This paper illustrates the need and importance for this shift in understanding, and suggest corresponding revisions to the ESA. It gives an overview of the recovery plan provisions and delisting process of the ESA. The paper gives an overview of a recently delisted species, the Western Great Lakes Distinct Population Segment (“WGL DPS”) of the gray wolf, which serves to illustrate in detail the more general analysis of ESA implementation. The paper also provides a framework for the importance behind identifying policy decisions and having greater accountability and consistency in decision-making.
The primary purpose of the Endangered Species Act (either “ESA” or the “Act”) is to conserve, protect, and assist in recovering endangered species.  While few doubt the importance of the Act to environmental law overall, many call for amendments to “fix” a host of perceived problems in listing species and creating and enforcing recovery plans.  Additionally, the Act is increasingly under pressure to produce results and show that the ESA works—by delisting species.  This paper maintains that the most important revisions to the ESA must include change to the manner in which agencies frame decisions so that agencies can no longer use science as a shield to hide inherently policy-based decisions.  This re-framing not only calls for Congress to acknowledge that “best available science” can never be the sole consideration in delisting, but also means that the myth of objective science feeding agencies the correct answers must be dispelled. Instead, the agencies must own up to the degree they decide society is willing to go in protecting endangered species. By facilitating honesty, these changes will eliminate ignorance in favor of conscious decisions, and will lead to accountability and better functioning of the democratic process in lawmaking.
This paper will illustrate the need and importance for this shift in understanding, and suggest corresponding revisions to the ESA. Part II will give an overview of the recovery plan provisions and delisting process of the ESA. Part III will give an overview of a recently delisted species, the Western Great Lakes Distinct Population Segment (“WGL DPS”) of the gray wolf, which will serve to illustrate in detail the more general analysis of ESA implementation. Part IV will provide a framework for the importance behind identifying policy decisions and having greater accountability and consistency in decision-making. Part V will explore the role science plays in the recovery plan decision-making, and correspondingly science’s inherent limitations in creating goals to “recover” a species to non-“threatened” status. Part VI will analyze in-depth the language used, scientific basis relied on, and policy decisions inherent in the recovery plan for the WGL DPS of the gray wolf. Part VII will similarly seek to find how the five factors of the delisting process can be based solely on best available science, and how this Congressional mandate is essentially an illusion. To further demonstrate these revelations in practice, Part VIII will analyze the delisting of the WGL DPS of the gray wolf. In Part IX, the paper will suggest potential remedies to the current structure and implementation of the ESA in recovery planning and delisting, focusing on decision-makers acknowledging policy-based decisions, giving transparency to the reasoning these decisions are based on, and embracing the resulting accountability once the “solely science” myth is eliminated.
II. The ESA: Recovery Plans and Delisting Process
Once a species is listed as threatened or endangered under the ESA, the Act requires the Secretary to create and implement recovery plans for the species’ conservation and survival, unless the Secretary determines in his or her discretion that a plan will not promote conservation of the species.  The Act requires the Secretary to give priority to those endangered species that will most likely benefit from a plan, in his or her discretion.  Courts have established that the Secretary has broad discretion to implement recovery plans.  Recovery is defined by federal agencies as “improvement in the status of listed species to the point at which listing is no longer appropriate under the criteria set out in section 4(a)(1) of the Act,” which essentially means to the point the species no longer satisfies the Act’s definitions of threatened or endangered.  The Fish and Wildlife Service (“FWS”) further explains recovery as “the process by which the decline of an endangered or threatened species is arrested or reduced so that species’ long-term survival in the wild can be ensured.”  The Secretary must include 1) a description of site-specific management actions as may be necessary to achieve conservation and survival of the species; 2) objective, measurable criteria that when met allow removal of the species in accordance with the Act’s provisions; and 3) estimates of the time required and the cost to carry out those measures.  The recovery plan decisions must also have “support” in science. 
To delist a species under the ESA, the Secretary must consider five factors based solely on the “best scientific and commercial data available:” 1) present or threatened destruction, modification or curtailment of a species’ habitat or range; 2) overutilization for commercial, scientific, or educational purposes; 3) disease or predation; 4) inadequacy of existing regulatory mechanisms; and 5) other manmade or natural factors affecting its continued existence.  Three reasons exist to delist a species: erroneous original classification, extinction, and, the point of interest of this paper, recovery. 
Usually, meeting recovery goals will lead to the Secretary evaluating whether to delist a species. If the recovery plan is what the Secretary relies on to show recovery, the plan must have addressed the five factors listed above.  After publishing notice of the proposal to delist in the Public Registrar, three independent species specialists may give their expert opinions to the agency; this is called peer review.  Once a species is delisted, the ESA requires agencies to cooperate with the appropriate states in monitoring the species for at least five years to ensure the species can sustain itself without the protection of the ESA.  If threats to the species arise in this period, the agency could extend monitoring or re-list the species. 
Since the Act gives the Secretary wide discretion in deciding the recovery plan, and even in whether to implement it,  judicial review of delisting in practice likely only forces the Secretary to actually decide that the five delisting factors were satisfied, with support of these decisions in “science.” 
III. Introduction to the Western Great Lakes Distinct Population Segment of the Gray Wolf
The ESA currently divides the gray wolf, canis lupus lycaon ,  into several distinct population segments, a unit recognized under the Act as meriting individual listing. Until recently, the WGL DPS was listed as endangered in most states and threatened in Minnesota. In 1967, due mainly to human-caused mortality, the gray wolf that had once occupied most of America was only present in Minnesota, Wisconsin, and Michigan—roughly three percent of its former historical range.  Biologists estimate that hundreds of thousands of gray wolves once inhabited America.  Prior to listing under the ESA in 1976, the gray wolf in Minnesota was the most significant wolf population and totaled at most one thousand individuals.  The populations in Wisconsin and Michigan were believed to be zero until about 1975 for Wisconsin, when individuals likely dispersed from Minnesota, and until about 1989 in Michigan, when a pair likely dispersed from Wisconsin. 
The latest survey relied on by the FWS estimates 2,301 to 3,708 wolves in Minnesota, 425 to 455 wolves in Wisconsin, and 405 wolves in the upper peninsula of Michigan.  Though annual increases in population and range are almost consistently reported in all three states, and any non-increase is not scientifically significant, the FWS indicated in its proposed rule to delist the WGL DPS that this data showed the WGL DPS had expanded its range into all suitable habitat and was reaching carrying capacity in these areas. 
The WGL DPS boundaries include all of Minnesota, Wisconsin and Michigan; the east portion of North and South Dakota; the northern half of Iowa; and the parts of Illinois, Indiana and Ohio north of Interstate 80.  These boundaries are based on including the core population, nearby areas where wolf packs may become established in the “foreseeable future,” and areas where wolves will disperse but are not expected to created persistent packs.  The proposal to delist the WGL DPS relied primarily on the recovery population goals being met and exceeded, and on the evaluation of Minnesota, Wisconsin and Michigan’s state management plans to alleviate present and future threats to the existing populations. The proposal also considered the five factors, as required by the ESA.
Currently, the wolf range includes northern Minnesota and Wisconsin, and the Upper Peninsula of Michigan.  While lone wolves have been found in other states, at the current population levels the wolf’s recognized range in the United States has not extended further than the northern parts of these three states.  However, the distribution of the packs has expanded consistently over the years, while under protection of the ESA.  FWS relies on studies finding the percentage of increase has been smaller in recent years, though this difference is not found in the studies to be significant. 
IV. The Importance of Transparency, Accountability, and Consistency
This paper will point out a myriad of ways that the recovery plan and delisting process is based on policy decisions, and encourage that agencies and/or Congress make such decisions transparent for greater honesty in boundary setting, accountability, and consistency. These are not just useless exercises for “better government,” but rather have real on the ground effects that matter.
To clarify, while this paper will point out how the agencies incorporate policy in the boundaries they set, it is important to realize that such policy choices can come from many sources. The applicable statutes, regulations, and court decisions should and do weigh heavily in policy choices by an agency. The agency also can make its own policy choices, though, often based on executive branch direction. In both contexts, especially the latter, the final source of policy choices is important: public opinion. The public can watchdog compliance with laws, but also can apply pressure, that the agency must balance, even if the public value is not in fact codified. Keep in mind this framework of value sources throughout the following discussion of the importance of clear reasoning behind boundary setting and policy choices in decision-making.
Agency analysis and explanation of policy and social value choices has at least a three-fold value. The first value is honesty within the agency and with the public. This may sound academic, but the apparent lack of honest or candor actually is often the root of much distrust and litigation. Beyond this practical importance, honesty leads to clearer decision-making, thereby greater efficiency, and forces the agency to make the necessary hard decisions rather than limping along on excuses.
The second and third values stem from honesty. Honesty leads to greater accountability to the public and to the legal system itself. Transparency in decision-making allows greater public education, which allows democracy to function properly. Meaning, while Congress and agencies get the first crack at determining policy decisions, the important feedback function of democracy is that if such decisions are not an acceptable value, the public vocalizes to let the government know. Only if the public has the ability to know such reasoning can they agree or disagree, however. For example, protection for ten years versus protection for fifty years runs differing risks of extinction for a given species. The public should know why one degree of risk is chosen over the other to fully participate in rather such determination is the actual amount of risk society is prepared to tolerate in species protection.
