Bill Edward SIMMONS, Maxine Dorothy Simmons, Steven J. Shaffner, Kathleen R. Shaffner, Laurel P. Touissant, Thomas Bik and Martha Crothers, Plaintiffs-Appellants, v. UNITED STATES ARMY CORPS OF ENGINEERS, Togo West, Secretary of the Army, Maj. Gen. Patrick Stevens, Chief of Engineers, U.S. Army Corps of Engineers, Ralph Greico, Commander and District Engineer, U.S. Army Corps of Engineers, and The City of Marion, Illinois, Robert Butler, Mayor, Defendants-Appellees.
Eight years have elapsed since the City of Marion, Illinois, first proposed building a new water reservoir in the southernmost tip of Illinois. In those eight years a tale has unfolded that is all too familiar. Lawsuits have stopped the project short; the case has visited the district court twice; and Marion still is no closer to a new water supply. As is routine in American administrative law, the litigation has little to do with what anybody really cares about. One side wants a dam built and a new lake created, and the other does not. Instead, the dispute, now in and out of federal court for five years, has centered on procedures-whether the U.S. Army Corps of Engineers fulfilled its procedural obligations under federal environmental law. All this is true. But the case provides a textbook vindication of the wisdom of Congress in insisting that agencies follow those procedures in the first place.
With passage of the National Environmental Policy Act, Pub.L. 91-190, Jan. 1, 1970, 83 Stat. 852, codified at 42 U.S.C. § 4321 et seq., Congress established the nation's central and unique environmental policy for (self-) regulating the federal government. Although policymakers and courts have cooked up enough acronyms under NEPA for a feast of officialese, the thrust of NEPA is simply expressed. For all “major Federal actions significantly affecting the quality of the human environment,” 42 U.S.C. § 4332(2)(C), federal agencies must articulate why they have settled upon a particular plan and what environmental harms (or benefits) their choice entails.
NEPA mandates a searching inquiry into alternatives, but says nothing about which to choose. It binds federal officials to justify their plans in public, after a full airing of alternatives. It thus blends a faith in technocratic expertise with a trust in democracy. Officials must think through the consequences of-and alternatives to-their contemplated acts; and citizens get a chance to hear and consider the rationales the officials offer. Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 349, 109 S.Ct. 1835, 1845, 104 L.Ed.2d 351 (1989). But, if a federal agency has heard all the objections to a plan and considered all the sensible options before it, the agency has fulfilled its duty. Robertson, 490 U.S. at 350, 109 S.Ct. at 1846.
When a federal agency prepares an Environmental Impact Statement (EIS), it must consider “all reasonable alternatives” in depth. 40 C.F.R. § 1502.14. No decision is more important than delimiting what these “reasonable alternatives” are. That choice, and the ensuing analysis, forms “the heart of the environmental impact statement.” 40 C.F.R. § 1502.14. To make that decision, the first thing an agency must define is the project's purpose. See Citizens Against Burlington, Inc. v. Busey, 938 F.2d 190, 195-96 (D.C.Cir.1991). The broader the purpose, the wider the range of alternatives; and vice versa. The “purpose” of a project is a slippery concept, susceptible of no hard-and-fast definition. One obvious way for an agency to slip past the strictures of NEPA is to contrive a purpose so slender as to define competing “reasonable alternatives” out of consideration (and even out of existence). The federal courts cannot condone an agency's frustration of Congressional will. If the agency constricts the definition of the project's purpose and thereby excludes what truly are reasonable alternatives, the EIS cannot fulfill its role. Nor can the agency satisfy the Act. 42 U.S.C. § 4332(2)(E).
We are confronted here with an example of this defining-away of alternatives. In 1989, the City of Marion applied to the U.S. Army Corps of Engineers (the Corps) for permission to build a dam and reservoir, as required by § 404 of the Clean Water Act, 33 U.S.C. § 1344. The dam would block up Sugar Creek, a free-flowing stream in southern Illinois running seven miles southeast of Marion. Marion envisioned that the resulting Sugar Creek Lake would supply water not just to Marion, but to the Lake of Egypt Water District, which encompasses six counties and 15,000 rural customers. Sugar Creek Lake would drown a substantial area, with the usual environmental effects of drowning, including the transformation or obliteration of the riverine habitats of several species.
