PLAGER, Senior Circuit Judge.

This case, brought under the United States Court of Federal Claims' Tucker Act jurisdiction, raises three separate though related issues. First, what must a plaintiff establish regarding the existence of a money-mandating law source in order for the Court of Federal Claims to have subject matter jurisdiction over the case under the Tucker Act? Second, assuming the trial court takes jurisdiction and addresses the merits of the cause, what are the consequences of a failure to prove the elements of the cause of action because the facts of the case do not bring it within the alleged source? And third, even assuming the cause of action is otherwise established, are there matters that are nonjusticiable because of their unique military implications?

The plaintiff in this case, Dr. Frank E. Fisher, a physician retired from the Air Force, filed a complaint in the Court of Federal Claims alleging that while he was on active duty he should have been found unfit for continued service because of a physical disability, and therefore under 10 U.S.C. § 1201 he should have been retired for disability, with appropriate retirement pay. The trial court at the behest of the Government dismissed Dr. Fisher's claim for lack of jurisdiction. The trial court indicated that, even if the court had jurisdiction, the matter was exclusively one for military determination, and thus nonjusticiable. Appeal was timely taken.

We conclude that, in light of the statutes on which the cause was based and the facts alleged, the court erred in dismissing the case. We further conclude that, despite the military origins, under controlling precedent the issue is justiciable. The case is remanded to the trial court for further proceedings in accordance with this opinion.

I. BACKGROUND

Dr. Frank E. Fisher served as a physician on active duty in the United States Air Force from 1989 to 1996. Thereafter, he served in the United States Air Force Reserves until September 7, 2001, when he was discharged for physical disqualification. While on active duty, Dr. Fisher appeared before three Medical Evaluation Boards (MEBs). The first MEB evaluation took place in May 1994 for Dr. Fisher's complaint of "persistent left shoulder pain." The MEB referred Dr. Fisher's case to an Informal Physical Evaluation Board (IPEB), which determined that his condition did not make him unfit for service. Accordingly, Dr. Fisher continued on active duty.

Dr. Fisher went before a second MEB in September 1995 for "chronic left wrist and hand pain." The MEB again referred the case to an IPEB, which found in October 1995 that Dr. Fisher was unfit for duty and recommended that he be discharged with severance pay and given a 10 percent disability rating. Dr. Fisher disagreed with the IPEB recommendation, alleging that the 10 percent disability rating was too low, and he appealed to a Formal Physical Evaluation Board (FPEB). The FPEB found in November 1995 that Dr. Fisher was "fully capable of performing all of his assigned duties and was doing so routinely." Accordingly, the FPEB found Dr. Fisher fit for duty and recommended that he be returned to duty. Dr. Fisher did not further challenge this recommendation at the time and returned to duty.

At about the same time that the FPEB was considering Dr. Fisher's case, Dr. Fisher was diagnosed with seronegative rheumatoid arthritis. Apparently that diagnosis was not considered by the FPEB during its deliberations.

In June 1996 Dr. Fisher went before a third MEB, which found him fit for continued military service. Dr. Fisher alleges that the third MEB was convened specifically to consider his diagnosis of seronegative rheumatoid arthritis; however, there is no documentation in the record regarding this third MEB. The MEB did not refer Dr. Fisher to an IPEB, and Dr. Fisher did not challenge this MEB's findings.

At the end of 1996, after completion of his active duty service commitment, Dr. Fisher was released from active duty in the normal course and became a member of the Air Force Reserve. During the time he was serving in the Reserves, he applied to the Department of Veterans Affairs (VA) for disability compensation. In March 1997 the VA awarded him a 40 percent service-connected disability rating for rheumatoid arthritis and a 20 percent rating for degenerative joint disease, which, using a combined rating table, established an overall rating of 50 percent. Dr. Fisher later was awarded an additional 10 percent disability rating for depression, which was increased to 30 percent in October 1997.

In December 1999, Dr. Fisher applied to the Air Force Board for Correction of Military Records (AFBCMR), alleging that the various medical boards before which he appeared should not have found him fit for duty and should have granted him a medical discharge. He requested that his records be corrected to show a medical disability based on rheumatoid arthritis and that he be discharged with a medical disability as of December 31, 1996, his last day of active duty. The AFBCMR denied Dr. Fisher's claim, determining that he had not demonstrated "the existence of probable material error or injustice."

Following the adverse decision of the AFBCMR, Dr. Fisher filed suit in the Court of Federal Claims. His complaint alleged that the actions of the Secretary of the Air Force and the AFBCMR were contrary to Air Force statutes, rules, and regulations. He asked the court to reinstate him to active duty, place him in disability retirement status effective January 1, 1997, award him medical retirement back pay from that date, and order the correction of his records to reflect such actions.

Dr. Fisher's complaint asserted that jurisdiction in the Court of Federal Claims was founded upon 28 U.S.C. § 1491 (the Tucker Act), 10 U.S.C. § 8011 et seq. (Air Force organizational statutes), 10 U.S.C. § 1201 et seq. (military disability retirement statutes), 10 U.S.C. § 1552 (military correction board enabling statute), and Air Force rules and regulations relating to medical standards and disability retirement. The Government filed a motion to dismiss for lack of jurisdiction.

After oral argument on the issue, the trial court granted the Government's motion. The trial court held that Dr. Fisher's claim was outside the scope of the Tucker Act because any monetary entitlement was dependent upon a declaratory judgment, which the court lacked authority to grant. Fisher v. United States, No. 00-740C (Fed.Cl. Jan. 7, 2002).

The trial court, citing Rice v. United States, 31 Fed. Cl. 156 (1994), aff'd, 48 F.3d 1236 (Fed. Cir. 1995) (summary affirmance), noted that in Rice a challenge to a determination regarding fitness for duty was deemed nonjusticiable even if the court possessed subject matter jurisdiction. Although the trial court in this case favorably cited that alternative holding of Rice, the court did not decide the case on that ground, stating that, because the court did not have jurisdiction under the Tucker Act to hear the case, it was unnecessary to address the Government's alternative argument that Dr. Fisher's claim was not justiciable. The trial court dismissed the complaint, and Dr. Fisher filed a timely appeal with this court. We have jurisdiction over the appeal pursuant to 28 U.S.C. § 1295(a)(3).

