In its time, the Belmont Report was invaluable. The United States had failed abysmally in protecting human research subjects. Despite the place of prominence of informed consent in the Nuremberg Code (1948), and discussion of consent in the Declaration of Helsinki (1964), by the early 1970s, the U.S. did not even have a requirement for informed consent.Footnote 1 Thus the Belmont Report (1979), and its practical implementation, moved the field of protection of human subjects considerably forward. Unfortunately, current standards for human subjects place far too great an emphasis on regulation, and not enough on the underlying ethic that informs research involving human subjects. The 20th century positivistic bias against teaching ethics in science persists even with the advent of new regulations.Footnote 2 Thus, for example, the new version of the Common Rule for human subjects is almost 600 pages long and does not talk about ethics. As Plato pointed out, it is not enough to have good rules.Footnote 3 Conceptually more important is to have good people who are ethically educated, and thereby understand the reasons for the rules. When the National Institutes of Health recently abandoned research on great apes, the reason given was societal concern, with no emphasis whatsoever placed on the underlying ethical issues that occasion such concern!
Thus, Hope Ferdowsian and colleagues are to be commended for a thought-provoking essay, in which they make a case for the application of the Belmont Report principles to animal research practices.Footnote 4 The aim is a commendable one; the development of a defensible and philosophically rigorous foundation for animal welfare ethics and the recognition of the moral status of animals. Given the paucity of either reflection or commitment to the rights of human research subjects, one can guess how limited ethical reflection on the use of animals has been. It was in good measure the moral imperative to introduce pain control into science that led my colleagues and I (BR) to draft what became the 1985 Amendments to the Animal Welfare Act requiring control of any research modalities causing pain and distress in animals in the course of research. Though the research community protested that they used copious amounts of analgesia, a literature search that I (BR) performed in 1982 through the Library of Congress was unable to find a single paper dealing with “analgesia for laboratory animals.” This fact was instrumental in convincing Congress of the necessity for such legislation.Footnote 5
While such a legislative and regulatory framework is obviously necessary, we submit that it is not sufficient, and must not be taken as such. A robust animal welfare ethic simply must attend to reasons that are more durable than either popular opinion or the notion that “the regulations say so”; we must work toward a broader understanding (among scientists, the public, legislators) about the why that undergirds our animal welfare ethos, not merely the what of any current regulatory framework. The current paper aims at this why, by proposing that we adapt the principles articulated in the Belmont Report, developed for human subjects, to animal subjects. On the surface, this sounds plausible, as the Belmont Report has the advantage of being ‘recognizable’; it has some historical cachet that in theory, could be leveraged as a framework for animal welfare. However, on closer examination, it is evident that there are enormous conceptual obstacles to effecting such a move.
The central point that must be made explicit in any discussion regarding animal welfare, animal rights, and our attendant duties as researchers, IACUC (Institutional Animal Care and Use Committees) members, scholars, or community members is simply this: we are speaking to one another about animals; we are discussing our responsibilities toward them, but can never meaningfully do so with them. This is a distinction that matters, and it certainly impacts the assumption that the Belmont principles are readily applicable to nonhuman animals. It simply does not follow that because Belmont ‘naturally’ fits the human research paradigm, it also ‘naturally’ transfers to an animal welfare ethic. We can recognize the many meaningful similarities between human beings and fellow members of the animal kingdom without blurring some very real, and, for the sake of robust ethics, very meaningful distinctions that are inescapable in practice. It is also essential to realize that one can acknowledge the rights of animals, without assuming that those rights must be the same as those of humans. One can affirm that animals have the right to live their lives in accordance with their biological and psychological natures, that they have a right not to suffer or be the subject of abuse at our hands, without assuming that they have the right to vote. Importantly, all of the discourse on the rights that attend animal welfare concerns are in fact duties we recognize for ourselves as responsible agents, not duties shared in a reciprocal fashion between human and nonhuman animals.
Given this, the most obvious defect in the proposition that the Belmont Report is a viable framework for animal welfare ethics is making sense out of extracting informed consent from an animal. Even in the case of those few animals that putatively understand some human assertions, none can respond to any question more complex than ‘are you hungry?’ or ‘do you want to go out?’ or ‘do you want to play?’ And rodents, constituting 90 percent of the animals used in research, do not even appear to be capable of such limited responses. Or, to be fair, perhaps we do not know rodents well enough to read such responses, but this is a distinction without a difference when it comes to the welfare issues at hand. Certainly, no animal can reply to any version of the question ‘do you consent to be a research subject?’ Most simply, there is absolutely no reason to believe that they can understand a concept as abstract as ‘research subject.’ Nor do they possess the syntax allowing them to articulate concerns about the future, or about counterfactual conditions or possibilities. It does not generally help to suggest appointing guardians or surrogates to represent the animal interests. Such guardians cannot be presumed to speak for the animals. If these guardians are scientists, they are likely to speak for science. If they are ‘antivivisectionists,’ we can presume the opposite. The philosopher Steve Sapontzis once suggested how one can get informed consent from an animal—open the cage!Footnote 6 It is difficult to see how one can get much beyond that.