We discuss the details of the project's history below, but the case boils down to no more than those bare facts. From the beginning, Marion and the Corps have defined the project's purpose as supplying two users (Marion and the Water District) from a single source-namely, a new lake. Accordingly, when the Corps prepared an environmental impact statement, it confined the analysis to single-source alternatives. And therein lies the difficulty. At no time has the Corps studied whether this single-source idea is the best one-or even a good one. Marion and the Lake of Egypt Water District share a common problem, a thirst for water. From this fact the Corps adduces the imperative for a common solution. We disagree. A single source may well be the best solution to the putative water shortages of Marion and the Lake of Egypt Water District. The Corps' error is in accepting this parameter as a given. To conclude that a common problem necessarily demands a common solution defies common sense. We conclude that the U.S. Army Corps of Engineers defined an impermissibly narrow purpose for the contemplated project. The Corps therefore failed to examine the full range of reasonable alternatives and vitiated the EIS. We reverse.
The City of Marion and the Lake of Egypt Water District both want more water. Since the 1920s, the City of Marion has drawn the bulk of its water from the man-made Marion City Lake. Marion's thirst for water has long since outstripped Marion City Lake's capacity of 1.1 million gallons per day, and Marion has had to go elsewhere for an additional 600,000 gallons per day. (The parties have not told us how or where.) Marion City Lake is a poor reservoir, to boot; its raw water requires costly chemical treatment to render it potable. Marion contends that a new reservoir-in particular, the proposed Sugar Creek Lake-would slake the City's thirst. The Lake of Egypt Water District, too, argues that it needs a new source of water. The Water District gets its water from the Lake of Egypt, another reservoir of marginal quality. The Water District turns out not to own the namesake lake, but instead must buy the water from an electric co-op, which does own the Lake of Egypt, and which limits diversion of the Lake's water.
Marion proposed to solve both problems in one stroke. A new reservoir was in order, and Marion thought a dam over Sugar Creek would work best. The dam would lie at the head of a valley about seven miles southeast of Marion. The new Sugar Creek Lake would be 2500 feet wide and 20,000 feet long, and would generate 8.9 million gallons per day of raw water. A 20-inch diameter pipeline would run the water to Marion. The Lake of Egypt Water District would then buy some of that water from Marion.
Then the federal government got involved. Sugar Creek falls under federal jurisdiction as one of the navigable waters of the United States. Section § 404 of the Clean Water Act, 33 U.S.C. § 1344, obliges anyone who would discharge “dredged or fill material into the navigable waters” to obtain a permit from the U.S. Army Corps of Engineers. 33 U.S.C. § 1344. A dam across Sugar Creek would require such a permit, see Van Abbema v. Fornell, 807 F.2d 633, 636 (7th Cir.1986) (noting that construction projects over navigable waters need a Corps permit), and in 1989, Marion sought the Corps' approval. The need for Corps approval makes the proposed dam a kind of federal action, and thus NEPA was invoked.
In 1991 the Corps completed a preliminary study of the proposed dam, termed an “environmental assessment” in the argot of NEPA. If a federal project has no “significant impact” on the environment, there is no need to go forward with a much deeper study-the EIS (environmental impact statement). In its environmental assessment, the Corps concluded that the four-mile long, half-mile wide Sugar Creek Lake, which would flood a square mile-and-a-half of wetlands, woods, fields and farms and would block up “one of the last free-flowing streams in southern Illinois”-the Corps' own words-would have no “significant impact” on the environment. A group of plaintiffs (affected landowners plus the Sierra Club) filed suit in the Southern District of Illinois, claiming that the Corps' determination was arbitrary and capricious and that NEPA obliged the Corps to prepare an environmental impact statement. On this first round through the Southern District of Illinois, the case fell to Judge Foreman. He ruled for the plaintiffs. In his well-reasoned, graceful (and unfortunately unpublished) opinion, Judge Foreman found manifold flaws in the Corps' conclusion of no significant environmental impact. Hundreds and hundreds of acres would be flooded. The bald eagle and two federally protected bats (the Indian bat and the gray bat) would lose habitats. Two aquatic creatures in Sugar Creek, the least brook lamprey and the Indiana crayfish, would be “extirpated,” in the Corps' ominous phrase. And every state and federal agency with environmental competence called at a minimum for the Corps to write up an environmental impact statement.
