Maieusis Through a Gated Membrane: "Getting the Science Right" in ...

Maieusis Through a Gated Membrane:

"Getting the Science Right"

in Public Decisionmaking

Deborah M. Hussey Freeland

I. INTRODUCTION ....................... ............... 375

II. CHANNELS THROUGH BOUNDARIES: PROFESSIONAL

PERFORMANCE AT A SCIENCE-PUBLIC DECISIONMAKING

IN TERFACE ................................................................................... 378

A. Professionalism and the Maieutic Ethic .......................... 3791. The Need for a Properly Permeable Boundary ........ 381

2. Passing as Science: The Maieutic Act of Boundary-Crossing as Performed by a ProfessionalTransm em brane Actor ............................................... 383

B. The Gated Membrane: Characteristics Necessary to aViable Science-Public Decisionmaking Interface ........... 3911. The Physical Basis of the Metaphor .......................... 3922. A Metaphorical Move to a Science-Public

Decisionmaking Interface .......................................... 393III. COMPREHENSIVE EVERGLADES RESTORATION UNDER THE

WATER RESOURCES DEVELOPMENT ACT OF 2000 ....................... 401A. The Everglades Ecosystem ............................................... 403B. The South Florida Economy ............................................ 406

Visiting Scholar, Center for Environmental Science and Policy, Freeman Spogli Institute

for International Studies, Stanford University. J.D. Stanford Law School; Ph.D.Interdepartmental Program in Biophysics, Stanford University. This paper is dedicated tomy parents, Bernard L. and Cora Hussey, who have always been so dedicated to me. I offermy deepest gratitude to them, and to Thomas A. H. Freeland, for their astoundinggenerosity of time and spirit. Many thanks to David L. Faigman, I. Lorraine Heisler,Lawrence Susskind, Barton H. Thompson, Jr. and Robert Weisberg for sharing theirthoughts in helpful discussion.

STANFORD ENVIRONMENTAL LA WJOURNAL

C. Federal Law and Administrative Regulations ................. 4121. The Central and Southern Florida Project:

Section 203 of the Flood Control Act of 1948 .......... 4132. T he R estudy ................................................................ 4153. The Comprehensive Everglades Restoration Plan:

Section 601 of the Water Resources DevelopmentA ct of 2000 .................................................................. 4 16

4. The Programmatic Regulations: 33 C.F.R. § 385 ...... 4225. RE C O V ER ................................................................... 4246. Legal and Ecological Definitions of "Ecosystem

R estoration" ................................. .............................. 427IV. GETTING THE SCIENCE RIGHT: AN INTERFACIAL ANALYSIS OF

ADAPTIVE MANAGEMENT IN PRACTICE ....................................... 431A. Uncertainties: The Need for Adaptive Management ..... 432

1. Characterizing Uncertainties for Management ........ 4322. Strategies for Managing Uncertainties ..................... 434

B. W hat Are the Actors Doing? ............................................ 441C. Direction for a Better Play ............................................... 447

1. To Ensure that Science Actually InformsIm plem entation ......................................................... 447

2. To Improve Project-Level Information Flow fromthe Policy Side to the Science Side ........................... 448

.3. To Provide Incentives for Individual Agencies toParticipate in Collaborative Management ................ 449

4. To Ensure Timely Reauthorization and Funding .... 451V. AN INTERFACIAL ANALYSIS OF SCIENCE-BASED DISPUTE

R ESO LUT IO N ............................................................................... 451

A . Joint Fact-Finding ............................................................. 452B. Joint Fact-Requisition ....................................................... 459

1. M ise-en-scine .............................................................. 4602. A Professional Mediator Performs as a

Transm em brane Actor ............................................... 4633. A G ood Play ................................................................ 464

C. Opportunities to Complement Adaptive ManagementUnder the Plan with Joint Fact-Requisition .................... 4671. The RECOVER Leadership Group ........................... 4682. The System Planning and Operations Team ............ 4683. The Programmatic Regulations ................................. 4694. The § 601(h) (2) Agreement ..................................... 469

[Vol. 26:373

MAIE USIS THROUGH A GA TED MEMBRANE

5. M od W aters ................................................................. 469V I. C O NFLU ENCE .............................................................................. 470

I. INTRODUCTION

The challenges of effectively informing the law with science ofthe highest quality persist and will intensify as more environmentalmatters require public action. This article presents a theoreticalframework for using science as a basis for public decisionmaking.Science enables the public to address pressing common problems,such as the sustainability of the natural resources and ecosystemservices on which our collective health and prosperity depend.When scientific procedures are thought to provide objectivity-maximized information, science also may serve as a commonlanguage, offering terms about which consensus among diversestakeholders-from individuals to nations-may form.

Legal scholars and other commentators have addressedproblems with integrating science into law-related decisionmakingprocesses by mechanism' and by discipline2 . The academic field ofLaw and Science appears to be in an exciting formative era, inwhich debate persists about whether it is an entire endeavor or acollection of questions that are more proper to a variety ofestablished legal fields.' Meanwhile, scientists face the challenges

1. See, e.g., Stephen G. Breyer, Genetic Advances and Legal Institutions, 28J.L. MED. &ETHIcs 23 (2000) (litigation); Wendy E. Wagner, Congress, Science, and Environmental Policy,1999 U. ILL. L. REV. 181 (legislation); SHEILA JASANOFF, THE FIFrH BRANCH (1990)

(regulation); John R. Ehrmann & Barbara L. Stinson, Joint Fact-Finding and the Use ofTechnical Experts, in THE CONSENSUS BUILDING HANDBOOK 375 (Lawrence Susskind et al.eds., 1999) (alternative disputte resolution).

2. See, e.g., David L. Faigman, Making the Law Safe for Science: A Proposed Rule for theAdmission of Expert Testimony, 35 WASHBURN L.J. 401 (1996) (evidence); Sheila Jasanoff, JustEvidence: The Limits of Science in the Legal Process, 34J.L. MED. & ETHICS 328 (2006) (criminallaw); Wendy E. Wagner, The Perils of Relying on Interested Parties to Evaluate Scientific Quality,95 AM. J. PUB. HEALTH S99 (2005) (public health); Rebecca S. Eisenberg, Patents and the

Progress of Science: Exclusive Rights and Experimental Use, 56 U. CHI. L. REV. 1017 (1989)(patent law); Arti K. Rai, Regulating Scientific Research: Intellectual Property Rights and theNorms of Science, 94 NW. U.L. REV. 77 (1999) (patent law); Dagmar Lohan, A Framework forAssessing the Input of Scientific Information into Global Decisionmaking, 17 COLO. J. INT'LENVTL. L. & POL'Y 1 (2006) (international environmental law); Kristin Carden, BridgingThe Divide: The Role of Science in Species Conservation Law, 30 HARV. ENVTL. L. REV. 165(2006) (reviewing literature in ecology law).

3. While the cycles of subject matter coalescence and fragmentation described byRobert P. Merges almost twenty years ago appear to persist in the field of Law and Science,it is emerging as a distinct discipline. See Robert P. Merges, The Nature and Necessity of Law

20071


of protecting the integrity of their professional work andreputations as the call for their participation in publicdecisionmaking becomes ever more urgent.'

So far, no solution to the problems of integrating science intopublic decisionmaking processes has demonstrated clear anduniform success. Rather than compile a bestiary of the variouscontext-specific institutional reforms that have been proposed andcontrast them, this work begins upstream of these developments,elucidating the properties necessary to any effective mechanismfor science-based decisionmaking. Once the rudimentary structureof such a mechanism is understood, we may use the modelstructure to evaluate and design specific institutional proposals.

The presented model draws inspiration from a biophysicalmeans for preserving the integrity of two very sensitive and distinctprocesses while maintaining their constant and effectivecommunication. I employ the metaphor of a membrane betweendifferent cellular environments as an analytical tool for tworeasons. First, the model captures the intricate balance ofcommunication and compartmentalization that is as necessary toscience-based public decisionmaking as it is to the functions of lifeat the cellular level. Thus, this scientific perspective can help usorganize our analysis of legal and political processes in a new anduseful way. Second, while this analysis is intended primarily for alegal audience, it should also be meaningful to scientists, offeringa platform for discussion among legal and scientific professionalsconsidering collaboration.

The interfacial model has two key features: a boundary topreserve the distinct mechanisms of different information-management compartments (a science-producing cell and a

and Science, 38 J. LEGAL EDUC. 315 (1988); Harold P. Green, The Law-Science Interface inPublic Policy Decisionmaking, 51 OHIO ST. L.J. 375 (1990); Steven Goldberg, The CentralDogmas of Law and Science, 36 J. LEGAL EDUC. 371 (1986). The work of Law and Sciencescholars tends to be multidisciplinary not only at the level of combining legal andscientific perspectives, but also of traversing many traditional disciplines within law. See,e.g., DAVID L. FAIGMAN, LABORATORY OFJUSTICE (2004) (constitutional law, history of law);SHEILA JASANOFF, SCIENCE AT THE BAR (1995) (toxic torts, family law, health law,intellectual property, litigation).

4. See, e.g., Donald Kennedy & Richard A. Merrill, Issues in Focus: Science and the Law,ISSUES SC. & TECH. (Summer 2000); Steven Schneider, Mediarology, available at http://stephenschneider.stanford.edu/Mediarology/MediarologyFrameset.html (last visited Apr.16, 2007) (discussing the "double ethical bind" of the scientist trying to report scienceaccurately in its complexity, while trying to communicate effectively with the publicthrough the sound-bite media); see also infra notes 7-8.

[Vol. 26:373

MAEUSIS THROUGH A GA TED MEMBRANE

public-decision-producing cell), and channels devised to passinformation between the compartments in forms that are useful toeither side. These theoretical features can be designed into actualscience-public decisionmaking procedures.

The model engages with the discursive situation of aprofessional at a science-public decisionmaking interface shapedby the ethics of professional performance. The metaphoric role ofa professional engaged in science-based public decisionmakingcarries an implicit and undertheorized breadth, which can beelaborated usefully by adapting terminology and concepts fromdiscussions about performance to explain the professional's dutiesand acts. Tying the theory to its realization in the work of peoplewho implement law and regulations, I explore the flow ofinformation through science-public decisionmaking professionalswithin the model structure.

I use the model to analyze two public decisionmakingprocesses at different levels of generality: a specific governmentalattempt to restore the Florida Everglades ecosystem, and a methodof science-based alternative dispute resolution that is applicable toa variety of environmental matters. These examples were chosen todemonstrate the analytical range of the model, and because bothendeavors are works-in-progress that might immediately benefitfrom recommendations afforded by this analysis. The model isapplicable to science-based public decisionmaking by anymechanism that involves the subject matter of any combination oflegal and scientific disciplines. Decisionmaking processesoperating at scales from the individual to the international aretractable.'

In the following Part, I apply the model to the integration ofscience in law and regulation, identifying the key features of asuccessful strategy for getting the science right in ecosystem

5. For example, Dagmar Lohan has developed criteria for evaluating theinstitutional arrangements for science to inform treaty negotiation and applied them tothe case of the United Nations Framework Convention on Climate Change. DagmarLohan, Assessing the Mechanisms for the Input of Scientific Information into the UNFCCC, 17COLO. J. INT'L ENVrL. L. & POL'Y 249 (2006). While the study provides a useful list ofinstitutional features that decisionmaking bodies should have, the list is incomplete, andLohan acknowledges that an institution can meet the criteria for "optimizing" scientificinput while failing to use science to inform decisionmaking. See id at 308. The gatedmembrane model could be used to deepen the analysis, not only to identify a morecomplete set of features, but also to examine why sets of model features are needed andhow they must interact so that scientists can maintain their professionalism and informour decisions with the highest quality science.

2007]

STANFORD ENVIRONMENTAL LA WJOURNVAL

management. In Part III I review federal water management lawsaffecting the Everglades ecosystem, focusing on theComprehensive Everglades Restoration Plan authorized by theWater Resources Development Act of 20006 as implemented by theSecretary of the Army through Programmatic Regulations. Such ananalysis has not been provided elsewhere. I also describe theseefforts' effects on the region's ecology and economy, and identifythe scope of the planned restoration. In Part LV I characterizeadaptive management pursuant to the Plan as a strategy fordecisionmaking under conditions of significant uncertainty. I usethe model to examine the current state of the developingadministrative structure for implementing the adaptive assessmentand monitoring mechanism authorized by § 601 of WRDA 2000,and I recommend improvements that could be effected throughthe procedure for refining the Programmatic Regulationspromulgated under this law.

I turn in Part V to an alternative dispute resolution context.Here, I use the model to analyze a formalized science-baseddispute resolution process known as 'joint fact-finding," and Ipropose 'joint fact-requisition," a development ofjoint fact-findingthat introduces a structure for maintaining boundaries betweenscientific and dispute-resolution fact-finding processes whilepromoting their cooperation. I then recommend formalconfluences between joint fact-requisition and adaptivemanagement for more effective science-based ecosystemmanagement.

II. CHANNELS THROUGH BOUNDARIES: PROFESSIONAL PERFORMANCE

AT A SCIENCE-PUBLIC DECISIONMAKING INTERFACE

I define "public decisionmaking" broadly as collectivedecisionmaking that affects matters of public concern. Under thisdefinition, legislation, the development of common law,policymaking, methods of alternative dispute resolution, andadministrative management all provide instances of publicdecisionmaking. For purposes of analysis, I distinguish sciencefrom other public decisionmaking processes, even though science

6. Water Resources Development Act of 2000, Pub. L. No. 106-541, 114 Stat. 2572(codified as amended in scattered sections of 16, 33, 42 U.S.C.) [hereinafter WRDA 2000].Title VI of WRDA, §§ 601-02, 114 Stat. at 2680-94, is devoted to "ComprehensiveEverglades Restoration"; Section 601 authorizes the Comprehensive EvergladesRestoration Plan ("the Plan").

[Vol. 26:373

MAIEUSIS THROUGH A GA TED MEMBRANE

can be studied as an institution for deciding what we know aboutcertain matters of public concern, such as the condition of anecosystem. I do this to focus on the problem of preserving theintegrity of empirically driven fact-production processes in theirinteractions with processes designed to prioritize other publicconcerns.

A. Professionalism and the Maieutic Ethic

Ineffective information flow between science and publicdecisionmaking processes persists, despite a mounting need toaddress this longstanding challenge.' Problems arise because thedivergent goals of scientific and public decisionmaking enterpriseseach entail peculiar procedures for fact-finding or production,yielding distinct articles of fact and inculcating conflicting sensesof duty in their respective practitioners. Further, distinctionsbetween the scientific and public-decisionmaking classes of fact,discussed infra Part II.A.2.a, are sometimes unappreciated orexploited, particularly with respect to their uncertainties.

Scientists are increasingly aware of the need for theirparticipation in public decisionmaking, but they are unsure ofexactly how to respond without compromising their research orcredibility! Legal scholars recognize the difficulty of using science

7. See, e.g., AM. ASS'N FOR THE ADVANCEMENT OF SCIENCE, ANNUAL REPORT 2003 10

(2003), available at http://www.aaas.org/publications/annual-report/2003/aaasannual_report2003.pdf ("Helping society achieve sustainable development is one of the greatscientific challenges of the 21st century. Progress will require . .. improving society'scapacity to generate and use knowledge for decision-making."); NAT'L RESEARCHCOUNCIL, THE SCIENCE OF RESTORING THE EVERGLADES 8-9 (2003), available at

http://dels.nas.edu/dels/rpt_briefs/everglades.pdf (noting the persistence of "asubstantial gap in both synthesis and integration" of science in project planning sincerestoration efforts began, and calling for the development of a "multidisciplinaryrestoration-wide mechanism" to address this problem); The National Academies Website,Strengthening Science-Based Decision Making in Developing Countries (last visited Apr. 11, 2007),http://www7.nationalacademies.org/sustainabilityroundtable/Type-l-Homepage.html.

8. See, e.g., David E. Blockstein, Integrated Science for Ecosystem Management: AnAchievable Imperative, 13J. SOC'Y CONSERVATION BIOLOGY 682, 684 (1999) ("As much as wemay feel ill-prepared or uncomfortable in new roles . . ., it is not sufficient for us ... simplyto be researchers and educators; we must work with managers..., and we must educatepoliticians about science needs and opportunities .... [T]he stakes re high and theconsequences of inaction are unacceptable."); see alsoJane Lubchenco, Entering the Centuryof the Environment: A New Social Contract for Science, 279 SCIENCE 491 (1998). The 2007Annual Meeting of the American Association for the Advancement of Science, "Science

and Technology for Sustainable Well-Being," included a dozen symposia on"Communicating Science," as well as dozens more on science and policy for specific topics(e.g., climate change, energy, fisheries, agriculture and other land uses). AM. ASS'N FOR

2007]


to inform legal decisions. As a result, bodies of scholarship havecontrasted the goals and working procedures of scientific and legalinstitutions in some detail.9 These literatures reflect the view thatour collective decisions would benefit from better scientificinformation, and describe some of the institutional and politicalproblems that arise in trying to use science to inform ourdecisions. I offer a new approach to the problems of science-basedcollective decisionmaking by beginning from the perspective of aprofessional engaged in its processes, mapping the situation ofsuch professionals in a larger decisionmaking structure, andrevealing how the structure mustbe aligned to allow theseprofessionals to perform effectively and with integrity.

To begin thinking in terms of the model, we might compare ascience-public decisionmaking professional to a gated channel in amembrane between two cells: when closed, the gated channelforms part of the boundary between the science and publicdecisionmaking cells; when open, the gated channel orchestratesthe flow of information between the cells. When the gated channelrepresents a professional scientist engaged in publicdecisionmaking, the closed state represents the scientist's responseto her ethical obligation to protect the integrity of scientific work.Thus, the boundary feature of the model reflects this responsibilityto her profession. The open state represents, for example, thescientist's conforming to her social responsibility to inform thepublic of her findings, overcoming the scientist's reluctance toappear to engage in political decisionmaking. When the scientistfunctions as an open gated channel, she ensures that properlyformed information flows between the cells.

In examining the roles of science-and-public decisionmakingprofessionals, I propose "maieutic ethic" as a term of art to refer toa particular principle that characterizes professionalperformance."0 The term "maieutic" derives from midwifery anddescribes a method of assisting one to realize that which is latentin the mind (as does the Socratic method). I use the term"maieutic" to describe the intermediary function served by a

THE ADVANCEMENT OF SCIENCE, PROGRAM BOOK, 2007 AAAS ANNUAL MEETING, SCIENCE

AND TECHNOLOGY FOR SUSTAINABLE WELL-BEING, 15-19 FEBRUARY 2007, SAN FRANcIsCo

(2007) (on file with the author).

9. See, e.g., references supra note 3.

10. For this discussion, a "professional" denotes one who is not only an expert in hisfield, but whose work also meets its ethical standards; thus, "professionalism" entailshaving and heeding one's sense of duty to perform one's job according to those ethics.

[Vol. 26:373

MAJEUSIS THROUGH A GA TED MEMBRANE

person playing the role of the gated channel in the model. I use"ethic" to denote a principle to which an act must conform to beappropriate to a given discourse. Thus, the maieutic ethic requiresthe gated channel to perform at least three distinct functionsproperly: to prevent improperly formed information from crossingthe boundary, to act on the information from one side of theboundary so that it may pass to the other side in an active form,and to push part of itself aside to allow information to cross theboundary with high fidelity.

Though the two responsibilities of protecting the integrity ofthe scientific fact-finding process and involving the public inscience-based decisionmaking might appear to be in tension, theiralignment is possible in a properly designed discursive structure.The proper performance of the gated channel (speaking in termsof the procedural architecture) -which may also be called the"transmembrane actor" (speaking in terms of the role played by anactual person)-has aspects of guardianship or border-patrol tomaintain the boundary, as well as translation or ambassadorshipfor boundary-crossing.

1. The need for a properly permeable boundary.

A science-public decisionmaking interface.that functions like agated membrane creates synergistic benefits: passing scientificresearch results produced with minimal stakeholder influence topublic decisionmaking professionals enables them to make betterdecisions, while developing scientific research questions based oninformation from the public decisionmaking arena allows scienceprofessionals to be more responsive to society's needs. Creativeenterprises attempt to use science as a seed crystal around whichstakeholders with apparently conflicting interests may align inproblem-solving and in the prospective management ofenvironmental or natural resource matters; particularly promisingefforts seek to involve all concerned parties in engaging scientificinformation early in a planning process. For example, the Planexplicitly relies "on the best available science . . . developedthrough an inclusive and open process that engages allstakeholders [.]"

11. U.S. ARMY CORPS OF ENG'RS (JACKSONVILLE DIST.) & S. FLA. WATER MGMT. DIST.,

FINAL INTEGRATED FEASIBILITY REPORT AND PROGRAMMATIC ENVTL. IMPACT STATEMENT 9-

1 (1999), available at http://www.evergladesplan.org/pub/restudy-eis.aspx#mainreport[hereinafter PROGRAMMATIC EIS] (Page numbers in the PROGRAMMATIC EIS are of the

3812007]


Despite the widely felt need to inform our legal and politicalpractices with our best science, formidable institutional barriersobstruct the emergence of a clear, generally applicable strategy fordoing so. In addition to problems ranging from lack ofinstitutional competence to intransigent parties or spoilers, aprofessional reluctance also operates at the level of individualscientists that can be understood in terms of the maieutic ethic.Attempts at opening scientific processes to political participationraise scruples for professional scientists, who need to preservescientific integrity while communicating with publicdecisionmakers (e.g., policymakers). In advanced education orprofessionalization as a scientist, one is socialized to perform fact-finding procedures in particular ways, and to view their results asparticular kinds of objects. As I discuss elsewhere, the scientificmethod, the practice of ensuring that results are reproducible, andthe peer review process are all designed to wash the subjectiveperspective of the scientist out of the process of science, so that inthe ideal limit, only nature remains to speak for itself.1 2 Thoughobjectivity is an unattainable scientific ideal, the effort tostrengthen objectivity-claims by adhering to certain proceduresshapes professional scientific behavior. The data resulting fromthose procedures are known to be subject to interpretation;acknowledgment of their provisionality and peculiar, attendantuncertainties is a hallmark of scientific rigor. Professional scientistsstrive to prevent political bias from creeping into their work, oreven from appearing to do so, because such bias undermines theobjectivity-claiming processes of their scientific research as well astheir scientific reputations. 3 A structured engagement process canavoid the appearance of impropriety by promoting the public

form: Section-page, so that "9-1" represents page 1 of Section 9.). The PROGRAMMATIC EIScontains the Plan at Section 9.

12. Deborah M. Hussey, What Are the Social Responsibilities of Scientists?, InauguralAddress for the U.S. Geological Survey Center for Science Policy (Oct. 19, 1999), availableat http://www.earthscape.org/pl/hud01/hud01.html (access may be obtained by visitinghttp://www.earthscape.org and following the link to "free trial").

13. Sheila Jasanoff describes the need for "drawing seemingly sharp boundariesbetween science and policy" as being "crucial to the political acceptability of [scientific]advice" to regulatory agencies. SHEILA JAsANoFF, THE FIFTH BRANCH 236 (1990). Theconcept of boundary work that she draws from the sociology of science literature mapsneady into the gated membrane analysis, which elucidates the structural and functionalfeatures necessary to such boundaries. See id. at 14 (citing Thomas F. Gieryn, Boundary-Work and the Demarcation of Science from Nonscience: Strains and Interests in the ProfessionalIdeologies of Scientists, 48 AM. SOCIOLOGICAL REv. 781 (1983)).

[Vol. 26:373


accountability of science-public decisionmaking professionals, sothat the public may develop trust in the decisionmaking process.This structured process can ensure that public decisionmakingand management properly use sound, world-class science.

2. Passing as science: The maieutic act of boundary-crossing asperformed by a professional transmembrane actor.

A maieutic ethic develops in those who see nature asobjectively real and devote themselves to studying it. This sense ofduty impels scientists to do their best to interrogate nature withoutdistorting it, and to articulate the data with which nature respondswith minimal interference from their own prejudices. This duty tobracket one's political values in one's work is the maieutic ethic.The maieutic ethic of the scientist requires her to serve as aconduit for nature's responses to scientific inquiries. While shewould be disingenuous to represent herself personally as thoughshe lacked political inclinations or allies, representation of herselfin a professional role signals to the audience that she is not aboutto speak as a private person, but as a representative of somethingelse.14 Were a professional actor named "Ian Holm" to portrayHamlet, he would be understood not to be lying to his audience bysaying, "This is I, Hamlet the Dane, ' 15 but to be representing thetruth of the character portrayed. Similarly, a scientist speakingprofessionally is expected to report faithfully the responses of thenatural world. Thus, her integrity as a scientist, as well as thequality of her science, depends upon the robust operation of themaieutic ethic; her reputation as a scientist, and the authorityattending a great reputation, depend upon the legiblemanifestation of this ethic in her scientific, performance.

a. Facts.

Scientific endeavors may be said to prioritize factual accuracyunder a correspondence theory of truth, while law, policy and

14. As when a professional actor speaks for a represented character, maieutic ethicsgovern professional performance in theatre and other fields. For example, a judge mustallow the law to work through him while withholding personal bias to the extent possible,and an attorney in litigation must speak for his client and also as an officer the court.Maieusis may occur even when the ideas that are to be communicated are one's own, aswhen a professional fiction writer strives to place his pen in the service of a character, or anonfiction writer to diarize faithfully.

15. WILLIAM SHAKESPEARE, HAMLET, act V, sc. 2.

2007]


administration emphasize dispute management, settlement andprevention. These divergent aims of scientific processes on the onehand and legal and political processes on the other result indifferences in the properties of bits of information called, "facts,"established by the operation of each endeavor. Being open-ended,scientific inquiry generates facts that are expected to beprovisional, having an understood and acknowledged potential forrevision as other facts are developed. Rigorous scientists treatscientific facts as placeholders in a sense, marking the best of ourknowledge at a moment, since our knowledge is understood to beever evolving. A responsible scientist details the provisionality ofher findings, reporting all she knows about its inherent limitationsand uncertainties, which often suggest new questions for furtherresearch.

Dissimilarly, public decisionmaking-which may also beunderstood as dispute management, resolution and prevention-does not strive to develop knowledge for its own sake. Instead,factual discovery serves a process that seeks relatively rapid closureand durability. Thus, in a legal or political process, the mostvaluable aspect of a fact is its ability to promote dispute settlementor consensus. One may hope that this ability will arise from thefact's correspondence with an objective reality, but a consensus-promotion ability may inhere in the fact even if it were poorlydeveloped, false, or readily falsifiable in scientific terms. One whoresolves disputes in which facts arise, and one who seeks to preventdisputes by establishing policies based on a factual body, hopesthat his management structure will continue to work undisturbed.