One possible exception in favor of using guardians is the degree to which researchers, particularly in the area of cancer, have begun to use animals naturally afflicted with the disease instead of creating the disease. This essentially means dogs, and to a lesser degree cats, who are being treated in veterinary hospitals, and where we can presume that owners intend the best interest of the animals.Footnote 7 Aside from the inherent problem of veterinary experts being able to utilize their Aesculapian authority to influence animal owners unfairly in their decisionmaking, we must recall that the majority of animals used in biomedical research are rats and mice. Who represents their interests? Importantly, the imposition of Belmont Principles to the animal welfare setting is further complicated by the latter being a setting that strikes us as inherently complicated for a ‘one size fits all’ approach. As alluded to above, animal welfare concerns arise in research settings that range from the ‘traditional’ lab animal welfare setting (most often, but not exclusively involving strains of rat and mouse), to companion animal research (as noted above), but also extends to field research on animals in the wild, which can often extend to more than mere behavioral observation. Field sterilization and/or population management efforts are far more complex, and raise a number of unique ethical issues than do not simply transfer to the lab animal setting, let alone the Belmont Report.
A major concept in the Belmont Report is justice. Thus, it is clearly unjust, as we have done in the past, to use slaves, prisoners, the disenfranchised, and myriad other oppressed people who do not benefit from research as experimental subjects. How do we apply this to animals? Consistent with what we note above, the notion of justice as applied to animal welfare invokes unilateral duties and responsibilities for the human actors; there is no shared understanding of justice between human and nonhuman animals, no shared deliberation on the contours of justice in theory or in practice. We do not expect the exercise of justice within animal communities precisely because the notion means absolutely nothing to any but the human animal. This arises from the obvious, and nonarbitrary, differences between human and nonhuman animals, and eliding such fundamental differences does not help. To their immense credit, Ferdowsian and colleagues note the injustice of such arbitrary distinctions within human populations, and rightly condemn them. Retaining an awareness of some essential, nonarbitrary differences between humans and animals need not entail the invocation of speciesism nor justify animal abuse; it can serve as a basis for pointing out our inescapable stewardship responsibilities as responsible moral agents to animals that, as the authors note, both have interests and can suffer. We, then, are bound to consider the exercise of justice in our interactions with nonhuman animals, but we must recognize that the application is not simple. Prudential deliberation about our responsibility for the care of animals must attend, as one of us (BR) has argued elsewhere, to the telos of the animal in question.Footnote 8 This interfaces with the environment in which we propose to study animals, the presuppositional work of considering how we either create or modify the environments in which the animals live, and the purposes, or ends to which our interactions with the animals are oriented.
Of the three central principles in the Belmont Report, the principle of beneficence (and nonmaleficence) is conceptually the easiest to apply to animals. If we do not cause the animals pain, fear, or any form of distress on the one hand, or on the other hand, deprive them of their ability to actualize their biological and psychological natures which, following Aristotle, I (BR) have called telos, we are adequately respecting their interests.Footnote 9 (Ironically, failing to respect animal nature will inevitably deform variables being investigated and confound the interpretation of the biological responses in question.) These principles, that are exemplary, would also put an end to the vast majority of animal research. It would effectively eliminate all caging or other housing done solely for human convenience.
We have no objection in principle to a blanket document enunciating the ethic that should underlie all animal research. Our problem is rather with the suggestion that such a document designed for research on human beings can be adapted to fit animals, especially all animals that could be utilized in biomedical research, from rats, to mice, to zebrafish, to armadillos, to horses, to nonhuman primates and birds. From a regulatory perspective, the best one can do in creating a general template is what we mentioned earlier—preventing pain, distress, fear, discomfort, and other negative modalities, and respecting the animals’ biological and psychological natures, their telos. In terms of the foundational role of education, placing a robust animal welfare ethic in a central position in STEM (Science, Technology, Engineering and Mathematics) and veterinary medicine education, and, arguably, even in a general education or ‘core curriculum,’ allows for the opportunity to develop the more sustainable reflection and understanding about our responsibilities for the care and welfare of animals that extend beyond the ‘culture of compliance.’