Not least among the flaws-and the most glaring in hindsight-was the Corps' failure to consider reasonable alternatives. Judge Foreman deemed the environmental assessment “incomplete and flawed,” and found no hint “that the Corps gave independent thought to the feasibility of alternatives.” To start, Judge Foreman faulted the Corps' explanation for dismissing other single-source ideas, including Rend Lake and Devils Kitchen Lake. But the problems ran deeper, to the roots of the Corps' thinking. The Corps had dismissed certain alternatives because they supplied Marion and the Water District from separate sources. Why the Corps assumed the imperative of a single-source project, however, remained unexplained. “[T]he Corps never questioned the need for finding a water supply for both Marion and Lake of Egypt Water District,” Judge Foreman wrote. “Yet there is no evidence that only Marion can supply Lake of Egypt Water District's water needs.” (As we explain below, Judge Foreman's remarks remain just as true, and the Corps' analysis just as faulty, five years later.) Based on the flaws we mention above, plus others we do not, Judge Foreman vacated Marion's permit and forbade the Corps to issue another until it completed an environmental impact statement.
By late 1994, the Corps had a draft EIS ready. The Corps held a public hearing in December 1994 and received over a thousand written comments. The final EIS came out in July 1995. More comments followed. The Corps learned it had incorrectly assumed that Rend Lake was unavailable. (Rend Lake was one of the existing unitary water sources that the Corps had cursorily dismissed as an alternative in its environmental assessment.) The Corps prepared a supplement to the final EIS. Finally, on July 29, 1996, the Corps made its decision. Colonel Ralph Grieco (the decision-maker and a defendant in this action) wrote that building Sugar Creek Lake would be “environmentally sustainable.” The project would not conflict with the public interest, he found. 33 C.F.R. § 320.4(a). The Corps re-issued the permit.
The plaintiffs in the first action (minus the Sierra Club) contested the Corps' decision. This time, the case went before Chief Judge Gilbert. On cross-motions for summary judgment, the plaintiffs argued three points in the district court: one, the final EIS was inadequate; two, the Corps violated NEPA and the regulations promulgated thereunder by the Council on Environmental Quality; and three, the Corps violated its own regulations. In a carefully drafted opinion, the chief judge resolved all three points in favor of the Corps and Marion. The plaintiffs appealed, but only on the second point: whether the Corps had breached the strictures of NEPA by failing to consider all reasonable alternatives in its final EIS. We consider the merits of their appeal next.
Logic and law dictate that every time an agency prepares an environmental impact statement, it must answer three questions in order. First, what is the purpose of the proposed project (major federal action)? Second, given that purpose, what are the reasonable alternatives to the project? And third, to what extent should the agency explore each particular reasonable alternative? See City of Carmel-by-the-Sea v. United States Dep't of Transportation, 95 F.3d 892, 903 (9th Cir.1996); City of Grapevine, Texas v. Dep't of Transportation, 17 F.3d 1502, 1506 (D.C.Cir.1994). These are questions for an agency (the Corps) to resolve in the first instance. We owe and accord deference to the Corps on whether the Corps resolved those three questions in a permissible way. “[I]t is not our role to second-guess” the Corps. Van Abbema, 807 F.2d at 636. We ensure only that “the Corps followed required procedures, evaluated relevant factors and reached a reasoned decision.” Id.
The Corps rigged the environmental impact statement on the question of purpose, contend the plaintiffs. The Corps defined the project's purpose as finding or creating a single source to supply both Marion and the Lake of Egypt Water District. (Actually, the Corps never did so explicitly; the district court had to “fill in the blanks” to ascribe this purpose to the Corps.) The Corps' imputed definition, the plaintiffs say, is unreasonable: the project's real purpose is supplying Marion and the Water District with more water, and there are reasonable alternatives to that beyond relying on a single reservoir. Marion, for instance, could hook its water-treatment plant up to an existing pipeline stemming from Rend Lake and lying only seven miles away. (Marion's water-treatment plant is twelve miles away from Sugar Creek Lake.)
As its first line of defense, the Corps essentially passes the buck to Marion. Marion wanted a single reservoir to supply it and the Lake of Egypt Water District. Since Marion is the proposer and will construct the project, the Corps must accept its definition. This is a losing position in the Seventh Circuit. Over a decade ago, we held that “the evaluation of ‘alternatives' mandated by NEPA is to be an evaluation of alternative means to accomplish the general goal of an action.” Van Abbema, 807 F.2d at 638 (emphasis added); 40 C.F.R. § 1502.13 (“The [EIS] shall briefly specify the underlying purpose and need”) (CEQ regulations) (emphasis added). The general goal of Marion's application is to supply water to Marion and the Water District-not to build (or find) a single reservoir to supply that water. Indeed, in Van Abbema, we specifically rejected the Corps' present position. An agency cannot restrict its analysis to those “alternative means by which a particular applicant can reach his goals.” Van Abbema, 807 F.2d at 638 (emphasis added); contra, Busey, 938 F.2d at 198-99. This is precisely what the Corps did in this case. The Corps has “the duty under NEPA to exercise a degree of skepticism in dealing with self-serving statements from a prime beneficiary of the project.” Busey, 938 F.2d at 209 (Buckley, J., dissenting).1 And that is exactly what the Corps has not shown in its wholesale acceptance of Marion's definition of purpose.