Accountability to the legal system itself requires asking whether the value choices made by the agencies are consistent with those of the ESA. This legal question is not answered with great clarity in the ESA, but the statute does embody some value choices that can guide this evaluation.
For instance, the agency could apply the ESA’s own policy choices on human valuation as indicating a preference for a common, consistent valuation of what “foreseeable” means in “foreseeable future.” Foreseeable future could feasibly refer to three different perspectives: foreseeable to the given species, foreseeable to humans, or factoring in both perspectives. While some may argue that not basing protection from any perspective other than from the wolves takes focus off of protecting wolves for their own inherent value, or that species differ too much to allow consistent determinations on what constitutes risk and time scale, value choices within the ESA indicate at least including the human perspective. The ESA in the purposes and policy section specifically states “these species of fish, wildlife, and plants are of esthetic, ecological, educational, historical, recreational, and scientific value to the Nation and its people.”  Additionally, the ESA allows economics to play a role in decision-making, ESA decisions occur over human life spans, and humans are a part of the ecology valued in biodiversity. By making the foreseeable future consistently take into account human life spans, not only would greater protection actually be given to most species because of how long our life spans are, it would also make species protection actually relate to the timescale humans value and understand, and lead to greater consistency throughout ESA protection.
As argued above, consistency should at least be the presumptive starting point for ESA decisions, and is supported within the ESA’s own value choices. Placing importance on consistency makes a better and more efficient process for innumerable reasons, including ease of monitoring policy choice changes, whether such changes reflect what society finds acceptable, and whether such decisions are scientifically consistent with a species ecological value. Additionally, besides ESA indications on how to make consistent decisions, discussed above, the ESA also importantly extended potential listing to all species that merit listing which are “fish, wildlife, or plants.” 
Transparency, accountability, and consistency are not the quick fix to making easy decisions under the ESA. These principles are a starting point, and a conversation starter, to making agencies accountable for the real reasons and having honest law making. The reality is that such changes discussed throughout this paper may not be necessarily “better” for a given species, or not necessarily easier for the agency and society, but most importantly will result in more upfront consciously made laws and regulations, dispelling back-room dealings and public ignorance. Hard decisions will result; we will face the future of each species together.
V. Recovery Plan: Clearly Not Based Solely on Science
The ESA gives little indication what recovery plans must include or be based on, which gives the Act great flexibility to deal with the wide range of species it covers, but does not guarantee that recover plans actually carry out the goal of the Act—conservation and recovery of species. For many listed species, recovery plans are essential if they are ever going to break out of their “emergency room status”  of threatened or endangered. If the delisting process (supposedly based solely on science) is to rely on recovery plans to show recovery of a given species, guidelines must be established to place boundaries on the Secretary’s discretion, and these boundaries must acknowledge that policy is inherent in this process and provide transparency for the public to know what these policy decisions are.
A. Science’s Role: Understanding Why Certain Policy Decisions are Detrimental to Species on a Given Timescale
Science is important to understanding how a species will be affected by given policy decisions, and thereby integral in deciding with policy to value in setting boundaries. For example, creation and implementation of a recovery plan can be essential to keeping many species, listed as on the brink of extinction, from actually becoming extinct.  To understand why, science must be consulted.
Depleted populations—including those of species listed as endangered or threatened—are subject to acute effects from selective pressures because of genetic impoverishment and compromised age structure. Long periods of population depletion further impede a species’ chances for survival because of genetic variance change over and leave the species vulnerable to additional pressures. 
Science helps the public understand that the Secretary’s decision not to create a recovery plan, or, if created, not to implement a recovery plan, can have negative consequences for species listed under the ESA. Science also helps the public or an agency understand whether a recovery plan actually helps the species reverse the trend towards extinction in certain ecological ways.
Thus, it is notable that science is limited to a “supporting” role, compared to the sole basis in the delisting process, in the Secretary’s creation and implementation of a recovery plan. These decisions only have to be “supported” by science. The Act’s phrasing of science’s role as needing only to support a decision is problematic because it does not require transparency in the range of scientific support considered, or available. By allowing agencies to cherry-pick scientific studies that support a given decision, without allowing the public to see the analysis behind valuations of conflicting studies, the Act currently invites other policy factors to weigh in through backroom dealing without public disclosure. An important improvement to the ESA would be to limit the Secretary’s discretion in recovery plans to being based partially on science, with disclosure and analysis of other scientific authority considered and reasons why the final scientific support was chosen. As indicated in this change, and discussed further below, science alone cannot inform the Secretary on what recovery plan criteria is “enough” to “recover” a species.
B. Limitations of Science: Determining How Precautionary Society is Willing to Be in Policy Decisions for Recovery Plans
Science does not have any mechanism for how scientific knowledge can answer policy questions of “how much conservation” or “ how much” risk of extinction is enough or tolerable.  For example, in determining how large a population of a listed species is “enough” for conservation under the ESA, the agency makes policy decisions like what degree of scientific uncertainty is tolerable to base decisions on, or how many years is considered the “future.” The answers to these policy questions put boundaries that scientists can input into equations and models to determine whether a recovery plan is adequate to recover a species based on these parameters. Importantly, these parameters are what we must recognize as integral policy-decisions that the Secretary makes. Correspondingly, the public has a right to see whatever real decisions led to the measurable criteria in a recovery plan, rather than cloaking the decisions as being mandated or approved by science. In other words, Congress and the public must no longer tolerate the hiding of economic values, societal values, and other factors as “answers” directly found through scientific modeling and field studies.
1. Inherent Uncertainty in the Science Itself
Science inherently does not find a final “truth.”  Scientists can empirically test theories, which can only be proven wrong, or observe the way the world works. These methods can only strengthen understanding about a species, though; they never resolve all uncertainty.  Science, therefore, even in its most base function of informing the Secretary on what effect his or her policy decisions will have on a species, requires a policy decision in understanding the effect of the information.
The Secretary needs to draw a clear line with regards to scientific uncertainty that is visible to the public, so that society can know what science the Secretary is willing to not act on from the start, or what science is enough to support an agency determination. The Secretary either can determine what amount of scientific uncertainty on a given issue means that a certain part of a recovery plan is not warranted, or that agencies should add measures to the recovery plan because society is willing to bear the costs in a more precautionary approach. For example, as mentioned earlier, in the WGL DPS of the wolves the FWS determined that the lower annual increases in populations, while not scientifically significant, indicated that the wolves were at carrying capacity within the three states.  Taking action without scientific support of the significance of a trend was a line drawn without analysis transparent to the public.
2. “Recovery” is Not a Scientific Principle
Recovery, as explained above, is not defined under the ESA, but in application agencies define it as a species no longer being able to be classified as threatened or endangered. Science cannot supply what this definition means; no biological definitions exist in science of what is an endangered or threatened species and what is not.  The Act defines endangered as “in danger of extinction throughout at least a significant portion of its range,” and defines threatened as “likely to become endangered within the foreseeable future.” 
The only word science can define in these definitions is “extinct;” science cannot inform the Secretary what is a “significant” portion of a species range, what “likely” means, or what part of the future is “foreseeable.” These decisions are value judgments, based on how much risk society is willing to take that a certain degree is enough to keep a species from going extinct within a time period society finds acceptable. Inherently, someone must determine limits on all these decisions, as otherwise the costs for protecting the species and habitat born by society would be enormous. Maximum protection could, in theory, place all listed animals with rights equal or greater than human rights, and recover populations at the fastest rate and to the largest extent biologically possible. The Act’s definitions “suggest the existence of some threshold viability level, but fail to specify what that threshold level is, how it should be determined, and over what period of time the risk should be assessed.” 
Additionally, when making the recovery plan, someone must make a range choice for what portion of the species historical range should be habited by these “viable” populations to be considered recovered. But nothing in science says what amount of land a species should cover. Science can inform on carrying capacity of a parcel of land, or the amount of land that individuals or groups might roam over, but not whether a recovery plan should include more or less land to “conserve” a species from extinction. That requires time and risk determination; otherwise, more would almost universally be better.
All these decisions can be masked, by framing the plan as being based on best science. What is left unseen is that science can only give an output based on what is put in; such decisions on timescale and range are essential to the Secretary’s determination that a recovery plan will actually recover a species under the Act. Changes to the ESA should emphasize that the importance of these decisions need to be recognized and visible to the public, and have incorporated boundaries in the Act’s regulations.
These changes would allow review of recovery plans and delisting to include whether the policy-based decisions, like the extent of range, degree of precaution, and risk choice, uphold the principles of the Act discussed earlier and are supported by the public, rather than the almost untrumpable science card.
C. Additional Limitations Created by Congress and Agencies
In determining the adequacy of a recovery plan and whether a species merits delisting, the agency is handicapped already in making a pure scientific determination because of the definition of a species, and especially because of the reality of determining a “distinct population segment.”  The fringe individuals of a “species” is by nature constantly changing due to adaptation, besides the fact that science itself does not as a community agree on one encapsulating definition of the concept.  A DPS is even more policy-based, and likely even politically- or whim-based, as demonstrated by the case of the WGL DPS in the current proposal including the three core states of the former Eastern DPS of the wolf. Each of these limitations deserve a paper in themselves to fully discuss the law behind them and implementation decisions, but further analysis is beyond the scope of this paper.