II. DISCUSSION A. Tucker Act Jurisdiction 1.

Separating the question of a federal court's subject matter jurisdiction over a cause from the question of what a plaintiff must prove in order to prevail in the cause is, in many areas of the law, not a difficult matter: a specific statute sets the court's jurisdictional parameters; a separate statute or regulation or common law rule establishes the right that allegedly has been breached. In Tucker Act jurisprudence, however, this neat division between jurisdiction and merits has not proved to be so neat. In these cases, involving suits against the United States for money damages, the question of the court's jurisdictional grant blends with the merits of the claim. This mixture has been a source of confusion for litigants and a struggle for courts. The Government's arguments, on which it prevailed at trial, and the trial court's view of the matter, require that we address this issue.

It is hornbook law that the Tucker Act (of which there are several versions — it is the Big Tucker Act with which we are here concerned) does two things: (1) it confers jurisdiction upon the Court of Federal Claims over the specified categories of actions brought against the United States, and (2) it waives the Government's sovereign immunity for those actions. See U.S. v. Mitchell, 463 U.S. 206, 212-18, 103 S.Ct. 2961, 77 L.Ed.2d 580 (1983) ( Mitchell II); United States v. Testan, 424 U.S. 392, 397-98, 96 S.Ct. 948, 47 L.Ed.2d 114 (1976). The causes to which the Act applies are claims for money damages against the United States "founded either upon the Constitution, or any Act of Congress or any regulation of an executive department, or upon any express or implied contract with the United States, or for liquidated or unliquidated damages in cases not sounding in tort." 28 U.S.C. § 1491(a)(1).

The Tucker Act itself does not create a substantive cause of action; in order to come within the jurisdictional reach and the waiver of the Tucker Act, a plaintiff must identify a separate source of substantive law that creates the right to money damages. Mitchell II, 463 U.S. at 216, 103 S.Ct. 2961; Testan, 424 U.S. at 398, 96 S.Ct. 948. In the parlance of Tucker Act cases, that source must be "money-mandating." See Mitchell II, 463 U.S. at 217, 103 S.Ct. 2961; Testan, 424 U.S. at 398, 96 S.Ct. 948.

Under the existing precedent of this court, the issue of whether a source is money-mandating is addressed in a two-step process. See Gollehon Farming v. United States, 207 F.3d 1373, 1378-80 (Fed. Cir. 2000) (citing Banks v. Garrett, 901 F.2d 1084, 1087-88 (Fed. Cir. 1990)). As a first step, and for purposes of satisfying the jurisdictional requirement that a money-mandating statute or regulation is before the court, the plaintiff need only make a non-frivolous allegation that the statute or regulation may be interpreted as money-mandating. The non-frivolous allegation satisfies the jurisdictional requirement. If, as a second step, the issue of jurisdiction is later pressed and it is subsequently decided that the statute or regulation is not money-mandating, then the case is dismissed for failure to state a claim upon which relief can be granted. Gollehon, 207 F.3d at 1379.

To the extent that Gollehon relied on Banks as precedent for such a two-step inquiry into the issue of whether a particular statute is money-mandating first for jurisdictional and later for merits purposes, Gollehon misplaced its reliance on Banks. In Banks, the issue was not whether the alleged statutes authorized monetary payment to persons who could establish the right to recovery under those statutes. The question was whether on the facts Captain Banks' claim fell within the terms of the statutes. On appeal we determined that the statutes alleged did not apply to his situation, and thus he failed to establish entitlement to back pay on the merits of his claim. Consequently, we held that the disposition of his case was properly a dismissal under Rule 12(b)(6) for failure to state a claim on which relief could be granted, rather than a dismissal for lack of jurisdiction, and as such was an adjudication on the merits. Banks is not support for a two-step inquiry into whether a statute or regulation is money-mandating; it remains good law on the issue before it.

Furthermore, nothing in Supreme Court opinions that address the Tucker Act suggests that a court should entertain and decide the jurisdictional and merits test in other than a single step. The single step would be one in which the trial court determines both the question of whether the statute provides the predicate for its jurisdiction, and lays to rest for purposes of the case before it the question of whether the statute on its merits provides a money-mandating remedy. (The Banks question, whether the facts of the case support a remedy, of course remains as a separate question). Because we read the Supreme Court cases to have approved a single test for deciding the money-mandating issue, we think the two-step process of Gollehon must be discarded. Gollehon is thus overruled.

When a complaint is filed alleging a Tucker Act claim based on a Constitutional provision, statute, or regulation, see 28 U.S.C. § 1491(a)(1), the trial court at the outset shall determine, either in response to a motion by the Government or sua sponte (the court is always responsible for its own jurisdiction), whether the Constitutional provision, statute, or regulation is one that is money-mandating.

If the court's conclusion is that the Constitutional provision, statute, or regulation meets the money-mandating test, the court shall declare that it has jurisdiction over the cause, and shall then proceed with the case in the normal course. For purposes of the case before the trial court, the determination that the source is money-mandating shall be determinative both as to the question of the court's jurisdiction and thereafter as to the question of whether, on the merits, plaintiff has a money-mandating source on which to base his cause of action.

If the court's conclusion is that the source as alleged and pleaded is not money-mandating, the court shall so declare, and shall dismiss the cause for lack of jurisdiction, a Rule 12(b)(1) dismissal — the absence of a money-mandating source being fatal to the court's jurisdiction under the Tucker Act.

The trial court's determination regarding the money-mandating character of the statute at issue is of course subject to appellate review as a question of law.

2.

