We argue that the key to understanding the relation between outcome and benefit of an experiment in the HBA lies in one’s reading of the language “is justified by the expected outcome (…) and may ultimately benefit”. Upon closer examination of these inconspicuous words, certain implicit premises become apparent.
2.2. The Heuristic of “May Ultimately”
As demonstrated above, outcome and benefit cannot be understood independently of each other in the Directive. We now aim at analyzing the nature of their relation. First, “may” in “may ultimately benefit” indicates that the outcome ought to have the potential to be of benefit. Second, the word “may”, especially in conjunction with “ultimately” indicates some time horizon for achieving benefit. Yet, what would count as timely or late remains entirely unclear. Cynics might even argue that it is logically impossible to preclude that any given experiment can “ultimately” (in the far unforeseeable future) yield some kind of benefit for humans, animals or the environment; nobody, one could argue, can say with certitude that a particular experiment will never—not even in 500 years—contribute to some sort of benefit. This would imply that projects eo ipso fulfil the criterion that they may ultimately benefit. Although it is clear that nobody can exclude the possibility that a project “may” benefit someone someday, the question remains how substantial the expected benefit is to be, and we are left wondering how such a thing is to be assessed.
The legitimizing power of a procedure’s outcome has been qualified by its expected benefit, and that benefit has been qualified as pertaining to humans, animals or the environment. According to recent publications, catalogues of criteria and guiding documents, the justifying power of benefit is dependent on a) the time frame within which the benefit shall be reached, and b) the probability of the projected benefit’s realization [
8,
9,
10,
11,
12]. For instance, the questions
“What” is the benefit, “who” will benefit, “how” will they benefit and importantly
“when” will they benefit? are highlighted in an EU guiding document [
13] and referred to in a recent publication by the “American Association for Laboratory Animal Science” and “European Federation for Laboratory Animal Science Associations” working group (AALAS-FELASA) [
9] as important parameters in assessing potential benefits. Prior to these recent publications, published work in the field had been arguing along these lines for years. In 1992 for instance, Porter [
5] qualified benefit by its importance for humans and the likelihood of its achievement. Scharman and Teutsch [
10] propose a checklist model that defines expected benefit as the improvement and/or development of diagnostics and/or therapies, and factors in the likelihood of its realization as well. Hirt, Maisack and Moritz [
12] restrict “benefit” to “benefit to humans with regard to what kind and what extent of benefit and the likelihood and time frame of realization” (authors’ translation). The “expert working group on project evaluation and retrospective assessment” suggests a modified Bateson’s cube [
13], an analysis model that dates from a 1986 publication [
14] and has served as an ethical evaluation tool for animal research ever since. The Bateson’s cube is a 3-dimensional grid that evaluates the degree of benefit, the amount of harm to the animals and the likelihood of benefit. Not only is the degree of benefit decisive, but also the likelihood of achieving it. Moreover, Bout et al. [
11] have developed another matrix based on the Bateson cube, which is used for the HBA in the Netherlands. In their model, the likelihood of the benefit (scientific and/or practical) is more than just a contributing factor in the weighing process; it is a stand-alone criterion. If the likelihood of success is too low, the proposal has to be rejected, regardless of its strengths. Again, preference is given to projects that yield foreseeable benefit. Stafleu et al. [
15] proposed a numerical algorithmic model with a detailed set of formulas for calculating harm and benefits. While the authors do acknowledge both the difficulty of prospective benefit assessment and the legitimate value of knowledge as an experimental outcome, only human health interests are capable of getting the maximum score in their evaluation tool. Interestingly, economic interests are accorded the same legitimizing power as the generation of knowledge in the Stafleu et al. model.
In this brief overview of relevant published work on the HBA, a pattern emerges in which the understanding of “benefits” is “foreseeable, expected, societal benefit.” The generation of direct, societal benefit factors more heavily in the HBA and is prioritized in literature on the evaluation process, at the expense of procedures whose intended and foreseeable outcome is of no foreseeable, tangible and immediate benefit to society. In that scheme, generating knowledge (the outcome of basic research) is typically regarded as a lesser benefit and weighs less on the scales of the HBA.
From this, it follows that knowledge per se as the
outcome of a project is only viewed as a necessary but not sufficient condition to the justification of harm done to the animals. In this dominant understanding of the terms, only an
outcome which is qualified by its potential to generate societal
benefit, as direct
benefit to humans, animals or the environment can succeed to legitimize harm and thus fulfil the legal requirements. Therefore, knowledge, as experimental outcome, is often considered of little justifying power. It follows that, e.g., severe harm cannot be justified solely on the grounds of expected gains in knowledge, as would be the objective of basic research [
16,
17]. As substantiated above, the prevailing consensus on the HBA evidently locates justifying power not in an expected outcome, but in expected societal and tangible benefit. Here, it is important to note that it is not only important that “benefit” may be generated but that “benefit” should, to some degree, be expected.
Our contention is that the wording “and may ultimately” with regard to “outcome” and “benefit” in the context of the HBA is interpreted in such a way that scientific benefit (knowledge) as an experimental outcome has no or little intrinsic value and legitimizes harm in procedures only through its potential to generate societal benefit (i.e., benefit to humans, animals or the environment beyond knowledge). Furthermore, the HBA prioritizes timely and likely (i.e., time frame and probability of success), direct, societal benefit over less tangible benefit, particularly knowledge. We would like to make it clear that we do not claim that knowledge is not a legitimate outcome of animal research according to the Directive. Nor do we claim, that an intention to generate societal benefit (e.g., health or environmental benefit) is a “sine qua non” requirement for project approval. Rather we want to stress that the HBA prioritizes and gives much more “weight” to societal benefits in project evaluation compared to knowledge and prioritizes potential societal benefits that seem more tangible and more likely. The main justifying power of a research project’s outcome is strongly linked to expected societal benefits and thus, gaining knowledge is on the bottom and, e.g., human health on top of the benefit hierarchy in the HBA. In the following, we will argue that this hierarchy is questionable with regards to prospective benefit assessment, due to logic, methodological and practical flaws. We argue that societal benefits of animal research should be assessed retrospectively, but consider it problematic to assess them prospectively in a HBA.