VI. To Illustrate: Policy Decisions Within the WGL DPS Recovery Plan
The WGL DPS Recovery Plan was originally published in 1978, and was revised in 1992.  As demonstrated below, policy was prevalent throughout the recovery plan process. The decisions themselves are not in dispute in this paper; however, characterizing these decisions as coming from the mouths of scientists is disputed.
A. On the Surface
The Recovery Plan for the WGL DPS of the wolf gives the “reasonable actions believed to be required to recover and/or protect” the listed species.  The objective of the plan is listed in two different ways: 1) the legal act of delisting the species, and 2) maintaining and re-establishing viable populations of the species in as much of its former range as feasible.  Seemingly, for the FWS to meet these objective(s), the recovery criteria must be achieved. The recovery criteria requires at least two viable populations that include 1) a stable, growing, and continued assured survival in the Minnesota population, and 2) a re-established population outside Minnesota with a minimum of 100 wolves if within 100 miles of the Minnesota population, or a minimum of 200 wolves if beyond 100 miles.  Additionally, these populations must be maintained for five consecutive years.  The FWS reveals that its basic approach to recovery plans is to “ensure” that two viable populations exist. 
The plan also includes a discussion of methods to achieve the plan’s goals. The main methods highlighted were 1) highly regulated management in Minnesota, including complete protection in Zone 1 (except live-trapping and transplanting of problem animals), and extensive prey habitat improvement in Zones 2 through 4; 2) re-establishment of at least one viable population outside of Minnesota; 3) an intensive public education campaign to enlighten public to wolf ecology and wolf management initiatives; and 4) strong research efforts to better understand wolf ecology, predation, population dynamics, and dispersal; causes of range restriction and mortality, including parasites and diseases; development’s effects on wolf populations; and re-establishment of wolves and augmenting low wolf populations. 
The recovery plan defines “viable,” as used in the recovery criteria, as either 1) an isolated population that averages at least 1 wolf per 50 square miles, has at least 200 wolves in a self-sustaining population, is distributed over at least 10,000 contiguous square miles, and maintains these criteria for five consecutive years; or 2) a population within 100 miles of the population listed in (1) that averages at least 1 wolf per 50 square miles or has 100 wolves within at least a contiguous 5,000 square mile area, and maintains this over five consecutive years.  The 1992 revision added in the characteristics of the second population and five consecutive years requirement.  The recovery plan then identified the 1992 population of Minnesota wolves and a Wisconsin/Michigan (W/M) population of 100 wolves as “viable.”  The W/M population was justified in the plan as viable because immigration from the Minnesota population would “supplement it geographically and genetically for the foreseeable future.” 
Even on the surface of the recovery plan, significant policy decisions are masked as scientific answers. The recovery plan limits required actions to recover a species to only those “reasonable.” “Reasonable” is a FWS-determined level of extinction risk, the amount of burden society will tolerate to try and recover the species. It follows that if FWS determines that the only methods for recovery of a species required actions that were unreasonable, the species would likely not have a successful recovery plan and possibly become extinct. The recovery plan does not expressly indicate that FWS, not science, made such decisions, and certainly does not disclose how FWS made the call.
The two supposedly synonymous objectives in the recovery plan also reveal policy decisions. For instance, the recovery plan equates recovery to population numbers and range square miles, rather than focusing on the degree and quantity of threats to the species’ continued existence like the delisting process does. The manner in which the recovery plan objective is framed has real effects on how scientists attempt to determine how the objective can be achieved. While increasing population numbers and increasing the range of populations can lessen the impact of threats, such reduction is a matter of degree, the extent of which is determined by FWS. This way of framing recovery does not address the possibility of actually tackling the threats themselves, possibly through active management requirements in the plan and set objectives for level of threat reduction. Also, while public education and further study were mentioned as methods, no requirements in these areas were set as criteria that must be reached to designate the species as recovered.
Either way, a policy decision of the amount of acceptable risk was inherent in the plan’s objective. For delisting, the acceptable amount of various threats that still allow delisting were policy-based levels of acceptable risk. Unless all threats were eliminated, some degree of risk is left. For the population and range goals, the concepts of “viable” populations and the limitation of “as much as possible” for re-establishment were policy-ridden. These concepts incorporate the hard truth that a limitation on recovery will be set somewhere. If no limitation was set, recovery efforts could continue indefinitely. For example, full establishment into all the former range would protect the species the most from extinction. By limiting the range to agency-determined possibilities, the status of “recovered” included an accepted level of extinction risk. As explained earlier, such decisions are a necessary reality, but should be disclosed for honest implementation of the ESA.
Additionally, a “scientific” determination of a “viable” population is not solely limited to population numbers and range miles. Many other factors contribute to the viability of a population, such as ecological and anthropogenic threats. The recovery plan did expressly recognize that “minimum viable population estimates are extremely subjective, based on different combinations of assumptions like economic costs, agricultural interests or ecological and anthropogenic threats, upon which reasonable biologists will disagree.”  The recovery plan nevertheless seemed to interpret this scientific disagreement as allowing FWS to define “viable” based solely on wolf and mile numbers, without revealing the other combination of assumptions inputted to reach those numbers.
The recovery plan didn’t truly define what viable meant, but instead gave the numeric definition for the two populations described in the recovery criteria. Since the recovery plan equates recovery with attaining viable populations, in effect recovery equals viable populations, which equals the number of individual wolves per certain square miles. The recovery plan does not reveal how these numbers are reached, nor why viable is expressed solely in individual wolf numbers and contiguous square miles, rather than their biological units, wolf packs, and suitable habitat square miles. Not integrating scientific knowledge of the wolves leads necessarily to the realization that FWS was working from a policy of valuing seemingly arbitrary numbers. In fact, the only scientific biological information cited with regards to number of individuals per square miles is the approximate carrying capacity of one wolf per ten square miles. While this was clearly an optimal level of recovery, no explanation is given for why the level of one wolf per fifty square miles for a total of 200 or 100 individuals was a recovered population.
Also, viable appears to have two definitions dependent upon whether the population is self-sustaining or is a population dependent upon a self-sustaining population. The acceptance of a dependent population as a viable population is disturbing because the approval of the first population essentially equates to a rubber stamp of the second population. The recovery plan also does not reveal what viable means for the future of the wolves—high chance of survival over 100 years, or over fifty years, or over just the five years that must be maintained to meet the definition of viable?
Additionally, while the recovery plan admitted that subjective assumptions formed the basis of these viable population estimates, it seemed to solely characterize this disagreement as scientific, rather than the result of different policy-influenced inputs.  Unfortunately, science alone cannot answer the question of how many wolves make a viable population without knowing these policy restrictions. Examples of policy restrictions used as inputs include the numeric definition of “foreseeable” years desired for the population to survive, the amount of acceptable risk that the model (if even based on a model) accurately portrays the effects of threats to the population (which likely includes an acceptable level of uncertainty in the quality and quantity of scientific studies that the model is based on), and other such factors.
The recovery criteria reflect a focus on the population and range objective as well, and begin to reveal the level of accepted risk. The number of populations required is limited to two and the number within each population is set at a “viable” minimum—a stable or growing Minnesota population with “assured” survival and a second population of 100 or 200 wolves, depending on location. “Assured” survival gives the illusion of permanent populations of wolves indefinitely, meaning an almost zero risk of extinction based purely on the number of individual animals in a given location. In actuality, “assured” means the level of risk over a certain number of years that FWS has determined is acceptable. While in theory this number could be a very protective policy, the requirement of these numbers only needing to be maintained for five years of monitoring cautions against such hopes. While possible re-listing does protect the species post-delisting, this also requires the species to reach the level of threatened or endangered again, likely a significantly more fragile condition compared to the recovered condition. Whichever the policy, this is precisely the information the public should know, and that FWS should have to support with reasoning. Also, by requiring only five years of stability or growth, FWS again gives the illusion that science says five years will “assure” recovery of the species, rather than detailing the policies that led to determining the acceptable risk.
The recovery plan explains that the requirement for multiple populations stems from a basic conservation biology concept that a species isn’t secure from extinction if only a single population exists.  While this statement is troubling from a scientific standpoint in that no species is really “secure from extinction,” but rather always has some level of risk, it also reveals the policy decision inherent in selecting the number of minimum viable populations for the recovery criteria. Requiring a single additional population in some part of the former range is clearly a minimum policy decision to expansion of the species to recovering in its historical range. Also, having more than one population minimizes the possibility of extinction through the impact of disease, loss of prey species, catastrophic habitat modifications, and the like.  Logically, the more populations in existence, the more the risk decreases to the species as a whole. In fact, the recovery plan states “the Recovery Team would have liked to have several viable populations prior to recommending delisting, but settled on two as the minimal acceptable number.”  Seemingly, two populations would be the highest risk possible in following the biological concept, especially since one of the populations is dependent upon the other for geographic and genetic supplementation. No explanation is given for why the Team determined that such risk was an “acceptable” policy.
