Michael Soulé, co-founder of the Society for Conservation Biology and its first President, is widely considered to be the founder of conservation biology (Sanjayan, Crooks, and Mills 2000). In setting out his vision for the field, Soulé argued that it is a crisis-oriented discipline like cancer-biology, which, he suggested, implies that ethical norms are an inherent part of conservation biology. He stated that the ethical norms include value judgments such as the postulate that the "diversity of organisms is good" which "cannot be tested or proven" (Soulé 1985, 730). In coming to these views, Soulé cited several scholars who influenced him, including Aldo Leopold, an early-mid 20th century hunter, forester, wildlife manager, conservationist, and professor who has likewise been extremely influential in conservation biology and related fields. Indeed, many of the ideas that Soulé described have precedents in Leopold's thinking. Leopold famously stated that the land "has value in the philosophical sense" and that "an ethical relation to land" should have "a high regard for its value" (Leopold 1949, 223). Moreover, Leopold explicitly compared health in humans and health in the land, suggesting that the science of doctoring the land had not really begun yet (Leopold 1949). Meanwhile, the ecologist typically "lives alone in a war of wounds" as a "doctor who sees the marks of death in a community that believes itself well" (Leopold 1947). Given his foundational status and his explicit commitments to value-driven science, Leopold is a promising figure to examine in trying to understand the role of values in conservation biology. I argue that Leopold's scientific beliefs and value beliefs co-evolved over the course of his life. Values were present in the very building blocks of Leopold’s understanding of the environment. From his father Carl Leopold, Aldo learned to appreciate and enjoy the natural world; his father also impressed upon him a hunter’s ethics (Meine 2010, 18). By the end of his life, the hunter of prey and the eradicator of predators had become a defender of predators and other species as well as land communities as wholes. He groped toward understanding land health as the goal of a land ethic as well as the ecological mechanisms that underlaid it. This extends the picture of the relation of science and values described by Elizabeth Anderson (2004), which suggests that "if values can legitimately influence empirical theories, then empirical theories can legitimately influence our value judgments” (Anderson 2004, 2). According to Anderson, facts can count as evidence for value judgments, and value judgements can help us see certain facts. For Leopold, this bidirectional influence occurred over time, which, although perhaps not Soulé's "test" or "proof," arguably offered an informed living laboratory in which both his (sometimes entangled) science and values could advance. This examination of Leopold's trajectory will inform how we might think of the role of values in conservation biology as well as how we might think of the role of values in science more generally.

The conservation ecologist Robert Lackey (2005, 2013) describes stealth policy advocacy as strategy deployed in the pursuit of “policy-based science.” As a proponent of the value-free ideal, Lackey argues that the adoption of ethical values by scientists (in a professional capacity) undermines their credibility and erodes public trust. Stealth policy advocacy is especially pernicious, he adds, because it presents scientific concepts as if they were purely empirical when, in fact, they “contain tacit policy preferences and thus, by extension, promote particular policy options” (2005). Lackey cites ecosystem health and alien species as examples of stealth-policy concepts that should be eliminated. The first part of this paper argues that stealth policy advocacy should be resisted by both proponents and critics of the value-free ideal. Critics like Longino (1996), Douglas (2009), and Elliott (2020) point to legitimate roles for values in hypothesis confirmation conceptual framing. However, in order to avoid systemic bias, these values must be transparent and open for debate, not tacitly disguised as value-neutral. The second part of this paper presents survey evidence (X-Phi) from a sample of practicing ecologists and conservation managers. I show that they employ two conceptions of biodiversity: a value-neutral conception that equates biodiversity with the diversity of units at some level (e.g., species richness), and a value-laden conception that equates biodiversity with the diversity of units plus integrity/naturalness—a normative property. Individual scientists’ reliance on a given conception was not explained by their subdiscipline, research focus, level of professional standing, their preferred explicit definition of biodiversity, nor their explicit reasons for valuing biodiversity. The coexistence of these two conceptions promotes equivocation and makes it practically difficult to tease apart the normative from the empirical. At the same time, the presence of a purely empirical concept alongside one that is normatively “thick” allows conservation scientists to deflect calls for elimination of the value-laden concept (Santana 2014). Specifically, it becomes possible to strategically deploy the value-laden concept while reverting, under scrutiny, to the value-neutral concept as a sort of alibi. I see this predicament as pointing to both the desirability and the impracticality of biodiversity eliminativism. Douglas, Heather. 2009. Science, Policy, and the Value Free Ideal. University of Pittsburgh Press, Pittsburgh. Elliott, Kevin C. 2020. Framing conservation: “biodiversity” and the values embedded in scientific language. Environmental Conservation 47: 260-268. Lackey, Robert T. 2003. Appropriate use of ecosystem health and normative science in ecological policy. pp. 175-186. In: Managing for Healthy Ecosystems, David J. Rapport, William L. Lasley, Dennis E. Rolston, N. Ole Nielsen, Calvin O. Qualset, and Ardeshir B. Damania, (eds), Lewis Publishers, Boca Raton, Florida. Lackey, Robert T. 2016. Keep science and scientists credible, avoid stealth policy advocacy. The Bulletin of the Ecological Society of America 46: 14-16. Longino, Helen. E. 1996. Cognitive and non-cognitive values in science: Rethinking the distinction. pp. 39-58. In Feminism, Science and the Philosophy of Science. L. H. Nelson (ed). Kluwer Academic Publishers: Great Britain. Santana, Carlos. 2014. Save the planet: Eliminate biodiversity. Biology & Philosophy 29: 761-780.

