September 25, 2003
Social Sciences And Humanities
Special Working Committee
Interagency Advisory Panel on Research Ethics
Dear Committee Members:
We are a national organization dedicated to academic freedom and scholarship (http://www.safs.ca). We are writing in response to a call for comment, as per http://www.pre.ethics.gc.ca/english/publicparticipation/commentondocuments.cfm.
As we understand it, PRE's Social Sciences and Humanities Research Ethics Special Working Committee (SSHWC) is presently engaged in a "consultation on evolving the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (TCPS) to better meet the needs of Canada's Social Sciences and Humanities (SSH) Communities."
We are pleased to see this discussion, because we have had concerns about the continued expansion of ethics regulations, especially given the lack of demonstrated effectiveness for such regulations. In our judgment, changes in the research ethics review system, in the past decade in particular, have greatly increased the burdens on individual researchers in the SSH and on local university administrations, with no documented benefit, and very likely no actual benefit, to the genuine protection of research participants. This cost-benefit problem is especially apparent in the SSH, and thus we feel modifications of the TCPS are very much in order, at least in the SSH. We will summarize some issues that we believe need to be addressed.
1. The medical model is being misapplied.
As we see it, many if not most of the problems were due to the attempt to create a single set of guidelines for all three granting agencies. This may have some veneer of efficiency in the eyes of bureaucrats, but in practice it has not worked. For one thing, what this blending has done is to allow the discussion of research safety to be completely dominated by what is a very limited biomedical model, the "clinical trials" paradigm.
This clinical-trials domination is directly observed in many ways, for example, the preoccupation in the TCPS and other documents with the benefits for the pharmaceutical industry. A further problem, and a more serious one perhaps, is that this clinical-trials mindset also indirectly shapes ethics reviews in SSH when, in fact, SSH research does not follow that format. That is:
- The clinical-trials paradigm is merely Consumer Reportsstyle testing, that is, product comparison. Thus the clinical-trials paradigm lacks the epistemological theory testing and theory building character of research in other traditions of basic scientific inquiry, such as SSH.
- The clinical-trials approach appropriately uses a fixed "protocol" for comparing products, a recipe. That is, there is a static way to do the comparisons. However, this attribute is not characteristic of research in SSH – nor basic research in general. That a "standard" method may exist in clinical trials has led Research Ethics Boards (REB) to think they can judge methodology in SSH proposals, a conclusion that is completely unwarranted.
- The commercial profit incentive in clinical trials is not typical of scientific inquiry, and particularly not research in the SSH. A profit agenda may be a reason to be suspicious about ethical and safety compromises, no matter how it is shrouded in concerns for relieving pain and suffering. That the further investment of scarce campus resources might result in further profits for the drug industry in itself seems worthy of debate, but such concerns definitely are not a reason to burden scholars in all academic disciplines on campus.
- The clinical-trials approach is accompanied by the expectation of inevitable and short-term therapeutic benefits. To quote from a recent article on research ethics (Lougheed, 2003, p. 10), "It made me think about where the line was drawn between therapy and research." This boundary line, and therefore this question, is simply not an issue in SSH research, where the objects of inquiry are not "patients," and "therapy" is not the end goal of research. In fact, this question is not relevant even in basic medical research, it is only relevant in the applied biomedical sector that includes clinical trials. This clinical commitment to producing a better therapy has provided a gratuitous platform for REBs to review SSH proposals for "worthwhile results" and "social desirability," that is, ideological meddling disguised as beneficence.
- The clinical-trials approach frequently involves a proprietary outcome, for example, the outcome of a drug comparison and associated benefits and side-effects. This view of knowledge, and the machinations needed to achieve it, is in complete contrast to the open sharing of research outcomes that characterizes research in the SSH.
Due to the acceptance of the assumption that the biomedical model provides a good prototype, all research areas have become implicitly burdened with the notion that the only good research is that which provides improved (medical) "treatment" (and the sooner the better). This concentration on desirable outcomes or "benefits" undermines the primary issue, subject safety, by giving prominence to agendas that make decisions about what is "socially desirable" research. That is, this clinical treatment emphasis shifts the role of research from decoding nature to producing positive outcomes, and it is a powerful mechanism by which inquiry in the SSH is being inappropriately constrained.