Furthermore, the general determination of placing boundaries on range for characterization as recovered is clearly a policy decision. The reality of the ESA is that federal protection is likely the most substantial and at least the most consistent over the entire historical range of a species that the species will have. As I will discuss later, state management likely halts further expansion by either maintaining current numbers or even decreasing current numbers when ESA protection is lifted. Therefore, satisfaction of recovery range under ESA protection is an important decision for the future of the species, and one that science cannot provide an answer for since species like the wolf, who FWS admits can inhabit a wide range of habitat, do not naturally limit themselves to a portion of their historical range.
While FWS can point to trends in the scientific population data of slowed growth, which in this case was not even scientifically significant, the truth is that with maximum protection and efforts to actively make more suitable habitat protected almost any species could continue to expand its population to areas larger than a recovery plan decides is “enough.” The decision not to require further expansion of range is not to further protect recovery of a species; it is rather a policy decision of what is enough. The reasoning behind this policy decision is what should be debated in determining range in the recovery plan, not whether scientific data shows carrying capacity in a few “important” areas.
B. Delving Deeper
The recovery plan then listed essential factors to maintaining a viable wolf population, which were largely identical to the listed factors deemed critical to long-term survival. Those factors were 1) large tracts of wild land with low human density and minimal human accessibility; 2) ecologically sound management; 3) adequate wild prey; 4) adequate understanding of wolf ecology and management; and 5) successful contention with the adverse effects of parasites and diseases. 
Again, policy choices are hidden within this list in a misleading manner. For instance, why these factors are critical while others are not, such as the ability to contend with the effects of global warming or effects of hybridization with coyotes, is not revealed. With respect to hybridization, despite scientific evidence of hybridization events and recognition of possible increases in chances for hybridizations, the plan concludes that no problem was apparent in 1992, or at least not a significant one.  The amount of precaution taken in guarding from risk is evident in the plan distinguishing between threats, as a policy line is drawn at some point for a certain level of risk being deemed “critical.”
What amount of time FWS means by “long-term” survival is hidden from public view, and similarly depends on how precautionary FWS is acting in protection of the wolves. Even beyond the terms and factors, having the definition of “viable” dependent upon the same terms as “survival” seems to indicate that the recovery criteria emphasizes solely survival, rather than a level of recovery beyond mere survival of the species. Tellingly, the factors that make up the determination of “long-term” aren’t part of the recovery criteria. In fact, no indication is given of the important boundaries in relation to each factor or how they should be balanced.
1. Policy Within the Factors
a. Large Tracts, Density, and Accessibility
The factor of “large tracts of wild land with low human density and minimal human accessibility” is replete with policy decisions. This factor cannot be characterized as solely science because of the use of “large,” “low,” and “minimal” as measurements. None of these words can in themselves lead to a scientific answer of whether a certain amount and quality of habitat fits these characteristics. These words leave substantial discretion to FWS in determining the policy of what is large enough, low enough, or minimal enough. The policies inherent in this factor are further revealed in the determinations of what the Recovery Team found as sufficient in each of these respects, and the methods they used to designate important areas.
The 1992 plan stated that the Team believed “sufficient suitable habitat exists in Minnesota, Wisconsin, and Michigan to achieve the Recovery Criteria.”  The plan later explained that the numbers of wolves in Minnesota would “ insure perpetuation at level optimal to various [Zones], with optimal including carrying capacity and compatibility with humans.”  The habitat in Minnesota was further divided into levels of suitability, Zones 1 through 5.  Zones 1-3 were designated as critical habitat in 1978 and were considered by the Team as primary wolf range, though what areas were considered part of each zone was revised in the 1992 recovery plan based on habitat conditions and wolf numbers.  Zone 1 included 4,462 square miles, Zone 2 included 1,864 square miles, and Zone 3 included 3,501 square miles.  Zone 4 was labeled a peripheral range, containing 20,901 square miles.  Zone 5 was considered as not part of the wolves’ range, and, in fact, was advised to exclude wolves.  Zone 1 was allowed to fluctuate naturally, Zones 2 and 3 had a maximum of 1 wolf per 10 square miles, Zone 4 had a maximum of 1 wolf per 50 square miles, and Zone 5 had a complete ban on wolves.  While still considering a few areas for re-introduction, the plan cautioned that investigations may not indicate re-establishment in those regions “prudent.” 
By finding the current habitat sufficient and therefore “large,” the recovery plan included a policy choice on the degree of protection afforded to the species. First, the plan incorporated a policy decision that the amount of habitat of suitable quality already in existence was enough for whatever time span the Team deemed foreseeable, and that the percentage of and location in the former range was enough. Second, the plan incorporates policy choices in finding what re-establishments are prudent, which clearly is not based on what habitat would help the species expand and decrease its risk of extinction the most. Correspondingly, the plan does not indicate what weight is given to the different factors involved.
“Insure perpetuation” is again misleading in appearing to give a scientific guarantee that wolves will survive indefinitely under the plan population numbers. Actually, these policy determinations include some level of risk as to what FWS determined was “insured” and what amount of time was “perpetual.” The optimal level at least discloses that besides carrying capacity, compatibility with humans is factored in, but fails to reveal the policy details concerning acceptable levels of humans. Science cannot give the answer as to how many humans in an area equals the threshold for incompatibility, or at least not without operating under a policy input of what level of interaction is acceptable within the definition of compatible. Clearly, science alone did not require that Zone 1 be allowed to fluctuate naturally while other zones had set numbers, or even a ban; FWS determined the more important policy for Zone 1 was natural habitation and the most important policy in the other zones was active management.
The plan generally outlines developmental and accessibility effects. For development, the effects are 1) more human development increases the chance of direct killing of wolves; 2) although undocumented, unnatural environment might deter wolves from inhabiting; 3) citing unpublished data, paved roads and the like might prevent and/or minimize dispersal; 4) increased human presence increases chance of pets introducing disease and parasites; and 5) reduced prey species abundance and diversity reduce wolf food supply.  The 1992 plan added the term “road density statement” to describe and quantify the importance of minimizing roads open to the public, with the guiding principle that human activity and exploitation increases with accessibility.  The statement includes a discussion of variables to consider in evaluating road density, such as proximity to den sites, as well as a critical density threshold.  The critical density threshold promoted by the plan was equal or less than 1 linear mile of road per square mile of land in the designated recovery areas for Michigan and Wisconsin, or where road density was limiting wolf recovery in Minnesota. 
Zones 1 through 3 were indicated as having lower accessibility and road density, while Zones 4 and 5 had higher degrees of each.  Zone 4 density and activities were “much” higher and “highly” accessible, with the only quantitative evaluation being “few areas not within 3 miles of roads.”  Zone 5 not only had greater accessibility and human density, but also a high proportion of intensely farmed areas. 
The plan also finds that total legal protection of the wolves has not led to “massive” increase in wolf populations in non-forested areas.  The plan reasons that illegal activities and accidental kill of wolves in accessible areas, and possible avoidance of open, settled areas, could explain the lack of repopulation in those areas. 
The developmental and accessibility effects are very disturbing in the manner they are presented, as they appear to the general public as scientific conclusions that FWS justifiably follows. Many of the “scientific” conclusions cited, however, rely on undocumented, unpublished, or no data whatsoever.
Again, loose terms such as “higher” accessibility and road density shade from the public an understanding of what level of these factors qualifies as higher or lower. Despite the quantification and citation to data, decisions that differentiate areas into Zones 1-3, from Zone 4, or Zone 5 are policy decisions that include factors outside the wolf’s ecology. While clearly the proportion of farms is considered relevant to these determinations, why one concentration is significantly different from another to justify different management is unknown. Clearly science does not answer this question, as the plan itself only cites to the rough determination that more of some factor that is human means more of some factor detrimental to the wolf. Even more indistinct is what FWS refers to with respect to the Zone 4 few “areas” not within 3 miles of roads. What constitutes a tract of land being considered an area, and therefore how FWS determined that “few” of such areas existed, is unclear and seemingly arbitrary.
The plan also shows policy determinations in its characterization of attempted wolf dispersal into areas beyond the range FWS determined the wolf should occupy. Despite citation to studies and data that have documented wolf dispersal into more developed areas, the plan characterizes these attempts as not a “massive” increase in wolf populations in these areas. In other words, despite the wolves documented actions that fit the ecology of their species cited earlier in the plan, FWS discounts the importance of these areas to wolves. FWS guesses that illegal and accidental kills and avoidance by the wolves might play into why these areas are not suitable habitat for the wolf. In truth, this characterization by FWS only serves to mislead the public into thinking wolves naturally don’t try to develop their range any further than FWS has determined important. Under this assumption, FWS doesn’t need to reveal that other activities are valued higher in these areas than recovery of the wolves.