Factual sedimentation occurs in science as well as in commonlaw, legislation and administration, in that older facts precipitateand subsequent facts are built upon them, so that the older factsare more firmly pressed in place. Nonetheless, science is not onlyopen to stirring things up, but also constantly prods at itsfoundations to see whether they will remain durable, since it is theduty of a scientist to do her best to remove distortions from ourviews of nature. In contrast, since dispute management processesbuild social order, dispute management institutions tend todisfavor the disruption of sedimented facts. If a sedimenteddispute-resolution fact were later falsified, rather thanexperiencing a scientist's thrill at learning more about nature, thedispute management professional would feel a duty to weigh thevalue to social order offered by correcting the structure against

[Vol. 26:373


that of protecting the established order.

b. Acts.

Metaphors of staged performance are useful in discussing theoperation of the maieutic ethic. For this analysis, the term"discourse" refers to that which influences and captures theexpectations of actors and their audiences regarding what islegible as a proper performance. I use "discourse" synonymouslywith "discursive structure," to evoke the image of a scaffold or netthat both binds and gives form to an actor's performance, and maybe worked by an actor to produce effects along the weft. I identifyethics, laws and professional norms as warp lines in the discursivenet, since these shape the expectations under which performanceoccurs.

Science and law meet as discourses in the performance of ascience-public decisionmaking professional. In terms of the gatedmembrane model, acts within the science cell are encapsulated bya membranous net of scientific ethics, laws and professionalnorms-a scientific discursive structure. Similarly, in the case ofEverglades restoration, the public decisionmaking cell is a policycell, encapsulated by a membranous discursive structure of law andregulatory policy; acts within the policy cell are bound up in thenet of legal ethics, codes, regulations and common law, as well aslegal and administrative professional norms. The science-policyprofessional is caught in and can work both nets, acting in thespace defined by the convergence of both structures.

In a given discourse, an act may or may not bear recognizablesignificance, i.e., it might not be legible, nor might it matter. To"matter" is to "materialize, and to mean."'6 An act that matters in adiscourse is like a signal that is rendered recognizable by thecontext of that discursive structure, while one that does not matterfails to materialize or to convey meaning, like stochastic noise orinsignificant silence that is not captured in the discursive net. Forexample, "ninety-point-eight percent chance" may be read as"equal to 0.908" by one audience, as "too likely to ignore" byanother, and as gibberish that goes unnoticed by a third."Maieusis," then, involves a form of translation and re-presentation-not simply the replacement of one jargon with

16. This useful definition of "matter" is drawn from Butler's discussion ofintelligibility. JUDITH BUTLER, BODIES THAT MATTER 32 (1993).

2007]


another, but the expression of material captured in one discursivenet into a new realm, in a form that will not be lost through gapsand breaches in the realm's delimiting discursive net, but thatinstead will materialize properly, be meaningful, and matter.

"Performance," or action, occurs within at least one (and oftenseveral simultaneous) discursive structure (s). An "act" is somethingthat a performer does that matters or is legible to an audience, orhas effects within a discursive structure. Acts constituteperformance. Performance pertinently comprises "speech acts," aterm used to denote utterances that do not only conveyinformation but matter in other ways as well.17 I generalize theterm "speech acts" to include written acts. For example, §601 (a) (5) of WRDA 2000 does not merely inform a reader that"South Florida ecosystem" denotes "the area consisting of the landand water within the boundary of the South Florida WaterManagement District in effect on July 1, 1999" and "includes ...the Everglades . . . the Florida Keys; and ... the contiguous near-shore coastal water of South Florida[,]" but also causes "SouthFlorida ecosystem" to have legal significance, or to matter in legaldiscourse, only as this area consisting of land and water, orderingthat it be so. The result of this speech act is that the only aspects ofthe Everglades ecosystem that manifest under the discursivestructure in which WRDA 2000 matters are the land and water inthe management district, although other aspects of the ecosystem(such as its interrelated species) matter in a scientific discourse.

i. Calling forth the actor.

Analogous to the provisionality of the scientific fact discussedsupra is the provisionality of an actor's identity. To explain thelatter provisionality, I use the term "interpellation" to refer to aniterative, bidirectional interaction between an actor and anaudience through a discursive structure, wherein their collectiveexpectations regarding a proper performance call them to act andattend, to speak and heed, to write and read, in particular ways.

17. See, e.g., JOHN L. AUSTIN, Performative Utterances, in PHILOSOPHICAL PAPERS (3ded., 1979);JOHN R. SEARLE, SPEECH ACTS (1969).

18. LOuis ALTHUSSER, Ideology and Ideological State Apparatuses, in LENIN ANDPHILOSOPHY AND OTHER ESSAYS BY LOUIS ALTHUSSER 127, 170-177 (trans. B. Brewster)

(1971). While Althusser developed the term "interpellation" as a tool for analyzingideology and the subject, I draw from a very limited sense of the term and redefine it fordiscussion of the role(s) of the transmembrane actor.

[Vol. 26:373


When an actor at a science-policy interface decides to speak as ascientist, her speech will employ a universe of terminology andforms, i.e., a code, that matters in a scientific discourse; herdecision to speak in this code may be informed by the audiencewhom she expects to address. Her performance as a scientist maybe said to be evoked by the situation in which her content is wellcaptured by the discursive structure and meets the expectations ofher audience; meanwhile, by choosing to speak in a scientific code,she catches the attention of an audience that is listening forscientific speech. I use "interpellation" to refer to both theevocation of a scientific performance from the actor, and to theactor's hailing the attention of an audience that reads science.

Similarly, an actor at a science-policy interface may decide tospeak as a manager, in a code that matters in a discourse ofadministration. For example, in written form, her interpellation ofan audience may involve her choice of writing style, herpresentation strategy for the document, her choice of its primaryrecipient, and her composition of its cover letter. Because bothscientific and administrative audiences may be present for theactor's performance, the actor may work the respective discursivestructures that interpellate these audiences iteratively, by code-switching, as a bilingual speaker may combine Spanish and Englishwords to convey her meaning. A subset of terms and forms may belegible (though with different implications) in both scientific andadministrative discourses, so that the same actor may attempt tospeak under both discursive structures simultaneously. Code-switching or superposition then involves her interpellating andbeing interpellated by audiences within at least two discursivestructures, the scientific and the administrative.

A professional. at a nexus of science and policy emerges intodistinct discursive structures simultaneously. This professional isinterpellated as a scientist to limit her speech to that of conveyingthe well supported findings of a scientific investigation in arigorous or circumscribed way, informing the audience of anyattendant uncertainties of which she is aware. As a scienceprofessional, she must practice a maieutic ethic as she speaks, akind of hygiene that requires her to strive for transparency, tobracket her personal opinions to the extent possible so that naturecan speak through her.

As the same professional operates within a managementstructure, the principles shaping her speech pull different

2007]

STAAORD ENVIRONMENTAL LA WJOURNAL

elements to the fore. A scientist employed by a governmentagency, for example, feels the interpellation of her public role tolimit her speech as a policymaker to that of performing accordingto her agency's mission. As a government employee, she mustpractice a maieutic ethic, mapping her scientific opinion onto apolicy question while serving as a channel for her agency's speech,rather than unwittingly presenting her own scientific opinion asthough it were her agency's position. The maieutic ethic of thepolicy professional is a hygiene analogous to, but differentlydirected from, that of the scientist; this representational ethicrequires her to strive for transparency, bracketing her personalopinions to the extent possible so that her agency can speakthrough her.

Divergent interpellations can be problematic for the science-policy professional, who must set aside her ego in two differentdirections in order to do her job. Exercising the performative skillof bracketing and framing one's thoughts in various ways forvarious audiences while communicating those thoughts as honestlyand completely as possible-i.e., with integrity at all times-is not atypical part of one's education or socialization as a scientist.Scientists who enter government service or other dispute-management roles may find themselves at odds-doubly bound ina now unfamiliar ethical net-and having to learn on the job howto negotiate the distinct lines of responsibility in performance.Sadly, a scientist in this bind may find some wiggle room in the(mis)framing and treatment of uncertainty, as discussed infra PartII.A.2.b.ii.

An administrative agency expected to perform in inconsistentways may respond to various expectations with different modes ofperformance, possibly offering rhetoric in satisfaction of one set ofresponsibilities, and other acts (such as more rhetoric or resourceallocation) in response to other interpellations. Thus, on aninstitutional level, divergent interpellations may evokeorganizational hypocrisy. 9

19. Nils Brunsson has used the term "organizational hypocrisy" to describe thebehavior of a political organization in its efforts to be recognized as legitimate. See NILSBRUNSSON, THE ORGANIZATION OF HYPocRisY 27-31 (2d ed., 2002). That is, theorganization may behave inconsistently on the whole, in its efforts to matter in differentdiscourses.

[Vol. 26:373


ii. A good performance conforms to the maieutic ethic.

The actor's legibility as a scientist-her audience's ability toidentify her as a professional scientific actor-depends upon herability to perform acts that matter in a scientific discourse. Heridentity as an administrator is similarly provisional. Whileparticular acts can be legible in both discourses, some subset ofthem will be accurate in only one. For example, a statement that astatistical uncertainty inheres in a class of scientific data may beaccurate under a scientific discursive structure, but in anadministrative context may be legible as a general and inaccuratestatement that the data are not reliable at all. This stretching of astatement's meaning onto two discursive frames affords an actoran opportunity to perform in bad faith, or to misrepresent hertext: an actor may hail an audience in scientific terms but thenattempt to act administratively, misconstruing data and then eitherdrawing authority from or devaluing it to support a managementdecision that she favors a priori, regardless of the strength of itsscientific basis. The actor's misrepresentation of her speech as thatof a professional scientist, or misinterpellation of her audience,fails to conform to the maieutic ethic.

In rhetorical performance, scientists and publicdecisionmakers cite or make calls to authority in an effort to gainacceptance and influence for their arguments. Argument occurs ina discursive framework with criteria for persuasiveness: forexample, those arguing in scientific terms cite sedimented science,while those persuading in legal terms cite legal precedent, andthose arguing in other terms appeal to what they think matters inother discourses. Each of these citations is a call to authority in theappropriate discursive structure, for instance, to the authority ofempiricism, of established and publicly recorded decisions, ofreligious texts, or of ostensible common sense.

In dispute resolution and in science, authority is oftenassociated with objectivity. Thus, citation to empirical data or tomodeling results generated by a computer using mathematicalalgorithms can function as a call to authority that bolsters anargument in either discourse, to the extent that the data or resultsare taken by the audience to represent objective information.Miscitation is the act of citing to an authority inaccurately, eitheraccidentally or with the intent to deceive. For example, deceptivemiscitation occurs when one cites to data in support of one'sargument, while realizing that the data do not provide strong

20071


support without selective interpretation or other distortion. Likemisinterpellation, the actor's misrepresentation, or miscitation, isan act that does not conform to the maieutic ethic.

In addition to an actor's misinterpellation of an audience,miscitation to another discursive structural element, and personalor organizational hypocrisy, failure to act in accordance with themaieutic ethic may take the form of inaction or postponement ofaction when action is appropriate, as occurs when a scientist has asocial responsibility to report research results but fails to do so, orwhen an administrator has a social responsibility to use science toinform a management decision, but fails to do so. These failuresare instances of irresponsibility, or failure to respond in a way thatmatters to the interpellative hail of the discursive structure towhich one's professional ethics are integral. Thus, uncertaintypresents a moral hazard, in that it may be cited in an attempt toexcuse irresponsibilities in performance.

The science-policy professional best engages the two (or more)discourses in which she is enmeshed when political fact-findingstructures can be aligned with scientific fact-finding structures. Forexample, the legal authority of § 601 of WRDA 2000 provides forscience to function as a source of authority in management. 2

1

Thus, in Everglades restoration under the Plan, the rigors ofscientific discourse should be prior to those of political discourse.Indeed, introducing adaptive management is a way of requiringmanagement to become more provisional in order to incorporatescience more effectively. In terms of the gated membrane model,this suggests the use of science to inform political decisionmakingby setting bounds within which its outcomes must fall-scientistsare not to be present as additional stakeholders on the publicdecisionmaking side of the model boundary, but as professionalswho inform, or shape, decisionmaking by setting these limits.

A failure to use science to inform the realm of permissibleoutcomes counts as a failure to engage in science-based publicdecisionmaking. Such a failure can be understood as acommunicative failure, whether arising from the science or publicdecisionmaking cell. Scientists hesitate to take or to appear to takea political position, yet their research results will likely supportsome courses of political action over others. Indeed, a scientist'sfailure to speak about potential environmental responses to

20. WRDA 2000, supra note 6, § 601 (h) (4) (A) (iii) (VII).

[Vol. 26:373


proposed courses of action to avoid "taking a side" would beanalogous to a physician's failure to render a diagnosis andprescribe a course of action to promote his patient's health.2

Taking responsibility for scientific research may entail makingrecommendations that clearly favor a stakeholder's position, sothat it would be irresponsible of a scientist to withhold or spininformation to avoid favoritism ex post. A public decisionmaker'sfailure to allow science to materialize properly in thedecisionmaking process would be analogous to a patient's failureto follow his physician's orders even if his health or life were indanger. Whether an individual decisionmaker has such a rightover himself is a distinct ethical question; the publicdecisionmaker, at least, must honor the public trust with which heis charged. When it is clear to the dual professional that it is herresponsibility to her agency to give a complete and accuratescientific answer to a query from the policy side, her speech canbecome coherent: the same speech can satisfy her maieutic roleswith respect to nature and with respect to government service.

At the scale of an individual, irresponsible acts harm thespeaker's personal and professional reputation. Although it isquite possible to bias one's scientific investigation andinterpretation to favor one's political interests, it is also possible todo much to prevent one's biases from distorting scientific work.However, precisely because it is quite possible to bias one's scientificinvestigation and interpretation to favor one's political interests, itis also imperative to take the greatest possible care to insulatescientific investigation and interpretation from political bias, sothat all political interests can obtain the fullest appreciation for theeffects of their decisions on the matter at hand. Thus, in additionto the scientific method itself, science-based administrationrequires a carefully orchestrated science-policy interface thatmaximizes the integrity of the resulting science and the efficiencyof communication between scientists and managers.

B. The Gated Membrane: Characteristics Necessary to a Viable Science-Public Decisionmaking Interface

Therefore, producing the best science and the best public

21. For example, the National Academy of Sciences has recognized the field of"sustainability science," which concerns "devis[ing] practical protections for the earth'skey life-support systems," including ecosystem services. William C. Clark, SustainabilityScience: A Room of Its Own, 104 PROCEEDINGS NAT'L ACAD. Sci. 1737, 1737 (2007).

2007]


decisions requires reliable separation between scientific and publicdecisionmaking arenas, as well as thoughtfully channeled,minimally distorting interaction between them. The requisitestructure is analogous to that of the membrane that surrounds acell within a larger organism, protecting the composition andfunction of the science-producing organs of the body politic.

1. The physical basis of the metaphor.

Though actual cell membranes are fascinating and complex,only the aspects sketched here and in Figure 1 are needed toillustrate the essential features of a workable science-publicdecisionmaking interface. Generally, water is the most abundantmaterial within and around the biological cell. Cells are boundedby lipid membranes that insulate one aqueous compartment fromanother, so that the solutions on either side can maintain distinctchemical properties. Specialized molecules form gated channels,pumps and other mechanisms for controlled transport throughthe membrane that protects the cell's internal biochemicalmachinery from the extracellular environment. Thus, for example,a hormone can bind to a protein embedded in the membrane, sothat the membrane protein's configuration changes to form atunnel that allows only ions of limited charge and size to pass; thetunnel closes when the hormone detaches."

22. See, e.g., GERALD KARP, CELL AND MOLECULAR BIOLOGY 122-82 (3d ed., 2002);STRUCTURAL BIOLOGY OF MEMBRANE PROTEINS (Reinhard Grisshammer & Susan K.Buchanan, eds., 2006).

[Vol. 26:373


Cell A

orael

orae i

GATED MEMBRANECell B

oorganell

Figure 1

Figure 1 is a diagram of a cell membrane, showing the physical

features forming the basis of the conceptual model of a properly

working science-public decisionmaking interface.

For simplicity, imagine two contiguous cartoon cells sharing asingle membrane, as depicted in Figure 1. Each cell contains achemically controlled medium and organelles that performspecific functions necessary to the cell's life. Flow of materialsthrough the structured, precisely regulated, insulating membraneallows the cells to communicate. Crucially, there are gatedchannels that, in response to a properly formed signal, can bind toparticular kinds of material in a cell and transport those materialsacross the membrane, while preventing other materials frompassing. Theoretically, the same factor could accept input fromeither side of the membrane. This membrane will represent aninstitutional structure that insulates the processes on its scientificand public decisionmaking sides from each other, while offeringspecific protocols for their interaction.

2. A metaphorical move to a science-public decisionmaking interface.

An example of a metaphorical move to a science-public

2007]


decisionmaking interface (pertinent to the discussion of large-scale ecosystem management pursuant to federal law infra Parts III-IV) is illustrated in Figure 2. This figure does not show the actualstate of affairs in Everglades restoration pursuant to federal law,which is diagrammed in Figure 4 and discussed infra Part IV.Rather, Figure 2 illustrates the key components in adaptivemanagement pursuant to WRDA 2000 in an architecture thatwould optimize their effectiveness in achieving the goals of thePlan. That is, Figure 2 represents the stage for effectiveperformance under the Plan, clarifying the roles of key actors.

Figure 2 (on the following page) shows a conceptual model of arecommended science-public decisionmaking interface for the case ofEverglades restoration pursuant to WRDA 2000 (discussed infra),showing the analogous functions of individual professionals andtheir agencies to those of the smaller-scale biological featuresdescribed in Figure 1.

a. A properly permeable boundary.

On the left side of the membrane is the realm of scientists.This realm is subject to professional ethics and other discursivestructural features influencing actors' and audiences'understandings of what counts as good science and how it isproduced. Analogs to the cellular organelles on the science sideinclude the science-performing entities within federal agencies,such as the United States Geological Survey, the United States Fishand Wildlife Service, and the National Park Service, and thosewithin state agencies, such as the Florida Department ofEnvironmental Protection and the South Florida Water

[Vol. 26:373

2007] MAlE USIS THROUGH A GA TED MEMBRANE 395

cn3jn3aq'

Q9 a

CD En

Enb

STANFORD ENVIRONMENTAL LAWJOURNAL [Vol. 26:373

Management District. 3 Notably, the particular science cell withwhich we are concerned contains a multi-agency science-producing organelle: the Restoration, Coordination andVerification (RECOVER) Group provided by the ProgrammaticRegulations.24 The RECOVER Group builds coherent,multidisciplinary scientific opinions in the form of technicalreports for use by policy actors outside the science cell. Crucialfactors in this cell are members of the RECOVER LeadershipGroup (RLG), who ensure that the produced science is in the bestpossible condition, and who preserve its integrity as it is developedand expressed from the science cell into the policy arena.5

The public decisionmaking side of the membrane is the realmof stakeholders. Various entities, represented for this case asorganelles within a policy cell, express their interests and interactwith each other in processes of conflict and negotiation. Ideally,the policy side would have a highly functional consensus-buildingorganelle analogous to the science side's RECOVER Group. TheRECOVER Group analog would facilitate negotiation processesand develop lists of properly framed scientific questions from thejustly harmonized interests of all of the stakeholders26 for delivery

23. Just as organelles have outer membranes and occasionally their own internalcompartments, each of these agencies can be modeled as containing its own, internalscience-policy interfaces. This article remains focused at the level of the Restoration,Coordination and Verification (RECOVER) Group established by the sponsoring agenciesas provided by the Programmatic Regulations for the Comprehensive EvergladesRestoration Plan, 68 Fed. Reg. 64,200 (Nov. 12, 2003) (codified at 33 C.F.R. pt. 385)[hereinafter Programmatic Regulations], which is most relevant to the implementation of§ 601 of WRDA 2000, supra note 6, § 601.

24. See 33 C.F.R. § 385.20 (Westlaw 2007).25. Further elaboration of the model is straightforward, in that various kinds of

scientific, political, or multiple-role actors (such as personnel, documents or evenparticular speech acts) can be loosely analogized to demonstrated or hypothetical cellularcomponents. For example, § 601 of WRDA 2000 may be framed metaphorically as agenetic code directing the expression of particular membrane proteins; 33 C.F.R. § 385may be analogous to the RNA transcribed from the code and translated to form theproteins; the RECOVER Group can be analogized to a mitochondrion within the cell,providing it with energy (here, in the form of scientific information), and members of theRECOVER Leadership Group can act as proteins translated from 33 C.F.R. § 385.20 thatassociate with the science organelle to bind information, insert themselves into themembrane to chaperone RECOVER science through the membrane to the policyorganelles, receive information from the policy arena, ensure that such signals arepackaged as proper scientific questions, and transmit those questions back to the science-producing organelle. Though various kinds of molecules have been shown to have someof these abilities, the accuracy of the biochemical details of the model is not necessary toits conceptual applicability.

26. For example, stakeholders in Everglades restoration include: individual citizens;


to the science cell. Because the scientific questions can function ascalls to authority that may reallocate the power of persuasionamong stakeholders, a transmembrane actor (such as a member ofthe RECOVER Leadership Group) must ensure that the sciencequestions address most fully the stakeholders' concerns and areframed properly for application of the scientific method. If thequestions are fit for processing by the science organelles, thetransmembrane actor ushers them through the membrane intothe science cell.

A suitably empowered RLG member could serve both sides as amembrane-crossing factor; however, I do not recommend this, forthe RLG member has duties tying her primarily to the science sideof the interface. Thus, requiring her to be responsible for trulyneutral coordination among stakeholders on the policy side wouldlikely compromise her ability to perform her boundary function. Ifscientific research results clearly favor the interests of a subset ofstakeholders, she would be required either to distort the science toavoid favoring particular stakeholders, or to fail to function as aneutral among stakeholders in preserving the integrity of thescience.

In both cells science crystallizes consensus. Though the role ofscience in consensus-building has been questioned, at least twopertinent observations support its applicability to consensus-building. First, the best science is itself produced and marked by aconsensus-building process. To be published in a peer-reviewedjournal and become part of the sedimenting scientific literature,scientific research results are reviewed by competing professionalsfor soundness; thus, the publication process is itself a consensus-forming process. Then, to gain wider acceptance and influence,scientific research results must be reproduced, usually in theprocess of their extension: as more science is built upon them,they sediment into harder facts. The process of extension andsedimentation is also a consensus-building process. If consensushad not formed about a point of science, it would not matter asscience at all: science materializes by crystallizing consensus aboutitself.

In addition, WRDA 2000 is an example of a law that explicitlyplaces science into the consensus-promotion function by using it as

federal, state, and local agencies; the Seminole Tribe of Florida; the Miccosukee Tribe ofIndians of Florida; urban utilities and agricultural, environmental and recreationalinterest groups. Programmatic Regulations, supra note 23, at 64,201.

20071


a basis for the adaptive management of the ecosystem. Citations toscience have been successful in promoting consensus, as can benoted in the formalization of joint fact-finding as a disputeresolution process, discussed infra Part V. In effect, science canfunction as an oracle, in that its association with objectivity allowsits citation to bestow authority upon an argument. Indeed, theoracular function of citation to science may be so strong thatconsensus among disputing stakeholders can form around theresults of a computational model-even when the uncertaintiesand other limitations on the model's accuracy are appreciated.This is possible because, at a minimum, the model is perceived tobe a relatively objective source of information.2" Objectivityoperates as an instance of neutrality, in that information or actsperceived to be unprejudiced in favor of any particularstakeholder can provide bases for negotiated agreement. Thus, themodel functions as a neutral, serving as a basis for agreement evenif it might be incorrect. Accordingly, a fact generated by ascientific model may have a dispute-resolution value that is greaterthan its scientific value. This is not to say that fresh, model-generated information is necessarily inaccurate, but that it has yetto sediment into fact in a scientific discourse. Modeling is but asingle source of scientific information; modeling alone will notteach us about an ecosystem, even though modeling providesinformation that cannot be obtained by any other means.Conceptual theories and biological and ecological measurementsare needed to complement modeling, and settle its results into adiscursive context.

b. A transmembrane actor.

A person performing as a transmembrane actor must haveprofessional properties that enable her to perform distinct roleswithin and on a side of the membrane: she may bind properlyformatted scientific results and transport them to the policy side,and she may accept properly formed scientific questions andtransmit them to the science side. Like the hormone-activatedgated channel described above, when the transmembrane actorreceives a properly configured signal, she allows proper forms of

27. Telephone interview with Dr. I. Lorraine Heisler, Supervisory Fish and WildlifeBiologist, U.S. Fish and Wildlife Service, and Member, RECOVER Leadership Group (May10, 2006) (notes on file with author) [hereinafter Heisler Interview].

[Vol. 26:373


information to cross the boundary. This actor effectively mediatesa dialogue between the cells, enabling science to progress andenabling the policy cell to perform its decisionmaking functions.In contrast, an actor who works only within a science-producingorganelle can restrict her focus to scientific performance,spending minimal energy developing expertise in communicationwith policy actors.

Thus, a transmembrane actor on the science side mustperform as a scientist, reporting her own investigations andperforming a quality-control service that entails reviewing others'science and managing disputes among scientists in the consensus-formation process of scientific research and peer review. Herscientific goal is to discern how the ecosystem will respond tochanges in experimental conditions. This is the work of maieusis:for this role she identifies herself as a scientist, alerting heraudience to expect the tentative language of the scientificprofessional as she does her best to speak for the ecosystem. Shemust avoid performing "advocacy science," which is working theoracle by having committed to a favored outcome a priori, thencrafting and interpreting science to conform to the desiredoutcome-a professional scientist is ethically bound to practiceempiricism, rather than fundamentalism.

Communicating from the science side to the policy side, thetransmembrane actor must perform another maieutic act inexpressing science into the policy arena. This performanceinvolves a separate set of skills from those involved in scientificperformance, and should be marked by a clear signal of a differentcharacter (such as a change of costume, "putting on a differenthat"). The actor now identifies herself as a dispute managementadvisor to alert her audiences that she will speak under a differentset of discursive rules and assumptions, addressing uncertainty in away that minimizes its vulnerability to rhetorical exploitation.

