DR. BRUCE ALBERTS
President, National Academy of Sciences
Subcommittee on Government Management, Information, And Technology
Committee on Government Reform
U.S. House of Representatives
Thank you for holding these hearings and for providing this opportunity to present testimony concerning the provision of Public Law 105-277 commonly referred to as "the Shelby amendment." As you know the Shelby amendment requires the Office of Management and Budget to amend OMB Circular A- 110, which applies to Federal grant awards, "to require Federal awarding agencies to ensure that all data produced under an award will be made available to the public through the procedures established under the Freedom of Information Act."
do not disagree with the basic objective of the Shelby amendment to the extent that it would result in making scientific data publicly available for reevaluation by other scientists. We have issued a number of reports through the National Research Council (NRC) stressing the importance of data sharing and openness for both science and federal decisionmaking. Such sharing is an important tradition in science. As emphasized in our 1995 booklet for students, ON BEING A SCIENTIST: RESPONSIBLE CONDUCT IN RESEARCH, "After publication, scientists expect that data and other research materials will be shared with qualified colleagues upon request. Indeed, a number of federal agencies, journal and professional societies have established policies requiring the sharing of research materials. Sometimes the materials are too voluminous, unwieldy, or costly to share freely and quickly. But in those fields in which sharing is possible, a scientist who is unwilling to share research materials with qualified colleagues runs the risk of not being trusted or respected. In a profession where so much depends on interpersonal interactions, the professional isolation that can follow a loss of trust can damage a scientist's work."
However, there are numerous problems with the Shelby amendment that arise from three of its fundamental aspects. First, the Shelby amendment relies solely on the Freedom of Information Act (FOIA) as the mechanism for disclosure. Valuable resources will be deflected from science into FOIA related administration, bookkeeping and legal battles. Second, the Shelby amendment does not define the key phrase "all data produced under an award." And third, the public data availability specified would not necessarily follow the completion of the researcher's scientific work, as signaled by its publication in a scientific journal.
Application of the Freedom of Information Act to Federally funded research grantees through the Shelby amendment will be extremely burdensome and costly to researchers and research institutions , and we see little evidence that the application of this new concept and its impact have been thought through. For example, researchers could be forced to make certain information publicly available, including their lab notebooks, draft manuscripts, electronic mail, and raw research data, even before its publication and analysis. We can predict that it will have a chilling effect on joint university-industry research collaborations, and that it will be used by various special interest groups to harass researchers doing research that these interests groups would like to stop. It will be exploited by both foreign and domestic concerns, as well as foreign military interests, as a new tool for scientific espionage.
For example, commercial interests that have a strong competitive interest in particular areas of research will now be able to use FOIA requests to obtain university-based research data for their own use and competitive advantage in an effort to dominate or control that area of research, ultimately discouraging independent university research in these areas. Where universities have industry partners for jointly sponsored research projects, commercial concerns can use FOIA requests to obtain research data from these projects to the detriment of the actual project sponsors, who are their competitors. Existing legal precedent is unclear on what protection, if any, would be available under FOIA to such industry sponsors, and as a result they are likely to be much less inclined to participate in research with universities. And foreign military interests acting through intermediaries will surely use FOIA requests to obtain Government-funded basic research data for use in their own research and development programs.
Another fundamental problem with the Shelby amendment is understanding exactly what the word "data" means in this legislation. Some of the definitions of the term "data" within the Federal government are extremely broad. However, we have suggested to OMB that statements by Senators Lott, Campbell, and Shelby on the Senate floor concerning the scope and purpose of the Shelby amendment make it clear that the "data" to which the Shelby amendment applies is "research data."
OMB Circular A-110 does not define the term "research data." We have suggested to OMB that the "research data" contemplated by the Shelby amendment are the raw data that result from research observations and experiments under Federal grant awards and that the U.S. Supreme Court's decision in Feist Publications, Inc. v. Rural Telephone Service Co., Inc., 499 U.S. 340 (1991), provides the legal definition of "raw data."
The Feist Publications case is a copyright case in which the Supreme Court explained the distinction between "facts" that cannot be copyrighted and "compilations of facts" in which there can be a copyright under appropriate circumstances. The decision equates the concept of "raw data" with "uncopyrightable facts." (499 U.S. at 361).
As the Supreme Court explained in Feist Publications, all facts -- scientific, historical, biographical, etc. -- are inherently part of the public domain and may not be copyrighted. They are available to everyone and merely await discovery. In scientific research, these "uncopyrightable facts" are the "raw data" which it is the purpose of scientific research to discover as part of the process of increasing our scientific understanding of natural processes and phenomena.