Many of these decisions are likely based on how much the agency decides society is willing to tolerate. For example, consider FWS finding that an average of one wolf per 50 square miles in the zone between the core wolf area and the more human populous region of the state is tolerable. This determination is based not only on the recovery of the wolf for the wolf’s sake, but also on just how far society is willing to tolerate having the wolf recover before the side effects of recovering the wolf are too much of a burden. While understandable, these determinations need to be explained for what they truly are, not touted as being what is best for the species. Once honest accounting of decision are revealed, the ESA can operate with the accountability described earlier. Specifically in the case of the wolves, this could lead to changes in regulations through public pressure, as the public weighs in on FWS’ policy decisions.
b. Ecologically Sound Management and Wild Prey
For the ecologically sound management factor, five prongs were stated in the plan: 1) protection where needed in order to restore to the original range and to preserve a naturally functioning population that can serve as a living museum, as a scientific subject, and as a reservoir to repopulate adjacent areas; 2) depredation control where wolves are killing domestic animals; 3) restocking into suitable areas of former range when feasible; 4) continued research and monitoring of wolf populations; and 5) provision of adequate prey diversity and numbers through habitat and population management and reintroductions where appropriate. 
These factors reveal how many of the decisions in the plan discussed above were based on policy influences other than “ecologically sound” science. Despite the importance of restoration to the original range and preserving naturally functioning populations,  the plan only requires two populations within the original range, actually bans the wolf from part of the original range, and only allows the population to naturally fluctuate in Zone 1 of Minnesota, if anywhere. Despite the stated ecological principles, clearly outside factors weighed in FWS decision-making to result in such a stark contrast between the plan’s management guidelines above and the actual plan requirements.
The first prong, protection, interplays with both the depredation control and prey population management prongs. Under the plan’s recommendation for depredation control, the amount of protection afforded depends upon what area or zone the wolf resides in. The plan recognizes that human development with livestock and pets may be subject to wolf attack wherever they exist.  While all zones include some level of depredation control for documented cases where likelihood exists for the occurrence of additional depredations, only in Zones 4 and 5 does the plan recommend preventative depredation control.  Preventative control in Zone 4 would allow lethal trapping for one-half mile around verified site, if that site had verified depredation in three of the previous six years and was likely to recur.  In Zone 5, trapping could occur in the same set of circumstances, but expands the radius to within five miles of the site.  Zone 5 also was determined in the plan as not suitable for wolves, thereby allowing depredations to be addressed by any legal means. 
The plan justifies these modifications to total protection by finding that, since the wolf itself is controversial, control may be necessary for the species’ own good in certain areas to maintain positive public attitude and/or decrease illegal takes.  The plan also finds that regulated takings will undoubtedly be necessary in any area that the wolf is reestablished for the same reasons.  The decision to even allow depredation control is a policy decision, as evidenced by the justifications given. The danger of these justifications is that any depredation program would seem to suit this policy as long as it wasn’t as detrimental to the species as increased illegal takes and destruction of legal protection. In other words, this seemingly justifiable policy decisions could actually gradually wear away the integrity of the protected populations as more trapping is allowed to “counteract” negative public attitude or illegal takes over time.
The depredation policy selected isn’t scientific either; in fact, it appears almost arbitrary in differing policies for the same wolf. In Minnesota, what level of depredation control a wolf undergoes depends upon what zone the depredation occurs. However, the way the zones were designated does not necessarily keep wolves from other zones from quickly entering banned zones. This means that an individual animal, of a species known to cross long distances, could travel a short distance from its pack range, designated critical habitat, in Zone 3 onto a Zone 5 ranch, and be killed by any legal means. This disregard of the species’ biology could lead to an arguably unjustifiable, and perhaps dangerous, lowering of the core population of the wolves.
These zone dependent policies could be, and seemingly were, chosen under the justification of ease of implementation, rather than any scientific basis for determining why some wolves that cause depredations should be killed, while others should only undergo non-lethal control. Furthermore, the determination of preventative control, with differing radius determination and the requirement of 3 over 6 years, can only be explained as a seemingly random policy decision that were determined as protective enough of wolf recovery. While enforcement must draw lines for implementation, the basis of such determinations must be transparent to the public so that they can hold FWS accountable for such policy decisions, especially when they appear to directly contradict the species’ biology.
The interplay between protection and prey species management is even more revealing. First, the plan recognizes the need for an exception to complete protection in two of the critical habitat zones, Zone 2 and 3.  The exception would allow consideration of artificial reduction of wolves if the deer population decreases over a three year period below the numbers necessary to support one wolf per ten square miles.  The plan specifically does not include Zone 1 in this exception, reasoning that in Zone 1 a more important value exists in allowing wolf numbers to fluctuate naturally, rather than try to maximize wolf densities.  The plan only contemplates that temporarily reducing or prohibiting harvesting of prey species to increase prey numbers is “biologically sound” in extreme circumstances, such as successive severe winters. 
This exception for Zones 2 and 3 is clearly policy driven by hunter interests, though washed as the biologically sound management practice. The first indication is that reduction of wolf numbers is advised if the number of deer is below the abundance necessary to support 1 wolf per 10 square miles. This is the number cited by the plan as the scientifically studied carrying capacity, or wolf maximum. Rather than implementing a policy that allows the species to naturally react to prey abundance and fall below the highest density of wolves, the policy is to actively, lethally control the numbers lower. This decision is disturbing in that human control does not follow the same guidelines as natural deaths from limited prey, thereby affecting natural selection, but also because this indicates hidden policy influences of possible revenue through controlled wolf hunting. Furthermore, the recovery criteria discussed above doesn’t even require wolf populations at carrying capacity to be delisted, yet this policy recommends killing wolves if there are less prey than would support the fictional maximum wolves.
The second indication is that Zone 1 is distinguished as having a more important value in natural fluctuation. This shows that different policies are integrated into management of these zones, despite all being labeled critical habitat, and despite the supposed value given to the ecological importance of naturally fluctuating populations. Apparently, other policy influences trumped these factors in Zones 2 and 3. While these policy reasons may be valid, they are not transparent and thus could avoid accountability if they are actually invalid, or even differing from what the public finds acceptable.
The last indication is that reduction of wolf numbers, which were still a listed species under the recovery plan, is encouraged before reduction or prohibition of human harvesting of the prey, even for sport. Limiting human harvesting of the prey in only “extreme” circumstances further cements the degree policy influenced this “biologically sound” exception.
Second, the recovery plan indicated one of the most important aspects was to maintain habitat in a high carrying capacity for prey.  The plan identified the most feasible method for accomplishing this as commercial and noncommercial sales of timber and habitat improvement projects.  Such sales and projects included creation of temporary roads. 
To discover the different policies implemented, this factor for prey management serves best as a comparison to the management of the wolf. Though different species, it cannot be denied that creation of new suitable habitat is beneficial to any species. However, only for the wolves’ prey does the plan contemplate active habitat creation. In comparison, wolves are considered for reintroduction into areas “as is,” based on whether “feasible” due to limitations on road density, accessibility by humans, and other policy considerations discussed above.
Of further note, the plan identifies as the most feasible method for new habitat creation for the prey as logging, which includes creation of temporary roads—the very culprits discussed at length as being so harmful to the wolves. While named temporary, these roads have the same lasting effects as regular roads for years in the future. However, FWS clearly had other policy inputs factoring in that trumped these concerns for the wolves’ recovery.
VII. Delisting Process: The Best Available Science Charade
Congress ideally envisioned the delisting process as mirroring the listing process in reverse and resting solely on the “best available science,” according to the courts.  Those words mask the fact that the best available science on a species can be minimal, or be flawed to a certain degree, but still allow agencies to make decisions in the name of that data.  Those words also do not allow room for recognition of the policy choices inherent in the five factor analysis; those choices determine the “scientific” answers agencies must tout as being objective to satisfy the Act.
A. How Much Science and Which Science?
In the delisting process, decisions about which scientific data to use, how many different studies to consult, and how much certainty is enough to indicate trends, has enormous consequences in the amount of protection a species has to continue to expand its population in range and numbers in the future. While admittedly some scientific uncertainty will always exist, and different methods often will produce at least slightly contradictory results,  the fact remains that the agency’s decision to make decisions of delisting based on a certain number of studies, with or without scientific findings of significance in trends, is a policy decision. Some limit must be placed on what science is enough, and at times an agency could validly act without a finding of scientific significance in the data. However, these decisions need to be recognized for what they are for the full, honest process of transparency and accountability discussed earlier.
Congress has tied the agency’s hands on this matter by requiring decisions based solely on science; to cure this deception, Congress must recognize that science is a means to understanding what ends these policy decisions bring us to. Agencies must allow transparency in their decision making and provide explanation for why, in their expertise, they determine the science before them is enough to base important decisions on. This transparency will allow accountability for the true decisions that have force in the delisting process. The reasoning must stay true to the Act, and, ideally, actually have limitations in the regulations implementing the Act.
B. The Five Factors: How Can Science Evaluate Them?
Science can never serve as the sole indicator to agencies on whether the five factors no longer show present or future threats to the species in question. As explained above, to delist a species, the Secretary must consider five factors based solely on the “best scientific and commercial data available:” 1) present or threatened destruction, modification or curtailment of a species’ habitat or range; 2) overutilization for commercial, scientific, or educational purposes; 3) disease or predation; 4) inadequacy of existing regulatory mechanisms; and 5) other manmade or natural factors affecting its continued existence.  Some of these factors are obviously policy based, like inadequacy of existing regulatory mechanisms. However, all of them require important policy decisions in interpreting scientific data before them and in programming the parameters of predictive models for the future.
