Saidman et al.1 describe 2 instances of ostensible plagiarism in serial editions of medical textbooks and also mention a third case. In all 3 cases, chapters in newly revised versions of these textbooks were found to contain over 80% of material from their corresponding versions written by other authors and published in their earlier editions. In the 2 cases described, a footnote was included acknowledging the contributions of the original authors, but these acknowledgements seem to minimize the more substantive contributions made by the original set of authors.
Certain features of these cases merit further discussion, but there should be no question that the transgressions described represent a breach of publication ethics. The authors of the revised chapters took substantial amounts of verbatim, text, data tables, etc., and presented these materials as their own work without giving appropriate credit to the original authors. Technically, this is plagiarism. That in 2 of the 3 cases offending authors were also editors of the textbooks makes these cases all the more noteworthy, in part, because they bring to mind a similar well-publicized case from the 1980s.2 In that episode, a highly respected physician, who had been chief of the department of medicine at a major university, had published a chapter in an edited medical textbook that contained unattributed portions of material from various chapters authored by different individuals. The plagiarized material had been published earlier in another medical textbook that the physician himself had coedited. As a result of an investigation by a university ethics committee, the physician was eventually censured resulting in his resignation as department chief.
As Saidman et al.1 point out, fundamental rules of scholarship (i.e., citation, attribution, and authorship) followed in journal articles are equally applicable to textbooks. It must be acknowledged, however, that there are certain characteristics of textbook writing that are unique to that genre. For example, many textbooks are revised periodically, and large portions of material, including text, figures, and data tables, are retained from one edition to another. Except perhaps for systematic reviews such as those from the Cochrane Collaboration, such degree of reuse is generally unacceptable in traditional journal articles. In fact, some journal editors have observed that instances of self-plagiarism are frequently found in traditional review articles.3,4 However, no one would accuse textbook authors of self-plagiarism for reusing relatively large amounts of their own material from one edition to the nexta. After all, no one would expect a new edition of a textbook to represent a completely new work because few, if any, areas of research change so profoundly in the relatively short period of time in which most textbooks are revised. But it is worth pointing out that, in many cases, authors of textbooks describe the nature and extent of the revisions in the preface to the new edition, thus providing an additional measure of transparency regarding the justification for a new edition. Unfortunately, not all textbooks follow this tradition.
Plagiarism, Inappropriate Acknowledgement, or Denial of Authorship?
Some edited textbooks present the unique situation in which the authors of the revised chapters may be different from those who wrote the original versions, and Saidman et al.1 discuss some of the reasons why this change in authorship occurs. Because revised chapters will typically retain portions of the original, the question arises as to how to handle attribution of the reused content. Standard scholarly conventions would dictate that reused verbatim material should be enclosed in quotation marks and given a citation to identify its origin. This would be an acceptable approach if the material consisted of, say, 2 or 3 paragraphs, but when the revisions consist of many paragraphs and/or many portions of paragraphs, as would be expected in a thorough revision, the repeated use of quotations and of citations to the same source becomes aesthetically undesirable and downright cumbersome for the reader. One option is to use a footnote, as was done in the cases in question, but such acknowledgements must be far more accurate in describing the relative contributions of each set of authors (and perhaps even each author). But herein lies the critical question posed by Saidman et al.1: “How much of a book chapter can ethically be republished under new authorship without including the original writer as an author?”
Regrettably, the evidence indicates that these types of inappropriate authorship practices in edited textbooks occur regularly. For example, the authors of an article on revising textbook chapters written by previous authors5 suggest that authors should contact the editor to determine how they will be credited in the book should such information not appear in the publisher’s contract that they are typically asked to sign. More to the point, these authors also write: “In most cases, when the new edition of a book is published, the writers for that edition are listed as the contributing authors. Individuals who wrote for the previous edition may be acknowledged as past contributors.”5 Thus, it seems that establishing new authorship in these cases is decided regardless of how extensive the revisions turn out to be, a practice that is not consistent with current guidance on authorship.