Philosophers agree that the value-free ideal is neither an accurate nor desirable model for science. Science typically requires non-epistemic value judgments. Debates remain as to what kind of non-epistemic values can legitimately influence science, and in what ways. One proposal is that “when scientists must appeal to nonepistemic values in the course of their work, they should appeal to democratic values—roughly, the values held by the public or its representatives” (Schroeder 2021, 2. See also Intemann 2015). In this paper, I argue that this view fails to solve the problems it aims to solve while raising other serious issues. The paper relies on two case studies: the controversy over wolf population management in the Yellowstone area and the long-lasting dispute over management of the National Bison Range on the Flathead Reservation. The democratic-endorsement view is supposed to solve two problems: 1. Legitimacy. Scientific research influences lay people’s daily lives, either through policy or direct, pervasive impact. Such influence qualifies as non-epistemic authority over the people. Legitimate authority in liberal democratic society is grounded in democratic endorsement. Hence, non-epistemic value influence in science is legitimate only if it appeals to democratically endorsed values. 2. Public Trust. The fall of the value-free ideal has eroded public trust. If non-epistemic value judgments influence science, then the people will and ought to trust science only if such values are representatives of their own values. Democratic endorsement is thus the best ground for warranted public trust in science. The democratic-endorsement view most likely fails to solve the problems above. It relies on the assumption that our society can find some overlapping consensus regarding non-epistemic values, which can then ground scientific consensus and public trust. Unfortunately, such consensus is likely lacking, especially in ecology and conservation biology. The controversy over Yellowstone wolf population management is a case in point (Smith et al. 2016). Value conflict and polarization undermine democratic consensus and the democratic endorsement view. The view also raises some serious issues, most prominently the issue of marginalization. Democratic consensus most likely ignores the needs and wants of historically marginalized communities. A study of the controversy over the management of the National Bison Range on the Flathead Reservation shows that democratically endorsed values may build upon a history of wrongful prejudice (Upton 2014). The democratic-endorsement view also faces the issue of non-neutral expertise. It is often impossible (and misguided) to be both informed and neutral. Expertise informs non-epistemic value judgments. The public is often in no position to make properly informed value judgments. An alternative to the democratic-endorsement view, one that respects both the public’s interest, including minorities’, and expertise is called for. Schroeder, S. Andrew. 2021. Democratic values: a better foundation for public trust in science. The British Journal for the Philosophy of Science, 72(2), 545-562. Intemann, Kristen. 2015. Distinguishing between legitimate and illegitimate values in climate modeling. European Journal for Philosophy of Science, 5, 217–232. Upton, Brian. 2014. Returning to tribal self-governance partnership at the National Bison Range Complex: Historical, legal, and global perspectives., 35, 51-146. Smith, Douglas.W., White, P.J., Stahler, Daniel.R., Wydeven, Adrian. and Hallac, David.E., 2016. Managing wolves in the Yellowstone area: Balancing goals across jurisdictional boundaries. Wildlife Society Bulletin, 40(3), 436-445.