With this background we can address the first issue: what must a plaintiff — in this case Dr. Fisher — establish regarding a money-mandating source in order for the Court of Federal Claims to have subject matter jurisdiction over the case under the Tucker Act. For decades the Supreme Court has applied what is known as the Mitchell test: a statute or regulation is money-mandating for jurisdictional purposes if it "can fairly be interpreted as mandating compensation for damages sustained as a result of the breach of the duties [it] impose[s]." Mitchell II, 463 U.S. at 217, 103 S.Ct. 2961. Recently, in United States v. White Mountain Apache Tribe, 537 U.S. 465, 123 S.Ct. 1126, 155 L.Ed.2d 40 (2003), the Supreme Court restated the test for determination of whether a statute is money-mandating for Tucker Act jurisdictional purposes. After repeating the test from Mitchell II, the Court stated that:

This "fair interpretation" rule demands a showing demonstrably lower than the standard for the initial waiver of sovereign immunity. . . . It is enough, then, that a statute creating a Tucker Act right be reasonably amenable to the reading that it mandates a right of recovery in damages. While the premise to a Tucker Act claim will not be "lightly inferred," . . . a fair inference will do.

White Mountain, 537 U.S. at 472-73, 123 S.Ct. 1126 (citation omitted; emphasis added). In dissent, four justices stated that in reaching their result (the statute was found money mandating) the majority established a new and different test for jurisdiction. The dissent described the majority's test as "a newly devised approach," id. at 482, 123 S.Ct. 1126, and stated: "The Court today fashions a new test to determine whether Congress has conferred a substantive right enforceable against the United States in a suit for money damages." Id. at 487, 123 S.Ct. 1126.

However, in a concurring opinion, id. at 479, 123 S.Ct. 1126, two justices noted that the majority opinion in White Mountain was not inconsistent with the opinion decided that same day in United States v. Navajo Nation, 537 U.S. 488, 123 S.Ct. 1079, 155 L.Ed.2d 60 (2003). In Navajo Nation the majority spoke only in terms of the established "fairly be interpreted" test, citing Mitchell II. No mention was made of the White Mountain test. (The dissent in Navajo Nation quarreled about the import of the facts, not the test to be applied.)

Both White Mountain and Navajo Nation were concerned with whether the United States owed fiduciary duties to Indian tribes under the laws relevant to the cases. In White Mountain a duty was found to exist; in Navajo Nation it was not. In White Mountain, as noted, a reformulated test was applied.

Whether White Mountain alters the Mitchell test, as suggested by the dissent in White Mountain, and whether the new test is less stringent in some respects or is the same, as suggested by the concurrence, is less than clear. Future opinions by the Supreme Court may clarify all this. For purposes of the case before us, however, this much is clear — under either the new or the old formulations Dr. Fisher has stated a claim that gives the Court of Federal Claims jurisdiction over his case.

Dr. Fisher alleges that 10 U.S.C. § 1201 provides the basis for his Tucker Act claim. Section 1201 enables the Secretary of a military branch to authorize disability retirement pay for service members on active duty. Subsection (a) provides in relevant part:

Upon a determination by the Secretary concerned that a member described in subsection (c) [i.e., on active duty] is unfit to perform the duties of the member's office, grade, rank, or rating because of physical disability incurred while entitled to basic pay . . ., the Secretary may retire the member, with retired pay . . ., if the Secretary also makes the determinations with respect to the member and that disability specified in subsection (b).

10 U.S.C. § 1201(a). Subsection (b) requires the Secretary to make certain determinations, including that "the disability is of a permanent nature and stable," and that "the disability is not the result of the member's intentional misconduct or willful neglect."

In this case, Dr. Fisher has little difficulty establishing that § 1201 is understood as money-mandating. Section 1201 was the statute alleged to be money-mandating in Sawyer v. United States, 930 F.2d 1577 (Fed. Cir. 1991). Despite the presence of the word "may" in the statute, in Sawyer we determined that the Secretary has no discretion whether to pay out retirement funds once a disability is found qualifying. Id. at 1580. Thus, we held that the statute is money-mandating because when the requirements of the statute are met — i.e., when the Secretary determines that a service member is unfit for duty because of a physical disability, and that disability is permanent and stable and is not the result of the member's intentional misconduct or willful neglect — the member is entitled to compensation. See id.

The Government here argues that Sawyer should be understood differently. According to the Government, § 1201 is money-mandating only for service members who qualify for benefits under the statute, i.e., those members who have been found by the Secretary to be unfit for duty. But that understanding turns the law on its head — according to the Government the only persons entitled to judicial relief are those who do not need it because they were awarded disability status; those who were denied that status cannot get relief because they were denied what they sought.

Such a perverse understanding of Congress's purpose cannot be the law; it is inconsistent with the literal language of the statute and with our construction of the statute in Sawyer. The fact that the statute imposes requirements for the payment of money does not mean that only claimants who have been determined by a Government official to meet those requirements have a right to the money the statute provides. It is the statute, not the Government official, that provides for the payment. If the Government official's determinations under the statute are in error, the court is there to correct the matter, and to have the proper determinations made.

In the case before us, Dr. Fisher contends that the Secretary's determination that he was fit for duty was arbitrary and capricious and contrary to law. He wants the money that would have been his due had he been discharged in the manner to which he claims he was entitled, and he wants the necessary steps taken to position himself for that result — i.e., reinstatement, etc. That is a classic Tucker Act suit for money and the related remedies the trial court is authorized to grant.

Furthermore, it is not, as the Government also insists, a declaratory judgment over which the trial court lacks jurisdiction. If Dr. Fisher were to succeed on his claim that the Secretary's decision was wrong and should be reversed, he would be entitled to disability retirement pay under § 1201, and whatever procedural remedies were necessary to achieve that result. The Court of Federal Claims is fully empowered to grant such remedies. Even if it can be said that the complaint was inartfully drafted, that does not change the basic thrust of the cause.

The resolution of the first issue, then, is that Dr. Fisher's well-pleaded complaint, clearly grounded on a statute that mandates compensation, gives the Court of Federal Claims subject-matter jurisdiction to address the case on the merits.

