of course, assumes a genuine desire to share one’s results rather than CV padding.

    • Sampath Parthasarathy January 24, 2015 at 11:59 am

      They may publish translation but they should not count that as two articles. If it is allowed, in how many languages I could publish? Is there a limit?

  • ferniglab September 13, 2013 at 10:00 am

    Republishing in the lingua franca of science, English is extremely valuable. There used to be translations of key papers in Russian made available a long time ago and this was very important, as it allowed access to ground-breaking science. So there clearly a need here, particularly with the growth of the science base world wide. If I recall there was a complaint that a leading Chinese microbiologist had published important work on SARS or bid ‘flu (I cannot recall which) in a Chinese journal and that the data were, therefore, not accessible to the rest of the world.

    One might turn this upside down and consider that in the context of open access and open data, publishing in any language reduces access (language barriers are a very effective “paywall”). This, of course cuts both ways, since english itself is a barrier to some.

    One route would be to contact the editor of the 9presumably) english language journal, explaining the importance of the work and ask whether the journal would consider a translation. There is the issue of copyright – the original publishers would have to agree, though if the original and the english target are both OA (good OA, not predatory), this may be simpler.

    Another route is to go to the original journal and discuss the importance of the work with the editor and offer to produce a translation, which can be associated with the original article. Readers can then choose their language and you attain the same end – a wider audience for your work.

    • cw September 13, 2013 at 10:11 am

      NO!!! Its freaking duplication!

      • ferniglab September 13, 2013 at 10:27 am

        Not it isn’t, it is translation. Many works are translated and we do not consider this to be duplication, because they are clearly identified as such. Without translators we would not be able to access huge swathes of human cultural achievements and science is part of this. So as long as the work is identifiable for what it is, a translation, then no problems. Having struggled and largely failed with early 20th century papers written in German, I certainly would have appreciated a translation. Duplication only if the paper is being passed off as something new. This has, correctly, been identified as plagiarism and resulted in retractions, featured here.

  • Bob Roehr September 13, 2013 at 10:21 am

    If the purpose of such “duplication” is to more broadly disseminate information, particularly for clinical care, then I see no problem with such publication so long as it is prominently identified by a tag along the lines of: A version of this paper was originally published in (language X) as (original title).

    However, with the growing sophistication of translation software provided by Google and others, there should be less need to do so.

    • Andrew September 16, 2013 at 4:19 am

      We’re a long way from machine translation good enough to translate academic papers.

  • tintin September 13, 2013 at 10:27 am

    Indeed it is considered as duplicate data publication. Ask Rioufol et al. who lost a paper in Circulation (best cardiology journal IF > 12) because they had published their data first in the French journal “Archives des Maladies du Coeur et Vaisseaux (IF < 1)….about IVUS of multiple atherosclerotic plaque ruptures in acute coronary syndromes…(by the way, now the French journal is published in English!!!)

  • Margaret Winker, MD September 13, 2013 at 10:28 am

    Publishing a translated work in medical journals is acceptable as long as conditions are met as reflected in the following policies:

    The World Association of Medical Editors policy states “Republication of a paper in another language, or simultaneously in multiple journals with different audiences, may be acceptable, provided that there is full and prominent disclosure of its original source at the time of submission of the manuscript. At the time of submission, authors should disclose details of related papers they have authored, even if in a different language, similar papers in press, and any closely related papers previously published or currently under review at another journal.”

    The ICMJE states:

    Acceptable Secondary Publication

    Certain types of articles, such as guidelines produced by governmental agencies and professional organizations, may need to reach the widest possible audience. In such instances, editors sometimes deliberately publish material that is also being published in other journals, with the agreement of the authors and the editors of those journals. Secondary publication for various other reasons, in the same or another language, especially in other countries, is justifiable and can be beneficial provided that the following conditions are met.

    1. The authors have received approval from the editors of both journals (the editor concerned with secondary publication must have a photocopy, reprint, or manuscript of the primary version).

    2. The priority of the primary publication is respected by a publication interval of at least 1 week (unless specifically negotiated otherwise by both editors).

    3. The paper for secondary publication is intended for a different group of readers; an abbreviated version could be sufficient.

    4. The secondary version faithfully reflects the data and interpretations of the primary version.

    5. The footnote on the title page of the secondary version informs readers, peers, and documenting agencies that the paper has been published in whole or in part and states the primary reference. A suitable footnote might read: “This article is based on a study first reported in the [title of journal, with full reference].”

    Permission for such secondary publication should be free of charge.

    6. The title of the secondary publication should indicate that it is a secondary publication (complete republication, abridged republication, complete translation, or abridged translation) of a primary publication. Of note, the NLM does not consider translations to be “republications” and does not cite or index translations when the original article was published in a journal that is indexed in MEDLINE.

    7. Editors of journals that simultaneously publish in multiple languages should understand that NLM indexes the primary language version. When the full text of an article appears in more than one language in a journal issue (such as Canadian journals with the article in both English and French), both languages are indicated in the MEDLINE citation (for example, Mercer K. The relentless challenge in health care. Healthc Manage Forum. 2008 Summer;21(2):4-5. English, French. No abstract available. PMID:18795553.)

