The ‘publish or perish’ phenomenon in science is more than a reality: it is important for scientists who seek longevity and status in their careers. The advantage, as clarified by the authors, is that ‘multiple publication of research results leads to greater reception of the research in the scientific community’. And the reception is even greater when the article is long. With length, citations increase.

The pattern that Dr Bornmann and Professor Daniel describe shows that multiple publication of research results leads to greater reception of the work in question by the scientific community. What’s more, if the article is long the number of citations it receives increase.

So it is the shrewd player who publishes lengthy articles in several journals. Such salami slicers are reacting to the heavy pressures of defending their positions. They are looking out for their future and assuring tenure and funding – in essence, they are safeguarding their careers.

The prestigious Nature magazine has called salami science a threat to the viability of scientific publishing (2) . It reports that the practise has led to tighter surveillance of scientific journals and a greater workload for unpaid peer reviewers.

Dr Bornmann of ETH Zurich’s D-GESS, Professorship for Social Psychology and Research on Higher Education’ backgrounds the study.

What are the adverse effects of salami slicing?

The state of research in any given area is summarized frequently by meta-analysis. If inter-dependencies exist in the data of different publications, they could lead to a distortion of the results of the meta-analysis, and thus to a false declaration of the research being carried out.¨

Do editors bear some responsibility for salami science?

No, I don’t think that editors in general have an overview of the author’s work who submits a manuscript to them. In the case of various ‘éminences grises’ there may perhaps be a certain awareness. But in most cases of multiple submissions from an author, the editor would have to turn to the Internet or appropriate data bases to check the different works by that author. However, this is very time consuming.

The 'Uniform Requirements for Manuscripts Submitted to Biomedical Journals' states that there can be no 'duplicate submission' and no 'redundant publication'. Salami science seems to come close to abusing this requirement without actually doing so. Is salami science an ethically sound practice?

Whether or not salami science is ethically defensible depends on how it is defined, and how it is practised. Authors have to follow the proper guidelines of journals, and be open with their editors about what they are doing. If scientists wish to reach different audiences, for example, then duplicate publications make sense. The problem is that the border between the ethical and unethical is sometimes porous. Under certain circumstances, a scientist has to repeat findings that have already been published in order to render newer findings understandable to the reader. Is this ethical or unethical?

What are alternatives for scientists who wish to avoid salami science?

The answer varies from study to study. With smaller studies, it is not a problem to publish all the findings in just one paper. However, if test series are done over a longer period it is not possible to wait with publishing until all results are available.

A researcher should always make sure that the publication of a study’s results will not create a bad impression with peers if that information is going to be printed in several publications. Otherwise, the author would appear to be more interested in making a name for him- or herself than in sharing his or her knowledge.