The resolution of the second issue — what are the consequences, once the court has taken jurisdiction, of plaintiff failing to establish all elements of the cause of action — follows from the answer to the first, though admittedly the answer has not always been stated in a consistent fashion. Assuming that the Court of Federal Claims has taken jurisdiction over the cause as a result of the initial determination that plaintiff's cause rests on a money-mandating source, the consequence of a ruling by the court on the merits, that plaintiff's case does not fit within the scope of the source, is simply this: plaintiff loses on the merits for failing to state a claim on which relief can be granted.

Certainly it does not follow that, after deciding the case on the merits, the court loses jurisdiction because plaintiff loses the case. Our cases explain that the law is to the contrary. Banks is one example. In Palmer v. United States, 168 F.3d 1310 (Fed. Cir. 1999), we held that a claim by a reserve officer that he had been improperly removed from his billet, thus denying him opportunities for pay, stated a cause of action under the cited pay statute, and thus conferred jurisdiction under the Tucker Act on the Court of Federal Claims. Id. at 1313. The ultimate conclusion in the case was that the money-mandating statute, applied to the facts proven, did not afford the remedy claimed. That was held to be a failure on the plaintiff's part to state a claim on which relief could be granted, and not a jurisdictional defect. We noted that, when the issue is raised by motion, the proper motion is under Federal Rule of Civil Procedure 12(b)(6), a dismissal for failure to state a claim on which relief can be granted — what the Court of Federal Claims formerly denominated an RCFC 12(b)(4) motion-and not a Rule 12(b)(1) dismissal for lack of jurisdiction.

B. Justiciability 1.

The third issue posed is whether, even assuming the cause of action is otherwise established, are there matters that are nonjusticiable because of their unique military implications? Justiciability has both constitutional and prudential dimensions, and encompasses a number of doctrines under which courts will decline to hear and decide a cause. Though justiciability has no precise definition or scope, doctrines of standing, mootness, ripeness, and political question are within its ambit. See generally 15 James Wm. Moore et al., Federal Practice § 101.01 (3d ed. 2003); 13 Charles Alan Wright, Arthur R. Miller and Edward H. Cooper, Federal Practice and Procedure § 3529 (2d ed. Supp. 2003).

One aspect of justiciability relates to the issue of whether deference in a given case should be given by the judiciary to the particular authority and competence of another branch of government. This can arise under basic separation of powers concepts or because Congress has dictated that such deference be given. When issues of Federal military authority are brought to the courts, the constitutional construct may dictate that, in limited circumstances, a particular controversy may be such that a decision maker other than the judiciary should have the final say. In such cases the authority of the Federal courts to protect individual rights and to decide controversies — "[t]he judicial Power shall extend to all Cases, in Law and Equity, arising under this Constitution, [and] the Laws of the United States . . ." — must be balanced against the authority of other constitutional decision makers.

An example is the Executive power vested in the President under Article II, and in particular the President's duties as Commander in Chief of the Army and Navy of the United States. When the question is one of physical or mental fitness for service in the military, courts are loath to interfere with decisions made by the President and his designated agents. It is the President who bears the responsibility for protecting the nation from harm, and the President has broad discretion in the selection of whom he chooses to perform this critical duty. This deference to Executive authority does not extend to ignoring basic due process considerations, however. When there is a question of whether reasonable process has been followed, and whether the decision maker has complied with established procedures, courts will intervene, though only to ensure that the decision is made in the proper manner.

This understanding of the law in military personnel cases by our court is well established. See, e.g., Murphy v. United States, 993 F.2d 871, 874 (Fed. Cir. 1993) ("[T]he merits of the Air Force's decision to release [an officer] from active duty are beyond judicial reach."); Sargisson v. United States, 913 F.2d 918 (Fed. Cir. 1990) (absent procedural error, the decision to release surplus officers and who should be released was a decision for the military to make); Voge v. United States, 844 F.2d 776 (Fed. Cir. 1988) (statute provided for special pay to medical officers; the decision whether to terminate the award was for the military and nonjusticiable); Heisig v. United States, 719 F.2d 1153, 1156 (Fed. Cir. 1983) ("[R]esponsibility for determining who is fit or unfit to serve in the armed services is not a judicial province.").

Though the question of fitness to serve may be nonjusticiable in various contexts, we have consistently noted that a challenge to a particular procedure followed by the military in rendering a decision may present a justiciable issue. Adkins v. United States, 68 F.3d 1317, 1323 (Fed. Cir. 1995). Even when Congress has given the military discretion in conducting its affairs, the military is bound to follow its own procedural regulations should it choose to promulgate them. Murphy, 993 F.2d at 873 (citing Sargisson, 913 F.2d at 921). A court may decide whether the military has complied with procedures set forth in its own regulations because those procedures by their nature limit the military's discretion. Id. at 873. Such a case presents a justiciable controversy because the `tests and standards' against which the court measures the military's conduct are inherent in the requirements of the applicable regulation itself. Adkins, 68 F.3d at 1323.

In one sense, the question before us is Dr. Fisher's fitness for continued service as a medical officer. At the time Dr. Fisher was released from active duty, the Air Force, following established procedures, determined that Dr. Fisher remained fit for duty, and discharged him without designating him as disabled. Thus the question could be considered to be his fitness for duty, and so considered could be understood as a question that is nonjusticiable.

In another sense, however, the question is not whether Dr. Fisher remains fit for duty as a military officer. It is clear that the Air Force does not wish to retain him in active service, and Dr. Fisher has made clear that his request for reinstatement is only for the purpose of positioning himself for obtaining a disability retirement. Rather the question is, once a military serviceperson is released from duty, are the terms and conditions of his release subject to judicial review? Specifically, can courts review the question of whether a former serviceman was entitled under the law to disability pay at the time of release from duty?

As previously noted, the trial court in the case before us decided the matter on jurisdictional grounds, and noted the non-justiciability issue only in passing. In their initial briefs on appeal, both Dr. Fisher and the Government focused primarily on the jurisdictional question, though both parties did address the justiciability question as well.

The Government argued that fitness determinations by statute are allocated to the discretion of the Secretary (of the Air Force), and that whether to retire a service member and award disability retirement pay upon any finding of unfitness is also left to the discretion of the Secretary. Further, argued the Government, there are no standards by which a court could review such findings.