  • Scot Wilcoxon September 13, 2013 at 10:34 am

    The headline mentions “identical data”, while most of the discussion assumes identical papers. Identical data but different papers on different aspects should not be an issue (ie, a lunar sample’s elemental composition being interpreted as lunar origin, isotope history, or geological formation). Identical, albeit translated, papers does run into copyright and duplication issues.

    • Miguel September 13, 2013 at 12:59 pm

      I think that there is something potentially troublesome when someone engages in acceptable (as per WAME or ICMJE standards) duplicate or redundant publications “as long as (they) clear it with the relevant editors”. The assumption there is that readers will be similarly informed about the nature and extent of data reuse beyond a simple author note such as that suggested by ICMJE. Do we really know that this type of full disclosure happens in every case where editors approve reuse? I am not so sure. No. it is not enough for editors to simply approve the reuse and for a note to indicate that reuse has taken place. What is most important is for readers to be fully informed about the nature and extent of the reuse. An editor may, of course, deny publication of a paper that reuses previously published data. But when any form of reuse is approved by editors, they must also compel the authors to similarly inform readers about the details of the reuse by providing them with a full explanation about the nature and extent of such reuse.

  • Dan Jenkins DDS September 13, 2013 at 1:04 pm

    I feel as long as it is clear that is a re-publication of the same information in another language there should not be any confusion as to its purpose. This would not be re-publishing to simply add to the number of articles an author has published. If it is clear to the first and second publisher as well as to the readers of the second publication it would seem to be ethical.
    Dan Jenkins DDS, AADE-CDE
    President, American Association of Dental Editors

  • KGM September 13, 2013 at 1:15 pm

    A thought for another poll or comment – what about an author(s) ‘republishing’ data/results in a journal that previously published in a patent or patent application?

    • Alpha September 13, 2013 at 5:34 pm

      Or what about an inventor (sole patent author) copying large segments of a paper without permission, acknowledgement or citation from a previously published with several collaborators? This has happened to me where a patent appeared a couple of years later containing the words and equations of one of my published papers, but without any of the other authors of the paper being listed on the patent. Both documents remain in the public domain.

  • Phronesis September 15, 2013 at 10:43 pm

    After I voted the results were 259 YES 259 NO. Very suspicious looking data to me. I demand an inquiry.

    • JATdS September 16, 2013 at 2:09 pm

      I feel VERY strongly about this topic but I wish to remain anonymous because I know that at least 50% of the bloggers will disagree (judging by the poll numbers). Whether most of you know this or not, the Impact Factor (IF) is used to generate funds (and profits) in countries like China. You may ask, what has the IF to do with this RW blog entry? Allow me to explain my ideas in more detail. A paper published in Chinese without an IF would generate XYZ Yuan. However, a paper published in an English journal without an IF would generate XYZ x 2 Yuan while a paper published in an English journal with an IF would generate XYZ x 10 Yuan x IF score. The factor of 2 or 10 I used could be anything from 2-40. In other words, there is a MASSIVE incentive, if not only financially, for Chinese scientists to re-publish their work in English journals, with or without an IF. This system applies to China, India, Iran, and goodness knows how many other non-Western countries who deify the IF. Herein lies the conundrum. I suspect that 99% of the scientists or editors commenting above have never encountered such a case, i.e., of a request to re-publish in their journal data that was previously published in another language. I personally think that it is unethical to re-publish a set of data in two languages for a few reasons.
      1) If a non-English scientist decides, at a moment in their career, that it would have been better to publish in English than in Hindi, Chinese or Farsi, then they should re-structure that information, for example, in the form of a review, part of which contains a deeper explanation of the data that was published in that original language. So, a review would be an appropriate way to represent the highlights of the finding.
      2) Where in fact a scientist wants to re-publish the full data-set, exactly as it had appeared in an original language, then, in addition to requesting and obtaining permission from the Editors-in-Chief, with a full consensus of the ENTIRE editorial board (to avoid bias by the EIC), as well as the publishers, a conflict of interest (COI) must be provided, specifying explicitly that no additional tangible financial benefits in terms of salary, position, funds, etc. will result from this re-publication. In this case, and in this case only, will re-publication serve for the ACADEMIC value of a journal and peers.
      3) Papers published earlier than 2000 will rarely present new advances to a 2013 audience, except for round-breaking advances. Ground-breaking advances, even in studies conducted earlier than 2000 would most likely have targeted to-level English (or Western) journals anyway, since these would have dominated the international publishing arena anyway until the start of the open access movement, at least.
      4) Most respectable foreign journals contain also the possibility of publishing an English abstract, so most important information can be exposed to the international public without the need to re-publish the data set.
      5) I personally believe that 99% of scientists who want to re-publish their full data-sets in English are involved in foul play. Most scientists nowadays understand that the lingua franca of science and publishing is English, and this has been a universal truth for at least the last 2 decades. French, Spanish, Portuguese, Arabic, German have all lost market ground to English, whether we like this or not. With the advent of the open access movement, we have seen a resurgence of these languages (e.g., the Scielo system by the Brazilian government, or Redalyc for the South American continent, serving Portuguese and Spanish communities). But, overall, most scientists know that when they publish, they want to reach the widest audience possible, globally. That means an English journal, in English. So, for any paper published in another language before 2000, i.e., before the start of the open access movement, does not deserve to be re-published.