In contact with the policy side, the transmembrane actor's roleis that of a teacher, responding to stakeholders' questions aboutwhat they can do to meet their various needs while contributing tothe restoration of the ecosystem subtending their economy. Shealso instructs those who are not science professionals in what kindsof questions scientific research can and cannot answer, explaininghow science can be used to address their questions and whatuncertainties remain beyond the scope of research. Her goal is toarticulate the best-supported interpretations of the science and its

2007]


attendant uncertainties in a form that will matter, or be legible tothe policy actors.

Working in the policy realm, a management professional whoparticipates in a multi-agency collaborative group plays at least tworoles: that of her home agency and that of the consensus ofagencies. When speaking for her agency, she must distinguish herown opinions from what she believes her agency's position to be.Similarly, when articulating a multi-agency consensus position, shemust be careful to express the consensus view rather than only herown or her home agency's positions. Though ultimately an actorcannot control her audience's interpretation of her performance,the actor bears the responsibility of representing her role to theaudience faithfully. Misrepresentation of her role constitutesunethical, unprofessional performance.

Communicating from the policy side to the science side, atransmembrane actor based on the policy side translates politicalconcerns into scientific questions. The goal of his performance isto map these concerns into scientific discourse so that researchfunds and personnel are allocated to answer them on the scienceside, while accommodating the interests of all of the stakeholdershe represents. The work of a transmembrane actor from thepublic-decisionmaking cell is examined in greater detail infra PartV.

Watching an actor perform various roles can be problematic,especially when different roles are performed through the samespeech acts, when role-changes are poorly signaled, and whenactors are not well trained. Professionalism for a science-policyactor demands awareness that she serves different functions andspeaks for different purposes in different situations. The science-policy actor must be able and willing to fulfill her duty to identifythese roles accurately as she performs them. One who is cast inscience-policy roles and is not aware of their multiplicity, who lacksformal training in managing her distinct duties, who has littletalent for speaking in multiple registers (often simultaneously), orwho lacks the integrity to avoid misrepresenting her ownperformance endangers the coherence of the science-policysystem. If the flow of question-and-answer through the membraneis obstructed, if science is distorted in the crossing, or if themembrane is breached, political decisionmakers' access to the bestof our knowledge is compromised. The integrity of the science cellis eroded, and we are left with an emulsion of politicized science

[Vol. 26:373


that clouds our view of nature and leads to ill-informed, untenablepolicies. The maieutic ethic calls for the transparency of thescience-policy professional, so that managers can teach scientists inpublic service what kinds of research would be most helpful intheir decisionmaking, and so that information from thescientifically understood system (such as an ecosystem) can reachthe public through a lens that is as clean and undistorting aspossible.

Though they are optimized for disparate purposes, science andpublic decisionmaking are not incommensurable: information canbe developed and configured to facilitate their cohesion. Whileprocedurally, law may be to science as oil is to water, both arefluid, and their interface can be designed carefully to allowbeneficial exchange while preserving their respective characters.Law and administrative regulations can provide for viable science-based public decisionmaking by establishing a structured science-policy interface, and by instituting intermediation procedures thatpreserve the integrity of their respective processes. Analogous to acell membrane with gated channels and pumps, this carefullydesigned interface allows the transparent coordination of scientificand political processes, preventing their degeneration into amurky, opaque emulsion.

III. COMPREHENSIVE EVERGLADES RESTORATION UNDER THE WATER

RESOURCES DEVELOPMENT ACT OF 2000

The world is watching and learning from our collectiveefforts to save the unique Everglades ecosystem for thepeople of Florida and the nation.28

The Comprehensive Everglades Restoration Plan under WRDA2000 is a unique and ambitious environmental law designed torestore a vast and complex ecosystem.2 ' The Plan's goal is to "get[]the water right"3 -to engineer the reallocation of over a billiongallons of freshwater per day1 to restore the severely damaged

28. U.S. ARMY CoRPs OF ENG'RS (JACKSONVILLE DIST.) & S. FLA. WATER MGMT DIST.,

CENTRAL AND SOUTHERN FLORIDA PROJECT COMPREHENSIVE EVERGLADES RESTORATIONPLAN 2005 REPORT TO CONGRESS xviii (2005).

29. WRDA 2000, supra note 6, § 601(b) (1) (A). For the Comprehensive EvergladesRestoration Plan, see Section 9: Recommended Comprehensive Plan, in PROGRAMMATIC EIS,supra note 11.

30. PROGRAMMATIC EIS, supra note 11, at 1-3.31. COMM. ON ENvr. & PUBLIC WORKS, WRDA 2000 REPORT OF THE COMMITTEE ON

2007]


Everglades, while sustaining the agricultural and urban economiesfor which Everglades water initially had been redirected.Tremendous uncertainty inheres in the management of entwinedhydrological and ecological systems to sustain genetically robustpopulations of diverse species in a naturally nutrient-poorenvironment, particularly when its periods of inundation vary overan 18,000-square-mile span. 2 Rather than allow these uncertaintiesto overwhelm Congressional efforts to recover the ecosystem,33

WRDA 2000's legislators commissioned an adaptive assessmentand monitoring program to engage with unforeseen circumstancesand new information arising as the Plan is implemented. Essentialto the Plan's success, this adaptive management program isintended to be a mechanism for developing an increasinglycomprehensive scientific understanding of the ecosystem'shydrological needs, and for using emerging science to improve thePlan's performance. As the multiple government agencies involveddetermine how to effect WRDA 2000's adaptive assessment andmonitoring strategy, other agencies and project teams are eager tolearn how they might similarly address large-scale ecosystemrestoration elsewhere, as are scholars and nongovernmentalorganizations around the world.

Section 601 of WRDA 2000 authorizes and orders theimplementation of the Plan. 4 The Plan is the most promisinglegislative effort yet to restore the Everglades, an ecosystem ofinternational and paramount regional importance that hassuffered extensive losses of biomass and function under the watermanagement regimes of the last century. This Part examines theeffects on the ecosystem of federal water management efforts,addressing the meaning of "restoration" under federal law. Thistreatment necessarily considers the major functions of theEverglades ecosystem, linking these to the economy of the state ofFlorida. These discussions clarify what is presently at hazard andwhat Title VI of WRDA 2000 seeks to restore.

ENVIRONMENT AND PUBLIC WoRKs To ACCOMPANY S. 2796, S. REP. No. 106-362, at 35 (2dSess. 2000), available at http://thomas.loc.gov/cgi-bin/cpquery/z?cp106:sr362: (followlink for "PDF") [hereinafter WRDA 2000 SENATE REPORT].

32. Id. at 41; PROGRAMMATIC EIS, supra note 11, at 2-2.33. PROGRAMMATIC EIS, supra note 11, at xi ("[I]t is acknowledged that all the

answers cannot be known at this time, and that inaction is not an option[.]").34. WRDA 2000, supra note 6, § 601(b) (1) (A).

[Vol. 26:373


A. The Everglades Ecosystem

While the land that is now south Florida lay below sea levelduring interglacial periods, thick, porous limestone sedimented,upon which the Everglades would eventually form. People arethought to have settled in the Florida peninsula between 12,000and 13,000 years ago, and rainfall patterns and species distinctiveof the Everglades ecosystem began to appear roughly 5,000 yearsago. A map of south Florida dated 1823 records the termcontemporaneously used to describe the sweeping open space ofits "Extensive Inundated Region, covered with Pine and HummockIslands, of all sizes," as, "the EVER GLADES." 6 Miccosukee peopleinhabiting the region called it, "Pa Hay Okee," translated as "grassywater," and journalist Marjory Stoneman Douglas attracted publicattention to its remarkable nature in her book, The Everglades: Riverof Grass.

37

The Everglades is a marsh, or grassy wetland, spotted with tree-bearing areas such as tree islands and hammocks of tropicalhardwoods."3 The climate of the Everglades is near-tropical, and itsrainfall is periodized on several timescales: there are cycles of wetand dry years, three-fourths of its average annual rainfall of 50 to60 inches occurs between May and October, and afternoonthunderstorms and occasional tropical weather events (such ashurricanes) lasting from hours to days sweep through with varyingfrequency.39 The Everglades retains rainfall in an extendedwatershed comprising the 4,000-square-mile Kissimmee Riversystem, which feeds the 730-square-mile, shallow Lake Okeechobeeto the south (the "Lake"), which historically overflowed furthersouthward into about 4,700 square miles of Everglades marshes. °

35. Previous two sentences informed by THOMAS E. LODGE, THE EVERGLADESHANDBOOK 11 (2d ed., 2005).

36. Henry S. Tanner, Map of Jorida in A NEW AMERICAN ATLAS CONTAINING MAPS OFTHE SEVERAL STATES OF THE NORTH AMERICAN UNION, PROJECTED AND DRAWN ON A

UNIFORM SCALE FROM DOCUMENTS FOUND IN THE PUBLIC OFFICES OF THE UNITED STATESAND STATE GOVERNMENTS, AND OTHER ORIGINAL AND AUTHENTIC INFORMATION(Philadelphia 1823), available at http://www.davidrumsey.com/insightredirector/insightredirector.asp?cid=8&iia=0&igDavid%20Rumsey%2OCollection&isl=0&gwisp=0%7CList%5FNo%7CList%2ONo%7Cl%7C5388.024%7C2&gwia=3&gc=O. In favored usage,"Everglades" is singular. LODGE, supra note 35, at 13.

37. H.R. DOC. No. 80-643, at 17 (1948); LODGE, supra note 35, at 13 (citing MARJORYSTONEMAN DOUGLAS, THE EVERGLADES: RIVER OF GRASS (1947)).

38. LODGE, supra note 35, at 15-17.39. Id.40. Id. at 18, 105; CENTRAL AND SOUTHERN FLORIDA PROJECT COMPREHENSIVE

2007]

STANFORD ENVIRONMENTAL LAWJOURNAL

The entire hydrologic system is very flat, having a gradient of onefoot per ten miles; thus, the unaltered flow of water advanced as asheet, filtering southward at rates ranging from virtuallyimperceptible to two feet per minute.4' Though at its highest theAtlantic Coastal Ridge is just over twenty feet above sea level, thisridge contributes to the impoundment of water in the marshes.Rivers through the ridge and the coastal areas of the marshesdelivered fresh water to estuaries and bays.42

Regional rainwater distribution varies spatially as well astemporally, and both modes of hydrologic variation supportheterogeneous habitats and webs of intricately interdependentspecies. Subregions of the wetlands differ in their hydroperiod, orthe duration for which land stands inundated, from permanentlyflooded to rarely flooded. Topographically, the Everglades is amosaic of sawgrass ridges that thrive in an environment of low-nutrient water, water-lily sloughs that anchor oxidation-prone peat,and tree islands that also accumulate peat and provide habitats-particularly, nesting and spawning sites-for vertebrate species.43

During wet periods the periphyton mat of algae, diatoms andother minute flora and fauna flourishes, forming the foundationof the food web; insects and other invertebrates that feed directlyon the periphyton abound. As water levels fall during drydown,this primary food production is concentrated in alligator holes andother depressions. These now-shallow pools provide refuge foraquatic species through the dry season, but also concentrate them,rendering them more accessible to higher-order predators, such aswading birds and alligators. In the 1930s regional hydrologicalpatterns could support supercolonies of up to 200,000 wadingbirds, which need to "read"44 or respond to rainfall patterns over

REVIEW STUDY (THE RESTUDY) FACTS & STATISTICS (1999), available at

http://www.evergladesplan.org/docs/facts.pdf [hereinafter RESTUDY FACTS]. The U.S.Army Corps of Engineers and the South Florida Water Management District, whichsponsor the Plan, host its official website at http://www.evergladesplan.org. The factsheetsavailable at this official website excerpt some useful information from the Restudy.

41. RESTUDY FACTS, supra note 40, at 2 (noting the gradient); LODGE, supra note 35,at 21 (noting the flow rate).

42. LODGE, supra note 35, at 20-21.43. Id. at 20.44. See PROGRAMMATIC EIS, supra note 11, at xi ("[I]n a healthy ecosystem ... water

patterns in one part of the system could be used to predict the patterns throughout thesystem. Animals living in the Everglades would 'read' the water patterns, and 'know' whereto go to find the food and water that they needed for successful reproduction and survivalunder a range of natural conditions .... [T]he combination of connectivity and space ...

[Vol. 26:373


large areas to find prey and nesting sites.43 Where the River ofGrass meets the sea, estuaries support species that mature at seabut return to lower-salinity environments to spawn. The influx offreshwater to Florida and Biscayne Bays has enabled commerciallyimportant marine and estuarine species to thrive.46

The Everglades has functioned as a capacious reservoir andfiltration system for the plentiful but punctuated rainfall of southFlorida. This freshwater reservoir has also sustained an ecosystemof remarkable population at multiple trophic levels, includinglarge populations of vertebrates that have been stable only throughtheir access to the variety of feeding and nesting possibilitiesafforded by their ability to forage over an expansive,interconnected geographic area."

Since the U.S. Army Corps of Engineers (the "Corps ofEngineers") constructed the elaborate plumbing system of theCentral and Southern Florida (C&SF) Project discussed infra PartIII.C.1, agriculture and urban population have mushroomed inand near the Everglades. As demands for reliable water suppliesand flood control have grown, the health of the ecosystem hasdiminished: slow, meandering rivers that retained water have beenchannelized; water levels in the central Lake are so low that half ofits area and the wetlands formerly at its banks have been lost; waterreleased from the Lake has damaged estuaries to the east and west;changes in the salinity of downstream bays have interfered withlocal species' habitats and breeding; soils exposed by unusually lowwater levels have burned or blown away; half of the Everglades nolonger exists, and its compromised water quality has resulted inchanges in floral species and harm to fauna. These changes areunderstood to result from the C&SF Project, the workings of whichare deemed unsustainable.4"

As a result, the Everglades is a "resource in peril, having beenseverely affected by human activities for over a hundred years."49 Itswading bird population has dropped to one tenth of its historical

created the range of habitats needed for the diversity of plants and animals."); see alsoLODGE, supra note 35, at 234 ("[T]he birds needed to follow long-lasting drying frontsthrough the Everglades on a predictable day-to-day basis.").

45. LODGE, supra note 35, at 198.

46. Id. at 156. The pink shrimp is an example of a commercially important species.Id. at 135.

47. PROGRAMMATIc EIS, supra note 11, at 5-32.

48. See PROGRAMMATIc EIS, supra note 11, at 3-1 to 3-2.49. Programmatic Regulations, supra note 23, at 64,200.

2007]


size, commercially significant bay and estuarine fish populationsare declining, and more than 1.5 million acres of the Evergladesare beset with exotic species that displace those characteristic of itspre-drainage ecosystem.5" While the restored ecosystem will lackthe expansiveness of the pre-drainage Everglades, Congress' goal isto restore principal hydrological and ecological features within theremaining area of the historic ecosystem, as elaborated infra PartIII.C.3-6.

B. The South Florida EconomyThe Governor's Commission for a Sustainable South Florida,

formed "to assure that a healthy Everglades ecosystem can coexist[with] and be mutually supportive of a sustainable South Floridaeconomy" has identified agriculture as one of three industriescritical to the region." Abundant rainfall and productive, organicsoils have supported prolific agricultural industries in southFlorida, which generated roughly $3.8 billion per year by a 1999estimate. 5

' Because the heavy rainfall of south Florida is highlyseasonal, the region's agricultural efforts require water storage andflood control works. For example, over the past century the fertilesoil immediately south of Lake Okeechobee has promptedagricultural enterprises to assume the challenges of watermanagement to venture crop production, and the development offlood control and railroads has promoted urban settlement on theAtlantic Coastal Ridge." The primary obstacle to depending uponswamped land for agriculture or urban settlement is water: watermust be drained from the land; more water must be preventedfrom flooding the land once it has been drained, and water mustbe stored in provision for droughts.

The Plan describes Lake Okeechobee as the Everglades' "liquid

50. RESTUDY FACTS, supra note 40, at 2.51. GOVERNOR'S COMM'N FOR A SUSTAINABLE S. FLA., The Governor's Commission for a

Sustainable South Florida, in THE INITIAL REPORT OF THE GOVERNOR'S COMMISSION FOR ASUSTAINABLE SOUTH FLORIDA (1995), available at http://fcn.state.fl.us/everglades/gcssf/initial/rpt~cont.html (follow links to relevant sections) [hereinafter GOVERNOR'SCOMMISSION REPORT]. The other two critical industries are international trade andtourism, which are here considered urban activities (since ecotourism depends on theurban infrastructure for transportation, lodging and support services).

52. PROGRAMMATIC EIS, supra note 11, at E-57.53. See MICHAEL GRUNWALD, THE SwAMP (2006) (offering a journalistic account of

the modernization of southern Florida).

[Vol. 26:373


heart."54 Historically, the Lake's water level would rise during thewet season and overflow its banks. This massive pulse of freshwaterslowly percolated southward, nourishing the ecosystem as it seepedacross south Florida into Biscayne and Florida Bays. If the Lake isthe heart of the hydroecological system, then the land immediatelysouth of the Lake is the system's aorta: obstructing southward flowfrom the Lake deprives the rest of the ecosystem of needed water,and tainting water in this region distributes toxins and excessivenutrients throughout the system.

This vital 700,000-acre region is now the EvergladesAgricultural Area, which is flood-controlled by a levee along theLake, by canals contributing to the diversion of 1.7 billion gallonsof water per day 5 from percolation through the Everglades toquick drainage through numerous canals, and by additional leveesbetween the Everglades Agricultural Area and the WaterConservation Areas to the south. 56 Pumps operate to irrigate theEverglades Agricultural Area with Lake water when the agriculturalarea is too dry for agriculture, and to remove water from theagricultural area back into the Lake and the conservation areaswhen the agricultural area is too wet.57 The latter "backpumped"water carries nutrients that have contributed to the eutrophicationof Lake Okeechobee and to the spread of cattails in the WaterConservation Areas.5 s Like algal blooms on the Lake, cattails thusserve as signals of deviation from the naturally phosphorous-limited conditions necessary to the Everglades ecosystem.

The rich soils that have been converted into agriculturalproducts during the last century had been deposited graduallyover millennia, and their rapid subsidence is not beingcounteracted. Agriculture has consumed so much soil in theEverglades Agricultural Area that the land has subsided severalfeet, so that if the area were to be inundated, it would now form alake rather than a riverbed.59 Most of the Everglades Agricultural

54. PROGRAMMATIC EIS, supra note 11, at 5-27. The report on the C&SF Project

authorized by the Flood Control Act of 1948 also noted that the Lake "and its controls arethe heart of any plan for flood control and water conservation in south Florida." H.R.Doc. No. 80-643, at 40 (1948).

55. RESTUDY FAcTs, supra note 40, at 2.

56. PROGRAMMATIC EIS, supra note 11, atJ-146,J-149 and J-156.57. Id.58. LODGE, supra note 35, at 230-31.

59. PROGRAMMATIC EIS, supra note 11, at 6-37: ("In the Everglades Agricultural Area,

soil loss has diminished the higher ground elevations that maintain the hydraulic head

2007]

STANFORD ENVIRONMENTAL LA WJOURNAL [Vol. 26:373

Area produces sugarcane,' ° though sugarcane growers cultivaterice, row vegetables and sod as supplemental crops.6 As the sugarindustry considers withdrawal from the area,6 2 additionalStormwater Treatment Areas are planned, and proposals for theremaining area have included the development of suburbanhousing.63 As Figure 3 suggests, its location renders watertreatment and management for the Everglades Agricultural Areacritical to Everglades restoration.'

[that] drives water south. In some areas more than eight feet of organic soil was lost by1984 .... Some areas in the southern Everglades Agricultural Area, where shallow soilsoverlie bedrock, already have less than two feet of soil remaining. As soils subside, themovement of stormwater out of the area requires increased pumping.").

60. Some discrepancy appears as to the exact size of the Everglades Agricultural Areaand the percentage of its land devoted to sugar production, even within the Plan;nonetheless, sugarcane is clearly its major crop. PROGRAMMATIC EIS, supra note 11, at J-146 ("Approximately 77 percent of the 1,122-square-mile EAA produces sugarcane (thepredominant crop) .... ); id. at E-58 ("Ninety percent of the roughly 593,000 acres of theEAA produce sugarcane."); LODGE, supra note 35, at 225 (stating that about 65 percent ofthe 700,000-acre Everglades Agricultural Area produces sugarcane). The U.S. Departmentof Agriculture (USDA) has valued Florida's 14,281,000-ton sugarcane production in 2004at $432,714,000 and its 12,746,000-ton production in 2005 at $356,888,000; productionrose to 14,488,000 tons in 2006. USDA National Agricultural Statistics Service,http://www.nass.usda.gov/StatisticsbyState/Florida/index.asp#.html (follow relevantlinks) (data on file with the author).

61. PROGRAMMATIc EIS, supra note 11, at E-60; see also id. atJ-146.62. See, e.g., Neil Santaniello, Florida Sugar Grower Refuses to Leave Land Slated for

Everglades Restoration, S. FLA. SUN-SENTINEL (Mar. 31, 2004), available athttp://www.everglades.org/033104a.html; Press Release, Neil Santaniello, Sun-Sentinel.com, U.S. Sugar Agrees to Leave 18,000 Acres in Everglades, (Apr. 30, 2004),available at http://www.flmnh.ufl.edu/fish/southflorida/news/ussugar2004.html;Suzanne W. Stuart, Coalition Hails U.S. Sugar Move: Lobbying Helped Grower Decide to VacateLand for Glades Project, S. FLA. SUN-SENTINEL May 1, 2004 at 10.B, available athttp://pqasb.pqarchiver.com/sunsentinel/access/626174211 .html?dids=626174211:626174211 &FMT=ABS&FMTS=ABS:FT&date=May+1 %2C+2004&author=Suzanne+Wentley+Stuart+News&pub=South+Florida+Sun++Sentinel&edition=&startpage= 10.B&desc=COALITION+HAILS+U.S.+SUGAR+MOVE+LOBBYING+HELPED+GROWER+DECIDE+TO+VACATE+LAND+FOR+%27GLADES+PROJECT.

63. PROGRAMMATIC EIS, supra note 11, at 4-8; see also Introduction, in GOVERNOR'SCOMMISSION REPORT, supra note 51 ("Even if the water management and water qualityproblems can be perfectly solved, the remaining Everglades will still fail if they areconsumed by suburban sprawl. Current land use trends are not sustainable and harshmeasures must be undertaken to curtail the suburbanization of what is left of SouthFlorida.").

64. LODGE, supra note 35, at 227.


Figure 3 shows historic (re-drainage), current and Plan flow ofwater through the Everglades. The Everglades Agricultural Arealies between Lake Okeechobee and the remaining Everglades to thesouth, and is represented by a gap in the Plan flow of water fromthe Lake to the rest of the Everglades.65

Water management for agriculture and urban life also affectsecology and human economies in other parts of the Everglades.For example, the human population of south Florida hasincreased from about 900,000 when construction of the C&SFProject began in 1950, to more than 5,500,000 in 1995, and isexpected to reach over 11,500,000 by 2050.66 As the Evergladesecosystem degenerates, recreational land uses that depend onecosystem health, such as fishing, hunting and ecotourism, declinein value. Thus, south Florida's projected population growth isexpected to exacerbate the loss in value of its recreationaleconomy.

67

Estuaries, in which freshwater from a land source meets thesaltwater tide, are the breeding grounds for numerous species thatprovide capital to the commercial fish and shellfish industries."Estuarine species are harmed by contaminants in freshwater inputs(such as nitrogen and phosphorous from fertilizers and exposed

65. For an animated version of this diagram, see Flowmaps,http://www.evergladesplan.org/education/flowmaps.html (last visited Apr. 7, 2007).

66. PROGRAMMATIC EIS, supra note 11, at 5-19.67. Id. at 4-36.68. Id. at 5-27.

2007]


muck soils), by hypersalinity due to insufficient freshwater influx,and by mistimed freshwater flows that are not coordinated withtheir breeding seasons. 69

Biodiversity diminishes as Everglades species become extinctand endangered. Not only do dwindling populations disappearfrom the gene pool, but their ecological niches become vacant aswell.7" The functions of missing species and genes may be revealedby their absence, or ever may remain unknown; valuation of thesefunctions is therefore especially problematic.7'

These examples may be taken to illustrate services provided bythe Everglades ecosystem. In addition to ecosystem services such asthose noted in the environmental impact statement for the Plan(climate, water and waste management, soil and food production,recreation and genetic resources),72 the Everglades has providedaesthetic inspiration to many who have encountered it-despite itshordes of mosquitoes, sharp sedges, poisonous snakes, alligators,crocodiles, and other sources of danger and discomfort. Just asworks of art and antiquities may be considered priceless nationalor cultural treasures, so are aspects of the Everglades. In the wordsof President Theodore Roosevelt:

Birds should be saved because of utilitarian reasons; and,moreover, they should be saved because of reasons unconnectedwith any return in dollars and cents.... The extermination ofthe passenger-pigeon meant that mankind was just so muchpoorer; exactly as in the case of the destruction of the cathedralat Rheims. And to lose the chance to see . . . a file of pelicanswinging their way homeward across the crimson afterglow of thesunset . . . - why, the loss is like the loss of a gallery of themasterpieces of the artists of old time.73

Accordingly, Everglades National Park has been designated aBiosphere Reserve within the Man and the Biosphere program ofthe United Nations Educational, Scientific and CulturalOrganization (UNESCO)," included in the List of Wetlands of

69. Id.

70. Id.71. Id.

72. PROGRAMMATIC EIS, supra note 11, at 5-21 to 5-28.73. THEODORE ROOSEVELT, A BOOK-LOVER'S HOLIDAYS IN THE OPEN 316-17 (Charles

Scribner's Sons 1916).74. UNESCO Website, UNESCO-MAB Biosphere Reserves Directory, Everglades &

[Vol. 26:373


International Importance under the 1971 Ramsar Convention onWetlands, for which UNESCO serves as Depositary, 5 and inscribedon the List of World Heritage in Danger pursuant to Article 11 ofthe 1972 UNESCO Convention Concerning the Protection of theWorld Cultural and Natural Heritage. v The Everglades is the onlyplace in the world bearing all three distinctions. 77

Services provided by the Everglades ecosystem areindispensable to the south Florida economy. Federal, tribal, stateand local governments and private organizations recognize thatthe economy is embedded in the ecosystem, and that theeco(n)system is not sustainable under the current managementregime.78 Valuation of ecological resources can be problematic, notonly because of uncertainty as to how to monetize a given aspect ofthe natural system, but also because of the potential forincommensurability between an ecological value and goods morereadily represented in computational economic models. Thus,under the policy of the Corps of Engineers, which serves as thefederal sponsor of Plan implementation, standard cost-benefit

Dry Tortugas, http://www2.unesco.org/mab/br/brdir/directory/biores.asp?code=USA+09&mode=all (last visited Apr. 14, 2007).