Dr. Fisher responded that the Secretary's discretion is limited by the Air Force's own rules and regulations, which themselves set forth `tests and standards' against which the Secretary's discretionary conduct may be measured. Dr. Fisher specifically cited Air Force Instruction (AFI) 48-123, Medical Examination and Standards, Attachment 2 (Medical Standards for Continued Military Service) (Nov. 14, 2000). In his complaint Dr. Fisher alleged that the Secretary violated this instruction by finding that his diagnosis of seronegative rheumatoid arthritis did not disqualify him from continued service.

2.

In the course of considering the appeal in this case, and having decided that the trial court erred in dismissing on jurisdictional grounds, we examined with some care the issue of justiciability. Our review revealed a line of cases decided by our predecessor court, the Court of Claims, not cited by either party, in which that court in military disability discharge cases did not limit its review, as was generally the case in military matters if any review was allowed, to the question of whether proper procedure was followed. Instead, the court reviewed the merits of the military's decision, albeit applying a deferential standard of review.

The line of cases begins with Towell v. United States, 150 Ct.Cl. 422 (1960), in which a former officer in the U.S. Army with a history of active duty medical problems and hospitalization was released from active service. The stated reason was not physical disability, and he was deemed not entitled to a disability retirement. Subsequent review boards, up to and including the Army Board for Correction of Military Records, reaffirmed that his physical disabilities at the time of discharge were not such as to entitle him to disability retirement pay.

When Towell brought his complaint in the Court of Claims, the only official action by the Army not barred by the statute of limitations was the decision of the Correction Board. The Court of Claims reviewed the record before that Board, "in order to determine whether its conclusions were supported by substantial evidence." Id. at 434. The court concluded that "[w]e find nothing in the record which permits us to say that the Army medical judgment was wrong and as a consequence that the action of the Correction board was erroneous." Id. at 435.

Furlong v. United States, 153 Ct.Cl. 557 (1961), is similar. When plaintiff sued for disability benefits the Army denied him, the Court of Claims undertook a review of his records, and concluded that "[w]e do not think that plaintiff has sustained his burden of showing by cogent and clearly convincing evidence that the Retiring Board was arbitrary or capricious in finding that [at the time of discharge] he was not incapacitated for active service." Id. at 563.

Ward v. United States, 178 Ct.Cl. 210 (1967), was the first of these cases in which the Government lost. Despite having had his right kidney surgically removed while on active duty, plaintiff was found physically qualified for release from active duty with no physical defects; that finding was affirmed by the Navy's review boards. After reviewing the record, the Court of Claims, citing to Towell and Furlong, concluded that "[o]n the whole record, it is found that plaintiff was not physically fit for active duty at sea or on foreign service at the time of his release to inactive duty and that the decision of the Board for the Correction of Naval Records to the contrary is not supported by substantial evidence and is arbitrary." Id. at 219.

Jordan v. United States, 205 Ct.Cl. 65 (1974), followed the Ward model, reversing the Army's refusal to grant disability retirement pay to an Army sergeant who, at the time of his discharge, was deemed to be physically fit for duty. In a lengthy decision, the Court of Claims reviewed the medical history plaintiff presented and the records before the Army review boards, including the qualifications of the various doctors. The court concluded:

Even though defendant's evidence in the instant case, considered of and by itself, might support the administrative decision by the Army to discharge plaintiff as physically fit, we find, as hereinafter discussed, that there is "opposing evidence [principally, plaintiff's medical record with the VA] so substantial in character" as to detract from the weight of the evidence in support of the Army discharge, and to render it "less than substantial on the record as a whole." Ward, supra.

Id. at 73 (quotes and brackets in original).

As late as 1982, the year this court was established as the successor to the Court of Claims, the Court of Claims was reviewing military disability retirement cases on the merits, applying the substantial evidence in the record/arbitrary and capricious standard. See de Cicco v. United States, 230 Ct.Cl. 224, 677 F.2d 66 (1982) (plaintiff loses); Hinkle v. United States, 229 Ct.Cl. 801 (1982) (plaintiff loses).

In light of this precedent, which we normally are bound to follow, see South Corp., 690 F.2d at 1371, we requested from the parties additional briefing on the following questions:

1. Whether the decisions cited above [ Jordan, Ward, Furlong, and Towell] require this court to treat as justiciable an issue of fitness for military duty involving, as it does in this case, a question of disability, and, if so, whether our standard of review of a decision by a Corrections Board on that issue is to determine whether the decision was arbitrary or without substantial evidence in the record to support it;

2. Whether the authority of the above-cited cases has been eroded or overruled by later statutes or regulations, or by subsequent decisions of this circuit, see, e.g., Adkins v. United States, 68 F.3d 1317 (Fed. Cir. 1995), or by decisions of the Supreme Court;

3. Whether the rationale and holding of Lindahl v. Office of Personnel Management, 470 U.S. 768, 105 S.Ct. 1620, 84 L.Ed.2d 674 (1985), applies to the scope of our review of fitness determinations made by the military departments, and if so, whether it replaces the standard of review otherwise applicable.

Fisher v. United States, No. 02-5082 (Fed. Cir. Oct. 7, 2003) (Order).

The parties have filed supplemental briefs, and we have further considered the matter. Our conclusions regarding the three questions are the following.

a.

With regard to the first question, it is clear from this review that the controlling precedents entitle a discharged service member to judicial review on the merits of the question of eligibility for disability retirement pay. The cases are consistent that this review is conducted under a deferential standard of review, essentially the standard under which administrative agency decisions are reviewed: whether the decision is arbitrary or capricious, unsupported by substantial evidence, or otherwise not in accordance with law.

The Government in its supplementary brief on the question of justiciability candidly acknowledges that "[b]ecause the decisions in Jordan, Ward, Furlong, and Towell are based upon general presumptions of reviewability, absent further developments in the law, they would preclude this Court from affording [conclusive] deference to a determination [by the military] of fitness for military duty." Appellee's Supplemental Br. at 4. Dr. Fisher candidly agrees, and notes that, in his view, the decision closest in point to Dr. Fisher's case is Jordan. Appellant's Supplemental Br. at 3.

b.