      Despite my arguments against the re-publishing of data, simply because I think it reflects more of a sinister plan to milk the system rather than to increase understanding of a topic, I also see some arguments in favor of the translation. An honest compilation of data, for example, of results published about a topic by Chinese, Vietnamese or Nepalese scientists would of course benefit Western scientists, no doubt. Similarly, work published in French, German, Spanish, or Portuguese would help advance our understanding of topics in the East. Translation of works, when done ethically and without conflicts of interest, draws benefits. For example, to use a non-scientific example. Had Paulo Coelho’s works not been translated into English, there would have been no way of appreciating this thoughts and ideas, which had originally been written in Portuguese. But science fiction is one thing. And hard-core science is another category altogether. Finally, just to irritate some of the editors and societies who are always so quick to establish ethical rules (using Western values and ideologies) without actually having a deep understanding of other languages and cultures: should an editor or editor board that approves the translation of a work, let’s say from Chinese into English, for re-publication, at least be bilingual? This is important, because one POSSIBLE reason why scientists may seek to re-publish, is to fix mistakes in the original publication, i.e., to mask a mess that may have previously been published, and to seek a second chance to correct the academic record. This last issue is worth a fiery debate, but only provided we have literate editors and bloggers who speak multiple languages fluently enough to understand, almost perfectly, the content of, for example, a Chinese scientific paper. In other words, Google translators or simply won’t do.

      Fait acompli.

  • Too many fools making too many rules September 16, 2013 at 8:35 pm

    4.7 Multiple submissions of research findings
    It is not acceptable to include the same research findings in several publications, except
    in particular and clearly explained circumstances, such as review articles, anthologies,
    collections, or translations into another language. An author who submits substantially
    similar work to more than one publisher, or who submits work similar to work already
    published, must disclose this at the time of submission.

  • JA Pascoe September 17, 2013 at 3:34 am

    I would prefer an extra answer on the poll: Yes, as long as it’s clear with the relevant editors * and the readers *

    I think author statistics should count the original and the translated article as one (this includes summing the citations for both versions).

  • R Lammey September 23, 2013 at 8:11 am

    A paper, by Helen Zhang (she writes a lot about matters of plagiarism etc) has recently covered some aspects of this: Bilingual (multilingual) publications and duplicate publications: for or against? (YH ZHANG) I found it interesting and I told Helen I’d post it on this site as she can’t access it.

  • Rogier Stuger November 23, 2013 at 10:24 am

    Any obstacle to spreading scientific data to the largest possible audience should be removed. That includes language barriers. If the authors disclose the original publication and promise in writing that they didn’t add, change, or remove anything in translation there’s nothing wrong with publishing a translated version. Sharing knowledge to the largest audience possible is way more important than nitpicking about copyrights, chinese scientist making a few extra yuans, or repeating the “thou shalt not publish more than once” mantra.

    The experiment that you’re gonna spend the next six months on may already have been published in chinese, russian, or swahili, but if the authors can’t publish a translation you’ll never know.

  • Cedric April 20, 2014 at 4:04 am

    I am an American physician working in China. I plan to publish data collected while in China in an English language medical journal. However, I know that many of the people who would directly benefit from the research are Chinese doctors, patients, and policy makers who can not read English-in effect it is not accessible to them. Therefore my plan was to also publish a translation of the English article into Chinese. If the editorial board from both journals agreed, and the translated work prominently cites the original article, is this proper? Thanks for any opinions.

  • Dr. B.J.C.Perera December 7, 2015 at 8:46 pm

    I am one of two Joint Editors of Sri Lanka Journal of Child Health and a Section Editor of the Ceylon Medical Journal.
    I used to have the conviction that multiple language publications would be unacceptable. However, in view of all the opinions expressed and the Guidelines from ICMJE etc., I have now changed my mind. One has to move with the times and the argument of making research information available to all is a compelling one.
    Incidentally, this problem does not directly concern the scientific journals published in Sri Lanka as all those are published in English. However, we do get submissions from outside our country and the question may arise with some of those submissions. So far this has not occurred but one never knows.
    Best regards to all who have expressed their views.
    Dr. B.J.C.Perera

  • Abdul Rauf Ch. June 23, 2016 at 8:40 am

    Very interesting and informative discussion about publishing in more than one language. My question is regarding the DOI for such published work. Will the translated versions of the parent publication have the same DOI number as that of the parent publication or a different number for each language?

  • Post a comment

    Threaded commenting powered by interconnect/it code.