To delist a species, existing regulatory mechanisms, not including the ESA, should be adequate to keep the species from satisfying the Act’s definitions of threatened or endangered.  Regulatory mechanisms are laws; science has no ability to evaluate the sufficiency of legal mandates and enforcement. Science cannot predict how enforcement in practice, based on certain man power and factoring prosecutorial discretion, will minimize human-based threats to a species. Switching to state programs to ensure species continue “recovered” status requires policy decisions on whether a legal scheme is protective enough, and policy decisions on what level the population needs to meet to still be considered recovered, especially if lower than its current level under the Act. Remember that determining what constitutes recovered status under the ESA in itself required important policy decisions; otherwise, full historical range at carrying capacity would be the default goal. 
Unless all threats in the other four categories are completely eliminated for the species after delisting, agencies must decide what degree of reduction of threat is enough to merit delisting, and for how far in the future is enough to satisfy “foreseeable.” The decision to delist before threats are reduced to zero is in itself a policy decision, as what level of risk is considered “recovery” of a species. These are risk decisions based on the cost the agency determines society is willing to bear to recover a species. However, they are framed, as mandated by statute, as scientific findings of sufficiently reduced threats.  Once again, determination of what science and what degree of scientific uncertainty is enough are also inherent policy decisions in evaluating these factors.
When these factors are not reduced below the threshold (whatever that is), future state regulatory mechanisms are likely to be heavily relied on as proof of reduction of threats.  This reliance is not evaluated scientifically, can only be predictive as long as the states’ proposed laws remain in effect, and even when in effect depends on the same legal parameters discussed above. Additionally, after five years, the federal monitoring period required in the Act,  the state could adjust those laws in ways that could resemble the very threats that instigated listing of the species in the first place. While the species could be re-listed again in the future, a time lag between harmful actions and visible threatening/endangering of the species would likely exist. This lag could cause the deterrent of possible ESA regulation to seem far enough in the future, or too uncertain, to actually affect present state decision-making and regulations. Additionally, yo-yoing between delisting and re-listing would weaken the integrity of ESA delisting, by not only making the ESA appear faulty, but also possibly leading to such permanent harm of the species populations that they become beyond recovery.
VIII. Evaluation of Five Delisting Factors for WGL DPS
The five factors were all discussed in the proposal for delisting, and were determined not individually or collectively “likely to cause the WGL DPS of the gray wolf to be in danger of extinction in the foreseeable future.”  The foreseeable future was defined in the proposed rule as thirty years, justified as the period FWS could make reasonable assumptions of threats to the wolf’s future.  FWS considered the areas where viable populations now exist as the range of the gray wolves in the WGL DPS.
According to FWS, the areas of the range considered significant was based on biological significance; the portion of the range where threats in that area could threaten the viability of the DPS as a whole, specifically for the WGL DPS meaning those areas “that are important or necessary for maintaining a viable, self-sustaining; and evolving representative meta-population or multiple separate populations in order for the WGL DPS to persist into the foreseeable future.”  Interestingly, the five factors were not specifically discussed in the final rule, which instead focused on upholding, meeting, and exceeding the recovery criterion set forth in the recovery plan. 
Since the delisting proposal relies largely on the recovery plan, it inherently contains all the policy decisions described above, most notable the reliance on “viability” of populations, as well as containing a few unique charades of its own. The overarching policy issue in the five factor delisting analysis is that when threats still exist to some degree even under federal protection, what current degree of reduction of threat is enough to merit delisting? This decision is inherently policy based, though Congress mandates delisting to be based on strictly the best science. This problem is compounded when zero to minimal threat reduction has been accomplished by federal protection, recovery goals of population increase have been met, and the legal tool of future state management plans is utilized to justify delisting. Putting aside questions of whether the ESA supports such practice, at the very least the practical application of the ESA requires recognition that delisting cannot be based solely on science in order to get policy decisions out of the basement and into the public’s plain view.
Furthermore, selection of thirty years as the foreseeable future is clearly a policy decision explained only as a period that FWS can make “reasonable” assumptions of threats. While at least this time period was defined and explained to some degree, the risk of “likelihood” of extinction was not revealed. Also, the explanation did not distinguish why 30 years could allow reasonable assumptions rather than 31, 35, 40, or 50, to name a few. FWS did not explain why this definition was consistent with the recovery policies of the ESA, either, such as protecting the “esthetic, ecological, educational, historical, recreational, and scientific values of the species,’ or conserve the species ‘to the extent practicable.’ 
A. Present or Threatened Destruction, Modification, or Curtailment of WGL DPS Habitat or Range
FWS recognizes that habitat for the gray wolf encompasses a wide range and that the species is highly able to disperse and extend its range, as evidenced by the forest population of Minnesota expanding into more agricultural portions of Minnesota, Wisconsin, and the Upper Peninsula of Michigan.  These abilities allowed the wolf historically to inhabit the entire Midwest, limited almost entirely because of inadequate prey density and human persecution.  FWS found that short of an unlikely, major shift in current deer management, prey density no longer threatens the “recovered” population levels in the states’ proposed management plans. 
Contrastingly, citing a scientific study, FWS determined a viable wolf population, meaning having less than ten percent chance of extinction over one hundred years, should have at least 175 to 225 individuals.  Most scenarios modeled in the study, including increases in population density, land ownership, and increased road density, allowed a “viable” population to survive in Wisconsin and Michigan.  Based on road density studies, FWS found that human-caused fatalities predominate when road densities are above 1.35 miles per square mile, and thus that almost all “suitable” habitat for the WGL DPS is currently occupied by packs in Minnesota, Wisconsin, and Michigan. 
Though FWS cites studies that indicate pup production in the Lower Peninsula of Michigan and other areas that dispersers appear to frequent, FWS discounts including these areas towards counting as significant suitable habitat for the wolf based on the recovery plan’s requirement of at least 10,000 square miles of contiguous suitable habitat being needed to support a viable isolated population.  Thus, FWS made a significant policy choice that threats to establishment of viable populations in areas outside the current range are not counted by FWS as threats to the significant portion of the wolf’s range and habitat, allowing the first factor to be met based on current occupation and proposed states’ management plans for the future (discussed below).
Any models that predict future threats to a species require policy decisions of the level of risk society is willing to tolerate and for what length of time. Additionally, decisions must be made on what level of uncertainty in the models is protective enough to base delisting decisions on. Here, at least, FWS gave the risk percentage and time frame used. But, not explaining why these numbers were chosen makes these determinations appear as scientifically determined rather than policy decisions.
B. Overutilization for Commercial, Recreational, Scientific, or Educational Purposes
FWS finds that removal of wolves from the wild have been very limited since the species was first listed under the ESA, averaging not more than two individuals per year, even though numbers of illegal kills are not known and exact figures in other areas is not available.  FWS relies on the states’ management plans (discussed in more detail below) as not allowing killing for these purposes for “several years” as sufficient to determine this threat is sufficiently minimized for the foreseeable future.  Though no recreational or commercial harvesting is allowed under the ESA, FWS finds sufficient that the “potential limited commercial and recreational harvest that may occur in the DPS will be regulated by the State.” 
While FWS recognizes that illegal takes by their very nature do not produce reliable data to determine the level of threat under ESA protection, because this threat under the federal protection did not halt population growth these takes are still not a significant future threat, even once that protection is lifted and left to untested state management plans that in some cases even allow takes in general manners. Some uncertainty is inherent in hypothesizing future threats once the federal protection ends, especially for those species that have been protected for decades. However, determining what degree is justified as reliable for delisting the species is a policy decision with reasoning that should be visible to the public.
C. Disease and Predation
Several diseases have had “significant impacts during the recovery of the species . . . [and] [i]f not monitored and control by States, these diseases and parasites, and perhaps others, may threaten gray wolf populations in the future,” however FWS finds that in light of the monitoring for significant outbreaks in states’ proposed management plans, these threats are sufficiently reduced. 
While many diseases are discussed in the proposed delisting, mange holds particular interest in light of recent climate trends. Though normally present in only low levels in the valuable core Minnesota population, recently, in the 1999–2005 period, mange appears to have grown more widespread.  FWS discounts this data as being based on small sample sizes or by hypothesizing observations as possibly being inconsistent over the years.  Scientists have hypothesized mange as being more widespread due to mange-mortality occurring mainly by winter cold, which has not been occurring in recent mild winters, allowing mange to be spread more by these survivors.  FWS also credits higher wolf population and density as being contributing factors to increased mange presence. 
For predation, intra-pack strife and human-caused fatalities are the major threats.  While recognizing the weight of human hunting of wolves, that wolves are commonly mistaken as coyotes outside the core three states and shot, and that the nature of illegal killing of wolves usually involves easily concealable evidence and occurs mainly in remote locations so that reliable estimates cannot be drawn, FWS relies on the states’ proposed management plans as being sufficient to alleviate this threat.  Additionally, FWS relies on the increase in the wolves populations when protected under the ESA despite however many illegal killings are perpetrated as evidence that illegal killings do not pose a substantial continuing threat any longer. 