In its time, the Belmont Report was invaluable. The United States had failed abysmally in protecting human research subjects. Despite the place of prominence of informed consent in the Nuremberg Code (1948), and discussion of consent in the Declaration of Helsinki (1964), by the early 1970s, the U.S. did not even have a requirement for informed consent.Footnote 1 Thus the Belmont Report (1979), and its practical implementation, moved the field of protection of human subjects considerably forward. Unfortunately, current standards for human subjects place far too great an emphasis on regulation, and not enough on the underlying ethic that informs research involving human subjects. The 20th century positivistic bias against teaching ethics in science persists even with the advent of new regulations.Footnote 2 Thus, for example, the new version of the Common Rule for human subjects is almost 600 pages long and does not talk about ethics. As Plato pointed out, it is not enough to have good rules.Footnote 3 Conceptually more important is to have good people who are ethically educated, and thereby understand the reasons for the rules. When the National Institutes of Health recently abandoned research on great apes, the reason given was societal concern, with no emphasis whatsoever placed on the underlying ethical issues that occasion such concern!
Thus, Hope Ferdowsian and colleagues are to be commended for a thought-provoking essay, in which they make a case for the application of the Belmont Report principles to animal research practices.Footnote 4 The aim is a commendable one; the development of a defensible and philosophically rigorous foundation for animal welfare ethics and the recognition of the moral status of animals. Given the paucity of either reflection or commitment to the rights of human research subjects, one can guess how limited ethical reflection on the use of animals has been. It was in good measure the moral imperative to introduce pain control into science that led my colleagues and I (BR) to draft what became the 1985 Amendments to the Animal Welfare Act requiring control of any research modalities causing pain and distress in animals in the course of research. Though the research community protested that they used copious amounts of analgesia, a literature search that I (BR) performed in 1982 through the Library of Congress was unable to find a single paper dealing with “analgesia for laboratory animals.” This fact was instrumental in convincing Congress of the necessity for such legislation.Footnote 5
While such a legislative and regulatory framework is obviously necessary, we submit that it is not sufficient, and must not be taken as such. A robust animal welfare ethic simply must attend to reasons that are more durable than either popular opinion or the notion that “the regulations say so”; we must work toward a broader understanding (among scientists, the public, legislators) about the why that undergirds our animal welfare ethos, not merely the what of any current regulatory framework. The current paper aims at this why, by proposing that we adapt the principles articulated in the Belmont Report, developed for human subjects, to animal subjects. On the surface, this sounds plausible, as the Belmont Report has the advantage of being ‘recognizable’; it has some historical cachet that in theory, could be leveraged as a framework for animal welfare. However, on closer examination, it is evident that there are enormous conceptual obstacles to effecting such a move.
The central point that must be made explicit in any discussion regarding animal welfare, animal rights, and our attendant duties as researchers, IACUC (Institutional Animal Care and Use Committees) members, scholars, or community members is simply this: we are speaking to one another about animals; we are discussing our responsibilities toward them, but can never meaningfully do so with them. This is a distinction that matters, and it certainly impacts the assumption that the Belmont principles are readily applicable to nonhuman animals. It simply does not follow that because Belmont ‘naturally’ fits the human research paradigm, it also ‘naturally’ transfers to an animal welfare ethic. We can recognize the many meaningful similarities between human beings and fellow members of the animal kingdom without blurring some very real, and, for the sake of robust ethics, very meaningful distinctions that are inescapable in practice. It is also essential to realize that one can acknowledge the rights of animals, without assuming that those rights must be the same as those of humans. One can affirm that animals have the right to live their lives in accordance with their biological and psychological natures, that they have a right not to suffer or be the subject of abuse at our hands, without assuming that they have the right to vote. Importantly, all of the discourse on the rights that attend animal welfare concerns are in fact duties we recognize for ourselves as responsible agents, not duties shared in a reciprocal fashion between human and nonhuman animals.
Given this, the most obvious defect in the proposition that the Belmont Report is a viable framework for animal welfare ethics is making sense out of extracting informed consent from an animal. Even in the case of those few animals that putatively understand some human assertions, none can respond to any question more complex than ‘are you hungry?’ or ‘do you want to go out?’ or ‘do you want to play?’ And rodents, constituting 90 percent of the animals used in research, do not even appear to be capable of such limited responses. Or, to be fair, perhaps we do not know rodents well enough to read such responses, but this is a distinction without a difference when it comes to the welfare issues at hand. Certainly, no animal can reply to any version of the question ‘do you consent to be a research subject?’ Most simply, there is absolutely no reason to believe that they can understand a concept as abstract as ‘research subject.’ Nor do they possess the syntax allowing them to articulate concerns about the future, or about counterfactual conditions or possibilities. It does not generally help to suggest appointing guardians or surrogates to represent the animal interests. Such guardians cannot be presumed to speak for the animals. If these guardians are scientists, they are likely to speak for science. If they are ‘antivivisectionists,’ we can presume the opposite. The philosopher Steve Sapontzis once suggested how one can get informed consent from an animal—open the cage!Footnote 6 It is difficult to see how one can get much beyond that.