"Social desirability" is an ideological matter. It has nothing to do with public safety. Extrapolated uncritically to SSH, this concept is now undermining the epistemological integrity of research efforts in SSH. REBs have become involved in trying to evaluate research outcomes, which they should not be doing, nor should they be trying to direct research efforts to "desirable" areas. These epistemological matters are of no consequence to the subject's experience in the experimental setting. There is no justification for assessing SSH proposals on the basis of positive outcomes, and this confusion, derived from the clinical-trials mindset, provides a serious constraint on inquiry in SSH research.
A second reason that the criteria in the review process have shifted to therapeutic benefits and "social desirability" appears to be that there is no evidence that SSH research has ever been an issue in terms of public safety (i.e., there is no actual evidence that these reviews are really needed in the first place re safety). The strategy of justifying ethics reviews routinely starts with the recitation of historical problem cases involving medical research, such as Nuremberg, Willowbrook, Tuskegee, and so forth. These sad events are used to create a "moral panic" to not let such unfortunate things occur again by instituting ever more regulations. In fact, however, in spite of the rhetoric from the bioethicists, nothing about the present REBs would have prevented the past problems even in medical research. That is, the perpetrators of such atrocities would have ignored the review formality, so there is no justification to reference them as a defense for present REB practices, certainly not in SSH research. It is intellectually dishonest, and unethical, to claim otherwise.
As a result, the general ideological agenda of social control now drives the review process for SSH proposals, rather than the issue of safety. That is, the SSH ethics reviews are not only indirectly driven by applied biomedical research values, that is, short-term therapeutic benefits repackaged as social engineering, but directly driven by the intrusion of professional ethicists to define the counterpart of "worthwhile" research in the SSH. This subterfuge of replacing the safety criterion with ideology has occurred uncritically, and there is certainly no evidence that the subject's research experience is improved by it.
With the ideology of "social desirability" as the REB's agenda, the behavioral difference between an REB and a censorship board disappears. Good intentions are not relevant, as censors always claim to be improving things. Public safety, no more and no less, must be reasserted as the focus of REB activity.
This muddle may have occurred as local REBs struggled to implement federal regulations, but the federal agencies could easily be characterized as having given passive consent by not intervening. Federal agencies need to explicitly reassert what REBs are to examine (safety), and not examine (ideology), and further assure that locals REBs do not feel justified to indulge in censorship of substance or methodology. The criteria for importance in SSH research must be allowed to once again be quality in epistemological theory testing and theory building.
3. Utopia has become a goal, even though a perfect world is not achievable.
The expectation of "zero risk" has replaced the notion of "everyday risk" in ethics reviews. The question was never to be "could something go wrong," but whether the likelihood of adverse consequences would exceed normal life, everyday risk.
Ironically, the distinction between these two notions, zero vs. everyday risk, is actually still quite clear in medical research where the concept of "side-effects" is well accepted by researchers, subjects, and review boards. Yet REBs in SSH now turn back proposals in pursuit of the fiction of "zero risk;" that is, REBs seem to require that absolutely no psychological discomfort could possibly occur. Of course, when directly confronted with such a silly notion, REB reviewers of SSH proposals may deny using this zero-risk value system, nonetheless people are viewed as far less robust psychologically than physically, judging by the trivial concerns that are often directed back to the SSH researcher.
Some of the reason for this likely follows from the nebulous nature of the "problem" that the SSH ethics review is now trying to solve, whereby hypothetical and unlikely incidents now are the only ones reviewers can identify. The task demands during the review meetings are such that if the reviewers do not object to something they fear that they will be construed as incompetent, so they grasp at something, no matter how inconsequential it is on the everyday-risk scale.
The world does contain everyday risk, everyday deception, and everyday ethics. There is no justification for the pursuit of perfection to have ever become a mandate, and allowing it to continue as the goal constitutes a serious, unrealistic, and unnecessary constraint on freedom of inquiry. With respect to this issue, it is not just a matter of rescuing SSH from the medical model but also a matter of re-establishing reality contact within REBs in the SSH. Ethics committees and reviewers need to be explicitly redirected to the "everyday risk" criterion, and how it differs from zero risk. They need to understand that finding no meaningful problem is likely the modal condition in a SSH research proposal, and not a sign of laziness or incompetence. They must also be encouraged to honor the concepts of exempt research, expedited research, proportionate review, waiver of consent, and waiver of documentation of consent.