Similar to the discussion of over-utilization, some uncertainty is inherent in hypothesizing future threats once the federal protection ends, especially for those species that have been protected for decades with only recently developing threats. However, determining what degree is justified as reliable for delisting the species is a policy decision with reasoning that should be visible to the public, especially when those new threats have some evidence of recently increasing. Public perception of such decisions can lead to accountability if not in line with the ESA or what society deems protective enough. While science has conventions of what trends are considered significant, policy decisions do not have to restrict themselves to such valuations. The decision not to value trends based on minimal data in a precautionary fashion is a policy decision of how precautionary FWS is in delisting.
D. Adequacy or Inadequacy of Existing Regulatory Mechanisms
FWS finds that the planned state regulatory mechanisms of Minnesota, Wisconsin, and Michigan are adequate to alleviate any remaining future threats to the WGL DPS. These plans at most require maintenance of the current population for the next five years, while the federal agency is still monitoring the wolves’ status, and in some areas allow immediate harvesting, not only of “problem wolves” in agricultural areas but consider also hunts to regulate population numbers.  What would be considered takes under the ESA are justified under the state plans as still maintaining population numbers above the “viable” levels determined in the recovery plan, and including plans for more restrictive protection if population numbers fall below those determinations.  At least the Michigan plan was still undergoing revisions even when the final rule was made. 
To begin with, as stated earlier, the reliance on legal regulations to ensure continuation of current species levels and threat levels cannot be characterized as scientific determinations. Some amount of risk is inherently involved in changing the legal structure from one that regulates the species on a regional rather than state level, as well. Additionally, when that reliance also involves expectations to alleviate emerging threats, and allows population numbers to vary from their current numbers, even more risk is inherently involved to the species survival. FWS needs to explain the risk over time associated with all these decisions, most significantly when state plans involve some of the very actions that led to the species decline cited in the recovery plan.
E. Other Natural or Manmade Factors Affecting the WGL DPS Continued Existence
The major other factor examined by FWS was public attitude.  Surveys have found strong public support if adverse impacts on recreational activities and livestock producers can be minimized; increasingly strong support, especially in the Upper Peninsula of Michigan, for lethal control of problem wolves (compared to the relocation efforts that first take place under the ESA); and that compensation for wolf depredation on livestock did not positively affect the attitudes of those compensated.  FWS found that continued education efforts in the states’ management plans sufficiently would continue to minimize negative views of wolves. 
Determining what degree of positive public support is enough to not merit consideration as a substantial threat, especially when current solutions and education have not garnered public support in the past major problem sectors like agriculture, is clearly a policy decision without explanation.
Clearly, reliance throughout the entire discussion of the five factors on the future and existing state legal frameworks only furthered masked the policy decisions made. Such reliance cannot be framed as only best science, nor should it have to be. These decisions cannot be hidden from public view, though, or not explained, if the delisting process is to truly uphold the underlying principles of the ESA.
IX. Focus on the Wolf: How to Improve the ESA
The above general description of the wolf recovery plan and delisting barely scratches the surface of all factors involved in the relevant decisions, but already provides numerous examples of phrasing policy decisions as scientific answers. The agency reasoning behind this falsity is clear: Congress has either given no guidance or restricted the agency to solely basing decisions on science, and the courts give extremely strong deference to agency decisions using their scientific expertise.  Meaningful changes to the ESA recovery plan and delisting process would focus on recognizing policy decisions, transparency in providing the reasoning behind these decisions, allowance for greater accountability without the shield of science, and streamlining the process so that policy decisions are consistent whether involving mega fauna or lesser known species.
A. Balancing Flexibility and Required Standards
Potential changes must allow the ESA to keep some of its flexibility in making policy decisions while also hopefully providing, or including a way to determine, starting points and baseline values for all species. This can occur actually in the statute itself, or through regulations. Congress or the agencies could unilaterally make such determinations, or, preferably, government could make a process for asking the public about societal values on the issues touch on throughout the paper. While such a process would certainly include a wide range of opinions, balancing these likely conflicting values into policy decisions that reflect a multitude of interests in an open manner would at least be an improvement over only allowing a limited set of opinions factor in through unknown ways.
B. Transparency, Accountability, and Consistency
While the above paper hints at many possible revisions to the ESA, the overall focus on any future amendment should include three considerations: transparency in admitting policy as the driving force behind ESA decisions and delineating the reasoning behind these decisions, accountability of the agency for making these policy decisions and not hiding behind scientific “mandates,” and consistency in application of the ESA once policy is recognized as a factor in ESA decisions.
Transparency in the recovery plan process is more easily accomplished than in the delisting process because of the restriction by Congress in delisting to strictly best science. Important to actually accomplishing this goal is shifting political attitude towards science from the objective answer-giver; only then will Congress recognize how such restrictions tie the agencies hands and how science has been manipulated through its current view.  While the involved agencies could themselves incorporate these goals in the recovery plan context, because of the lack of specificity in the ESA on what must be involved in the recovery plan process, this same vagueness creates problems in the accountability context.
For accountability, changes should include more structure in the recovery plan process; for the delisting process, structure that recognizes the role of policy decisions must be provided by Congress. Without this structure the transparency behind the policy decisions reasoning merely informs the public and allows them to comment; it provides no remedy to hold the agency accountable to consistent implementation of policies, and could still allow unjustifiable politics to direct decisions. Revealing what are the driving forces behind agency decisions in their discretion is only the first step; creating a process with accountability so that transparency allows reform of the process is the most important step.
Finally, accountability should include a measure of consistency across species so that certain species, like mega fauna, are not protected to a higher degree than less aesthetically pleasing species. Similarly, consistency would likely filter out some political pressure in protecting those species, like wolves, that may have significant opposition to recovery in large ranges. Consistency would also significantly increase efficiency in the long run, though admittedly may take more work to begin with, by creating a system for what a recovery plan should include based on certain factors that adjust for the contours of each species. Important in this development of this process will be recognition of policy decisions made to determine these contours. An example of such a process is the International Union for Conservation of Nature and Natural Resources Red List, which operates on an international scale.  The Red List, for instance, uses a unified system for classifying species at a “high risk for global extinction.” 
C. Examples of Application of These Reform Principles: WGL DPS of the Gray Wolf
The analysis above was replete with examples of policy decisions throughout the recovery plan and the delisting process. Each of these examples should invoke application of the reform principles. This section will highlight one example of how to attempt this reform, so that these processes are consistent with the ESA, and include public involvement to ensure this consistency. The example chosen involves the recovery plan, but considering the extent to which the delisting process is shown in this paper to rely on the recovery plan, it is abundantly clear how important application of these principles is in the planning process as well.
The use of “viable” in the recovery criterion masked a wide array of policy decision while still appearing as a scientific term to the public. This term is a prime example of where application of the principles of transparency, accountability, and consistency could lead to a more public and balanced application of the ESA.
To start, FWS needs to spell out in the recovery plan the policy decisions involved in selecting and defining this term. This includes identifying the amount of years FWS is attempting to have the population survive over, the amount of risk acceptable that the desired results will not be achieved, and the level of scientific certainty that will be deemed acceptable to operate under for both of these factors. This also includes identification of the policy decisions involved in selecting the methods to reach these populations, such as protecting current habitat, actively expanding habitat, what level of scientific knowledge is needed to decide what makes a “suitable” habitat, what other considerations are balanced against the wolves’ needs, the weight that is given to each factor, and other such factors listed above. This also importantly includes an explanation of how, based on these policy decisions, scientific knowledge suggests the objective goals.
Merely identifying policy decisions in citing to scientific studies and final determinations isn’t enough. FWS must then provide the public with clear reasoning behind the policy decisions made, as such explanation increases the agency’s accountability for their own decisions. Such reasoning would explain why X number of years was determined as foreseeable, why X amount of risk was still protection consistent with the ESA, why X amount of scientific uncertainty was enough to act on in the situation. This would also include discussion of why X amount of weight was given to competing factors beyond optimal recovery of the wolf (carrying capacity in all of the historical range and perhaps beyond), and what effect that has on the future of the wolf in terms of risk over time.
FWS needs to explain every policy decision on what acts to take and what acts not to take in protection of the wolf, and reconcile these decisions with the Act’s requirement for recovery of the species. Here, for example, FWS would need to explain why two populations are redundant “enough,” in other words how much risk over how long exists that this criterion will still result in extinction of the wolf and why this amount was selected. FWS should also explain why the population numbers selected for Minnesota as viable represented enough diversity in genetic material, so that the public knows how much risk is involved of depletion of the genetic pool over so many years for the species and why this amount was selected.
Ideally, this identification and explanation would not occur under the current system either. Otherwise, the loose language of the Act in this area begs for inconsistency and arbitrary valuation. While flexibility in implementation is important, it is not important in this manner. Protection of species for X number of years is not dependent upon the life cycle of a species, or the habitat required, or the prey needed. This number is dependent on how long into the future the ESA protects species from extinction once listed. The amount of risk acceptable that the species will not actually survive over those X years is similarly universal. While the amount of individuals, populations, and extent of range will differ from species to species to reach X risk, a set acceptable risk can still be a goal for all species. Similarly, acting on a minimal degree of scientific certainty is either precautionary or less protective of a species depending on what decision is reached, but applies the same to each species.