75. See Wetlands International Website, Ramsar Sites Information Service,http://www.wetlands.org/rsis/ (follow the "Search the Ramsar Sites Database" hyperlink,then search the "Wetlands International Site No." box for "4US005") (last visited Apr. 14,2007).

76. UNESCO World Heritage Centre - Everglades National Park,http://whc.unesco.org/en/list/76 (last visited Apr. 7, 2007).

77. UNESCO Website, List of Biosphere Reserves Which Are Wholly or Partially WorldHeritage Sites and Ramsar Wetlands, http://www.unesco.org/mab/BRs/brs whc ramsar.shtml (last visited Apr. 14, 2007).

78. See, e.g., PROGRAMMATIC EIS, supra note 11, at 5-28 ("[T]he health of the southFlorida economy is closely linked to the health of all of theseinterconnected/interdependent ecosystems."); id. at E-15 ("The economic system is

connected with the natural ecosystem and in general is ultimately dependent upon it forsurvival."); id at Summary ii to iii ("The Everglades has molded the regional character ofcentral and southern Florida and sustains the economic and cultural growth of the region.• ..These natural systems will not recover their defining characteristics under currentconditions and will not be sustained into the future."); Introduction, in GOVERNOR'SCOMMISSION REPORT, supra note 51 ("[I]t is easy to see that our present course in SouthFlorida is not sustainable. The inextricable link between the human community and thenatural system is obvious. The natural system is the basis for our public health, safety,recreation, welfare, and aesthetic activities. 'Many who live in South Florida do not realizethe benefits they receive continuously from a functioning natural ecosystem and whatecosystem collapse would mean to them.' The current trends cannot continue. Time is ofthe essence. If we are to curtail the deterioration and evade further catastrophe, urgentstrategic action is needed." (internal citation omitted)).

2007]


analyses are not performed for ecosystem restoration projects.79

WRDA 2000 allows the Secretary of the Army to determine that alegally authorized ecosystem restoration, preservation orprotection activity is justified by its environmental benefits to theecosystem; then, as long as the Secretary also finds that the activityis cost-effective, he need not provide any further economicjustification for it."0 While the economic effects of structural andoperational modifications to the C&SF Project under the Plan arenot monetized, they are generally expected to be beneficial overthe sum of effects on agricultural and urban water supplies,commercial navigation on Lake Okeechobee, flooding potential,recreational value and Lake, estuary and bay fishing."'

C. Federal Law and Administrative Regulations

Originally, Congress enlisted the aid of the Corps of Engineersto control the Everglades' hydrology to facilitate agriculture andsettlement, while providing for navigation, for the conservation ofwildlife, water and soil, and for the amelioration of saltwaterintrusion into freshwater supplies. 2 The U.S. Fish and WildlifeService had estimated the value of the C&SF Project's benefits tofeatures aiding fish and wildlife at $291,000 per year; the bases forthe estimate are not cited, though the estimate is described ashaving been reviewed, and as reasonable. 3 One might infer thatthe estimate is based on consideration of industries associated withfishing and hunting, since the Fish and Wildlife Serviceinvestigated "preserving . . . fish and game resources" andpresented plans for the Water Conservation Areas, accordingly. 4

Decades later, finding that the Corps of Engineers'engineering had "unintended consequences" for the Evergladesecosystem, Congress called for adjustments to the structures and

79. PROGRAMMATIC EIS, supra note 11, at E-10; id at E-13 ("Corps of Engineersecosystem restoration policy has been formulated in recognition of the practical limits ofavailable economic tools to value environmental resources. . . . [E]cosystem restorationprojects are not subject to traditional benefit-cost analyses.").

80. WRDA 2000, supra note 6, § 601 (f) (2); 33 C.F.R. § 385.3 defines "cost-effective"in WRDA 2000 § 601 (f) (2) (A) (ii) as "the least costly way of attaining a given level ofoutput or performance .... " Thus, the Secretary must find that the activity is to beperformed as inexpensively as possible.

81. PROGRAMMATIC EIS, supra note 11, atE-7 to E-9.82. H.R. Doc. No. 80-643, at 13 (1948).83. Id. at 50.84. Id. at 36; id. at 29 (mentioning the "game-refuge features" of the Park).

[Vol. 26:373


operations of the C&SF Project for "the restoration, preservation,and protection of the South Florida Ecosystem while providing forother water-related needs of the region, including water supplyand flood protection." 5 Reflecting an understanding of theembeddedness of the human economy in the greater Evergladesecosystem, Congress found that, "(1) the Everglades is anAmerican treasure and includes uniquely[ ]important and diversewildlife resources and recreational opportunities; (2) thepreservation of the pristine and natural character of the SouthFlorida ecosystem is critical to the regional economy .... ."6 Iexamine the efforts of Congress through the Corps of Engineers tomanage Everglades water, upon which the economies of all of itsspecies depend.

1. The Central and Southern Florida Project: Section 203 of the FloodControl Act of 1948.

After a major drought, the hurricanes of 1947 overwhelmedsoutheastern Florida, leaving ninety percent of its land inundated.for months and resulting in losses estimated shortly thereafter bythe Corps of Engineers at $59,000,000.87 In response, Congressauthorized the Central and Southern Florida Project of the Corpsof Engineers, "for flood control and other purposes.""s The C&SFProject included constructing three water conservation areas thatwould serve as reservoirs for and mitigate saltwater intrusion intothe freshwater supply of east-coast communities and agriculturallands, and "benefit fish and wildlife in the Everglades." 9 Amongthe C&SF Project's other purposes, flood control for agricultureand eastern settlement received priority. Ironically, Congress in1948 recognized that "[o]verdrainage of certain lands is now aserious problem which must be rectified."9° Accordingly, the C&SFProject was intended "to restore the natural balance between soiland water" that had sustained human, other animal, and plant lifein the Everglades until water management projects anddevelopment before 1948 had destabilized it, resulting in parched

85. WRDA 2000 SENATE REPORT, supra note 31, at 35; WRDA 2000, supra note 6, §601 (h) (1).

86. WRDA 2000, supra note 6, § 602(a) (2).87. H.R. Doc. No. 80-643, at 47 (1948); PROGRAMMATIC EIS, supra note 11, at 1-25.

88. Flood Control Act of 1948, Pub. L. No. 858, 62 Stat. 1176, § 203.

89. H.R. DOC. No. 80-643, at 42-43 (1948).

90. Id. at 34.

2007]


and burning lands, extensive flooding of human settlements, andsaltwater intrusion.9' Congress believed that the C&SF Project"afford[ed] an excellent example of the coordination ofimprovements for flood control, water control, and relatedpurposes with requirements for preservation of fish and wildlife. 92

Now, the C&SF Project is effectively an enormous plumbing systemthat extends over 18,000 square miles in sixteen counties fromOrlando to the Florida Keys, comprising 1,000 miles of canals, over720 miles of levees, sixteen pumping stations, water storage areas,and other water management structures and operations to speedthe movement of freshwater from the wetlands to the sea,maintaining drainage for agricultural and metropolitan activities.93

The C&SF Project has accomplished significant drainage:ground water levels are lower, annual flows have decreased, landsthat had stood inundated for long periods are drier and lands thathad been drier are flooded, so that habitats associated with thesevarious hydroperiods have been relocated or destroyed, andsalinity levels in estuaries have changed. Results of this drainageinclude: losses in ecosystem area, water storage capacity andresiliency; increases in water pollution, sedimentation of formerlyflowing areas, and infiltration by exotic species; and thecompartmentalization of hydrological features that had beencontinuous.94

Even as the C&SF Project was constructed, its uncountenancedeffects necessitated its modification. For example, in the 1960s alengthened levee and construction of new Water ConservationAreas 3A and 3B just north of Everglades National Park (the"Park") obstructed natural water flows into the Park, even thoughthe Secretary of the Army had believed that the C&SF Project andthe new Park would be "complementary features of Federal activitynecessary to restore and preserve the unique Everglades region."95

Because managed water inputs subsequently introduced to thePark were inadequate, the ecosystem within the Park began todegrade. In 1971 the Park received a schedule of minimum

91. Id. at 33 (italics added).

92. Id. at 56.93. Programmatic Regulations, supra note 23, at 64,200; CENTRAL AND SOUTHERN

FLORIDA PROJECT COMPREHENSIVE REVIEW STUDY (THE RESTUDY) UPDATE ANDBACKGROUND (1999), http://www.evergladesplan.org/docs/backgd.pdf; RESTUDY FACTS,supra note 40.

94. PROGRAMMATIC EIS, supra note 11, at 5-4.95. H.R. DOC. NO. 80-643, at 57 (1948).

[Vol. 26:373


monthly water inflows that included infelicitous inputs during thedry season, and ecosystem deterioration accelerated. Mostrecently, the Park has secured improved water deliveries underfederal law that authorizes the Corps of Engineers to alter theC&SF Project pursuant to its General Design Memorandum for the"Modified Water Deliveries to Everglades National Park" project("Mod Waters"). Mod Waters is designed to provide water flows tothe Park that will better sustain its wildlife, and to attempt torestore its natural hydrological conditions while guaranteeingcontinuing flood protection for nearby residential and agriculturalareas. 6 Mod Waters must be completed before the components ofthe Plan that most directly affect the Park may be implemented.97

In view of these unintended effects, Congress decided toreassess the C&SF Project. Based on the results of this "Restudy,"Congress is now attempting to modify this extensive watermanagement system through WRDA 2000.

2. The Restudy.

In 1992 Congress required the Corps of Engineers to reviewthe C&SF Project to determine whether modifications wereneeded to improve environmental quality, aquifer protection andthe urban water supply.9" The Water Resources Development Actof 1996 authorized the Secretary of the Army to develop "aproposed comprehensive plan for the purpose of restoring,preserving, and protecting the South Florida ecosystem."99

Accordingly, the Corps of Engineers conducted The Central andSouthern Florida Comprehensive Review Study (the "Restudy").The Restudy determined that the Everglades had lost half of itsgeographic area and seventy percent of its input flows, that itswater quality had deteriorated (most notably with the introductionof toxic levels of mercury and excessive phosphorus), that its

96. Everglades National Park Protection and Expansion Act of 1989, § 104, 16 U.S.C.§ 410r-8(a)-(d) (Westlaw 2007).

97. WRDA 2000, supra note 6, § 601 (b) (2) (D) (iv); PROGRAMMATIC EIS, supra note11, at 5-32 to 5-33.

98. Water Resources Development Act of 1992 § 309(l), Pub. L. No. 102-580.99. Water Resources Development Act of 1996, Pub. L. No. 104-303 §

528(b) (1) (A) (i) (1996); see also § 528(0 (establishing the South Florida EcosystemRestoration Task Force, which was then to coordinate ecosystem restoration); WRDA 2000,supra note 6, § 601 (j) (1) (requiring the Task Force to provide consultation in establishingan independent scientific review panel); 33 C.F.R. § 385.10(e) (elaborating the consultingrole of the Task Force in Plan implementation).

2007]

416 STANFORD ENVIRONMENTAL LAWJOURNTAL

habitats were being destroyed and at least sixty-eight of its specieshad become endangered, 00 and that exotic species that alter theecosystem had invaded much of the remaining Everglades.

The Restudy therefore recommended the ComprehensiveEverglades Restoration Plan.' The Plan includes sixty-eightindividual components, such as those creating roughly 217,000acres of water storage and treatment areas, and others removing240 miles of levees and canals so that water in the Everglades canbe decompartmentalized in the hope of restoring adequatenatural sheetflow. 1 2 Roughly eighty percent of the over 1.1 millionacre-feet of water per year that the Plan is expected to yield isearmarked for ecosystem recovery.0 3 The Plan has been called "theworld's largest ecosystem restoration effort[.]"104

3. The Comprehensive Everglades Restoration Plan: Section 601 ofthe Water Resources Development Act of 2000.

Prioritizing ecosystem restoration, WRDA 2000 approves thePlan as a framework for improvements to the C&SF Project, andorders its implementation. 5 While four pilot projects employingwater storage and reuse technologies are authorized forimplementation,10 6 eleven initial projects, including an AdaptiveAssessment and Monitoring Program, are authorized subject tofour conditions.0 7 Three of the conditions concern thepreparation, submission, and approval-based funding of individualproject implementation reports.' s Notably, the fourth condition

100. CENTRAL AND SOUTHERN FLORIDA PROJECT COMPREHENSIVE REVIEW STUDY(THE RESTUDY) KEY POINTS (1999), http://www.evergladesplan.org/docs/key.pdf. Sixty-eight species are listed pursuant to the Endangered Species Act; other species may be nearextinction as well. Id.

101. Programmatic Regulations, supra note 23, at 64,200. The Plan, dated April 1,1999, was submitted to Congress in the results of the Restudy on July 1, 1999.

102. Id. at 64,200; RESTUDY FACTS, supra note 40.103. RESTUDY FACTS, supra note 40.104. CENTRAL AND SOUTHERN FLORIDA PROJECT COMPREHENSIVE REVIEW STUDY

(THE RESTUDY) UPDATE AND BACKGROUND (1999), http://www.evergladesplan.org/

docs/backgd.pdf.105. WRDA 2000, supra note 6, § 601(b) (1) (A), describing the approved changes to

the Project as those "that are needed to restore, preserve, and protect the South Floridaecosystem while providing for other water-related needs of the region, including watersupply and flood protection." Similar language appears in WRDA 2000 § 601 (h) (1).

106. Id § 601 (b) (2) (B).107. Id. §§ 601(b) (2) (C)-(D).

108. Id. §§ 601(b) (2) (D) (i)-(iii).

[Vol. 26:373


prevents decompartmentalization and restoration of sheetflow inthe Water Conservation Areas immediately upstream of EvergladesNational Park until the Mod Waters project is complete.0 9

The Plan's Adaptive Assessment and Monitoring Program isinitially authorized for ten years and budgeted $100 million."0

Because it is the means by which the Plan is to incorporate the bestscience into administrative decisionmaking, its implementation istreated in more detail infra. Working partially in parallel with theAdaptive Assessment and Monitoring Program, an independentscientific review panel is to assess the Plan's progress in restoringthe ecosystem and to report its findings to Congress every otheryear."' The Programmatic Regulations implementing § 601 ofWRDA 2000 indicate that "independent scientific review is crucialfor ensuring that the best available science is used" in Evergladesrestoration." 2 Assembled by the National Academy of Sciencesgenerally pursuant to 33 C.F.R. § 385.22(a) (3), the panel first meton October 24, 2004, and has produced its first biennial review.' 3

The approved Plan groups its water management components,such as levees, canals, pumps, and their removal, into about forty-five projects. Except for those that may be implemented at thediscretion of the Secretary of the Army under his additionalprogrammatic authority to expedite Plan execution, projects otherthan those fifteen discussed supra and two approved withlimitations under § 601 (g) (3) 14 require a new act of Congressbefore they may be implemented.' 5 In particular, any project thatis to deliver the water newly made available by the Plan is expresslynot approved for implementation, because although this water

109. Id. § 601 (b) (2) (D) (iv).110. Id. § 601 (b) (2) (C) (xi); WRDA 2000 SENATE REPORT, supra note 31, at 40.111. WRDA 2000, supra note 6, § 601(j).112. Programmatic Regulations, supra note 23, at 64,206.113. See National Academies, Project: Independent Scientific Review of Everglades

Restoration Projects, (2004-07) (2007), http://www8.nationalacademies.org/cp/projectview.aspx?key=WSTB-U-03-04-A; COMM. ON INDEP. SCIENTIFIC REVIEW OF

EVERGLADES RESTORATION PROGRESS, NAT'L ACADEMIES, PROGRESS TOWARD RESTORINGTHE EVERGLADES: THE FIRST BIENNIAL REVIEW, 2006, available athttp://www.nap.edu/catalog/I1754.html [hereinafter BIENNIAL REVIEW].

114. These two projects primarily involve land acquisition rather than water-controlengineering; thus, although they were deemed meritorious, they were consideredinappropriate for execution by the Corps of Engineers. One is to be funded by theDepartment of the Interior, and the other is beyond the scope of the Plan. WRDA 2000SENATE REPORT, supra note 31, at 51.

115. WRDA 2000, supranote 6, §§ 601(c), 601 (d)(1).

20071


would likely benefit the national parks at the southernmost edgesof the Everglades, its use may have detrimental effects upstream.116

No Water Resources Development Act has been passed since theyear 2000. Thus, additional projects have yet to be authorized, andany water saved by the Plan cannot yet be used for ecosystemrestoration.

The Corps of Engineers is the federal sponsor of the Plan; thePlan's non-federal sponsor is the South Florida WaterManagement District (SFWMD). The costs of Evergladesrestoration are to be shared equally by the federal government andthe non-federal sponsor."7 The Secretaries of the Army and of theInterior are jointly to report to Congress on Plan implementationat least every five years from October 1, 2005 to October 1, 2036,specifically providing an expense analysis for the AdaptiveAssessment and Monitoring Program."' The law also requires theSecretary of the Army and the Governor of Florida to execute anagreement establishing a mechanism for resolving disputesbetween the Corps of Engineers and the state."9 This provisionmay be expected to streamline the cooperative efforts of the Corpsof Engineers and the SFWMD, reserving time and funds forproject implementation.

Section 601 (h) provides for public assurances that the Planmeets the Congressional goal of ecosystem restoration whileproviding for south Florida's other water needs, such as floodcontrol and water supply for human endeavors. These assurancesinclude: (1) a binding agreement to be executed by the Presidentof the United States and the Governor of Florida; (2)Programmatic Regulations to be issued by the Secretary of theArmy to ensure that the Plan serves its purpose, discussed infra;and (3) assurances specific to each individual project under thePlan. The project-specific assurances consist of a contract and anoperating manual for project implementation, along with a ProjectImplementation Report that is based upon the best availablescience, 2 complies with the National Environmental Policy Act of

116. Id. § 601 (g) (1) (A); WRDA 2000 SENATE REPORT, supra note 31, at 50.117. WRDA 2000, supra note 6, § 601(c)(3). The estimated cost at the 1999

presentation of the Plan to Congress was $7.8 billion. RESTUDY FACTS, supra note 40.118. WRDA 2000, supra note 6, § 601(l).119. Id. § 601 (i); WRDA 2000 SENATE REPORT, supra note 31, at 57.

120. WRDA 2000, supra note 6, § 601 (h) (4) (A) (iii) (VII).

[Vol. 26:373


1969 (NEPA),'12' and represents the results of project planning.2 2

The law prevents the Secretary of the Army from executing thecontract, which is a Project Cooperation Agreement to be signedby himself and the SFWMD, until the state has reserved orallocated water for the natural system according to the ProjectImplementation Report. 23 Thus, the sponsoring federal agencycannot act to update the C&SF Project for Florida until the statehas exercised its regulatory authority in accordance with both thePlan-level § 601 (h) (2) agreement and the individual project-level §601 (h) (4) (B) agreement.

Accordingly, the binding, federal-state agreement required by§ 601(h) (2) obliges the state of Florida to conduct its consumptivewater use permitting process so that water arising from a Planproject is neither allocated nor "otherwise made unavailable" bythe state until a sufficient amount is first reserved for ecosystemrestoration. 24 President George W. Bush and Governor Jeb Bushexecuted the requisite agreement on January 9, 2002.125 The Planprovides for enforcement of the agreement by "[a]ny person orentity that is aggrieved by a failure" of the parties to comply; theaggrieved may obtain injunctive relief from a federal court. 26

The assurances also include a savings clause, prohibiting Planmodifications in the C&SF Project from removing "existing legalsources of water" until equivalent, new water sources areavailable. 12 As examples of existing water supplies that are to beprotected, the statute enumerates those for agricultural and urbanuse, those allocated to the Seminole Indian Tribe of Florida and tothe Miccosukee Tribe of Ihdians of Florida, those for Everglades

121. National Environmental Policy Act of 1969, P.L. 91-190, 83 Stat. 852 (1970),codified at 42 U.S.C. § 4321 et seq. (Westlaw 2007); WRDA 2000, supra note 6, §601 (h) (4) (A) (iii) (111).

122. WRDA 2000, supra note 6, § 601 (h) and WRDA 2000 SENATE REPORT, supranote 31, at 53.

123. WRDA 2000, supra note 6, § 601 (h) (4) (B) (ii).124. Id. § 601 (h) (2) (A); Comprehensive Everglades Restoration Plan Assurance of

Project Benefits Agreement 2 (2002) [hereinafter Agreement] (copy on file with theauthor; text available at http://frwebgate.access.gpo.gov/cgi-bin/multidb.cgi?WAIStemplate=multidb results.html&WAISqueryRule=%24WAISqueryString&WAISdbName=2002-presidential documents+Weekly+Compilation+of+Presidential+Documents+%282002%29&WAISqueryString=everglades+restorat'.on&Submit.=Submit&WAISmaxHits=50&WrapperTemplate=wcomp-.wrapper.html).

125. Agreement, supra note 124, at 3.126. WRDA 2000, supra note 6, § 601(h) (2) (B) (i).127. Id. § 601 (h) (5) (A).

.20071


National Park, and those for fish and wildlife generally. 21 In

addition, existing, legally recognized rights to flood protection areto be honored. 129 Because the quoted language had not been"defined in WRDA 2000 or elsewhere in Federal or Florida Statelaw[",]"' ° the public has sought further clarification, expressingconcerns about established uses that have not been subject to statepermitting requirements; these include preserving urban wetlandsand maintaining sufficient freshwater in urban water supplies toprevent saline intrusion.' The Secretary of the Armycommissioned a Guidance Memorandum to establish criteria fordetermining whether a water source is an existing legal source andwhen such a source has been removed, and ordered that existing-legal-source analyses be included in each Project ImplementationReport.

132

The law identifies the "overarching objective of the Plan" as"the restoration, preservation, and protection of the South FloridaEcosystem while providing for other water-related needs of theregion, including water supply and flood protection."'3 Ratherthan simply providing for water-related needs of the region otherthan those of the ecosystem, the savings clause effectivelyprioritizes the extant water allocation regime, the establishmentand maintenance of which has degraded the ecosystem in the firstplace. Fortunately, it may be possible for C&SF Project services topersist under the Plan, since the 1.7 billion gallons of water perday 3 4 currently evacuated to tide theoretically may be capturedand retained in the ecosystem.

The Plan provides for a report on the applicability of existingenvironmental laws and executive orders, and on its compliancewith them; suggested modifications to the Plan must abide by allapplicable federal and state laws. 135 The Secretaries of the Army

128. Id. § 601(h)(5)(A)(i)-(v).129. Id. § 601(h) (5) (B); WRDA 2000 SENATE REPORT, supra note 31, at 57.130. Programmatic Regulations, supra note 23, at 64,203.131. PROGRAMMATIC EIS, supra note 11, at 10-12, 10-15.132. 33 C.F.R. §§ 385.36, 385.5(a)(2)(vi) (Westlaw 2007); Programmatic

Regulations, supra note 23, at 64,203-04, 64,215.

133. WRDA 2000, supra note 6, § 601(h)(1).134. RESTUDY FACTS, supra note 40.135. See Section 12: Environmental Requirements, in PROGRAMMATIC EIS, supra note 11;

33 C.F.R. §§ 385.10(c), 385.32(a) (2) (Westlaw 2007). These sections of the ProgrammaticRegulations, governing Plan implementation and modification, respectively, list examplesof applicable federal laws, including the Endangered Species Act, the Clean Water Act, theNational Historic Preservation Act, and the Marine Mammal Protection Act. The water

[Vol. 26:373


and of the Interior are also to fulfill their legal obligations to thetribes as they perform their duties under § 601 of WRDA 2000."6For example, the operating manuals for individual projects mustinclude drought contingency plans that are required by otherfederal regulations and are consistent with the Seminole tribalcompact and the Florida Administrative Code.'37 In particular, theDistrict Engineer of the Jacksonville District of the Corps ofEngineers is responsible for ensuring that Plan implementationcomplies with NEPA, though he may invite the assistance of otheragencies in preparing NEPA documents. 8 A NEPA process is to beperformed for each individual project under the Plan, subject tocategorical exclusions.139

Synergy between federal and state laws is desirable to avoidredundant or counterproductive ecosystem restoration efforts. Tospeed the execution of the Plan, the SFWMD funds an "Acceler8"program comprising eight state projects predominantly involvingpublic lands, which the agency has determined would provideimmediate benefits complementary to those of the Plan.14 ° OneAcceler8 project is devoted primarily to improving water quality,an issue with which the state has achieved some success. Florida'sEverglades Regulatory Program for the Everglades AgriculturalArea required farming concerns to adopt Best ManagementPractices to reduce phosphorus concentrations in agricultural

rights of the Seminole Tribe of Florida are defined by a compact among the tribe, thestate and the SFWMD that carries the force and effect of Federal law. 25 U.S.C. § 1772e(Westlaw 2007). The rights of the federally recognized tribal governments in the region toset water quality standards, as manifest in the Seminole Tribe of Florida Tribal Water Codeand the Miccosukee Tribe of Indians of Florida Environmental Protection Code, isacknowledged in the Restudy, PROGRAMMATIC EIS, supra note 11, at H-1 6 to H-19, andconsultation between the sponsoring agencies and the tribes is to occur on a government-to-government basis. 33 C.F.R. § 385.10(b)(i) (Westlaw 2007). Through Florida'sEverglades Forever Act the state legislature began to address the problem of excessivephosphorus in the Water Conservation Areas, the Park and the Arthur R. MarshallLoxahatchee National Wildlife Refuge. FLA. STAT. ch. 373.4592 §§ (1) (d), (2) (i) (Westlaw2007).

136. 33 C.F.R. § 385.10(b) (ii) (Westlaw 2007).137. 33 C.F.R. § 385.28(a) (6) (vii) (Westlaw 2007).138. 33 C.F.R. § 385.14(a)(1)-(3) (Westlaw 2007).139. Programmatic Regulations, supra note 23, at 64,216; 33 C.F.R. § 385.14(d).140. See Acceler8 Homepage, https://my.sfwmd.gov/portal/page?-pageid=

54,926712&_.dad=portal&_schema=PORTAL (last visited Apr. 14, 2007). While the stateprogram could work synergistically with the Plan, Acceler8 has also "raised concerns aboutdisproportionate funding and control by the state over [Plan] implementation[.]"BIENNiAL REVIEW, supra note 113, at 9.

