Volume 21, No. 1, Fall 1999
Scientific integrity is more than a matter of avoiding the “big three”: falsification, fabrication, and plagiarism. It is a complex moral ideal, calling upon the character of the scientist, to be sure, but also on institutional culture and structure. One scientist in our community has talked about “the relentless opportunity for error in data reporting. It is so hard to keep track of every side effect, adverse event, outcome, that many things go unnoticed.” Pressures for profit (in industry) and for grant funding or stature (in the university) can combine with simple human fallibility in dangerous ways. In addition to these deforming external pressures, institutions can fail to provide the stable infrastructure that is a crucial safeguard against error.
Nonscientists may not realize, as another veteran researcher has remarked, the extent to which research depends on a finely-tuned network of support operating somewhat automatically. When units are restructured — combined or split in new ways — that essential foundation can crack. One of our associates, discussing this problem, speculated that the new “Life Science Corridor” within Michigan could serve to buffer inevitable institutional restructurings: perhaps investigators could be temporarily assigned to established appropriate units in other universities.
Another facet of scientific integrity involves the development of standards in new areas. One scientist remarked, “My work has been called ‘cutting-edge.’ While the label sounds wonderful, what it means in practice is taking significant scientific risks which could mean getting the brass ring or falling off the merry-go-round. It also means being on the “edge”—in fact, the brink—of many regulations because there are no precedents.” How does a community best work together to develop standards in previously uncharted areas?
So, talk about scientific integrity could be more far-ranging than it often is. Below, two of the Center’s Adjunct Faculty talk about other unexplored aspects of that moral ideal.
—JA
We tend to think about misconduct in terms of irresponsibility. But in doing research we can also be ir-responsive: we can fail to respond to the concerns of those our work will affect. Should standards of conduct include a notion of responsibility for its consequences?
Martha (“Marty”) Crouch first raised this question for me when I was a graduate student. Marty, a biologist at Indiana University, discovered that her well-intentioned botanical research was being used to propagate economically oppressive and environmentally devastating forms of agriculture. She decided she had to abandon her research program. She now works on sustainable agriculture and teaches about the biology of food.
In my own work on human intersexuality, I’ve tried to be aware of how my research and publication affects the medical professionals and intersexuals on whom I focus. The trick has been to think about how to be responsive without sacrificing intellectual freedom and an ideal of objectivity. In this arena, I have found especially helpful this article, written by two philosophers: Lisa M. Heldke and Stephen H. Kellert, “Objectivity as Responsibility,” Metaphilosophy, vol. 26 (1995), pp. 360-378.
Alice D. Dreger is an Assistant Professor
in Lyman Briggs School, MSU
An essential part of science is publication, and unfortunately there are some gray areas about the integrity of the peer review process. I’ve had three experiences that made me raise my guard. (Identifying details of these stories are changed.)
First, a referee thoroughly trashed a piece I submitted, and then published something very similar himself. (The editor of the journal inadvertently told me the reviewer’s identity.)
To make the story still more complicated, the writer in question sent me a copy of the manuscript he was writing and asked me to comment on it. He had forgotten that he had reviewed mine!
There’s an ethical double-edged sword in reviewing manuscripts. On one hand, someone knowledgeable in the area is the best qualified to review the manuscript; on the other hand, there can be a temptation to incorporate an argument into one’s own work. I think this is especially true in philosophy where one writes thoughts rather than doing experiments.
Along this same line, I mentioned to someone at a conference that I was working on a grant proposal. She was eager to hear more; I somewhat hesitantly went into more detail. Later this person (now director of an ethics center) submitted almost exactly the same topic for funding. I have a copy of the proposal, so I can assure you that this story is not hearsay. Of course, how I got a copy of the proposal is another issue...
Another ethical problem area concerns the use of references one has not read. In one manuscript, the authors quoted almost verbatim from a piece I had written, including references so obscure that I think it is impossible the authors would have been able to find them. In fact another writer had contacted me about these references, and told me that her paper was in press in a journal whose editors had asked her to provide proof of the references. She contacted me to get a copy of them. She had never read the references, yet included them in the manuscript, and submitted it. This is probably more serious, as well as more likely, in medical science journals because of the pressure to cite all the previous literature. It is not uncommon to have manuscripts with hundreds of references. In philosophy on the other hand, the “meat” of the manuscript is often an extensive examination of what one or several previous authors have addressed. It would be difficult to write a critique if one hadn’t read it.
A third gray area concerns delays in publishing. I had a manuscript accepted at a journal two years ago; it’s still not in press. Do editors have an obligation to publish in a timely fashion? This editor refuses to publish the date accepted because the journal will “look bad.”
The ethical issue here is one of the enormous power the journal holds over authors. To balance this, the journal ought at least to indicate what their turnaround rate is. The paper in question includes a review of the literature that is now more than two years old. When it appears it will suggest that I did a poor literature review, which is not the case. I suggested that the journal do a special edition or a supplement to remedy their backlog, and was told that the publisher wouldn’t pay for it. Unsuspecting authors, whose careers may be at stake, are completely at the mercy of journals.
It should be possible to develop ethical guidelines about these matters. In the meantime, I’ve become uncomfortably cautious about sharing my ideas with others.
Mike Fetters, MD
Family Practice, University of Michigan