In response to the second question asked, whether the authority of these precedents has been eroded or overruled, the Government argues that the force of the cited decisions of the Court of Claims clearly has been undermined by subsequent cases, in particular Adkins v. United States, 68 F.3d 1317 (Fed. Cir. 1995). Dr. Fisher denies that Adkins had such effect, and cites to several cases decided by the Court of Federal Claims (various trial judges), and in particular Haskins v. United States, 51 Fed. Cl. 818 (2002), in which a discharged veteran's suit against the Army for medical retirement with full disability pay was found justiciable (though the veteran lost on the merits).

The Government is correct that our more recent precedents have articulated a standard of judicial review of military service decisions broadly indicating that courts will not address the merits of such decisions. See, e.g., Adkins, 68 F.3d at 1323. As we stated earlier, that is a proper standard to apply to the basic question of an individual's eligibility to serve the nation as a war fighter. In Adkins, the question was whether the Secretary of the Army acted properly when he removed Adkins' name from the Colonel Army Promotion List. When Adkins took his case to the Court of Federal Claims, that court held it had jurisdiction over the cause, but that the matter presented was nonjusticiable. On appeal, we agreed that the merits of the decision whether to promote or not were nonjusticiable. We further concluded, however, consistent with earlier doctrine, that Adkins was entitled to judicial review of the procedures by which the decision was reached, to ensure that there were no violations of applicable statutes or regulations. The matter was remanded to the trial court for appropriate further proceedings.

Adkins was not a disability retirement situation, but a case that addressed directly who should be allowed to serve on active duty, and in what capacity. Our precedents leave little doubt that, absent procedural or due process issues, that issue is for the Executive to decide, not the courts. Thus we find nothing in the specifics of the Adkins case, or its outcome, or in other cases on which it relied, to undercut the cited precedents regarding disability retirement pay, even if such "eroding" of prior precedent were to be recognized. Nor are we aware of any statutes or any more recent Supreme Court authority that would dictate a different result.

The Government's argument did not rest entirely on Adkins and its broadly-stated propositions. The Government further supported its argument by reference to general principles of respect for military decision-making, and the importance of constitutional grants of authority to the President and Congress regarding the waging of war. In particular, in its petition for rehearing, the Government argues that when it comes to military decisions, there is no room for balancing; it is the court's duty to uphold the power of the military to govern its own affairs.

The Government supports its position by invoking two early Supreme Court decisions, Reaves v. Ainsworth, 219 U.S. 296, 31 S.Ct. 230, 55 L.Ed. 225 (1911) and Denby v. Berry, 263 U.S. 29, 44 S.Ct. 74, 68 L.Ed. 148 (1923), neither of which, the Government argues, were properly considered in the cases cited in support of this court's precedents. According to the Government, these Supreme Court decisions make both the process and substance of fitness for duty determinations nonreviewable, and dictate that this court may not substitute its judgment for the Secretary's as to who is fit to serve.

It is true that language in these cases, read as sweeping proclamations, could be taken to say that military decisions are simply nonjusticiable, and that appeals to the courts should be dismissed out of hand. The difficulty with this reading of the cases is that that is not at all what happened either in Reaves and Denby themselves, or since then.

Reaves involved a first lieutenant in the artillery who was honorably discharged from the service. A military examining board had first concluded that he was at the time of the examination physically incapacitated, and recommended his retention on sick leave. During a subsequent review, however, the board concluded that he was physically fit but that he was mentally unable to carry out his assigned duties, which led to his discharge pursuant to Presidential order. Had the lieutenant been discharged by reason of physical disability contracted in line of duty, he would have been retired with a lifetime benefit. Instead, since he was found otherwise not fit to serve, he was discharged with only one year's pay, as the law then provided. The argument of the serviceman was that the findings of the first board were final, that he should have been retained and promoted and retired in due course with a pension, and that the findings of the board could not be undone by the President.

The Supreme Court in its review of the case disposed of the finality argument by finding that the first board's conclusions were "not a final order, but a provisional one." Reaves, 219 U.S. at 300, 31 S.Ct. 230. After careful review of whether the statutes that govern such discharges had been complied with, the Court concluded that the ultimate question of who was to be retained in service was the President's, and he was fully entitled to accept the later board's recommendation.

In Denby, a Naval Reserve officer on active duty was found by a naval board of medical survey to be under permanent disability incurred in line of duty; the board recommended that his case be referred to a retiring board. The Secretary of the Navy disapproved this recommendation, and the officer was ordered released from active duty. The officer challenged the authority of the Secretary to make such an order, claiming that under the law he was entitled to have his case considered by a retiring board in the same manner as regular officers. The Court of Appeals of the District of Columbia agreed with the officer.

The Supreme Court disagreed with the Court of Appeals, finding that reserve officers came under a different statutory structure, under which the decision to order to inactive duty reserve officers serving on active duty was within the discretion of the President and his alter ego in the Navy Department, the Secretary. "Nowhere is there found any limitation upon the discretion of the Executive in this regard. The orders in such cases were in the nature of military orders by the Commander in Chief in the assignment or withdrawal of available forces to or from duty for the good of the service." Denby, 263 U.S. at 34, 44 S.Ct. 74.

In neither of these cases did the Supreme Court issue a blanket denial of judicial reviewability. On the contrary, in both cases the Court examined the statutory fabric within which the decision was made, and concluded that on the facts before it the military's decision was authorized and proper. This approach was followed in later cases. See, e.g., Harmon v. Brucker, 355 U.S. 579, 78 S.Ct. 433, 2 L.Ed.2d 503 (1958); Bell v. United States, 366 U.S. 393, 81 S.Ct. 1230, 6 L.Ed.2d 365 (1961); see also Darrell L. Peck, The Justices and the Generals: the Supreme Court and Judicial Review of Military Activities, 70 Mil. L.Rev. 1, 33 (1975).