2007]


runoff.' Though the program targeted a 25 percent reduction inreleased phosphorus, the industry has attained a 51 percentreduction. 4 2 Mercury may be converted into a toxicbioaccumulative form more quickly in the presence of highernutrient concentrations; if so, phosphorous pollution mitigationcould contribute to the reduction of mercury poisoning in theecosystem.' Despite this project's success, additional phosphorousmitigation is required for ecosystem maintenance."'

Many modifications to the C&SF Project that may beecologically desirable involve privately held land in the Everglades,or might deprive private citizens of legally recognized floodcontrol rights. Therefore, land acquisition is prerequisite to thefull realization of the Plan's restoration goals. By 2005, the statehad acquired just over half of the land needed for Planimplementation (over 207,000 acres) for over $800 million thatincluded federal funds. 4 5 Thus, ecosystem restoration costs exceedthose of altering the C&SF water management system itself.

4. The Programmatic Regulations: 33 C.FR. § 385.

The legislative history of WRDA 2000 describes the Plan as"unique in Federal environmental law,' ' 146 in that § 601(h)(3)requires the Secretary of the Army to obtain the concurrence ofboth the Secretary of the Interior and the Governor of Florida inissuing programmatic regulations to implement the Plan.'47 The

141. FL. ADMIN. CODE ANN. r. 40E-63.011 to .165, Part I: Everglades RegulatoryProgram: Everglades Agricultural Area (Westlaw 2007), available at https://www.flrules.org/gateway/chapterhome.asp?chapter=40E-63.

142. PROGRAMMATIC EIS, supra note 11, at 5-12.143. Id. at 5-19.144. U.S. ARMY CORPS OF ENG'RS (JACKSONVILLE DIST.) & S. FLA. WATER MGMT.

DIST., CENTRAL AND SOUTHERN FLORIDA PROJECT COMPREHENSIVE EVERGLADESRESTORATION PLAN 2005 REPORT TO CONGRESS A-7 (2005), available at http://www.evergladesplan.org/pm/program-docs/cerpreport.congress_2005.cfm#docs.

145. Id. at 32.146. WRDA 2000 SENATE REPORT, supra note 31, at 55.147. WRDA 2000, supra note 6, § 601 (h) (3). Other United States, Florida, tribal and

local agencies are also to be consulted, but their concurrence is not required. Thelegislated partnership between the federal and state governments is designed to facilitatecoordination between agencies with diverse missions and jurisdictions as projects areadministered. WRDA 2000 SENATE REPORT, supra note 31, at 56. Writing before theregulations were promulgated, Doyle and Jodrey discuss the Plan's assurances provisionsgenerally, recounting the evolution of the federal-state partnership as manifest in theconcurrence requirement. See Mary Doyle & Donald E. Jodrey, Everglades Restoration:Forging New Law in Allocating Water for the Environment, 8 ENVTL. LAW 255 (2002).

[Vol. 26:373


Programmatic Regulations establish management processes.Though they were to be issued no later than December 11, 2002,48the regulations were promulgated on November 12, 2003 andmade effective on December 12, 2003.149 While the front-loadedconcurrence requirement might have slowed the regulatoryprocess, the regulations may well function more smoothly for theresulting support of pertinent parties. The concurrencerequirement is to operate only at the programmatic level; the lawexpressly prohibits this requirement from applying to individualprojects under the Plan. 5'

The regulations are intended to fulfill the requisites of §601(h) (3), assuring that Plan goals are achieved by guiding therealization of individual Plan projects.' Of particular interest isthe requisition of a mechanism for integrating "new informationresulting from changed or unforeseen circumstances, newscientific or technical information or information that isdeveloped through the principles of adaptive managementcontained in the Plan" into the Plan as it is executed.152

Accordingly, the Programmatic Regulations must be reviewed atleast every five years so that new information will be assimilatedinto Plan administration. 5 ' This mechanism for updating projectguidance renders the Programmatic Regulations "a centralcomponent in the adaptive assessment and management processon which success of the Plan, and this bill, depends."154

The regulations also establish procedures for composing the §601 (h) (4) project-specific assurances, and for setting interim goalsfor use as benchmarks in evaluating the effectiveness of ecosystemrestoration as the Plan is implemented. 55 The interim goals are toidentify water quantity, quality, timing and distribution objectivesand desired ecosystem responses to be achieved by the year 2010and every five years thereafter.'56 Since these benchmarks are to beused by multiple agencies for planning, monitoring andassessment at the individual project and comprehensive Plan levels

148. WRDA 2000, supra note 6, § 601 (h) (3) (A).149. Programmatic Regulations, supra note 23, at 64,200.

150. WRDA 2000, supra note 6, § 601(c) (2).151. Programmatic Regulations, supra note 23, at 64,200-01.

152. WRDA 2000, supra note 6, § 601(h) (3) (C) (i) (II).153. Id. § 601(h) (3) (E); see also 33 C.F.R. § 385.6 (Westlaw 2007).154. WRDA 2000 SENATE REPORT, supra note 31, at 56.

155. WRDA 2000, supra note 6, § 601(3) (C) (i) (I) and (III).156. Programmatic Regulations, supra note 23, at 64,209.

2007]


as well as for overall budgeting, and are to be made available forindependent scientific review, 57 they may well serve asconvergence points that enable administrative actors with differentroles to perform from the same text. The regulations create aprocess for revising the interim goals at least every five years if newinformation suggests the need, and adaptive managementmechanisms are to be engaged if the interim goals are unlikely tobe met.158

Analogous to the interim goals for ecosystem restoration arethe interim targets for south Florida's other water needs. Toprioritize the ecosystem's needs, the interim targets were to beconsistent with and to be established after the interim goals.159

Otherwise, the targets' function parallels that of the goals.

5. RECOVER.

"RECOVER" is an acronym for "Restoration[,] Coordinationand Verification," naming the functions of the multidisciplinaryteam of science professionals described in the Plan and establishedby the sponsoring agencies to incorporate the highest qualityscience into Plan implementation while maintaining an ecosystem-wide focus. 60 In an effort to overcome barriers to ecosystemrestoration at the level of individual agency missions, RECOVERconvenes experts from federal, state and tribal governmentagencies to facilitate the development of consensus, and to"establish an open process that promotes scientific contributionsto the Comprehensive Plan.' 16' RECOVER's mission is to synthesizethe best scientific research from various disciplines (e.g.,hydrology, atmospheric science, the biology of particular classes oforganisms, and ecology) so that individual Plan projects areinformed by a unified body of research and coordinated at theecosystem level. 2 RECOVER empirically assesses the actual

157. 33 C.F.R. § 385.38(b) (2) (Westlaw 2007).158. 33 C.F.R. §§ 385.38(b) (3), 385.38(b) (5) (i) (Westlaw 2007).159. Programmatic Regulations, supra note 23, at 64,210.160. 33 C.F.R. §§ 385.3, 385.20 (Westlaw 2007).161. U.S. ARMY CORPS OF ENG'RS & S. FLA. WATER MGMT. DIST., PROGRAM

MANAGEMENT PLAN FOR RESTORATION COORDINATION AND VERIFICATION (RECOVER)FIscAL YEARs 2004-2006 at 9 (Version 5.0, Sept. 2004), available athttp://www.evergladesplan.org/pm/recover/recover-docs/mg tplan/rec-pmp fna-a

ug_2004.pdf [hereinafter RECOVER PROGRAM MANAGEMENT PLAN].162. Programmatic Regulations, supra note 23 at 64,207 ("The role of RECOVER is

to promote an integrated view within the implementing agencies on matters relevant to

[Vol. 26:373


performance of the Plan on a system-wide scale, evaluates plannedproject designs using mathematical modeling of hydrologicalchanges and their ecosystemic effects, and informs planning bydeveloping recommendations to improve the effectiveness of thePlan.16 RECOVER's assessments are subject to external peerreview; external peer review of other RECOVER activities occurs atthe discretion of the sponsoring agencies or of the RECOVERLeadership Group."6

RECOVER currently consists of three task-centered teamscoordinated by a leadership group. The RLG comprises twelveactors: one program manager from each of the Corps of Engineersand the SFWMD, as well as one representative from each of theEnvironmental Protection Agency, the National Oceanic andAtmospheric Administration, the United States Fish and WildlifeService, the United States Geological Survey, the National ParkService, the Miccosukee Tribe of Indians of Florida, the SeminoleTribe of Florida, the Florida Department of Agriculture andConsumer Services, the Florida Department of EnvironmentalProtection, and the Florida Fish and Wildlife ConservationCommission.'65 The RLG is to provide policy information toRECOVER's science professionals, and scientific or technicalguidance to consulting management professionals. RECOVERexpressly "is not a policy making body," and its work products"shall not be self-executing."'" This means that the RLG'sauthority does not extend beyond RECOVER, and that RECOVERhas no power other than to provide information for thesponsoring agencies to consider in consultation with other federal,tribal, state and local agencies.'67 However, information producedby RECOVER "shall be available to the public.""6M This provisionaffords some transparency to the public, and promotes the publicaccountability of agencies for their actions with respect to the

the implementation of the Plan in order to ensure that the goals and purposes of the Planare achieved."). RECOVER aspires to use an inclusive, consensus-building process thatpromotes transparency and professional responsibility in taking a science-based approachto ecosystem restoration. RECOVER PROGRAM MANAGEMENT PLAN, supra note 161, at 9.

163. RECOVER PROGRAM MANAGEMENT PLAN, supra note 161, at 2; see alsoRECOVER - Who We Are!, http://www.evergladesplan.org/pm/recover/recover.cfm (lastvisited Apr. 15, 2007).

164. 33 C.F.R. §§ 385.20(e)(1), 385.20(g) (Westlaw 2007).165. 33 C.F.R. §§ 385.20(d) (1)-(2) (Westlaw 2007).

166. 33 C.F.R. §§ 385.20(a)-(b) (Westlaw 2007).167. 33 C.F.R. § 385.20(b) (Westlaw 2007).

168. Id.

20071


reported information.The $100 million that WRDA 2000 provides for the Plan's

Adaptive Assessment and Monitoring Program has been budgetedinto $10 million per year for ten years and supports RECOVER'sactivities.16 Agencies that participate in RECOVER must fund theirown participation, 7 ° diverting personnel and other resources fromtheir primary missions to cooperative service. The Corps ofEngineers and the SFWMD each contain a unit that allocates fundsand staff to RECOVER. Thus, although RECOVER is a multi-agency body, the sponsoring agencies effectively steer RECOVER'sacts through their ability to provide or withhold needed funding.1 7'

The sponsoring agencies are to establish an adaptivemanagement program that provides for informed innovation inimplementing the Plan.172 RECOVER is charged with developingan assessment program to identify and measure ecosystemicresponses to changes in the C&SF Project under the Plan. 17

RECOVER is to monitor progress toward ecosystem restoration, todetermine whether corrective action should be contemplated, andto present its assessments in a published technical report to thesponsoring agencies at least every five years.74 The RECOVERtechnical report is to inform an assessment report that is created inturn by the sponsoring agencies, which also determine thequantity of water that the Plan is expected to capture. Theseagencies may act to improve Plan performance based on theirassessment report, independent scientific review of Planeffectiveness, and other appropriate information. 175 Thus,members of the RLG serve as guardians of the adaptivemanagement process as well as directors of RECOVER's technicalprojects.

The Corps of Engineers directs RECOVER to develop andrecommend the interim goals and interim targets for Planmanagement, providing a quantitative benchmark for assessing thePlan's effectiveness based upon the "best available science and

169. WRDA 2000, supra note 6, § 601(b) (2) (C) (xi); WRDA 2000 SENATE REPORT,supra note 31, at 40; 33 C.F.R. § 385.20 (Westlaw 2007) (defining the role of RECOVER inadaptive management).

170. 33 C.F.R. § 385.20(c)(1).171. Previous two sentences informed by Heisler Interview, supra note 27.

172. 33 C.F.R. § 385.31 (Westlaw 2007).173. 33C.F.R. § 385.31 (b)(1).174. 33 C.F.R. §§ 385.31(b) (2)-(4).

175. 33 C.F.R. §§ 385.31(d) (1), 385.31 (c).

[Vol. 26:373


information [.] ,,7' RECOVER's responsibilities under theProgrammatic Regulations also include reviewing andrecommending improvements to the system-wide analysis andsimulation tools used for Plan assessment and evaluation; 77

analyzing proposed revisions to the Master ImplementationSequencing Plan that sets forth the order in which all of the Plan'sindividual projects are to be executed,17 and analyzing proposedmajor modifications to the Plan itself.179 With respect to individualprojects, RECOVER is to inform the discussion of how theecosystem would behave with and without the project, evaluatehow alternative project plans would contribute to the overallperformance of the Plan, assist in developing a project's operatingmanual and analyzing its effects on water allocation, and integratemonitoring at the project level with the overall monitoring of thePlan's environmental effects.' s RECOVER's principal actions todate are examined infra Part 1V.B.

6. Legal and ecological definitions of "ecosystem restoration."

a. Ecosystem.

As used by ecologists, definitions of "ecosystem" varydepending on the focus of their study. In the context ofCongressional discussions of ecosystem management, a" [c] ommunity of organisms interacting with one another and withthe chemical and physical factors making up their environment," isa useful starting point, provided that certain implications areborne in mind.' Because an organism may play a role in manyecosystems at once, attempts to distinguish ecosystems from eachother may have arbitrary or scientifically meaningless results. Thus,a boundary drawn around an ecosystem might best be regarded as

176. 33 C.F.R. §§ 385.38(c) (1), 385.39(b) (1) (Westlaw 2007).177. 33 C.F.R. § 385.33(b) (Westlaw 2007).178. 33 C.F.R. §§ 385.30(a), 385.30(b) (2) (Westlaw 2007).179. 33 C.F.R. § 385.32(a)(4) (Westlaw 2007). A major modification is one that

requires a supplemental Environmental Impact Statement for the Plan under NEPA;Project Implementation Reports are appropriate vehicles for making minor modifications.See 33 C.F.R. § 385.32(c).

180. 33 C.F.R. § 385, App. A, Fig. 2 (Westlaw 2007); 33 C.F.R. § 385.26(c) (Westlaw2007). Project Implementation Reports are to include RECOVER analyses. 33 C.F.R. §385.26(a) (3) (xiv).

181. M. Lynne Corn, Ecosystems, Biomes and Watersheds: Definitions and Use, CRSREPORT FOR CONGRESS (1993), available at http://www.cnie.org/nle/crsreports/biodiversity/biodv-6.cfm.

2007]


a fiction useful for a particular analytical or management purpose;for example, an ecosystem analysis focusing on the roles ofmigrant roseate spoonbills may span a larger area than onefocused on the narrow-ranging Cape Sable seaside sparrow,though both species may also be studied as components ofidentical ecosystems. While they may differ, ecological framings ofecosystems tend to be based upon some behavioral feature of atleast one of an ecosystem's organisms. Accordingly, in its scientificmeaning the land area associated with an ecosystem is likelydynamic, changing with rainfall patterns, for instance.

One limit to the definition of "ecosystem" with specialsignificance for Everglades restoration is that an ecosystem almostalways exists over a continuous area. An historical ecosystem thathas been divided into disjoint regions is generally understood tohave been divided into distinct, though perhaps similar,ecosystems. I"2 This definition implies that thecompartmentalization of the pre-drainage Everglades ecosystemhas not only made it smaller, but broken it into several pieces.Considering the Everglades ecosystem from an ecologicalperspective highlights the importance of decompartmentalizationto earnest attempts at its restoration.

Because of the need to draw boundaries for managementpurposes, the legal definition of the Everglades ecosystem under §601 of WRDA 2000 differs from ecological ecosystem definitions.Legally, the south Florida ecosystem is defined as "the areaconsisting of the land and water within the boundary of the SouthFlorida Water Management District in effect on July 1, 1999,"expressly including the Everglades, the Florida Keys and coastalwaters."8 3 Neither particular species nor sets of species matter inthis definition; thus, it may not be well suited to ecologicalanalyses. Lands outside the SFWMD may be very important torestoration of the ecosystem fragmented within the watermanagement district. However, tying the definition to land overwhich a sponsoring agency has management authority makespossible the implementation of Congress' ecosystem restorationPlan.

182. Id.183. WRDA 2000, supra note 6, § 601(a) (5).

[Vol. 26:373


b. Restoration.

To restore something is to return it to a former or originalcondition. True restoration of the Everglades ecosystem cannot beachieved. Not only does the law prevent the removal of establishedwater and flood control services, but much of the originalecosystem's land and many of its species are now irretrievable. Thisimpossibility leaves room for expert opinions to vary over whatpotential states of the ecosystem should be considered restoredstates."4 Though an ecologist might suggest that restoration couldbest be achieved by removing the C&SF Project works andreconnecting the compartmentalized elements of the Kissimmee-Okeechobee-Everglades watershed to restore sheetflow throughthe Everglades, one could define restoration of the ecosystem invarious terms, such as population levels for given species, orrecovery of pre-drainage ecosystem functions. Ideally, any set ofplausible performance measures (such as the interim goals) will bemet by a truly restored ecosystem, so that restoration is robust tothe choice of what to monitor."5

The Plan treats the uncertainty about the best meaning for"restoration" not only as an opportunity for difference of opinionand conflict, but also as a kind of room for play available toecosystem managers. There is a very low-resolution agreement thatas long as the ecosystem gets wetter in general, its health willimprove. 18 6 This play allows for the flexible, iterative planning thatcharacterizes adaptive management, so that administrative acts canfunction as queries to the ecosystem, the reactions of which caninform the next query, in a kind of call-and-response between theecosystem and management authorities representing the public.

Indeed, defining restoration has been a politically chargedprocess. Individual stakeholders have been concerned that thedefinition might not reflect their interests. Some have felt thatunless the definition were to "assign a proper role to science infulfilling the objectives of the Plan," the Plan's implementationwould "be driven by political compromises."" 7 On the other hand,some were concerned that Congress' intent to address water needsother than those for restoration would be "ignored in an effort to

184. PROGRAMMATIC EIS, supra note 11, at 5-36.

185. Id. at 5-30.186. Id. at 5-37.

187. Programmatic Regulations, supra note 23, at 64,204.

2007]


advance an elusive and constantly changing vision of restorationapproved by the scientists[.]' s88

The definition of "restoration" announced in theProgrammatic Regulations is "the recovery and protection of theSouth Florida ecosystem so that it once again achieves and sustainsthose essential hydrological and biological characteristics thatdefined the undisturbed South Florida ecosystem."'" 9 Restorationunder the law treats only a subset of the ecological values at stake:for instance, the natural sheetflow from the Lake through theEverglades Agricultural Area-from the ecosystem's heart throughits aorta-will not be restored. However, the Plan springs from thehope that flows sufficiently similar to natural Everglades hydrology.can be recreated. This hope persists even though the detrimentaleffects of human settlement are largely irreversible, and theattempts to provide more sustaining managed flows areincompletely informed: for example, the model of the originalEverglades hydrology that serves as a restoration target is based ononly 36 years of rainfall data, and since the actual pre-drainagelandscape is unknown, the model uses a post-drainage landscapewith all water management structures removed and an inferredpre-drainage topography. Thus, "restoration" here does not entaila return to a pre-drainage state, but an effort to preserve whatremains and prevent further harm.

In effect, the Plan's goal is to preserve and restore the health ofthe remaining Everglades while continuing to support humansettlement and economic activity, and the instruments to be usedin achieving that goal will continue to be the orchestratedoperation of water storage and movement structures, along withefforts to streamline our ability to rework them as ecosystemrestoration and human settlement progress. "As authorized byCongress, the restored South Florida ecosystem will be significantlyhealthier than the current system; however it will not completelyreplicate the undisturbed South Florida ecosystem."' 90 Thus,Congress attempts not literally to restore the pre-drainageecosystem, but to engineer a new approximation to it. As stated bythe Secretary of the Army, the definition of "restoration" in theProgrammatic Regulations "acknowledges that, as authorized byCongress, the restored ecosystem will be different [from] the

188. Id. at 64,205.189. 33 C.F.R. § 385.3 (Westlaw 2007).190. Id.

430 [Vol. 26:373

AIMEUSIS THROUGH A GA TED MEMBRANE

historic ecosystem.... [T]he definition affords flexibility to allowfor adaptive management" and for other regional water needs asindividual Plan projects are authorized. 9 '

Clearly, this regulatory definition of "restoration" differs fromwhat an ecologist might envision for the Everglades, particularly inthat a restoration effort focused on the community of Evergladesspecies would likely entail decompartmentalization of theEverglades-especially restoration of sheefflow from the Lakethrough the Everglades Agricultural Area to the shores-and mayideally include the decompartmentalization of the Kissimmee-Okeechobee-Everglades watershed. Since ecosystem restorationefforts must also secure south Florida's development, restorationgoals are more limited. The physical and ecological characteristicsof the pre-drainage Everglades that have been chosen forrestoration are: wildlife access to a mosaic of habitat typesspanning a large area and affording multitudes of opportunitiesfor feeding and nesting to extant species, heterogeneity of habitattypes along a spectrum of spatial scales, oligotrophy (the almostundetectable concentrations of phosphorus that enable theperiphyton that supports the Everglades food webs to survive), anddynamic water flow and storage patterns-the alternating floodingand drying that allows food to be produced at the base of theecosystem during wet periods, and to be concentrated as prey forhigher predators during drydown.'92

The Secretary of the Army stated that the definition"recognizes implicitly that science will be the foundation ofrestoration," while realizing that all of Congress' goals are to beachieved.'93 Fully explicit details of the mechanisms for basingpolitical decisionmaking on scientific research remained below theresolution of the Programmatic Regulations, though theregulations did establish the RECOVER Group to conduct theadaptive management process.

IV. GETTING THE SCIENCE RIGHT: AN INTERFACLAL ANALYSIS OF

ADAPTIVE MANAGEMENT IN PRACTICE

The adaptive management strategy for using science to informpublic decisionmaking is the Plan's primary response to the

191. Programmatic Regulations, supra note 23, at 64,205.192. PROGRAMMATIc EIS, supra note 11, at 5-30 to 5-32.193. Programmatic Regulations, supra note 23, at 64,205.

2007]


challenges of management under uncertainty. Therefore, thedevelopment of specific mechanisms for adaptive assessment andmonitoring under the promulgated regulations merits detailedanalysis. I apply the conceptual analytical tools introduced supraPart II to assessing an adaptive management process, identifyingthe essential aspects of a viable administrative structure and theroles performed by those operating within such a structure.

A. Uncertainties: The Need for Adaptive Management

In terms of the discussion of facts supra Part II.A.2.a, theuncertain consists of unsettled questions. Some uncertainties arefree-floating and likely ever-elusive, such as those regarding thequantitative aspects of pre-drainage Everglades ecology. Others arehighly resolved questions about facts that are precipitant, but havenot yet hardened into the strata of our knowledge.'94 Uncertaintiesalso vary in origin; for example, some are born of the limitations ofa measurement method that can be complemented by othermethods, while others arise from a lack of prescience and can onlybe answered in the flow of time. Like water, uncertainties can bemanaged to some extent.

1. Characterizing uncertainties for management.

The different aims of scientific and dispute-resolution fact-finders contribute to differences in their approaches to managinguncertainty, as discussed supra Part II. Though science mayembrace uncertainty too avidly to be of practical use in politicaldecisionmaking, and management may abhor uncertainty toomuch to allow itself to be informed by the best availableempirically limited knowledge, scientific and managementprocesses can be performed in concert, if their performance isguided by the following principles: as our knowledge base withrespect to a matter grows, our uncertainty about it subsides;uncertainties of differing magnitudes and directions are tractablein various ways; since uncertainty characterizes the state of ourknowledge at every level of its growth short of our omniscience,"uncertainty" can denote every state from utter confusion toknowledge of all but the subtlest details of an expected outcome;

194. The topography of a given area is such a question, in that regularly updating atopographical database with new data theoretically could provide us with as precise a set ofmeasurements as needed for a given purpose.

[Vol. 26:373


because uncertainty is ever-present, it is irresponsible to citeuncertainty as a reason for failure to act.

Both quantifiable uncertainty and risk are expressible in thelanguage of probability, so some uncertainties may be framed interms of risk on various scales., On the system-wide scale, forexample, some uncertainty as to whether efforts at ecosystemrestoration will succeed may be treated as some risk that thoseefforts will fail, or even accelerate ecosystem degradation. Toaddress this kind of uncertainty, the Restudy team performed ananalysis of the "future without plan condition."'95 This analysiscompared estimates of future hydroecological conditions in southFlorida in two cases: one in which no changes to the C&SF Projectwere made, and another in which the changes contemplated inthe Plan were effected. Reporting the results of this analysis to thePresident of the Senate, the Assistant Secretary of the Armyconcluded that:

[T]he Everglades are dying. And if we do not act now, we mayvery well lose the opportunity to save them for futuregenerations. Today, I submit to you on behalf of theAdministration, a scientifically and economically sound plan thatwill allow us to rescue this unparalleled natural resource fromextinction. 1

96

This recommendation reflects the opinion of the Corps ofEngineers' uncertainty analysts, that:

The existence of uncertainty is no excuse to avoid makingdecisions about Florida's future. . . . Deciding not to proceedwith the Restudy because uncertainty exists is foolish as long asthe key uncertainties are recognized, acknowledged and arebeing addressed in a rational and appropriate manner .... Totake no action, to make no decisions because the Restudy is notfree of uncertainty may doom central and south Florida to a farmore uncertain future than they might face in a well-reasonedand carefully monitored new direction.' 97

195. PROGRAMMATIc EIS, supra note 11, § 4.196. Letter from Joseph W. Westphal, Assistant Secretary of the Army (Civil Works),

to Hon. Albert Gore, President of the Senate 1 (July 1, 1999), available athttp://www.evergladesplan.org/docs/comp-plan-apr99/asachief.pdf.