Over the past twenty years, philosophers of science have given sustained examination to the role that values play in science, identifying positive roles for both cognitive and noncognitive values. These include value judgments made during research, such as the conceptualization of phenomena, data selection and analysis of results, as well as in applications of scientific knowledge for policy. Consequently, many philosophers have concluded that although there is a risk that dogmatically held values may bias scientific research, there is also a legitimate role for values in scientific judgment. The distinction between the two is whether scientists are transparent about their value assumptions and methodological choices and whether the social and ethical values at play are widely shared and have been subjected to deliberation (Elliott 2017, 14-15). Conservation science overtly rests on normative postulates, and its journals have long made space for normative debates about the fit between conservation practices and the normative goal of preserving biological diversity (Soulé 1985). In this paper I identify a gap between, on the one hand, philosophers’ endorsement of transparent deliberation about values and, on the other hand, the practical methods, shared norms, and concrete guidance required to move from discussing values in a theoretical way to incorporating them into practice. Although conservation scientists discuss values openly, norms have not developed for systematic moral deliberation and, after deliberation, action. I argue that, while it is necessary, at times, to talk about scientific research and values separately (in philosophical terms, to distinguish facts and values), it is also essential, following discussion, to systematically incorporate values into conservation research. I investigate two specific techniques for incorporating values into science with regard to debates in conservation science about managed relocation. First, I identify how knowledge gaps block conservation practice. Knowledge gaps are created and maintained by a bias in favor of theory over practice and a conservative bias toward existing frameworks for conservation practice rather than innovation. Because theoretical research carries more prestige than applied research, the majority of new research in conservation science describes species and ecological relationships—states, causes, and mechanisms—rather than designing, implementing, and evaluating conservation interventions (Williams, Balmford & Wilcove 2020). The consequent lack of practical knowledge entrenches value-based arguments against novel interventions on the grounds that their consequences are unknown. A second technique to encourage discussions of values involves reforming risk assessment frameworks. Typically, risks posed by interventions are measured against the status quo or a past benchmark rather than being compared to likely future states given rates of biodiversity loss and ecological degradation. By ignoring the urgency of conservation needs, this practice systematically biases risk assessment frameworks toward inaction. In sum, I use examples from managed relocation to demonstrate how an ethos of restraint follows from uncertainty about how to integrate values into conservation research. In contrast, an ethos of responsible action follows from a more thoroughgoing analysis of the role values can play in action-oriented conservation research.

One of the most important proposed examples of a trophic cascade concerns the reintroduction of grey wolves into Greater Yellowstone Ecosystem (Ripple et. al. 2001, 2014, 2015). As the story goes, the reintroduced grey wolves have reduced elk populations, and this has encouraged a variety of plant and animal species to increase. However, some have argued that this example’s success in the academic and public imagination has not been the result of the empirical evidence in its favor (Peterson et. al. 2014; Marris 2018). Rather, it is because it promotes various environmental values at play. These biases in fact have led to methodologically problematic science. Philosophers of science have been exploring values in science, and I will do so in this debate. Following the work of Helen Longino (1990, 2002) and Solomon (2010) in particular, I argue that the presence of values is not especially problematic provided diverse evaluative commitments are manifest in the research process. I explore this case study to see to what extent transformative criticism has taken place and where improvements can be made to the epistemic structure of these debates. Longino, H.E., 1994. Science as social knowledge. Princeton university press. ______. 2018. The fate of knowledge. Princeton University Press. Marris, E., A good story: Media bias in trophic cascade research in Yellowstone National Park. In Effective Conservation Science, pgs. 80-84. Oxford University Press. Peterson, R. O., Vucetich, J. A., Bump, J. M., and Smith, D. W. (2014). Trophic cascades in a multicausal world: Isle Royale and Yellowstone. Annual Review of Ecology, Evolution, and Systematics 45, 325–45. Ripple, W. J., Beschta, R. L., Fortin, J. K., and Robbins, C. T. (2014). Trophic cascades from wolves to grizzly bears in Yellowstone. Journal of Animal Ecology 83, 223–33. Ripple, W. J., Beschta, R. L., Fortin, J. K., and Robbins, C. T. (2015). Wolves trigger a trophic cascade to berries as alternative food for grizzly bears. Journal of Animal Ecology 84, 652–4. Ripple, W. J., Larson, E. J., Renkin, R. A., and Smith, D. W. (2001). Trophic cascades among wolves, elk and aspen on Yellowstone National Park’s northern range. Biological Conservation 102, 227–34. Solomon, M., 2007. Social empiricism. MIT press.