Furthermore, neither Reaves nor Denby when considered on its facts is directly relevant to the case before us. In each of the cases that the Government cites the serviceman was trying to get the courts to order the military to take particular action — in Denby to order his reinstatement, in Reaves to order the board to place him on the retired list. Neither case presented a claim against the Government for money in the Court of Claims; neither arose under the statute involved here; neither held that a claim for retirement benefits brought in the Court of Claims was nonjusticiable. The case before us involves solely the issue of a claim for money from the United States.

We are aware of and sensitive to the admonitions contained in Supreme Court cases, such as "judges are not given the task of running the Army," Orloff v. Willoughby, 345 U.S. 83, 93, 73 S.Ct. 534, 97 L.Ed. 842 (1953), and "[t]he complex, subtle, and professional decisions as to the composition, training, equipping, and control of a military force are essentially professional military judgments," Gilligan v. Morgan, 413 U.S. 1, 10, 93 S.Ct. 2440, 37 L.Ed.2d 407 (1973). Cases like Adkins attempt to strike a careful balance between honoring those professional judgments and protecting the due process rights of our citizens.

But the case before us stands outside these admonitions — the issue is not the composition of the military, but the society's legal obligations to those who are no longer within the military forces. We are compelled by logic and the force of precedent to conclude that this question is properly subject to judicial review, and that a deferential standard of review strikes the correct balance here. (Of course, courts can provide review only so long as there are tests or standards by which the decision can be measured.)

c.

This leads to the answer to the third question asked, whether the rationale and holding of Lindahl v. Office of Personnel Management, 470 U.S. 768, 105 S.Ct. 1620, 84 L.Ed.2d 674 (1985), applies to this case, and should it in effect replace the standard of review utilized by the Court of Claims. Lindahl involved the issue of eligibility for disability retirement by civilian personnel working for the United States Government. Under the Civil Service Retirement Act, Congress early on had mandated that decisions regarding disability retirement for Federal civilian employees would be made by the Civil Service Commission; subsequently the Supreme Court concluded that such disability retirement decisions could be reviewed by district courts under the Tucker Act. Id. at 772, 105 S.Ct. 1620 (citing Dismuke v. United States, 297 U.S. 167, 172, 56 S.Ct. 400, 80 L.Ed. 561 (1936)).

Congress later amended the civil service laws to include a finality provision that applied to the Commission's disability retirement decisions. The current version of that provision, not substantively different from the original, provides that "decisions . . . [of the designated administrative agency] concerning these matters are final and conclusive and are not subject to review." 5 U.S.C § 8347(c). Interpreting this proviso in 1968, the Court of Claims in Scroggins v. United States, 184 Ct.Cl. 530, 397 F.2d 295 (1968), ruled that the statute strictly limited judicial review. Courts could not weigh the evidence or even determine whether disability determinations were supported by substantial evidence in the record. The court went on to hold, however, that there remained a residual level of review despite the statute: disability decisions could be reviewed under a highly deferential standard to determine whether there had been "a substantial departure from important procedural rights, a misconstruction of the governing legislation, or some like error going to the heart of the administrative decision." Id. at 297 (internal quotation marks omitted).

Congress in 1978 replaced the Civil Service Commission with the Office of Personnel Management (OPM). A new Merit Systems Protection Board was to review OPM's decisions in retirement cases. No modifications were made to the finality clause of § 8347(c).

Subsequently, the question arose whether the Scroggins rule still applied to disability decisions by OPM. In view of the new civil service statutory provisions, this court concluded that Congress now intended § 8347(c), the finality provision, to mean exactly what it said: there would be no judicial review of civilian employee disability retirement decisions whatsoever; the Scroggins rule was deemed no longer operative. Lindahl v. Office of Pers. Mgmt., 718 F.2d 391, 394 (Fed. Cir. 1983) ( en banc).

The Supreme Court disagreed, and reversed this court. See Lindahl, 470 U.S. at 782-83, 105 S.Ct. 1620. The Supreme Court held that the changes to the civil service system did not express a clearly stated Congressional intention to overturn the Scroggins residual review standard, and that it remained in effect. Id. As a result, the Scroggins review standard re-established in Lindahl is sometimes now referred to as the Lindahl standard. In absence of a more rigorous review standard, it does provide a residual basis on which a court could examine both the process and outcome of an administrative decision.

In its supplemental brief, the Government urges us to adopt the Lindahl test as the appropriate standard of judicial review for military disability determinations, arguing that this standard conforms with the deference owed to military decision-making. The Government argues that judicial review of military members was not intended by Congress to be subject to more searching review than that applied to civilian members of the Federal workforce. Furthermore, in the Government's view, since Dr. Fisher does not raise any claims of procedural error as such, if we were to apply the Lindahl test there would be no need for a remand for further proceedings.

Dr. Fisher sees no place for the Lindahl test in his case, pointing to the fact that Scroggins and Lindahl turned on an interpretation of § 8347(c), and that there is no parallel statute applicable to the military cases. Dr. Fisher notes that, contrary to the Congressional mandate of nonreviewability in the civilian cases, the broad language of the Tucker Act, specifically 28 U.S.C. § 1491(a)(2), argues in favor of judicial review.

The factual issue in Scroggins, in Lindahl, and in the case before us is the same: whether the individual is entitled to a disability retirement. The parallels cease at that point. First, Dr. Fisher is correct that the legal framework is quite different. There is no express statute governing military disability case review, as there is in the case of civilian disability cases. Further, as we have explained, the established fabric of judicial review of military disability decisions is fully woven: the military disability cases have their own established standard of review, a standard that strikes a balance between allowing the military to control its membership, while preserving the individual's right to earned retirement pay as provided by law.

Under these circumstances, we see no basis in the case before us under which we could substitute the Lindahl residual review standard in place of the review standard established by our precedents. We recognize the anomaly in applying a more deferential standard to review of civilian disability cases than we do to military disability cases. However, in absence of further Congressional guidance, if a more deferential standard of review such as the Lindahl test is to be applied in the military disability cases in place of the established substantial evidence/arbitrary or capricious test, that change would have to be made by the Federal Circuit sitting en banc or perhaps by the Supreme Court, if review is granted by either of those bodies.