197. PROGRAMMATIc EIS, supra note 11, at 0-2.

2007]


However, there is a great deal of room between utter refusal toact in the face of uncertainty and proceeding with caution. In anenvironmental context the precautionary principle favors buildingup a knowledge base regarding the effects of a technology beforeattempting to introduce it to an ecological system, and using thatknowledge to manage potential adverse effects of the technologyon the ecosystem. Implicit in the adaptive management strategyunder the Plan, which serves as a mechanism for managementcourse-correction, is the recognition that a management decisionintended to improve ecosystem health instead may result in asurprising detriment. Indeed, the original C&SF Project wasexpected to provide water management for human endeavors andto conserve wildlife.198 The future-without-Plan analysis weighs theprobability and consequences of the Plan's failure, forming thebasis for the overall management decision that the ecosystem hasthe greatest chance of survival if we attempt its recovery.

To take action, any ecosystem management process requiresmaking at least two considered decisions with respect to eachsignificant uncertainty (i.e., each uncertainty the resolution ofwhich could confound the management plan): what levels ofuncertainty to tolerate as actions progress (such as deciding to usethe best knowledge available at the moment to informdecisionmaking), and how management steps will be made robustto the uncertainties that remain (for example, by contingencyplanning). These decisions may be informed by analysis andtesting, and may be iterated, as they are in an adaptivemanagement regime.

2. Strategies for managing uncertainties.

Uncertainties ranging from the primarily scientific to thepredominantly political attend Plan implementation. The Plancombines several strategies for guiding its implementationthrough uncertain terrain, notably: analysis of scientificuncertainties, prioritized testing of technological uncertainties,contingency planning, and adaptive management. In Part V, Idiscuss joint fact-requisition as a dispute-resolution procedure thatcould afford an additional strategy for managing uncertainty.

198. H.R. Doc. NO. 80-643, at 13 (1948).

[Vol. 26:373


a. Analysis of scientific uncertainties.

The Corps of Engineers treated those uncertainties mostsignificant to the execution of the Plan in part by commissioningan uncertainty analysis on the scale of the entire Evergladesecosystem, the scale at which the Plan itself operates.'99 Sinceadditional uncertainties attend subcomponents of the Plan at itssublevels, uncertainty analyses also may be run for individualprojects.

Uncertainties associated with the Plan were categorized asperceived uncertainties (perceived by those who are unaware thatexpert knowledge on a question has been developed), routineuncertainties (routine to planning processes, arising from theoutcomes of decisions that have yet to be made, such as where apump will be sited; these are clearly greatest early in a planningprocess), and unique (or other, specific) uncertainties."'Uncertainties arising from the subjective judgments of particularplanners were deemed "inevitable and appropriate," and thusignored; instead, the analysts focused on unique uncertaintiesaccompanying four key aspects of the Plan: modeling, linkagesbetween hydrological and ecosystemic restoration, applications oftechnologies on new scales, and risk assessment. 0 ' The analystssurveyed the uncertainty landscape for each aspect, reportingwhether and how pressing questions were being addressed. Inaddition to gathering key questions so that planners couldcoordinate their efforts, the analysts' report recommended that:modeling improvements prioritize exploring rainfall patternsdifferent from those represented in the 1965-1995 data set; theuncertainties inherent in each level of modeling (e.g.,hydrological, ecosystemic) be analyzed cumulatively as well asseparately; a management process be developed for meeting thenon-hydrological requirements of ecosystem restoration; a team ofexperts investigate in detail uncertainties associated withtechnologies for preventing water seepage, and that a conceptualrisk assessment consider whether and how the Plan may adverselyaffect the ecosystem or human water usage.

199. PROGRAMMATIc EIS, supra note 11, App. 0.

200. Id. at 0-3.201. Id. at 0-5.

2007]

436 STANFORD ENVIRONMENTAL LA WJOURNAL

b. Pilot projects: testing water management technologies.

As noted supra Part III.C.3, pilot projects to test the suitabilityof water storage and reuse technologies for ecosystem restorationwere immediately authorized by WRDA 2000.02 Aquifer storageand recovery, seepage management, and wastewater reusetechnologies are to be tested first.2

13 These promising technologies

have yet to be used at the scale required by the Plan.2 4 Aquiferstorage and recovery, which involves injecting water intounderground spaces for storage and pumping it back out forregulated use, offers the advantage of protecting stored water fromthe significant loss to evaporation occurring from open storageareas such as surface reservoirs; however, pumping requires fuelexpenditure. Seepage barriers improve the effectiveness of leveesand in-ground reservoirs, and may also mitigate saltwater intrusioninto fresh water. Currently available seepage managementtechnologies are to be tested to identify any potentiallyundesirable effects. While wastewater treatment could capturemore water for ecosystem restoration, the cost of achievingacceptable reclaimed water quality is high; funds used to treatwastewater may be more beneficially spent on other Plancomponents. Tests of wastewater treatment technologies includeobserving the effects of treated wastewater on the species balanceand health of a small area of marsh.

c. Robustness and contingency planning.

i. Robustness.

"Robustness" denotes resiliency, or the ability to function wellin spite of an imperfect start. RECOVER defines "robustness" as"the sensitivity of key design parameters to operate effectivelygiven the variability and uncertainty of future events." °5 RECOVERattempts to design robustness into the scientific assessment of Planprogress by examining the interactions of multiple variables asindicators of ecosystem status, rather than relying on a single kind

202. WRDA 2000, supra note 6, § 601 (b) (2) (B).203. See PROGRAMMATIc EIS, supra note 11, at 10-50 to 10-51.204. Id. at 10-70.205. RECOVER, COMPREHENSIVE EVERGLADES RESTORATION PLAN ADAPTIVE

MANAGEMENT STRATEGY 3 (2006), http://www.evergladesplan.org/pm/recover/recover docs/am/rec-am-strategy-brochure.pdf [hereinafter ADAPTIVE MGMT.STRATEGY].

[Vol. 26:373


of measurement to report the ecosystem's condition.2 16 It is

possible that an ecosystem in need of restoration may not besufficiently resilient to withstand any management interventions,except for the withdrawal of management. The state of anecosystem should be assessed before attempting any managementaction, to determine the tolerable range of testing formanagement decisions and for the technologies to be used in theirimplementation.

ii. Contingency planning.

In addition to discerning which variables can serve as the mostreliable monitors of ecosystem health, managers can buildresiliency into their decisions by developing contingency plans, oralternative courses of action to pursue in case a managementdecision does not yield a desired outcome. For example, in case awater reuse project component fails to provide water of sufficientlyhigh quality to recharge a wellfield, resort to conservation and toother in-ground or surface water supplies are plannedalternatives."7 The Programmatic Regulations require contingencyplans to be included in the implementation reports for individualprojects, so that management actions reflecting the goals of theproject and the Plan will be available in case a project fails." 8

iii. Beyond the plan: robustness to the exploitation ofuncertainty.

As a complement to robust designs for scientific tools andcontingency plans for scientific and technological shortfalls,science-policy interactions also can be designed to be robust to themanipulation of uncertainty for political purposes. Appreciationthat uncertainties can be very large or minute, and can arise frommore and less tractable sources, can assist those engaged inpolitical decisionmaking in recognizing when an actor isattempting to devalue research results that tend to disfavor hisinterests. Such an attempt might involve miscitation (e.g., framingscientifically sound research results as though they were

206. See, e.g., id. at 4 ("The hypothesis-based approach recognizes the complexities ofthe ecological responses being detected by the ... project-level monitoring and attemptsto capture the mechanistic interactions of multiple stressors rather than relying on a singlemetric to characterize ecological complexity." (internal citation omitted)).

207. PROGRAMMATIC EIS, supra note 11, at 7-40 to 7-42.

208. 33 C.F.R. § 385.35(c) (Westlaw 2007).

2007]


controversial), or misinterpellation (e.g., taking rhetoricaladvantage of facts presented in their scientific form with ananalysis of attendant uncertainties). Some political robustness canbe built into an adaptive management process through a detailed"uncertainty meeting," as ought to occur in any related joint fact-requisition process, as discussed infra Part V.

d. Adaptive management.

The Programmatic Regulations define "adaptive management"as a process for elaborating our understanding of the ecosystemand trying to improve the Plan in response to new information, toensure the Plan's success."9 Adaptive management can beunderstood as a means for testing management decisions: amanagement action is taken, and ecosystem responses aremonitored and studied so that the next management action has agreater chance of being well informed and effective. RECOVERrecognizes adaptive management as a strategy for dealing withuncertainty.210

Adaptive management in Everglades restoration is a legislativeand regulatory course-correction process that coordinatesindividual agency actions with their congressional mandate underthe Plan, as brain ffinctions responsible for walking motions maybe coordinated with those for navigation. As noted supra PartIII.C.4, the mandatory update of the Programmatic Regulations atleast every five years gives effect to the adaptive managementprocess.211

The Restudy team valued adaptive management not only as ameans for incorporating predictably evolving knowledge into long-term management, but also as a way to prepare for unexpectedcrises so that existing administrative regimes would not be

209. 33 C.F.R. § 385.3 (Westtaw 2007).210. See ADAPTIVE MGMT. STRATEGY, supra note 205, at 1 ("The overall purpose of

[adaptive management] is to substantially improve the chance of success in achievingecosystem goals when there is significant uncertainty about how this is to beaccomplished."). The independent scientific review recognizes this as well. See BIENNIALREVIEW, supra note 113, at 107-08 (finding the Adaptive Management Strategy a "soundorganizational model for the execution of a passive adaptive management program ... foraddressing uncertainty" and recommending a more active adaptive managementapproach to "reduce the likelihood of... management mistakes and.., the overall cost of... restoration").

211. WRDA 2000, supra note 6, § 601 (h) (3) (E); 33 C.F.R. § 385.6 (Westlaw 2007).

[Vol. 26:373


stymied. 12 The Restudy team realized that "there always will begaps in knowledge regarding the relationships within and amongnatural and social systems and that these information gaps requirethat plans be modified as technical knowledge improves and socialpreferences change";21

1 thus, they recommended adaptivemanagement in their attempt to avoid "developing a fixed goaland an inflexible plan to achieve the goal[.] '1" Because adaptivemanagement "involves an iterative process of developingmanagement tactics, [it] provides a process for breaking impasseswhere agencies are unwilling to proceed because of an inadequateknowledge base. 21 5 Also, since uncertainties of the magnitudesinvolved in large-scale ecosystem restoration allow manyopportunities for differences of opinion, adaptive managementbenefits from an associated dispute resolution process like thatrequired under WRDA 2000. 621

An associated dispute resolution process may be particularlyhelpful in adaptive management, because uncertainty can bepolitically exploited as discussed supra Part II.A.2.b.ii. For example,consider the ongoing controversy over the effects of inundation ontree islands. The National Park Service (the "Park Service") ischarged with protecting the part of the Everglades withinEverglades National Park (the "Park") as a "permanent wildernesspreserving essential primitive conditions including the naturalabundance, diversity, behavior, and ecological integrity of its floraand fauna. 21' To fulfill its mission, the Park Service has longsought more water from the Everglades upstream of the Park.2 8

The Park boundary circumscribes less than half of the extantfreshwater Everglades, so that in performing their official roles,agency scientists must focus on the water needs of a portion of theecosystem. This emphasis on only part of the ecosystem couldinterfere with the ability of agency scientists to participate

212. PROGRAMMATIC EIS, supra note 11, at 9-38.213. Id. at5-32.214. Id.215. Id. (internal citation omitted).216. ADAPTIVE MGMT. STRATEGY, supra note 205, at 7; WRDA 2000, supra note 6, §

601 (i).217. NAT'L PARK SERV., EVERGLADES NATIONAL PARK STRATEGIC PLAN 2001-2005 7

(2000), available at http://www.nps.gov/ever/current/strategicplan.pdf.218. U.S. OFFICE OF THE INTERIOR, OFFICE OF INSPECTOR GEN., MODIFIED WATER

DELIVERIES TO EVERGLADES NATIONAL PARK: AUDIT REPORT, REPORT NO. C-IN-MOA-0006-2005 8 (2006) [hereinafter AUDIT REPORT].

2007]


effectively in a multi-agency system-wide restoration process.In the post-C&SF Project hydrological regime, the water flows

desired by the Park Service can only be achieved by raising thewater levels in the Water Conservation Areas immediatelyupstream of the Park. Research indicates that additional floodingof these areas would be detrimental to their tree islands, which relyon lower levels of properly periodized inundation. Excessive,untimely flooding would impair not only the tree islands'functions as habitats, but also their potential functions as nutrientsinks."9

Two agencies within the Department of the Interior (DOI)disagree as to the effects of higher water levels on the tree islands:the United States Fish and Wildlife Service (FWS) takes theposition that higher water would damage the tree islands, whilethe Park Service "insists that the tree islands can survive withhigher water[.] ' 2 ° The DOI's Inspector General conducted anaudit of the DOI's participation in the Mod Waters project, findingthat the project has suffered delays and cost overruns for reasonsincluding that "[t]his argument has persisted for many yearswithout resolution." '221 Noting that an independent governmentbody-the Corps of Engineers-estimates that extant water depthsare already damaging up to 246 acres of tree islands per year, andthat the costs of their restoration could be as high as $123 million,the Inspector General found that the DOI needed a method fortimely resolution of such disputes.222 Meanwhile, optimizing thehealth of only the downstream part of the Everglades impoverishesthe ecosystem as a whole, so that an agency's smaller-scale missioneffectively frustrates the progress that the public desires inrestoring the ecosystem more broadly.

The tree-island controversy provides an example of how anagency may strive to advance its individual mission exclusive ofother objectives, such as its effects on the collective work ofecosystem-wide recovery. In such a case, an independent agency orother body could be employed as a source of neutral opinion,

219. See, e.g., Paul R. Wetzel et al., Maintaining Tree Islands in The Florida Everglades:Nutrient Redistribution Is the Key, 3 FRONTIERS ECOLOGY & ENV'T 370 (2005).

220. AUDIT REPORT, supra note 218, at 8.221. Id.222. Working in parallel, the independent scientific review recommends that "Mod

Waters should be completed without further delay." BIENNIAL REVIEW, supra note 113, at135.

[Vol. 26:373

] MAEUSIS THROUGH A GA TED MEMBRANE

effectively functioning as an oracle to be consulted in disputeresolution. Joint fact-requisition could be useful in resolving thistype of controversy.

B. What Are the Actors Doing?

In this Part, I apply the gated membrane model to the adaptivemanagement process for Everglades restoration under WRDA2000. Figure 4 represents the current state of the adaptivemanagement process in the form of the conceptual model.Comparison with Figure 2 indicates that although the science-producing structures are well developed, the gated membrane hasnot been clearly established. The fate of the technical reports thatare to inform the adaptive management processes that would takeplace on the policy side is uncertain. The technical reportproceeds to the sponsoring agencies to inform theirdecisionmaking, and to the public for viewing, but it need notnecessarily be used to facilitate consensus-building amongstakeholders in ecosystem management. There also appears to beno uniform process accessible to the stakeholders equally andjointly for developing requests for scientific research regarding theimplementation of particular projects or to address other jointlydeveloped concerns. Stakeholder rejection of research reported tothe policy side may result in requests for revision of the research,or for research designed to address the argument of a particularstakeholder against another; such improper research requests aredepicted in unshapely forms in Figure 4.

Figure 4 (on the following page) is the conceptual model of a

science-policy interface as applied to the current state of the

implementation of the adaptive management program pursuant to

§ 601 of WRDA 2000. Key components of the model shown in

Figure 2 are missing or not well established.

Most significantly, RECOVER is responsible for developing thePlan's overall adaptive management strategy." The perspicuouspurpose of this approach is to grapple with uncertainties thatmight otherwise cripple efforts to restore a dying ecosystem, byallowing its managers to learn from their attempts to save it."'

223. ADAPTIVE MGMT. STRATEGY, supra note 205.224. Id. at 1 (2006); see also PROGRAMMATIc EIS, supra note 11, at 5-33.

2007]


CD,

CD

k0o

[Vol. 26:373

MAIEUSIS THROUGH A GATED MEMBRANE

In effect, under adaptive management each attempt at.ecosystemrestoration is treated as an experiment, in that its projected effectsare to be evaluated exhaustively before its implementation, itsactual operation is to be as exhaustively assessed, and subsequent,prospective evaluations are to be informed by these assessments.The feedback mechanisms that advance learning also makescientific investigations more transparent to those on the policyside, since project managers communicate with scientists at variouspoints in the adaptive management process.225

The first update of the ecosystem-wide mathematical model ofEverglades hydrology offers an initial example of RECOVER'sperformance in the adaptive management process. Since thehydrological conditions of the pre-drainage Everglades are lost inthe flow of time, the version of the mathematical model that wasused in the original Plan relied on rainfall data from 1965 through1995 as a baseline for the input of water to the hydrological system.When the model was updated to include rainfall data through theyear 2000, changes also were made in response to newtopographical data. The sponsoring agencies performed atechnical overhaul of the model pursuant to the ProgrammaticRegulations using a variety of new data, and RECOVER produceda technical report evaluating the Plan based on the revisedmodel.226

Clearly, this update improves the programming of the modeland captures with greater accuracy more of the actual hydrologicinputs to the system, though rainfall from the currently measuredperiod still is not expected to coincide perfectly with rainfall datathat would have been obtained in the pre-drainage Everglades, northat could be measured in the future. Programming artifacts andthe potential differences between unmeasured and recentlymeasured rainfall are some of the known sources of uncertainty inour understanding of Everglades hydrology.227 Rather than decide

225. See ADAPTIVE MGMT. STRATEGY, supra note 205. Multiple feedback loops in theadaptive management process are mapped at an overview level in Figure 1 of the AdaptiveManagement Strategy, and in greater detail in Figures 2-5. See id. at 2-6.

226. Heisler Interview, supra note 27.

227. PROGRAMMATIC EIS, supra note 11, at 5-29. Regarding topographicalrefinements, see id at 10-16 ("Additional data will need to be collected to further designthe 'next' tools needed to implement the Comprehensive Plan. These data will includeitems such as topographic ... data. The southern portion of Florida has unique features

such as a very flat topography; a large, highly concentrated human population; and a veryunique and fragile ecosystem. Because of this flat topography . . . [t]he current lack of

2007]


that nothing can be done to restore the ecosystem becausemodeling can always be improved and additional actual rainfalldata are unavailable, scientist-managers obtain results from thebest available model run with the best available information.Though this approach is reasonable and improvements should bematters of course, agencies that are displeased with modelingresults may be tempted to attack the uncertainties associated withmodeling rather than to reevaluate their project designs.

Evaluation of the Plan with the updated model yielded resultsthat did not meet the original expectations for the Plan. 22

' Theresults revealed a need for refinements in the plannedconfigurations of structures and operations that had been thoughtto optimize restoration based on the older model, run with morelimited rainfall data, which had been approved by Congress. Thisplanning revision in response to new information is exactly whatthe adaptive management approach is designed to accomplish. Asdiscussed in the legislative history of WRDA 2000:

The committee recognizes the levels of uncertainty involved inthe Plan and fully intends for the adaptive assessment andmonitoring process to account for such as the Plan is executed.... The committee does not expect rigid adherence to the Plan

as it was submitted to Congress. This result would be inconsistentwith the adaptive assessment principles in the Plan. Restorationof the Everglades is the goal, not adherence to the modeling onwhich the April, 1999 Plan was based. Instead, the committeeexpects that the agencies responsible for project implementationreport formulation and Plan implementation will seekcontinuous improvement of the Plan based upon new information,improved modeling, new technology and changed circumstances.229

Congress recognized that unexpected results represent neitherfailure nor institutional incompetence; the adaptive managementapproach attempts to expect the unexpected.23 Changes in project

precision in existing vertical control can result in erroneous estimates to importanthydrologic variables.").

228. Heisler Interview, supra note 27.229. WRDA 2000 SENATE REPORT, supra note 31, at 41 (emphasis added).230. PROGRAMMATIC EIS, supra note 11, at 9-38 to 9-39 ("Adaptive assessment is a

process for learning, and for incorporating new information into the planning andevaluation phases of the restoration program. Adaptive assessment is valuable in that ittreats all responses, expected or not, as major learning opportunities. An unexpectedresponse does not represent a failure for the program if it can be used to substantiallyimprove our understanding of a complex system.").

[Vol. 26:373

MIEUSIS THROUGH A GA TED MEMBRANE

planning called for by the updated model should be welcomed asopportunities to learn, and for implementing agencies to practiceoperating in an adaptive management framework.

The new adaptive management approach is much more front-loaded than traditional agency action, in which projectconstruction precedes assessment and error-correction.231 Like themastery of any unfamiliar skill, developing expertise in adaptivemanagement takes practice, and some agencies may have greateraptitude for it than others. For example, when RECOVERproduced the new modeling results that seemed negative,management at the sponsoring agencies proposed delaying thereport. In response to this proposed delay, an RLG member statedthat the RLG-rather than a sponsoring agency-approvesRECOVER reports for release to the public. Although RECOVERhas authority only over itself,2"2 RECOVER was able to protect theintegrity of its work by refusing to withhold or revise any of itspublications, all of which are available to the public pursuant tothe Programmatic Regulations.2 3 Were it not for this assertion, it islikely that the report would not have been released until thesponsoring agencies had revised the document.3 4 This citation toproper procedure-to the gated discursive membrane-protectedRECOVER's science production and publication protocols, whileensuring undistorted transmission of the research results thusprotected to the decisionmaking public.

A proper request for RECOVER science takes the form of ascientific question that has not yet been addressed in one of itsearlier reports, or of an update extending an earlier report. Theonly proper bases for revision of a report would be clearlydemonstrated errors that any professional scientist wouldrecognize as such.23' If a request for document revision can bereformulated as a request for technical assistance with a newmatter, then it may prompt not a revision, but the production of anew document reporting new information. However, individualagencies have the option of producing their own reports on topics

231. ADAPTIVE MGMT. STRATEGY, supra note 205, at 3.

232. See 33 C.F.R. § 385.20(b) (Westaw 2007) (RECOVER work products "shall notbe not self-executing, but shall be provided as information for consideration" by thesponsoring agencies.).

233. Id. ("Technical information developed by RECOVER shall be available to thepublic.").


235. Id.

2007]


that RECOVER has addressed; in such a situation, the greaterprestige of the multidisciplinary, multi-agency group may confergreater authority and persuasiveness on the RECOVER document,but the individual agency's interpretations will also be recorded.Each management professional must perform his duty to avoiddevaluing science, carefully avoiding misinterpellation ormiscitation that creates the appearance of uncertainty or scientificcontroversy where only political controversy exists.

Government agencies have different missions and accordinglydevelop differing expertise. Ideally, the work of diversegovernment agencies is complementary. In practice, however, thepriorities of an agency can interfere with its ability to cooperatewith other agencies, as discussed supra Part IV.A.2.d regarding thetree-island controversy. Joint fact-requisition, as discussed infra PartV, may be effective in a dispute in which each side resorts toscience for authority in support of its position, though joint fact-requisition could fail if a stakeholder ultimately decides to rejectscientific findings that disfavor its position.

In addition to protecting its professional integrity, RECOVERhas succeeded in contributing to Plan implementation strategies atthe system-wide scale. For example, RECOVER has producedseveral significant system-level technical documents, such as theinterim goals for use in evaluating the effectiveness of the Plan inecosystem restoration as required by § 601 (h) (3) (C) (i) (III),236 theinterim targets required by the Programmatic Regulations,237 andthe Monitoring and Assessment Plan.2 3

' Through their respectiveagencies, members of the RECOVER Group have contributed tothe development of the Programmatic Regulations and of the Planitself.

9

In contrast, RECOVER has remained largely uninvolved inplanning at the level of individual projects. In part, this occursbecause RECOVER's system-wide evaluation methods may not besufficiently sensitive to reflect the smaller-scale effects of

236. RECOVER, THE RECOVER TEAM'S RECOMMENDATIONS FOR INTERIM GOALSAND INTERIM TARGETS FOR THE COMPREHENSIVE EVERGLADES RESTORATION PLAN (2005),available at http://www.evergladesplan.org/pm/recover/igit-subteam.cfm.

237. Id.238. 33 C.F.R. § 385.20(e) (1) (ii) (Westlaw 2007); RECOVER, CERP MONITORING

AND ASSESSMENT PLAN: PART 1 MONITORING AND SUPPORTING RESEARCH (2004), availableat http://www.evergladesplan.org/pm/recover/recover-map.aspx.


[Vol. 26:373


alternative configurations for a particular project.24 However,RECOVER's insufficient involvement in project planning alsooccurs because the Guidance Memoranda required by theProgrammatic Regulations remain incomplete (leaving projectmanagers without detailed guidance), and because the priorities ofproject managers are strongly driven by project deadlines andbudgetary concerns.241 To implement adaptive management at theproject level, the professional performance evaluations of projectmanagers should depend in part on their success in incorporatingRECOVER recommendations.

C. Direction for a Better Play

Applying the gated membrane model to engineering thenascent adaptive management program for Everglades restorationpursuant to WRDA 2000 would result in a more effective science-based management program and contribute to the success of thePlan. The model offers science-policy professionals responsible forregulatory implementation a map orienting them to the situationof their performance(s) as transmembrane actors at a science-public decisionmaking interface. In addition, the followingrecommendations flow from this interfacial analysis.

1. To ensure that science actually informs implementation.

RECOVER is assigned a lead role in adaptive managementthrough its responsibility for providing managers with thescientific bases for decisionmaking. Thus, ensuring that scienceactually informs Plan implementation entails ensuring thatimplementing managers actually read and consider RECOVERtechnical documents. Tying the performance reviews of managersand agencies to their responses to RECOVER documents providesa means of measuring RECOVER's effectiveness as well as anincentive for the science-based decisionmaking required forProject Implementation Reports by § 601 (h) (4) (A) (iii) (VII) ofWRDA 2000.

A complement or alternative would be to require RECOVER toreview and approve the plans that managers have produced afterreceiving RECOVER input, to ensure that the goals of the Plan arenot being defeated by failures to incorporate technical

240. Id.241. Id.

2007]


information that could result from an implementing agency's rushto meet project deadlines. Allowing RECOVER to veto plans mightinappropriately begin to extend RECOVER's authority into thepolicy side. However, RECOVER may be allowed to give plans ascore based on the scientific soundness of their approach tomeeting the Plan's goals, and agencies and individual managerscould be rewarded based upon the scores earned by their work.