We turn then to probe the reasonableness of the Corps' looking only to single-source alternatives to damming Sugar Creek. Here the Corps argues that Marion and the Lake of Egypt Water District share the same problem-a shortage of water-and that they are contiguous. Hence, a single source is the obvious solution. Maybe, even probably, the Corps is right. Alternatives might fail abjectly on economic grounds. But the Corps and, more important, the public cannot know what the facts are until the Corps has tested its presumption. It is axiomatic that the Corps need not examine every conceivable alternative. Identifying, assessing and comparing alternatives costs time and money, Metropolitan Edison v. People Vs. Nuclear Energy, 460 U.S. 766, 776, 103 S.Ct. 1556, 1562, 75 L.Ed.2d 534 (1983), and an agency should focus its energies only on the potentially feasible, not the unworkable. 40 C.F.R. §§ 1502.14(a)-(c), 1508.25(b)(2). As a matter of logic, however, supplying Marion and the Water District from two or more sources is not absurd-which it must be to justify the Corps' failure to examine the idea at all. In fact, the plaintiffs have advanced at least one concrete alternative that seems reasonable: for Marion to build a shorter pipeline that would ultimately feed water from the extant Rend Lake. Alaska Wilderness Recreation & Tourism v. Morrison, 67 F.3d 723, 729 (9th Cir.1995) ( “ ‘The existence of a viable but unexamined alternative renders an environmental impact statement inadequate.’ ” (citation omitted)). What other alternatives exist we do not know, because the Corps has not looked. Perhaps the Corps is relying on a contract between Marion and the Water District for Marion to supply the Water District with water if it succeeds in damming Sugar Creek. But this condition depends on meeting environmental requirements, which, in turn, demand exploration of alternatives free of contractual arrangements. The public interest in the environment cannot be limited by private agreements.
Perhaps cognizant of the weakness of the Corps' stance, co-defendant Marion argues that the Corps did look into separate-source alternatives-which the Corps itself does not even claim. This last ditch effort is unavailing. Marion cites to the EIS and the Corps' record of decision, but all we find are conclusory statements. Two separate projects would cost more than one, we are told, with no support: no mention of concrete proposals, no cost figures, no specifics at all. Even if we put aside that the Corps disclaims making this inquiry, the record offers precious little to show that the Corps ever paused to test its foundational assumption.
If NEPA mandates anything, it mandates this: a federal agency cannot ram through a project before first weighing the pros and cons of the alternatives. In this case, the officials of the Army Corps of Engineers executed an end-run around NEPA's core requirement. By focusing on the single-source idea, the Corps never looked at an entire category of reasonable alternatives and thereby ruined its environmental impact statement. We presume that the Corps' failure to satisfy NEPA was inadvertent, notwithstanding the Corps' failure to heed the admonitions of Judge Foreman on this point. We believe that Judge Foreman was right at the time of his decision, and remains right.
We regret that eight years have passed since the City of Marion first proposed to remedy its water shortage. Still no decision has been made. Better-oiled machinery for environmental policy-making would have long since decided yea or nay on the single-source rationale for Sugar Creek. Either way, Marion would probably have its extra water by now, from whatever source. But the specter of further delay cannot in itself justify setting aside the mandate of the law. The Corps failed to comply with the most basic terms of NEPA. If fault must be found, it lies with those who refused to consider patently reasonable alternatives and who ignored the explicit directions of the federal bench.
The opinion of the district court is Reversed and the case Remanded with instructions to enter summary judgment for the plaintiffs and to vacate the permit.
1. The Corps' own regulations state as much. “The Corps is neither an opponent nor a proponent of the applicant's proposal ․ [T]he Corps [ ] will in all cases[ ] exercise independent judgment in defining the purpose and need for the project from both the applicant's and the public's perspective.” 33 C.F.R. Pt. 325, App. B, §§ (9)(b)(5), (4).
CUDAHY, Circuit Judge.