4. There is no objective evidence for the effectiveness of, nor need for, REBs.
Is the world a safer place now that SSH researchers have to get ethics approval for everything they do? Answering this question requires (a) knowing what the reviews are trying to achieve (the problem definition), (b) knowing what the baseline (pre-regulation) level of the problem was, and (c) monitoring the change. This is simple pre-post assessment logic, and yet somehow the ethics industry has grown continuously for three decades with no one ever being concerned to document the effectiveness of any of the many new regulations that have been enacted! Further, as noted in points 2 and 3 above, the problem has been so redefined and obfuscated in the SSH reviews that now we do not really know what is actually considered "wrong" nor what the goal looks like. "Respect, beneficence, and justice" are commendable aspirations, but if self-defined by an ethics reviewer, unmonitored and undocumented, then they become an open-ended license for at least manipulating the research agenda by irrelevant ideological dictates, and quite possibly explicit censorship.
On campus we try to measure performance through annual reports, course evaluations, all sorts of things. It is, therefore, quite remarkable that REBs apparently are exempt, self-policing bodies. It is an extreme of irony that for a service directed toward research there are no data (research) on the effectiveness of any aspect of the enterprise! Review committees may think that requesting a cosmetic change in the consent form, for example, justifies the enterprise, but the only meaningful measure of value is the subject's experience, and we just do not have any data on what regulations have produced benefits, if any, there.
In fact, the failure to document both the necessity and effectiveness of REBs is irresponsible and unethical in itself. The present plan for running REBs would not survive submission to a peer-review journal (and certainly not to an REB!). We now have three decades of accumulated constraints on freedom of inquiry and freedom of association in SSH research, all of it without any evidence that any of these regulations has changed the subject's research experience for the better. Taxpayers deserve better, we all do. The lack of concern with data on the cost-benefits of the ethics enterprise no doubt contributes to the feeling of alienation from the process among researchers compared to regulators. That is, researchers reasonably ask: "Why are we doing this if there is no evidence that it improves anything?"
5. The researcher has become a nuisance.
Given the lack of evidence on effectiveness of REB actions, researchers understandably have been inclined to opt out of the process in a number of ways. However, the fact is that there are very few mechanisms for meaningful input from researchers even if or when they want to participate. The communications and regulations involving research ethics have not only come from the medical area to the rest of us in SSH, but communications and review mechanisms now operate on the presumption that the researcher is unethical and must prove innocence. Not only do researchers not feel interested, the sentiment that their involvement is not wanted, and even that the researcher is the problem, is quite clear. As sad as that is, there is every reason to believe that the relationship will deteriorate even further in the future.
Problems with this perspective have been revealed in recent ugly incidents whereby research ethics issues have been used to harass researchers, for example:
- Dr. E. Loftus, U. Washington (Tavris, 2002), investigated to suppress data on "false memory syndrome." This was clearly not about public safety, but instead an ideological attempt to suppress an unwanted finding.
- Dr. L. Pagliaro, U. Alberta (Gillis, 2000), investigated after reporting drugs in schools. This was also an outcome someone did not want to hear, not a threat to public safety.
Given the lack of consequences to the institutions for such abusive treatment of researchers, there is no reason to expect such malicious witch-hunts to diminish in number or severity in the future. To the extent that (a) we continue to permit the fuzzy goal state (ethics and social engineering instead of safety) and (b) fail to document effectiveness, it seems likely that in the future there will be more such incidents rather than fewer. The most parsimonious and most accurate description of these activities is "censorship."
6. The existence of REBs themselves sets up many double standards.
Why is it that academics must be so restricted in terms of human interaction, compared to others? An anthropologist wanting to conduct interviews in Central America spent two years getting clearance, whereas CBC newsreader Peter Mansbridge could be there in 24 hours or less. Just imagine how long reporting about the war in Iraq would have been delayed if an academic had wanted to be embedded with the military rather than Geraldo and Christiane Amanpour. Journalists can study hand-washing behavior in airports, without clearance and without consent forms. Are we less ethical than journalists? Why can a telemarketer simply ring up and ask "Would you mind answering a few questions?" whereas an academic would require an 8-page legalistic consent form, signed and witnessed? It is not just a restriction of freedom of speech, but restricted freedom of association for academics, with, once again, no evidence that the subject's experience is affected in any way. Would it really be an unreasonable attribution here to say that some people (and some REBs) just need to control other people's behavior, which is censorship in most dictionaries, not safety.
7. The future of academic research.
As it becomes increasingly difficult to do SSH research from an academic platform, with more and more regulations of undocumented validity, it seems reasonable to expect a faculty selection process to occur. Ironically, the tactic of being obliged to deal with minor and apparently meaningless demands is a key part of the process whereby prison guards establish authority over prisoners, such as in Zimbardo's infamous prison experiment. Some faculty may capitulate and carry on, but it would not be surprising that many senior faculty may move their efforts re scholarship to consulting activity, books, or other venues that avoid confronting unwarranted constraints on their intellectual inquiries.