In clarifying the distinction between plagiarism and copyright, other authors caution that because publishers typically own the copyright of these works, they can transfer ownership of that work to others, authorize others to produce a derivative work, sell and distribute such work, etc.6 It is perhaps here that the real problem lies. A publisher who owns the work has the legal right to select individuals and assign them to revise others’ work and grant those new individuals authorship,6 and it seems that in some cases such assignments may be based on considerations other than those that merit authorship designation, such as which names will sell more books.7 But I am also convinced that these types of situations occur because some authors,8,9 journal editors10 and, I suspect, even some in the publishing industry who oversee these types of arrangements, are just not well acquainted with the ethical subtleties of plagiarism and other matters related to publication ethics.
Well-established guidance on plagiarism and authorship has been articulated by a number of professional organizations, and it is fairly consistent. For example, the guidance on authorship proposed by the International Committee of Medical Journal Editors,11 admittedly designed for establishing authorship in journal articles, is followed by most medical journals and can easily be adapted to most other forms of collaborative scholarship, such as textbooks. For example, whereas the current International Committee of Medical Journal Editors guidance requires that substantive contributions need to be made to the conception of design and to acquisition, analysis and interpretation of data in order for authorship to be earned, authorship in textbooks should equally be based on substantive contributions to the review and synthesis of the target research literature.
Grounded, in part, on these fundamental principles of authorship assignment, Wager and Wiffen12 have made a number of relevant suggestions for those who prepare systematic reviews that I believe to be equally applicable to new authors and editors of edited textbooks. For example, if the revisions produced by the new authors are minor, then they recommend that the original contributors should be kept as lead authors. If, instead, major revisions are produced, the original contributors may be retained as authors, assuming that they have reviewed and approve the new version. As subsequent revisions occur and the extent of the original contributions is minimized, Wager and Wiffen11 suggest that the original authors may be dropped from the authorship list, but they should still be acknowledged as having contributed to the earlier editions. Perhaps their contributions can also be briefly described.
In my view, many instances of plagiarism, inappropriate authorship, and other unethical (possibly inadvertent?) publication practices could be avoided if we all became more mindful of a few fundamental principles intrinsic to the reader-writer contract; an implicit contract between us, authors, and our readers/audience. Accordingly, and unless we specify otherwise, readers will always assume that our work is original. That is, the material that we present is always expected to be the product of our own individual efforts. In instances where others’ ideas, processes, designs, figures, data, text, etc. are being used, we have an ethical obligation to alert the reader of the provenance of such material in a clear and unambiguous manner. This should be done using standard scholarly conventions (e.g., citations, footnotes, quotation marks) in a manner that is consistent with the particular style of writing used in the discipline. Proper assignment of authorship (primary versus secondary author) and acknowledgement of others’ contributions is subsumed under this first principle. In addition to the originality of our work, readers tend to expect that the material presented or published is new and has not been disseminated previously. Any past dissemination, whether partial or complete, should be clearly communicated. Finally, and perhaps most importantly, the reader-writer contract specifies that our communications are accurate to the very best of their ability. The aforementioned principles are applicable to all scholarly/scientific activity, whether a journal article, a book, or a conference presentation.
In sum, Saidman et al.1 have brought to our attention yet another problematic practice occurring in the biomedical sciences that is in need of our immediate attention. That these malpractices in authorship assignment may be common in the textbook publishing industry cannot be used as an excuse. At a time when misbehavior in science is increasingly visible and the public’s trust in science is diminishing, we cannot sit idly by. Aspiring biomedical professionals and the public at large deserve better. We deserve better. Therefore, I urge authors, editors, and publishers to be more attentive to the issues raised by these cases and for all stakeholders to observe sound publication practices.
Name: Miguel Roig, PhD.
Contribution: This author wrote the manuscript.
Attestation: Miguel Roig approved the final manuscript.
This manuscript was handled by: Steven L. Shafer, MD.
a The appropriateness of authors’ reuse of their own text from one textbook to another, even when copyright concerns are not an issue, is a different matter but one that is beyond the scope of the present commentary. Cited Here...