One possible exception in favor of using guardians is the degree to which researchers, particularly in the area of cancer, have begun to use animals naturally afflicted with the disease instead of creating the disease. This essentially means dogs, and to a lesser degree cats, who are being treated in veterinary hospitals, and where we can presume that owners intend the best interest of the animals.Footnote 7 Aside from the inherent problem of veterinary experts being able to utilize their Aesculapian authority to influence animal owners unfairly in their decisionmaking, we must recall that the majority of animals used in biomedical research are rats and mice. Who represents their interests? Importantly, the imposition of Belmont Principles to the animal welfare setting is further complicated by the latter being a setting that strikes us as inherently complicated for a ‘one size fits all’ approach. As alluded to above, animal welfare concerns arise in research settings that range from the ‘traditional’ lab animal welfare setting (most often, but not exclusively involving strains of rat and mouse), to companion animal research (as noted above), but also extends to field research on animals in the wild, which can often extend to more than mere behavioral observation. Field sterilization and/or population management efforts are far more complex, and raise a number of unique ethical issues than do not simply transfer to the lab animal setting, let alone the Belmont Report.
A major concept in the Belmont Report is justice. Thus, it is clearly unjust, as we have done in the past, to use slaves, prisoners, the disenfranchised, and myriad other oppressed people who do not benefit from research as experimental subjects. How do we apply this to animals? Consistent with what we note above, the notion of justice as applied to animal welfare invokes unilateral duties and responsibilities for the human actors; there is no shared understanding of justice between human and nonhuman animals, no shared deliberation on the contours of justice in theory or in practice. We do not expect the exercise of justice within animal communities precisely because the notion means absolutely nothing to any but the human animal. This arises from the obvious, and nonarbitrary, differences between human and nonhuman animals, and eliding such fundamental differences does not help. To their immense credit, Ferdowsian and colleagues note the injustice of such arbitrary distinctions within human populations, and rightly condemn them. Retaining an awareness of some essential, nonarbitrary differences between humans and animals need not entail the invocation of speciesism nor justify animal abuse; it can serve as a basis for pointing out our inescapable stewardship responsibilities as responsible moral agents to animals that, as the authors note, both have interests and can suffer. We, then, are bound to consider the exercise of justice in our interactions with nonhuman animals, but we must recognize that the application is not simple. Prudential deliberation about our responsibility for the care of animals must attend, as one of us (BR) has argued elsewhere, to the telos of the animal in question.Footnote 8 This interfaces with the environment in which we propose to study animals, the presuppositional work of considering how we either create or modify the environments in which the animals live, and the purposes, or ends to which our interactions with the animals are oriented.
Of the three central principles in the Belmont Report, the principle of beneficence (and nonmaleficence) is conceptually the easiest to apply to animals. If we do not cause the animals pain, fear, or any form of distress on the one hand, or on the other hand, deprive them of their ability to actualize their biological and psychological natures which, following Aristotle, I (BR) have called telos, we are adequately respecting their interests.Footnote 9 (Ironically, failing to respect animal nature will inevitably deform variables being investigated and confound the interpretation of the biological responses in question.) These principles, that are exemplary, would also put an end to the vast majority of animal research. It would effectively eliminate all caging or other housing done solely for human convenience.
We have no objection in principle to a blanket document enunciating the ethic that should underlie all animal research. Our problem is rather with the suggestion that such a document designed for research on human beings can be adapted to fit animals, especially all animals that could be utilized in biomedical research, from rats, to mice, to zebrafish, to armadillos, to horses, to nonhuman primates and birds. From a regulatory perspective, the best one can do in creating a general template is what we mentioned earlier—preventing pain, distress, fear, discomfort, and other negative modalities, and respecting the animals’ biological and psychological natures, their telos. In terms of the foundational role of education, placing a robust animal welfare ethic in a central position in STEM (Science, Technology, Engineering and Mathematics) and veterinary medicine education, and, arguably, even in a general education or ‘core curriculum,’ allows for the opportunity to develop the more sustainable reflection and understanding about our responsibilities for the care and welfare of animals that extend beyond the ‘culture of compliance.’