Optimally, Congress should set objective limits in these areas which the agencies would have to adhere to in making the policy decisions discussed above. Alternatively, the agencies could set these requirements in the regulations themselves. While certain circumstances could make reaching X amount of risk “unfeasible” because of the extreme hardship it would create for humans, or other such exceptions may occur, provisions could also be included to allow outs in specifically limited circumstances. As these would require extensive explanation by the agency on why the limitations could not be adhered to, greater consistency among all species and less inclusion of unwarranted outside influences would infect the recovery plan and delisting processes.
Important revisions to the ESA will focus on transparency, accountability, and consistency in the recovery plan and delisting process decisions, so that agencies can no longer use science as a shield to hide inherently policy-based decisions. This re-framing not only will call for Congress to acknowledge that “best available science” can never be the sole consideration in delisting, but also means the myth of objective science feeding agencies the correct answers must be dispelled, and instead the agencies must own up to the degree they decide society is willing to go in protecting endangered species. Recognizing the true extent of science’s role in such decisions is only the first step, so that the public can understand the driving forces behind decisions and so that the administrative system can properly function in this area. Once policy is recognized as the major factor in ESA decision-making, however, more extensive work by Congress must be undertaken to create a process that limits agency discretion and results instead in agency accountability and consistency.
 Michael A. DiSabatino, Validity, Construction, and Application of the Endangered Species Act , 32 A.L.R. Fed. 332, § 2[a] (2006).
 See generally Ashling P. McAnaney, Remembering the Spirit of the Endangered Species Act: A Case for Narrowing Agency Discretion to Interpret “Significant Portion” of a Species’ Range , Golden Gate Univ. L.R. 431, 435 (Spring 2006) (calling for transparency in determinations of what is a “significant portion” of a species range for listing decisions); Kristin Carden, Bridging the Divide: The Role of Science in Species Conservation Law , Harvard Env. L.R. 165, 194 (2006) (finding that Congress’ “strictly science” mandate forces agencies to make listing decisions in the name of science even though the decisions inherently involve value choices); Eric Helmy, Teeth for a Paper Tiger: Redressing the Deficiencies of the Recovery Provisions of the Endangered Species Act , Env. L. 843, 845 (Fall 2006) (finding the lack of deadlines and too much discretion for recovery plans has fatally limited the ESA’s ability to recover species and fulfill the Act’s goal).
 Erika Kuester, The Proposed Delisting of Canis Lupus Lycaon in the Eastern United States: A Recovery Story in Progress , Univ. of Toledo L.R. 781, 793 (Spring 2006).
 Carden, supra note 2 at 185.
 16 U.S.C.A. § 1533(f)(1); William Lindsley & Tim A. Thomas, Endangered Species Act , AMJUR FISH § 64 (2006).
 16 U.S.C.A. § 1533(f)(1)(A).
 Helmy, supra note 2 at 853.
 Kuester, supra note 3 at 789.
 U.S. Fish & Wildlife Service, Endangered Species Recovery Program, http//www.fws.gov/endangered/pubs/ES%20Recovery%20Program.09.05 (last visited Dec. 29, 2006).
 16 U.S.C.A. § 1533(f)(1)(B).
 50 C.F.R. § 424.11(d).
 Id. at 788–91; 16 U.S.C.A. § 1533(a)(1), (b)(1); 50 C.F.R. § 424.11(c); U.S. Fish & Wildlife, Delisting a Species, http//www.fws.gov/endangered/recovery/delisting [hereinafter Delisting a Species ] (last visited Dec. 29, 2006).
 Kuester, supra note 3 at 791.
 Delisting a Species , supra note 12.
 Helmy, supra note 2 at 855.
 50 C.F.R. § 424.11(d).
 Kuester, supra note 3 at 782–83.
 Id. at 782.
 Id. at 783.
 Proposed Rule, 71 Fed. Reg. 15266, 15269 (Mar. 27, 2006), available at http://www.fws.gov/midwest/wolf/archives/2006pr_dl/index.htm [hereinafter Proposed Rule ].
 Id. at 15270–71.
 Id. at 15269–71.
 Id . The WGL DPS includes the core population of what was previously named the Eastern Distinct Population Segment (EDPS) of the gray wolves. U.S. Fish & Wildlife Service, Questions and Answers: Proposed Removal of the WGL DPS, http//www.fws.gov/midwest/wolf/2006pr_dl/2006pr_dl_qanda [hereinafter Questions and Answers ] (last visited Dec. 29, 2006). In 2003, the FWS proposed delisting the EDPS, however two courts stuck down the validity of this delisting based on encompassing too large a range. Id. The new proposal includes a smaller range for delisting. Id.
 Proposed Rule , supra note 23 at 15274–75.
 Id. at 15274.
 Id. at 15275.
 Id. at 15272. The known dispersers to other states have mostly been discovered through human-caused fatality.
 Id. at 15269–72.
 Id. at 15270.
 Final Rule Removing WGL DPS From the ESA List, 72 Fed. Reg. 6,052 (Feb. 8, 2007).
 16 U.S.C. § 1531(3) (2006).
 16 U.S.C. § 1531(1).
 Carden, supra note 2 at 204.
 Helmy, supra note 2 at 848.
 Id. at 847.
 Carden, supra note 2 at 202.
 Id. at 174.
 Id. at 170.
 Proposed Rule , supra note 23 at 15270–71.
 Carden, supra note 2 at 200.
 Id . at 201.
 Id. at 197–98.
 U.S. Fish and Wildlife Service, 1992, Recovery Plan for the Eastern Timber Wolf , Twin Cities, Minnestoa, p. 1 [hereinafter Recovery Plan ].
 Id. at 2.
 Id. at 4, 24.
 Id. at 4.
 Id. at 4.
 Id. at 24.
 Recovery Plan , supra note 49, at 27.
 Id. at 25–26. An additional definition is given for an isolated second population, but since both Wisconsin and Michigan are within 100 miles I will not go into discussion of the separate numeric goals set in that parameter.
 Id. at 5.
 Id. at 24–25, 4.
 Id. at 4; 26 (finding 100 wolves essential to maintenance of an acceptable level of diversity).
 Id. at 26.
 Recovery Plan , supra note 49, at 26.
 Id. at 25.
 Id. at 24–25.
 Id. at 24.
 Id. at 16, 26.
 Id. at 12-13.
 Recovery Plan , supra note 49, at 4.
 Id. at 29 (emphasis added).
 Id. at 14.
 Id. at 5.
 Id. at 14.
 Id. at 14.
 Recovery Plan , supra note 49, at 20, 29.
 Id. at 29.
 Id. at 17.
 Id. at 5, 17.
 Id. at 18–19.
 Recovery Plan , supra note 49, at 17, 19.
 Id. at 13-15.
 Id. at 15.
 Id. at 15.
 Id. at 16.
 Recovery Plan , supra note 49, at 20.
 Id. at 24.
 Id. at 17.
 Id. at 32.
 Id. at 32.
 Id. at 32.
 Id. at 20.
 Recovery Plan , supra note 49, at 29.
 Id. at 20.
 Id. at 20–21.
 Id. at 21–22.
 Recovery Plan , supra note 49, at 21.
 Carden, supra note 2 at 190.
 Id. at 191.
 Id. at 189.
 Delisting a Species , supra note 12.
 Cf. Carden, supra note 2 at 200.
 Id. at 201.
 See eg. Proposed Rule , supra note 23 at 15302.
 Delisting a Species , supra note 12.
 Proposed Rule , supra note 23 at 15302.
 Id. at 15277.
 Final Rule Removing WGL DPS From the ESA List, 72 Fed. Reg. 6,052–57 (Feb. 8, 2007).
 16 U.S.C. § 1531(3)-(4).
 Proposed Rule , supra note 23 at 15302.
 Id. at 15281.
 Id. at 15278.
 Id. at 15279.
 Id. at 15281.
 Id. at 15283.
 Id. at 15285.
 Id. 15285–86.
 See id.
 Final Rule Removing WGL DPS From the ESA List, 72 Fed. Reg. 6,053–57 (Feb. 8, 2007).
 Proposed Rule , supra note 23 at 15287.
 Final Rule Removing WGL DPS From the ESA List, 72 Fed. Reg. 6,056 (Feb. 8, 2007).
 Proposed Rule , supra note 23 at 15301.
 Carden, supra note 2 at 191.
 See id. at 203.
 International Union for Conservation of Nature and Natural Resources, International Union for Conservation of Nature and Natural Resources Red List, http://www.iucnredlist.org/info/introduction (last visited Jan. 11, 2007).
 International Union for Conservation of Nature and Natural Resources, International Union for Conservation of Nature and Natural Resources Red List, http://www.iucnredlist.org/info/categories_criteria2001 (last visited May 26, 2008). The Red List emphasizes that individual life histories may lead to over- or under-estimation of extinction risk. Nevertheless, such methods are a good starting point for facilitating comparison of common extinction risks across taxa.