There are also more subtle selection processes. For example, in the case of graduate students, it's not enough just to meet local REB criteria. We are aware of instances, where the external examiner on a PhD thesis piously objected that "This would not have been approved by my REB," even though it had been approved by the student's local REB. So, on the basis of the examiner's mere assertion (as opposed to a formal review by the second REB), the students have had to redo their thesis. The lengthy processing time for ethics proposals itself has added months in many cases to the normal time to complete a degree, even for innocuous research that takes less time to execute than the review delay, and now this added trap lies in wait for the student at the last moment.
Do such experiences encourage students to consider continuing a career in academic research? Not likely. Blend in observations of their faculty mentors being treated à la Loftus, plus the constraints of speech codes and related manifestations of political correctness in coursework, and it becomes fairly easy to see a selection process at work in defining the nature of future academic researchers. Specifically, subject matter expertise will become less important, and instead a critical trait for academic survival will be deference. Research activities will be restricted to conventional "safe" and popular questions, using non-controversial methods that fit within the proscribed limits of right-thinking ethicists. Policy bureaucrats will have prevailed.
As the meek inherit the universities, where will break-through research get done? Where will controversial thinking occur? But, most importantly, where is the evidence that the subject's research experience is improved by any of these constraints on inquiry on campus?
- The clinical-trials paradigm is a disaster as a model for SSH research. For this reason alone, if no other, SSH research should not be treated the same as the other research covered by the TCPS. Perhaps there is bureaucratic convenience to "one size fits all," but the cost to SSH research has been and will continue to be immense. This problem is not soluble by allocating yet more local resources and "educating" researchers. The clinical-trials model just doesn't apply.
- The research ethics industry seems to have been driven by Health Canada, as well as the attendant commercial interests of the health-care sector. There must be greater representation of the SSH sector, and there must be meaningful recognition of the limitations of the medical model for campus-wide scholarship.
- Public safety must be reasserted as the agenda for ethics reviews in SSH, not "socially desirable outcomes." This is a matter of re-educating local REBs, not researchers, and requiring the REBs to adhere to the regulations (i.e., safety), and explicitly discouraging REBs from engaging in social engineering and censorship. Likewise the REBs must discard the preposterous pursuit of “zero risk,” there is no utopian risk-free world.
- We must have some evidence of effectiveness for the whole REB enterprise, but especially in SSH research. Changes have been implemented for over 30 years, and we have no evidence that the subject's experience has been affected at all, much less improved.
- REBs, and institutions, must somehow be held accountable for harassment of scholars. Federal funding may be cut off when a subject dies in a clinical trial, why not cut off federal funding to the institution in a case such as that of Loftus? Some penalty to the institution is required, otherwise the censorship and researcher harassment will no doubt grow.
- Somehow the "agenda of inquiry" must be restored to its preeminent status over the "agenda of control." It has become chic in some quarters to try to deflect criticism of the ethics industry with an observation such as "Research is a privilege, not a right." This fatuous thinking simply conceals an effort to maintain control at all costs. Research is a job requirement for faculty, and research is degree requirement for students. Freedom of inquiry is widely accepted and respected in everyday life, it is a truly just part of the natural order of human existence. That inquiry is so much more constrained on campus than in the everyday world, without good cause, is something we should all decry.
We realize that this is a strong commentary. However, the constraints on inquiry have been great, ill-conceived, and yet, as far as anyone can document, unnecessary as far as the participant's experience is concerned. Correcting this will require more than simple tweaking, and it must start with the honest admission that this 30-year "experiment" has produced null results at considerable cost to many disciplines and researchers.
We hope these thoughts are helpful to your committee. If we can provide further input, or clarify any of the points here, please let us know. We look forward to following the progress of your committee's work, because it is important that change occur.
John Mueller, University
Steve Lupker , University of Western Ontario
John and Steve are members of SAFS Board of Directors.
Gillis, C. (2000).
Professor again in hot water over findings. National Post, November
Lougheed, T. (2003) A question of ethics. University Affairs, June/July 2003, 10-13.
Nature, editorial (2001). Time to cut regulations that protect only regulators. Nature, 2001, 414, 379.
Tavris, C. (2002). The high cost of skepticism. Skeptical Inquirer, 26, 14(4), 41-44. http://www.csicop.org/si/2002-07/high-cost.html