2. To improve project-level information flow from the policy side to thescience side.

Notably, the adaptive management process so far lacks amechanism for attending to policy issues that emerge whentechnical teams try to address individual projects. TheseRECOVER subteams need responses from the policy side toqueries such as: should we treat all kinds of ecosystem servicesequally and try to optimize them jointly, or should we prioritizecore areas while other areas suffer? Are we to prioritize theecosystem services of tree islands or of water lilies?242

The model suggests that to perform the needed function, aRECOVER analog should be created on the policy side of themembrane, as shown in Figure 2. This policy organelle would becharged with developing and articulating a comprehensive policyperspective for overall ecosystem restoration, and applying thatpolicy perspective to management decisions at the project level.RECOVER can advise the public regarding what is expected tohappen to a given ecosystem service provider depending on how itis prioritized, but it is the role of the public, and not of its scientificadvisors alone, to determine which services it cares most to restoreand sustain when tradeoffs are inevitable.

The System Planning and Operations Team (SPOT) describedin RECOVER's Adaptive Management Strategy could be framed asa policy-side organelle that analyzes scientific, technological,political, legal and budgetary inputs to develop qualitative optionsor assessments for the sponsoring agencies to consider inimproving the Plan.243 The SPOT is designed to complementRECOVER's efforts when the sponsoring agencies must generatedocuments under NEPA or make policy decisions: the SPOT isassigned the role of "provid[ing] the vital link between

242. Id.243. ADAPTIVE MGMT. STRATEGY, supra note 205, 5-6, Fig. 4.

[Vol. 26:373


RECOVER's work [on the one hand] and policy and/or NEPAprocesses [on the other]."244 When NEPA processes are triggered,RECOVER is unable to act, because the sponsoring agencies retainthe responsibility of fulfilling NEPA requirements.

The "link" mentioned in the quote above apparently refers tothe integration of RECOVER technical reports in planning, butdoes not actually entail the bidirectional communication betweenthe science and policy sides that would be afforded by the additionof a transmembrane actor on the policy side, analogous to an RLGmember, as shown in the upper part of Figure 2. While RECOVERis to "elevate" issues that affect planning at the overall system levelto the SPOT,245 there is no transmembrane actor who structuresand facilitates communication between the science and policysides over smaller-scale rna.tters. Thus, a necessary functionality ismissing from the science-policy interface.

Considering its obligation to protect the integrity of scientificdecisionmaking processes, it is fitting that RLG authority extendsonly over the RECOVER organelle on the science side; providingthe RLG with a decisionmaking function on the policy side maycompromise the separability between science-production andmanagement functions. The adaptive management processencompasses adaptation to new management information as wellas to new scientific information.246 Thus, it should be possible forthe Programmatic Regulations to be augmented to provide for apolicy-side organelle analogous to the RECOVER Group of thescience side, as well as for transmembrane actors based on thepolicy-side (analogous to RLG members), who could work with theRLG to improve the flow of project-level information from thepolicy side to the science side.

3. To provide incentives for individual agencies to participate incollaborative management.

Incentives for individual agencies or agency scientists toparticipate in collaborative processes like the RECOVER Group

244. Id. at 5.245. Id. at 8.246. The Programmatic Regulations are "to ensure that . . . information that is

developed through the principles of adaptive management contained in the Plan, orfuture authorized changes to the Plan are integrated into" Plan implementation, alongwith new scientific or technical information. WRDA 2000, supra note 6, §601 (h) (3) (C) (i) (II).

2007]


could be improved. Participation in itself does not guaranteegreater influence in the overall process to any participatingagency. Nor should it, in that as discussed supra, RECOVERimportantly serves to balance and integrate individual agencymissions with a focus on ecosystem-wide restoration. However,since the wholehearted participation of individual agencies in theoverall effort is needed,247 something should be offered to thosewho participate, and more to those who participate well and in aspirit of collaboration. Perhaps individual agencies' missions couldbe expanded to include participation in multi-agencycollaborations for large-scale ecosystem management.

The Plan envisioned that agencies would be encouraged toadjust their approaches to project design and implementation byvirtue of participation in an adaptive management process.24 s

While the Inspector General's Report criticizes agencies for failureto collaborate, the rewards of collaboration for a contributingagency remain unclear. Even if increasing the access of successfullycollaborating agencies to staff, funding or other centralizedresources (perhaps through the sponsoring agencies) werepossible, such an approach may be problematic, in that acollaborative process that came to be dominated by a particularparticipant may more likely reflect the dominating agency's goalsrather than those of the collective. Instead, providing additionalresources for general use to individual agencies which, after a longperiod of participation (or at the end of the collaborative process),have demonstrated sustained efforts to contribute to collaborativescience-based management may become an effective means ofpromoting agency cooperation. Providing such incentives toindividual agencies would promote the adaptive managementprocedure's robustness to political spoilers.

247. See, e.g., BIENNIAL REVIEW, supra note 113, at 72 ("[S]uccess will requirecooperation among a disparate group of organizations with differing missions as the broadgoal of getting the water right is more precisely defined.").

248. See PROGRAMMATIC EIS, supra note 11, at 9-39 ("Overall, adaptive assessmentprovides a much needed framework for... encouraging agencies to be flexible about thedesign and implementation of the Comprehensive Plan. The bottom line is that adaptiveassessment substantially improves the probability that a complex, regional ComprehensivePlan will be successful, by providing a structured, well-focused process for evaluating andrefining the design and performance of that program on a continuing basis throughout itsimplementation.").

[Vol. 26:373


4. To ensure timely reauthorization and funding.

As discussed supra Part IV.B, the cooperative adaptivemanagement process is front-loaded in that multiple agencies willengage in iterative planning before project construction begins.Furthermore, project completion may be slowed if assessmentsindicate that changes should be made while implementation is inprogress. Meanwhile, the adaptive assessment and monitoringprogram, including RECOVER, requires reauthorization after itsfirst ten years, and multi-year agency contracts with participatingbusinesses may depend on the allocation of funds for specificprojects that Congress has yet to approve. Just as actual cells needenergy to maintain their structure and function, the modeledscience and policy cells need a steady stream of input resources tokeep their science-based public decisionmaking functions alive.

In an initial construction schedule for Plan components, it wasassumed that a Water Resources Development Act would be passedevery other year."' Congress has enacted no Water ResourcesDevelopment Act since the year 2000. The nascent momentum ofthe Plan's adaptive assessment and monitoring program-andpotential Everglades restoration under the Plan-depend ontimely renewal of Congressional authority and funding.2z0

V. AN INTERFACIAL ANALYSIS OF SCIENCE-BASED DISPUTE RESOLUTION

I use the conceptual model to analyze the role of science injoint fact-finding, a method of science-based dispute resolutionoften used for environmental matters. As a result, I propose thedevelopment of joint fact-requisition, a procedural structure for thedialogue between stakeholders and scientists regarding thepotential effects of the stakeholders' decisions on a scientificallyunderstood matter. Joint fact-requisition emphasizes theimportance of distinguishing between the stakeholders and thescientists involved in negotiation or collaborative decisionmaking.I then explore the potential for incorporating joint fact-requisitionexplicitly into a legal and regulatory framework for the adaptivemanagement of an environmental matter, offering specificrecommendations for the use ofjoint fact-requisition in Everglades

249. PROGRAMMATIC EIS, supra note 11, at 10-44 to 10-45.250. Indeed, the independent scientific review states that if federal funding for the

Plan does not increase, major restoration projects may not be completed timely. BIENNIALREVIEw, supra note 113, at 72.

2007]


restoration under WRDA 2000.

A. Joint Fact-Finding

The term 'joint fact-finding" refers to a consensus-buildingmethod of alternative dispute resolution in 'which stakeholdingparties work together to use scientific research to inform theirnegotiation and promote collective decisionmaking. 251 As recentlydeveloped and formalized, joint fact-finding can be an importantstep in an effective approach to dispute resolution forenvironmental matters that, at its most successful, can alsoestablish institutions for dispute prevention and ongoingcooperative management.

252

Currently, joint fact-finding processes seek to involvestakeholders in scientific research as much as possible. Ajoint fact-finding process may happen to involve scientists conductingresearch somewhat independently and educating stakeholdersabout the environmental matter at stake based on their researchresults. However, the goal is for the stakeholders themselves toconduct research, or at a minimum to agree on research methodsand interpretations, and to oversee the work of scientists-ideally,stakeholders are to be involved in scientific research at all stages,from defining research questions and designing studies toanalyzing and interpreting data. Involving stakeholders inproducing scientific facts is expected to increase the transparencyof the dispute-resolution process to the stakeholders, therebyincreasing their trust in the process so that it will more likely resultin an agreement.254 Stakeholder involvement is also expected to"[i]ncreas[e] the legitimacy and salience of research[.] 1255

The joint fact-finding process helps negotiating parties to avoidspending their resources on a battle of experts, in which individualstakeholders commission research into questions that emphasizetheir peculiar interests and often return to the table with

251. John R. Ehrmann & Barbara L. Stinson, Joint Fact-Finding and the Use of TechnicalExperts, in THE CONSENSUS BUILDING HANDBOOK 375 (Lawrence Susskind et al. eds., 1999).

252. See MIT-USGS Science Impact Collaborative (MUSIC) Website,http://web.mit.edu/dusp/epp/music/index.html (last visited Apr. 18, 2007). MUSICconducts research on joint fact-finding.

253. See Consensus Building Institute Website, Strategies, at 2 (formerly available athttp://www.cbuilding.org/projects/music/strategies.pdf, on file with the author).

254. See id. at 1.255. See id

[Vol. 26:373

] MAEUSIS THROUGH A GA TED MEMBRANE

conflicting understandings of the natural phenomena at issue.Instead, a neutral mediator works to ensure that all stakeholdersgain equal access to scientific information pertinent todecisionmaking, facilitating stakeholder education on morecomplex topics. 256 Stakeholders' joint efforts to learn about theenvironmental matter at stake promote their cooperation inreaching agreement, both in that stakeholders begin to cooperateas they develop technical knowledge, and in that their negotiationis based on more similar understandings of relevant technicalmatters.

While this approach offers many advantages as a means ofdispute resolution, stakeholder involvement in scientific researchcan be problematic for the resulting science. If the boundarybetween scientific fact-finding and dispute-resolution fact-findingprocesses is not intact, the legitimacy and value of scientificresearch in dispute resolution is in danger of compromise.

Despite its reliance upon scientific fact-finding, "fact-finding"in joint fact-finding more closely resembles fact-finding in otheralternative dispute resolution processes or in litigation. Even whenthe scientific facts are actually found by professional scientists, as isthe case when stakeholders jointly opt to commission theperformance of scientific research by independent scientists, thestakeholders are said to engage in joint fact-finding.5 7 Meanwhile,the scientists advising the process perform two roles: that ofconducting scientific research, and that of communicating withthe stakeholders about the research.

In performing these dual roles, scientists are interpellated intodifferent discursive structures that require them to perform inmodes that do not necessarily cohere. For themselves and othermembers of their scientific audiences, they must perform researchin a manner consistent with scientific norms; for example,research methods should be chosen to optimize the scientific valueof the results, as may be accomplished by combining approachesthat complement each other's insensitivities, and not bygenerating data using a single method that is most likely to favor agiven theory. If a scientist is simultaneously performing for

256. See Consensus Building Institute, Joint Fact Finding: Key Steps in the Process (2005),available at http://www.cbuilding.org/projects/music/5_JFFKeySteps2005.pdf.

257. See MIT-USGS Science Impact Collaborative, What Is Joint Fact-Finding?,http://web.mit.edu/dusp/epg/music/wwd/jointFactFinding.html (last visited Apr. 18,2007).

2007]


stakeholders, the scientist may also feel obliged to answer to thestakeholders by providing results tailored to their interests, ratherthan having the independent authority to communicate withstakeholders in a way that does not allow the mechanics of theresearch process to be dictated by stakeholding non-professionals.This potential pressure to perform in a scientifically inappropriatemanner in order also to perform as a participant in disputeresolution is strongest when a scientist's research is too closelydirected by stakeholders. The professional scientific actor can bestparticipate in a joint fact-finding process if she is clearlyunderstood by all of her audiences to function as an advisor to thedispute-resolution process as a whole, conducting research in adistinct scientific arena and communicating about the researchthrough a procedural membrane that is gated by the mediator. Itshould be clear that producing science is not necessarily a jointfact-finding process, and it is not the joint fact-finding process thatwill resolve the dispute. 25

1

Moving from consideration of the professional scientist's actsto the facts he produces, note that two kinds of facts matter inscience-based dispute resolution: the scientific research resultsproduced by the scientists, and the dispute-resolution kind of factused in political decisionmaking. Even when the same piece ofinformation is the locus of both kinds of fact (and in an ideallyinformed political decisionmaking process, it would be), the factbears distinct scientific and dispute-resolution aspects. Forexample, consider a scientific research finding that a tree islandcan tolerate only a certain range of inundation levels for a givenduration, qualified by attendant uncertainties. In a politicaldecisionmaking process, this information might matter only in theoverstated form, "do this, and tree islands will be fine" or "do that,and we will lose tree islands." As the scientific fact freezes into adispute-resolution fact, its grey areas become black or white. To ascience professional accustomed to the gradual sedimentation of

258. When a dispute occurs among scientists, then scientists are parties to (i.e.,stakeholders in) a dispute. In the process of producing science, scientists typically resolvedisputes through methods such as peer review, as described supra Part II.A.1. If scientistswere to engage in joint fact-requisition to resolve a dispute among themselves, they wouldperform joint fact-requisition when they convene to decide which issues would beaddressed and which other scientists would conduct the investigation. The other scientistswho actually carry out the investigation would be advising the joint fact-requisition process,but would not themselves be engaging directly in joint fact-requisition while conductingtheir scientific research.

[Vol. 26:373


scientific research results into facts as they remain subject toscientific testing, this transformation can seem sudden andarbitrary.

In practice, it has been profoundly difficult to engageprofessional scientists in joint fact-finding, for reasons beyondtheir unfamiliarity with such processes.2 59 I propose that thisdifficulty arises from the scientists' maieutic ethic: scientists areconcerned that at worst, contact with political decisionmakers mayimproperly influence their work, and at best, their reputationsmight be tarnished because contact with stakeholders will besuspected to have influenced their work. A joint fact-requisitionprocess in which it is openly known that scientists advise the process(rather than any particular stakeholder) through a neutral canprotect scientists' work and reputations while enabling them toparticipate in a dispute" resolution, dispute prevention, ormanagement process.

Consider a professional scientist asked to involve a recreationalorganization in research on optimal water levels in a lake. What isconsidered an "optimal" water level depends on the goal ofoptimization. For instance, a professional ecologist answering thequestion as independently as possible (with public funding forbasic ecological research, for example) would likely develop atheory for optimizing overall ecosystem health in relation to lake-water levels, and would report her findings and projectionsaccordingly. In contrast, an ecologist working with a sport-fishingclub to answer the question will likely develop a theory foroptimizing the populations of particular game fish in comparisonwith water levels, and report her findings and projectionsaccordingly. It would not be surprising for scientists working withdifferent stakeholders to develop different results-thisphenomenon characterizes adversarial science, to which joint fact-finding seeks to offer an alternative.

Even science that is conducted not for a single stakeholder, butfor a group of stakeholders, remains problematic. Unless thegroup of stakeholders in negotiation represents every interest,including those of the public at large, future generations, and thevalue of learning about an environmental matter for its own sake,research results sought by the group reasonably may be expected

259. Telephone interview with Lawrence Susskind, Founder and Senior Advisor,Consensus Building Institute (Mar. 29, 2006).

2007]


to reflect a subset of interests more than any independentproperties of the natural system. 2

1 Indeed, the objectivity-claimingability of science is greatest when scientific research is produced inas disinterested a manner as possible. However, the goal of adispute-resolution process is not .to develop an objectiveunderstanding of the natural world per se, but to promote thesettlement of a dispute. Thus, the facts produced in joint fact-finding are developed for their dispute-resolution value, and notfor their scientific value-though ostensibly, the former dependsfrom the latter.

In addition to any actual bias arising in the science fromintensive stakeholder involvement in its production, suchinvolvement may reasonably form the basis for perceived bias.Even if the scientific research results were to correspond well tothe natural system's properties, a scientist engaging in researchdriven by a non-professional stakeholder group may sufferreputational degradation due to the possibility of bias in herresearch arising from the management of her work by a non-professional. For these reasons, it would be understandable for aprofessional scientist to shy away from participation in joint fact-finding.

Joint fact-finding processes remain vulnerable both toproducing and to being informed by biased or unprofessionalscience. Note that a joint fact-finding process may be highlysuccessful in dispute resolution, even if agreement is developed onthe basis of faulty science: as discussed supra Part II.B.2.a withrespect to the ability of potentially imperfect model results to serveas bases for consensus, a fact may have a dispute-resolution valuethat is greater than its scientific value.

The current joint fact-finding formalism is represented in aninset to Figure 5. Scientists are among the stakeholders: while they

260. A recent independent scientific study of the ailing Sacramento-San JoaquinDelta suggests that ineffectiveness of CALFED, a stakeholder-driven effort to manage thisvast, multiple-use wetland that is in ecosystemic decline, results in part from this problem.SeeJ. LUND ETAL., ENVISIONING FUTURES FOR THE SACRAMENTO-SANJOAQUIN DELTA 10-11(2007) (The independent scientists "purposely [took] a broader view of the [watermanagement] options than those commonly under discussion in stakeholder circles[,]"because options acceptable to the stakeholder group were not sustainable.); id. at 157("[Our] approach ... differs from the underlying assumption of CALFED that everyonecan 'get better together.' Stakeholders whose land and water interests cannot be directlysatisfied may be compensated by financial or other means. Even with such mitigations, onecannot reasonably expect universal satisfaction" if the wetland is to be managedsustainably.).

[Vol. 26:373


help other stakeholders to develop information as a basis foragreement, scientists lack any explicit authority to limit the field ofacceptable agreements based on their environmental effects. Infact, scientists may be subordinated to "other" stakeholders, whooversee and limit their research. 261 A goal ofjoint fact-finding is "toensure that an appropriate balance between science and politicscan be maintained[,]" rather than to use science to define limitswithin which dispute-resolution agreements have tolerable effectson the environment.

262

In other words, the role of a scientist in dispute resolutionshould be that of an advisor to the process. This advisor must beable to perform scientific research with at least sufficientindependence to exercise his professional judgment. Otherwise,the research results are not likely to be good science, nor a worthybasis for public decisionmaking. The independent researchproduced should matter in the process-it must be able tocontribute to public decisionmaking by bounding the universe ofacceptable stakeholder agreements. Otherwise, even good scienceproduced by the public decisionmaking process will bemeaningless in that process, and fail to materialize in its outcome.

This point bears emphasis. In discussion of joint fact-finding asit is currently practiced, science has been described as one interestamong many, and scientists as stakeholders around the negotiation

261. See Consensus Building Institute Website, Florida Everglades [TIechnical[W]orkshop on [Hiabitat [Rlestoration (formerly available at http://www.cbuilding.org/projects/everglades/index.html; on file with the author) (reporting that agency "scientistsand other stakeholders" participated in a recent (2005) workshop on accelerating therestoration of regions of the remaining Everglades affected by excessive nutrients(emphasis added)); Consensus Building Institute, Joint Fact Finding: Key Steps in the Process(2005), available at http://www.cbuilding.org/projects/music/5_JFFKeySteps2005.pdf(describing stakeholders as performing scientific research functions typically performedby professional scientists); see also Herman A. Karl et al., A Dialogue, Not a Diatribe, 49 ENVr.20 (2007). In the recent article by Karl et aL, a new awareness of the problems associatedwith treating scientists as (or subordinating them to) stakeholders appears to be modifyingthe authors' views; they state that a "well-designed and managed joint fact-finding]process does not result in 'science by committee' or allow science to devolve to lowestcommon denominator thinking." Id. at 23. However, the authors fail to explain how suchdevolution may be prevented. Elsewhere, they indicate that the mediator must ensure that"all stakeholders-including scientists and technical experts" engage in conversation thatshould not end with the scientists' technical report, though the scientists "should probablynot advocate a particular policy outcome." Id at 24 (emphasis added). Thus, the scientists'role and distinction from the stakeholders remains unclear in joint fact-finding.

262. See MUSIC, What [I]s MUSIC? (formerly available at http://www.cbuilding.org/projects/music/index.html; on file with the author) (emphasisadded); Karl et al., supra note 261, at 23.

2007]


table. A canonical introduction to joint fact-finding indicates thateitherjointly commissioned experts or the stakeholders themselvesmay perform the scientific fact-finding if they have the expertise.263

However, scientists conducting research on the part of astakeholder may be or may appear to be biased in favor of theiremployer. Three recommendations are offered regarding how todesign a work group that will perform the scientific fact-finding:(1) all stakeholders jointly are to articulate the group's goals; (2)the scientific fact-finders collectively are to represent the interestsof all stakeholders, and (3) the scientific fact-finders are to haveadequate time and funds for their work.2

1 It is clearly necessary tothe success of a joint fact-finding process for the stakeholders toparticipate in it equivalently, and for its consulting scientists tohave sufficient time and resources to conduct the requiredresearch. However, while the equal representation of allstakeholders among the scientific fact-finders would be preferableto their unequal representation, the association of individualscientific fact-finders with particular stakeholders remainsproblematic. The recommendation against unequalrepresentation of the stakeholders in the scientific fact-findingwork group reflects the understanding that unintentionalsubjective influence (or even advocacy science) is possible-thatthe questions the researcher asks and the methods she chooses foraddressing those questions can affect the results of scientificresearch, just as interpretation can construe results variously. Ifthis were not the case, the membership of the scientific fact-finding work group would not matter, because the resultsgenerated would be robust to the scientists' affiliations.

To remove the likelihood or appearance of bias among thescientific fact-finders, I recommend that the scientists engaged asadvisors to a joint fact-finding process be unaffiliated with anystakeholder. This important structural feature of thedecisionmaking process is captured in the model for joint fact-requisition, discussed infra Part V.B. Where such independencefrom the stakeholders is not possible, at a minimum the scientistsfrom the stakeholding institutions should not be among thoseadvocating their institution's position in negotiation. Rather, theyshould be decoupled from the political process to the extentpossible, perhaps through the use of an ethical wall between the

263. Ehrmann & Stinson, supra note 251, at 386, 398.264. Id. at 386.

[Vol. 26:373


consulting scientists and their dispute resolution-side colleagueswithin the stakeholding organization for the duration of theiradvisorship to the collective group of stakeholders. A boundaryand structured communication process such as joint fact-requisition must be established to mitigate any tendency towardbiased performance on the part of the professional scientist, orperception of bias by the scientist's audiences-to protect theintegrity of the science in a science-based dispute-resolutionprocess as well as the participating scientists' reputations.Consultation of independent scientists in the joint fact-requisitionstep of a dispute resolution or management process also helps toprevent the scientific fact-finding process from becoming amicrocosm of the stakeholders' negotiation process on the publicdecisionmaking side. That is, the gated membrane keeps theprocess from emulsifying: requisitioning the scientific facts fromindependent scientists helps to prevent the carefully structuredjoint fact-requisition process from devolving into a wasteful andcounterproductive battle of experts.

Thus, 'joint fact-finding" would more properly refer tostakeholders jointly finding facts of the dispute-resolution form, andnot to scientists finding facts under their commission. Thepossibility that scientists and stakeholders will work together tofind scientific facts troubles the scientists' maieutic ethic. A scientistconsidering participation in a joint fact-finding process andconstruing 'joint fact-finding" correctly to refer to the finding ofscientific facts may well be scrupulous about opening hislaboratory to nonprofessionals who will want him to change theway he conducts research, hoping for results that support theirfavored outcome. To perform his boundary-maintenance functionfaithfully, he must do his best to prevent stakeholders frominfluencing or otherwise accessing the scientific research outsideof the dispute-resolution procedure-from tinkering withprocesses in the science cell directly, without going throughproper channels. To signal clearly the proper roles of stakeholdersengaged in science-based dispute resolution, I use the term 'jointfact-requisition."

B. Joint Fact-Requisition

The joint fact-requisition process introduces a gatedmembrane as a formal structural distinction between stakeholdersand scientists, thus preserving the advantages of joint fact-finding

2007]


in a dispute-resolution process while mitigating its disadvantages asa scientific fact-production process. To emphasize this distinction,I call the structured process 'joint fact-requisition," becausestakeholders jointly request scientific facts from professionalscientists on the science side, and then use those as a basis fordeveloping dispute-resolution facts on the public decisionmakingside (in this case, the dispute resolution side) of the gatedmembrane.

1. Mise-en-scene.

A joint fact-requisition process can be framed in terms of thegated membrane model, as shown in Figure 5. Here, the publicdecisionmaking side is an alternative dispute resolution cell, andthe mediator is a transmembrane actor. Originating on the publicdecisionmaking side, the mediator brings the stakeholderstogether and serves as a repository for expressions of theirconcerns. Moving to the membrane, he works in consultation withthe stakeholders and the scientists to formulate the stakeholders'concerns in scientifically tractable terms to the extent possible.Stakeholders and scientists continue to function on theirrespective sides of the gated membrane, while the mediator bothpromotes and structures their repeated interactions so that therole of each is clear to all. Since the mediator is not charged withperforming scientific research, his neutrality is not compromised:he remains able to serve the stakeholders equally on the disputeresolution side, while facilitating communication between thestakeholders and their scientific advisors at the interface.

Figure 5 (on the following page) shows the conceptual model of ascience-public decisionmaking interface as applied to joint fact-finding (inset) to yieldjoint fact-requisition.

More specifically, a professional mediator consults all of thestakeholders (iteratively, individually and jointly) in developing asingle list of questions for coherent scientific investigation, and asingle list of independent scientists whose conduct of the researchwould be acceptable to all; the research is jointly commissionedand funded. While stakeholders and scientists remain instructured contact as research is performed, decisions integral tothe scientific fact-finding process are made by professional

[Vol. 26:373

2007] MAIEUSIS THROUGH A GATED MEMBRANE 461

0

00

rC

00-.

oil.

0 \

-- o;

aMD

' [21

" ;;

Q..


scientists, who conduct appropriate scientific research. Scientistseducate stakeholders regarding applicable research methods andthe nature of the resulting data, so that the stakeholdersunderstand what kinds of questions science can and cannot answerfor them, and what kinds of uncertainties arise from both thenature of the matters at stake and the sensitivities of applicableresearch methods. In turn, stakeholders teach scientists abouttheir goals and concerns, so that the scientists realize what kinds ofinformation would be most helpful in public decisionmaking.When scientists report their results to the stakeholders, theyfurther advise the stakeholders regarding which interpretations ofthe data are and are not well supported, and what kinds ofadditional research could address remaining or new concerns.