Relying upon this concept, we have suggested to OMB that "research data" should be defined to mean "facts, which are in the public domain and may not be copyrighted that result from scientific observation, experiment, or similar methods of research." These facts in the public domain are the raw data of scientific research.
We have also suggested to OMB that the definition of "research data") should contain a provision that would, for research involving human subjects, define "research data" in a way that would require that any information that would identify any specific individual be aggregated or redacted. Research data on human subjects are currently routinely shared, transferred and used by researchers in aggregated or redacted formats that reasonably prevent the identification of specific individuals.
The third fundamental problem with the Shelby amendment is that the Freedom of Information Act clearly does not protect a scientist's right to publish the results of the scientist's own research. In Burka v. U.S. Dept. of Health and Human Services, 87 F.3d 508 (D.C. Cir. 1996), the U.S. Court of Appeals for the District of Columbia squarely held that research data could not be withheld under Exemption 5 of FOIA (which applies to certain inter-agency and intra-agency materials) to allow researchers to publish scientific articles because there is no "established or well-settled practice of protecting research data in the realm of civil discovery on the grounds that disclosure would harm a researcher's publication prospects." 87 F.3d at 52 1. And we know of no other generally applicable basis under FOIA for protecting a researcher's publication
Thus, Federally funded research grantees now face the threat under the Shelby amendment of having their research data made public under the Freedom of Information Act before they have had the opportunity to publish the results of their research. But publication of research results in peer-reviewed scientific journals is one of the most critical elements of the entire research process. It is the means by which new discoveries are communicated to others in the scientific community and to the public at large. Permitting the researcher who actually collected the data to be the first to analyze and publish conclusions concerning the data is an essential motivational aspect of research.
Requiring public release of research data prior to publication in scientific journals would seriously short-circuit the scientific research process that has been so effective in the United States. It would severely disadvantage Federally funded scientists while providing unreasonable advantages to their competitors, both in the United States and internationally. And premature release of research data before careful analysis of results, and without the independent scientific peer review that is part of the normal process of publication of scientific research, would also increase the risk of public disclosure of erroneous or misleading conclusions, thereby creating unnecessary confusion and loss of confidence in science on the part of the public.
To address this problem, we have suggested to OMB that Circular A-110 might be amended, for example, to provide that when research findings produced under an award are published, the principal investigator would promptly send a copy of (or citation to) the publication to the Federal awarding agency. Thereafter (assuming the Shelby amendment remains in place) in response to a Freedom of Information Act request directed to the Federal awarding agency for the publication and the underlying research data, the Federal awarding agency would (1) send a copy of the publication to the requester and (2) obtain from the principal investigator the underlying research data identified by the principal investigator as the basis for the research findings in the publication so that these research data can be made available to the requester and the public under FOIA.
This approach would make "publication" the triggering event for application of the Shelby amendment -- although this may be controversial because there is no explicit reference to "publication" in the Shelby amendment. It would also introduce the problem of defining what it means to "publish" the results of scientific research.
We have pointed out to OMB that the scientific community would generally regard publication in the peer reviewed scientific literature to constitute "publication" and not press reports, seminars or talks at scientific meetings. However, the world of scientific publishing is changing as it becomes increasingly electronic, and any precise definition of "publication" of research may need to be modified in the future. While we recognize that a dispute over the definition of "publication" could easily lead to litigation in the future, this seems to be the only available method of protecting the essential publication rights of researchers under the Shelby amendment.
Finally, in my opinion, FOIA is fundamentally flawed as the mechanism here, because it fails to require evidence from the data requester that the disclosure of the data in question is in the public interest. Congress needs to do more investigation of this concern.
The Freedom of Information Act was designed to provide public access to Government records, not to all of the research data of Federal research grantees. I welcome this hearing today and believe that Congress should hold additional hearings to gain a better understanding of the problems that would be created by the application of the Freedom of Information Act to Federal grantee research data. Then Congress can craft specific legislation to provide for public access to Federally funded research data that balances the interests of the public in access to data with other important public interests.
Such legislation should, at a minimum, provide specific statutory language for the protection of the publication rights of researchers and the proprietary interests of participants in joint university-industry collaborative research projects. In the meantime, the Shelby amendment should either be set aside or at least temporarily suspended to permit a congressional study of these issues and the enactment of specific legislation.
The National Academy of Sciences would be pleased to work with the committee and the Congress to help develop a workable system -one that would properly balance the interests of the public in having research data made public with the interests of the scientific community, so that our scientists and engineers can continue to contribute to the improvement of our living conditions and our economy.
I offer for the record copies of two letters that I have sent to OMB concerning the Shelby amendment. I would also be pleased to answer any questions that you may have.