As in the current practice of joint fact-finding, contact betweenthe scientists and the stakeholders can maximize the relevance ofscientific research efforts to problem-solving and promoteappreciation of and trust in the scientific process. Trust amongstakeholders is also fostered, as they work from the outset to unifytheir questions about nature and together develop anunderstanding of how it will respond to their actions. In particular,working with the scientists, the mediator ensures that all of thestakeholders become equally informed about the sources andkinds of uncertainty attending the science they commission, sothat their decisionmaking will be more robust to the rhetoricalmanipulation of uncertainty. Once their informed decisionmakingregarding the original dispute is complete, stakeholders andscientists may opt to continue to work together to manage theresource in question, preventing future wasteful disputation.Indeed, such an ongoing institution may be viewed as asedimented process, a process that persists after having well servedits initial purpose.

Distinguishing joint fact-requisition, which occurs on thedispute resolution side, from scientific fact-finding, which occurson the science side, suggests a use for joint fact-requisition in theinitial management of uncertainty. In a structured subprocess runby the mediator, stakeholders may be interviewed as to theirperceptions of uncertainties associated with the matter at issue,and scientists can address these perceptions (which may bepresented to them anonymously, through the mediator). Thissubprocess would afford a valuable, shared educationalopportunity for the stakeholders, who could be briefed on which

[Vol. 26:373


perceived uncertainties are actually settled or otherwiseunproblematic and which have yet to be studied. Stakeholdersmight learn about additional uncertainties, and gain assistance inprioritizing remaining uncertainties for scientific fact-finding. This"uncertainty meeting" may be a suitable followup to an initial,general briefing between the stakeholders and the scientificadvisors to the process.

2. A professional mediator performs as a transmembrane actor.

Clearly, the professionalism of the mediator, who plays the roleof the gated channel, is crucial to the integrity of the joint fact-requisition process, which is in turn fundamental to the integrityof the science produced to inform the process. The mediator mustfunction in a politically neutral manner, bracketing his personalfeelings to protect the structure of the negotiation process. Thisduty to bracket his own values in the service of the process is againthe maieutic ethic. The maieutic ethic of the mediator requireshim to serve as a conduit for the stakeholders' communications,while packaging them to reduce any toxicity they may bear and toincrease their effectiveness in dispute management.

One of the mediator's responsibilities is to ensure that all ofthe stakeholders are represented in the dispute-managementprocess. Thus, an appropriate way for the mediator to enablehimself to serve the process rather than his personal position is toensure that his interests are competently represented among thestakeholders.26 Then, he can trust a conservationist or a land-development group, for instance, to advocate his personalinterests, freeing himself professionally to treat all the stakeholdersfairly in seeking the best management regime for all. Ideally, themediator would be an outsider to the dispute, though he may haveworked on other matters with some of the stakeholders. Themediator's reputation for treating all parties fairly and achievingoutcomes with which all parties are satisfied can mitigate mistrustthat may arise in other stakeholders from his history with a givenstakeholder.

The model makes clear that as a transmembrane actor, theprofessional mediator contacts both sides of the membrane and istherefore best positioned to inform actors on either side of their

265. Telephone interview with Scott T. McCreary, Principal, CONCUR, Inc. (Apr.13, 2006).

2007]


roles in the dispute-management process. Both stakeholders andscientists may benefit from education concerning these roles. Forexample, an ex perienced mediator of environmental mattersrecounts the story of a scientist consulted regarding themanagement of potent polychlorinated biphenyl toxins in the NewYork Bight: the scientist presented several dozen slides onpolychlorinated biphenyl chemistry, without explaining exactlywhy the stakeholders should be concerned about polychlorinatedbiphenyls nor how the stakeholders might handle them.266

Scientists are not routinely taught how to communicate theirfindings to audiences outside of their fields, nor for purposesother than scientific progress. Therefore, it is likely that scientistswill not be aware of the different skills involved in informing apublic decisionmaking process rather than reporting theirresearch results to peer scientists. The mediator must teach thosewho are new to science-public decisionmaking interaction how theinteraction is structured, alerting them as to the presence andfunction of the membrane and of their particular roles as eitherscientists or stakeholders. Indeed, this orientation helps tomaintain the membrane by interpellating performances from all ofthe actors that are appropriate for the stage afforded by thisstructure. Put another way, this direction teaches the actors how tomatter in their various roles on either side of the membrane-howto communicate using the gated channel.

For example, just as scientists are not stakeholders withnegotiable interests in a dispute, neither are they mediatorscharged with supporting all of the stakeholders equally. Thefunction of a scientific advisor to a joint fact-requisition process isto identify boundaries within which the stakeholders may crafttheir agreements; to the best of their ability, the scientists informthe stakeholders as to the tolerances of the environmental matterto the stakeholders' proposals.

3. A good play.

A successful joint fact-finding process was a key component ofthe work of the Guadalupe River Flood Control ProjectCollaborative (the "Collaborative") in northern California. 67 The

266. Id.267. See Scott McCreary & Austin Mclnerny, Negotiating Agreements for Integrated Flood

Control: Guadalupe River Food Control Project Collaborative, Santa Clara Valley, California, inPROCEEDINGS OF THE 1999 APA NATIONAL PLANNING CONFERENCE (American Planning

[Vol. 26:373


floodplain of the Guadalupe River includes substantial areas of thedensely settled Santa Clara Valley, which have been inundatedroughly three times per decade over the past sixty years. Inconjunction with local authorities, the Corps of Engineers hadbeen working to develop a comprehensive flood control projectfor the area. The project entailed widening the river's channel tospeed the flow of water to the San Francisco Bay. This approachwould have removed shade plants from the river bank, so thatwater temperatures would no longer be low enough to sustain theriver's cold-water fish populations. The project also would haveremoved shelter for juvenile fish, which would be carried intomore saline water prematurely. The adversely impacted fishspecies included already endangered populations. Citizens sued toprevent the foreseeable adverse environmental effects of theplanned project.

To avoid litigation, the Corps of Engineers and localauthorities agreed to engage in a collaborative dispute-resolutionprocess with national and state agencies and conservationists. TheCollaborative consulted a professional neutral environmentalmediation service to structure the negotiation process, whichincluded the establishment of an independently workingTechnical Fact-Finding Subcommittee. This joint fact-findingprocess was structured essentially as ajoint fact-requisition process,with the Technical Fact-Finding Subcommittee in the science celladvising the stakeholders in the public decisionmaking cellthrough the professional mediator. The Subcommittee evaluatedalternative proposed water-management schemes with respect totheir potential for flood protection, timely completion, cost-effectiveness, and environmental impact mitigation, and advisedthe Collaborative accordingly. As the self-imposed deadline forreaching agreement approached, the frequency ofcommunications between the Collaborative and the Subcommitteerose, so that increasingly detailed technical questions could beaddressed. This negotiation process resulted in a DisputeResolution Memorandum that reflected unanimous support forone of the proposed water management schemes, which involvesthe construction of an underground bypass culvert that couldremove floodwaters while leaving the river banks intact.

Assoc., ed., 1999) available at http://design.asu.edu/apa/proceedings99/MCINER/MCINER.HTM. Other examples of successful joint fact-finding processesinclude those mentioned in THE CONSENSUS BUILDING HANDBOOK, supra note 251.

2007]


The flood control project in accordance with the DisputeResolution Memorandum has been approved and is beingimplemented. 2

' An Adaptive Management Team has also beenestablished to monitor the project's effectiveness and actualenvironmental impacts. 269 Notably, the Collaborative continues tomeet to manage potentially controversial matters as they arise,such as the planned construction of a River Walk that couldcompromise the project's environmental mitigation. 27

" Thus, theinitial investment in a collaborative process involving the jointrequisition and assimilation of relevant technical informationappears to have prevented continuing wasteful disputation andineffective management. 271 The value of such an initial investmentis also recognized in the adaptive management strategy used forEverglades restoration pursuant to WRDA 2000.272

An additional significant benefit of the participation ofrepresentatives from individual agencies in the collaborativeprocess is the representatives' work within their home agencies todevelop support for the multi-agency agreement.273 As I suggestsupra in Part IV.B.-C., to encourage individual agencies to allocatetheir staff and other resources to cooperative managementprojects, performance reviews of individual staff members by theirhome agencies should reflect credit for effective participation inmulti-agency management, and agencies that perform well in suchcollaborative processes should also be credited.

Of particular significance is the improved interagencycommunication that can result from an effective collaborativeprocess. The Inspector General's Report discussed supra in Part

268. See U.S. ARMY CORPS OF ENG'RS, USACE San Francisco District: Upper GuadalupeRiver, http://www.spn.usace.army.mil/projects/upperguadalupe.html (last visited Apr. 18,2007).

269. See U.S. ARMY CORPS OF ENG'RS, Guadalupe River Flood Control Project Collaborative,http://www.spd.usace.army.mil/cwpm/public/plan/pdact/O2workshop/hannah.htm(last visited Apr. 18, 2007) [hereinafter Guadalupe River Collaborative].

270. McCreary & McInerny, supra note 267.

271. This conciusion is supported by an opinion expressed in a workshop on theCollaborative: "Initially, the process is time [-]consuming[;] however, in the long term, thetime required for the design, approval and review of the project is shortened[.]"Guadalupe River Collaborative, supra note 269.

272. ADAPTIvE MGMT. STRATEGY, supra note 205, at 7 ("Although collaborationrequires more time and effort to cultivate at the beginning of the restoration process, amore sustainable and effective management approach results and the benefits to thesystem outweigh the initial investment.").

273. McCreary & McInerny, supra note 267.

[Vol. 26:373


IV.A.2.d noted the need for improved interagency communicationfor the successful implementation of Mod Waters. 274 Relationshipsthat are established between contacts in different agencies duringa mediation process involving joint fact-requisition can promotethe successful implementation of a planned project beyond itsinaugural agreement. Accordingly, a joint fact-requisitionapproach may be helpful to the completion of Mod Waters, asdiscussed infra Part V.C.5.

C. Opportunities to Complement Adaptive Management Under the Planwith Joint Fact-Requisition

Like adaptive management, joint fact-requisition may be auseful tool to engage in Everglades restoration pursuant to thePlan. As discussed above, joint fact-requisition provides analternative to the engagement of experts by individualstakeholders, which may result in wasteful and counterproductivescientific research on overly narrow questions, allowing particularstakeholders to gain information that emphasizes their interestsdisproportionately. It is preferable not only for dispute resolution,but for scientific integrity, for experts to provide the stakeholderswith a unified, balanced view that addresses coherently thescientific aspects of the matter at stake. Adversarial engagement ofexperts can also erode the public's trust in science, both withrespect to the matter at hand and more broadly. For example,relatively settled science may appear excessively uncertain throughthe representation of scientific opinion in the dispute-resolutionprocess in ways that do not reflect the proportions of scientificopinion among scientists generally. Science that "washes out" ofthe dispute-resolution process becomes useless at best.

Adaptive management under the Plan depends upon theeffective collaboration of multiple federal and state agencies andtribal governments. RECOVER, the institution primarily involvedin implementing adaptive management under the Plan, engages inand commissions scientific fact-finding, but does not conductdispute-resolution processes such as joint fact-requisition.Everglades restoration pursuant to the Plan may benefit from theexplicit introduction of joint fact-requisition at various points tostrengthen collaborative management relationships.

274. AUDIT REPORT, supra note 218, at 9.

2007]


1. The RECOVER Leadership Group.

Since multiple government stakeholders are represented in theRLG, it may be a suitable unit for joint fact-requisition on thescience side. Joint fact-requisition would be useful for settlingdisputes among scientists that arise because of the influence ofagency perspectives on interpreting scientific research results. Forexample, observational data sets (such as those that will bedeveloped in the monitoring program for the Plan) are especiallyvulnerable to "cherry-picking"-to an interpreter's improperlyfocusing on data that supports an a priori view, rather thananalyzing the data empirically and as a whole." If a disagreementbetween scientists appears to spring from their political roles, as itwould if one side's analysis were demonstrably scientifically faulty,the entire RLG may convene to address it. A procedure for seekingindependent scientific review of particular questions should bespecified, so that, for example, the RLG could readily consult aneutral to submit a question to an independent research body thatcould advise the group in its dispute resolution if necessary." 6 TheProgrammatic Regulations could be amended to establish ascientific dispute-resolution process accessible to RECOVER,paralleling the external peer review for addressing specificscientific questions that is available to individual agencies under 33C.F.R. § 385.22(b) (1).

2. The System Planning and Operations Team.

Similarly, the SPOT may be a suitable platform for joint fact-requisition on the policy side. The SPOT is responsible forconducting "a structured dialogue involving scientists, managersand stakeholders[,]" the goal of which is "for experts and agencymanagers to develop a common interpretation of the scientificand technical information which may have implications for

275. Heisler Interview, supra note 27.276. To promote its independence, the WRDA 2000 § 601(j) independent scientific

review panel is barred from accepting any tasks not specified under that section. 33 C.F.R.§ 385.22(a) (5) (Westlaw 2007). The panel is "to review the Plan's progress towardachieving the natural system restoration goals of the Plan," and to produce reports toCongress including "assessment[s] of ecological indicators and other measures of progressin restoring the ecology of the natural system, based on the Plan." WRDA 2000, supra note6, § 601(j) (1)-(2). Though the level of detail of the panel's review is not further specified,it would be inappropriate for RECOVER to consult the panel regarding a particular studyor data-set; under the exclusive reading of WRDA 2000 § 601 (j) required by 33 C.F.R. §385.22(a) (5), the panel is to report only to Congress.

[Vol. 26:373


management decisions" impacting the Plan.2 7 7 From thisdescription it sounds as though the SPOT effectively performssomething like joint fact-requisition already, even though it wasnot designed with the joint fact-requisition formalism in mind.Therefore, the SPOT may benefit from the conscious applicationofjoint fact-requisition theory to its decisionmaking processes.

3. The Programmatic Regulations.

The process for developing the Programmatic Regulations "wasdesigned to identify the major concerns of the agencies andvarious groups, prior to publishing the proposed regulations andsoliciting formal public comment"; then, the public was invited tocomment on the draft outline and the initial draft of theregulations.27 As future refinements to the ProgrammaticRegulations are proposed, the process for identifying stakeholders'major concerns may be extended explicitly to include formulatingcommon questions for scientific inquiry that may be addressedthrough formal joint fact-requisition.

4. The § 601(h)(2) Agreement.

The Agreement requires the federal government "[t]o workwith the State of Florida on developing information jointly tosupport the adaptive assessment component of the Plan[.] '279 Thiswording reads like an explicit invitation to federal and stateimplementers to engage in joint fact-requisition if needed, todevelop a unified body of research to inform their collaborativeadaptive management efforts.

5. Mod Waters.

Before the Plan modifications to the Water Conservation Areasmay begin, the Mod Waters project must be complete."' Asrecommended by the Inspector General of the DOI, coherentcommunication and a unified approach to Everglades restorationmust be established for the DOI to participate effectively in ModWaters. 8' In response to the Inspector General's

277. ADAPTIVE MGMT. STRATEGY, supra note 205, at 5.278. Programmatic Regulations, supra note 23, at 64,201.279. Agreement, supra note 124, at 3.280. WRDA 2000, supra note 6, § 601(b) (2) (D) (iv).281. AUDIT REPORT, supra note 218, at 8-9.

2007]


recommendations, the DOI has hired an Assistant DeputySecretary to serve as an arbitrator among DOI agencies and as aliaison between the DOI and external entities.2 2

Successful joint fact-requisition or joint fact-finding has led tothe development of ongoing institutions that continue to manageenvironmental matters in which stakeholders have conflictinginterestsY. A consensus-building process may fail in spite of jointfact-requisition, if stakeholders who are unhappy with the effects ofthe requisitioned information on their position in the negotiationdecide to challenge the science instead of proposing newsolutions. The DOI should consider engaging the relevantsubteams of the Park Service and the FWS in joint fact-requisition,to promote the commitment of these agencies to basing theirrecommendations under Mod Waters on a unified, comprehensiveview of the water needs of the Park and the effects of variousdelivery alternatives on upstream areas.

VI. CONFLUENCE

I have used a viable water-management structure on a smallscale as a metaphorical basis for a conceptual model of a viableinformation-management structure for science-based publicdecisionmaking. Just as individual living cells can optimize theirinternal environments for their distinct productive processes whilecommunicating with each other effectively, distinct professionalentities can optimize their internal procedures to produce,respectively, science and science-based collaborative decisionsaffecting matters of public concern while engaging in effectivecommunication.

The work of harmonizing science- and public-decisionmakingprocesses requires a kind of professionalism that exceeds theformal training involved in becoming either a scientist or, forexample, a government manager. I have identified a professionalethic that is troubled when a specialist is asked to perform outsideof his constitutive institutions, and I have examined its effects onthe performances of individuals working at the confluences ofscience and policy, and of science and alternative dispute-

282. Letter from P. Lynn Scarlett, Deputy Secretary of the Interior, to Earl Devaney,Inspector General (Mar. 8, 2006), in AUDIT REPORT, supra note 218, at 27-29.

283. See, e.g., Guadalupe River Collaborative, supra note 269; Sarah McKearnan &Patrick Field, The Northern Oxford County Coalition, in THE CONSENSUS BUILDINGHANDBOOK, supra note 251, at 73840.

[Vol. 26:373


resolution. This maieutic ethic requires the professional to serve asa faithful translator, speaking for those whose voices have beenunintelligible and thus have not mattered (such as ecologicalentities), or whose message is overdetermined (such as agovernment agency or group of agencies) and therefore not in aform that lends itself readily to articulation by a coherent,accountable actor. I have used staged performance as a metaphorin examining the structured flow of speech and writing in the workof science-public decisionmaking professionals.

Analysis with this conceptual model has yielded certain insightsand recommendations for integrating science and publicdecisionmaking procedures generally, as well as specifically for twocases. The first case concerns Congress' ComprehensiveEverglades Restoration Plan. The natural hydrological patterns ofthe Everglades supported a remarkably productive ecosystemspanning a vast territory, gathering water into its heart, andfeeding the ecosystem with a pulsed, sheeting flow. To provideflood control and a more stable water supply for agriculture andurban settlement, governmental management efforts diverted andcompartmentalized Everglades water. Realizing that theengineered management regime had harmed the ecosystem andwould not long sustain regional economies, Congress authorized are-engineering of water-management system structures andoperations, to be informed by state-of-the-art multidisciplinaryscientific study. To meet the challenge of science-based adaptiveecosystem restoration and management, the re-engineered systemwill need to provide an effective approximation ofdecompartmentalized water flow to the remaining ecosystem, andto establish an effectively compartmentalized science-policymanagement structure for the flows of properly formedinformational requests to scientific consultants, and of undistortedscience to administrators. The flow of speech must be bettercompartmentalized, so that the flow of water can be betterdecompartmentalized.

The major mechanism for effecting the Plan's adaptivemanagement process is the mandatory update of theProgrammatic Regulations implementing the Plan. Update of theregulations could promote better science-based management byincorporating some of the results of this analysis.

To ensure that science generated for the adaptive assessmentand monitoring of Everglades restoration actually informs

2007]


ecosystem management, the Programmatic Regulations may beupdated to help provide incentives to individual agencies. Forexample, RECOVER, the institution responsible for gathering,consolidating and making recommendations based on science foradaptive management, may be empowered to evaluate plannedprojects for their expected contributions to Plan implementation.

Beyond the scope of the regulations, RECOVER evaluationsmay be used by individual agencies and by Congress'to inform theperformance reviews of managers and agencies, respectively,responsible for implementing projects under the Plan, so that thequality of the managers' individual and collective responses to thetechnical reports produced for adaptive management may matterto their reputations and funding. Further, Congress may provideincentives, such as access to additional funds and other resources,to agencies with consistently excellent performance incollaborative science-based ecosystem management. Congress mayalso provide incentives to individual scientists for participation inscience-based management and dispute-resolution processes; forexample, public service through an institution like the NationalAcademy of Sciences may be required of scientists receivinggovernment research funding. Alternatively, a percentage ofresearch grants could be contributed to a funding pool forcontracting independent scientists to participate in theseprocesses.

To provide needed functionality to the political processesengaged in adaptive management, the Programmatic Regulationsmay be updated for the formal creation of analogs to RECOVERand its Leadership Group. Such a multi-agency institution on thepolicy side may facilitate the collaboration of diverse stakeholdersand entities in developing detailed policies that collectivelyadvance Plan implementation. Structured communicationbetween RECOVER and its policy-side analog, likely through theirrespective Leadership Groups, could ensure that unified publicviews of ecological priorities inform scientific research.

An update of the Programmatic Regulations could also provideexplicitly for joint fact-requisition functions to be associated withthe RECOVER Group and its analog. The SPOT may effectivelyperform joint fact-requisition on the policy side, so thatformalizing its science-based consensus-building function in thismanner would be relatively straightforward.

On the science side, the RLG may conduct a joint fact-

[Vol. 26:373


requisition process to settle interagency disputes that are framedin terms of scientific research results, such as the tree-islandcontroversy discussed supra Part IV.A.2.d that must be resolved sothat appropriations for Plan projects that are contingent on thecompletion of Mod Waters pursuant to § 601(b) (2) (D) (iv) may beauthorized. At a minimum, the RLG may conduct a multiagencyuncertainty meeting in which the scope and nature of theuncertainties associated with such a controversy are addressedjointly, since a common understanding of these uncertaintiescould promote the robustness of the adaptive managementprocess to an agency's potential refusal to cooperate based on abiased interpretation of commonly understood research results. Amechanism may be made available to the RLG for consultingscience neutrals, analogous to the external peer review available toindividual agencies under 33 C.F.R. § 385.22(b) (1). Alternatively,the Programmatic Regulations could provide explicitly forindividual agencies involved in a scientific dispute pertinent toPlan implementation to access a joint fact-requisition mechanismthat operates independently of RECOVER.

In addition to the update of the Programmatic Regulations,Congressional action is required to ensure timely funding andauthorization for continuing adaptive assessment and monitoring,as well as for the initiation of most Plan projects. Though adaptivemanagement is a front-loaded process, especially when theinstitutions undertaking it are still developing their abilities toengage in it, this strategy for more preventive management isexpected ultimately to be more effective and efficient, and lessexpensive, than management that cannot correct its course.Indeed, had adaptive management been implemented in theC&SF Project in 1948, governmental responses to its unexpectedadverse ecosystemic effects may have begun decades earlier, whilethose effects were more tractable.

Everglades restoration pursuant to the Plan could fail for lackof sustained funding and authorization of the RECOVER Group.8 4

Diverse examples of other threats to successful Evergladesrestoration include insufficient staffing by skilled and ethical

284. PROGRAMMATIC EIS, supra note 11, at 9-38 ("To be successful, an adaptiveassessment process requires . . . that it be possible to make changes in the design andsequencing of the plan in response to information learned from the monitoring programand from new research and modeling, and that a specific protocol for conducting the adaptiveassessment process be in place throughout the life of the program." (Emphasis added.)).

2007]


science-policy professionals, inadequate restoration of sheetlikewater flow through the Everglades Agricultural Area, unmitigatedinvasion by exotic species that alter habitats, and loss ofcooperation from participating agencies. Just as the adaptivemanagement strategy promotes learning from unexpected results,the world that is watching this attempt at the adaptivemanagement of a large-scale ecosystem could learn a great dealfrom the efforts expended pursuant to § 601 of WRDA 2000-evenif the Plan itself has surprising, unintended effects.

The theoretical analysis has also yielded insights andrecommendations specifically applicable to science-based disputeresolution. To optimize the integrity of scientific and public-decisionmaking processes in cooperation, scientists must have theindependence to exercise their professional judgment inperforming research. Scientists also must inform thedecisionmaking process, working with a mediator to advise thedecisionmakers collectively rather than individually, just asmediators must serve stakeholders collectively rather than favoringsome over others. Unlike the work of mediators, however, wellperformed scientific research may clearly favor the positions of asubset of stakeholders-the integrity of science does not dependon the neutrality of its findings ex post, but on the neutrality of itsresearch approach and conduct ex ante.

Scientists who become advisors to management and dispute-resolution processes should be trained to communicate scientificresults to various audiences. Just as decisionmakers withoutscientific training would benefit from education about scientificuncertainties generally as well as those specifically pertaining tothe matter at stake, scientists and all those participating in science-based management would benefit from the explicit discussion ofthe roles of scientists in decisionmaking processes, and of theduties of scientists and other professionals involved in publicdecisionmaking.

Potential participants in joint fact-finding should be aware thatthis dispute-resolution process and scientific research are properlydistinct activities-that stakeholders in a joint fact-finding processought to engage in joint fact-requisition, commissioning scientificexperts to conduct research on jointly developed questions.Having learned about the stakeholders' informational needs, thescientists may teach the stakeholders about what kinds of questionscould be addressed with what kinds of uncertainty. Collective

[Vol. 26:373


stakeholder education about uncertainties attending the requisitescience can contribute to the robustness of the dispute-resolutionprocess to a potential spoiler, who may otherwise have been able toexploit the uncertainties rhetorically to devalue the science if itdisfavors his interests. After iterated, structured communicationamong stakeholders and scientists to develop research questions,scientists conduct research to address the stakeholders' questionsin a caesura of stakeholder involvement. Then, the scientificconsultants to the joint fact-requisition process report their resultsto the stakeholders jointly, gathering questions for additional studyas may be required. The boundary between scientific fact-findingand dispute-resolution processes preserves the integrity of thescience-which in turn enhances its consensus-building power-while the structured protocols of communication betweenscientists and stakeholders in ajoint fact-requisition process enablethose working on either side of the boundary to interacteffectively. The gated membrane protects the reputations ofparticipating scientists and the integrity of the science theyproduce, and enables the science to function as a neutral ororacular seed for consensus. The membrane prevents the dispute-resolution process from emulsifying.

The conceptual model of the science-public decisionmakinginterface introduced here can be used to explore science-baseddecisionmaking on small scales, such as that of the resolution of aland-use issue concerning a small tract and a few stakeholdersusing ajoint fact-requisition process, as well as on large scales, suchas that of Everglades restoration through adaptive management.The modeled structure continues to apply to even larger-scalesituations, such as that of international negotiations overenvironmental security issues. The model may also be applied toother circumstances in which scientific or technical knowledge ispertinent to dispute resolution or prevention, for example, withrespect to issues in public health and national security beyondtheir environmental aspects. Those working at the confluences ofscience and public decisionmaking are charged with protectingthe boundaries between scientific research and publicdecisionmaking processes, while ensuring their benefit to eachother through properly structured flows of communication.

2007]