III. CONCLUSION

The Court of Federal Claims has jurisdiction to hear the case brought by Dr. Fisher, and the issue raised by Dr. Fisher is justiciable. Accordingly, the matter is remanded to the trial court for further proceedings consistent with this opinion.

REVERSED AND REMANDED.

PLAGER, Senior Circuit Judge, additional views, joined by PAULINE NEWMAN and GAJARSA, Circuit Judges.

The court's opinion, with the support of the full court acting en banc, straightens out an area of confusion concerning the jurisdiction of the courts to hear and decide Tucker Act claims and the relationship of the jurisdictional issue to the Tucker Act's requirement for a money-mandating source. Clarity, especially when accompanied by simplicity, is to be valued.

I regret that the en banc court did not choose to address the justiciability issue as well. The Government sought to have us resolve the matter in its favor by applying two old Supreme Court cases that will not stretch that far; given the current state of the law, the conclusion the panel reaches supporting a level of intrusive judicial review is correct. However, I believe the state of the law is less than satisfactory.

I do not see a rational basis for imposing a more intrusive level of judicial review in these military disability cases than is imposed in the same cases involving civilian workers in the federal government. The issues and problems are indistinguishable. At most, the military disability cases should be subject to the same residual due process assurance standard imposed by Lindahl. I believe such a deferential standard better reflects the deference due the military in these cases, and brings them into line with the civilian cases. Why I believe this to be the right answer can best be explained by looking more generally at the question of judicial review of military decision making.

The question of whether actions by administrative officers of the Government, acting under Article II of the Constitution, are subject to review by judicial officers, acting under Article III, has roots back to the Nineteenth Century. One of the earliest cases addressing the question, Decatur v. Paulding, 39 U.S. (14 Pet.) 497, 10 L.Ed. 559 (1840), was a military case involving the application of a federal pension statute to the widow of a member of the Navy. The Court's opinion, denying judicial reviewability of the executive branch decision, did not limit itself to military concerns. From this case and others of that era developed a doctrine of non-reviewability of executive branch decisions.

The doctrine of non-reviewability held sway for a considerable period of time. It was still strong when the two cases cited to us by the Government, Reaves v. Ainsworth, 219 U.S. 296, 31 S.Ct. 230, 55 L.Ed. 225 (1911) and Denby v. Berry, 263 U.S. 29, 44 S.Ct. 74, 68 L.Ed. 148 (1923), were decided. They followed in time a series of cases involving petitions for review of military courts-martial decisions, cases in which the Supreme Court made clear that it would not allow civil court review of the merits of such military tribunals. See, e.g., Dynes v. Hoover, 61 U.S. (20 How.) 65, 15 L.Ed. 838 (1857); Ex Parte Reed, 100 U.S. 13, 10 Otto 13, 25 L.Ed. 538 (1879).

The Reaves case was not itself a criminal proceeding, but an administrative decision by the Army involving the discharge from the service of an officer who claimed he should have been discharged with a physical disability pension. The issue in Denby was whether a federal court could force the Secretary of the Navy to return to active duty an officer in the Naval reserve whom he had ordered released from active service, the officer seeking a return to active duty so he could be evaluated by a retirement board with a view to being retired with disability pay.

It is important to recognize that in both Reaves and Denby the Supreme Court did not simply say "this is a military service case, and these are cases we judges do not review." Instead, in both cases, and despite its protestations of respect for Executive authority, the Supreme Court undertook a careful review of the applicable statutes and regulations, and examined whether the military had stayed within "the scope of its lawful powers" in its decision-making. Even so, if the broad language of these cases represented the last word on review of Executive branch decision-making in general, and on review of military authorities in particular, there would be some force to the Government's argument. That of course is not the case.

It would extend unduly this opinion to recite in detail the evolution of judicial review of administrative action during the more than 160 years since Decatur v. Paulding, including the impact of Congress' enactment in 1946 of the Administrative Procedure Act (the APA). Today the doctrine of non-reviewability of Executive action is understood to have lost much of its force; although the doctrine has occasional resurgences, it is now generally accepted that there is a presumption in favor of judicial review of administrative action except when Congress has expressly mandated otherwise.

Judicial review of administrative decisions made by an agency of the military regarding service members has likewise undergone significant change. The APA itself provides exemption only for specified military functions, leaving the general run of military administrative decisions presumably subject to the Act. An exhaustive study based on an extensive review of Supreme Court decisions in the years since Decatur v. Paulding concluded that there are four established categories of legal challenges to military administrative personnel actions that are judicially reviewable: (1) lack of jurisdiction over the person; (2) violation of statutory authority; (3) violation of the military's own regulations; and (4) violation of the Constitution. In each of these categories the issue is the one posed in Reaves v. Ainsworth: did the military stay within the scope of its lawful powers in its decision-making.

The question of judicial reviewability becomes more difficult when the issue is whether the military decision, though within the scope of the authorized power, is on the merits arbitrary or capricious, or involves a denial of due process, or is unsupported or otherwise unlawful. There are indeed good reasons why aspects of the President's conduct of military affairs should not be subject to challenge on their merits in the civil courts — separation of powers argues for leaving the conduct of war and the requirements needed for war fighting to the President and Congress, both as a Constitutional construct and as a matter of relative competence. At the same time, when important individual rights are at issue and are allegedly the subject of administrative abuse, it can be argued that some room must remain for effective judicial review — the fact that an administrative personnel decision is made by an agency that is part of the military as distinct from an agency in some other part of the Executive branch should not per se immunize it from judicial scrutiny.

One suggested approach to the merits-review question contemplates a careful balancing of the interests of the individual with those of the military interests at stake. Factors affecting the individual are the nature and importance of the right asserted, and the impact of the injury resulting from the action. Factors of importance to the military include the type and amount of discretion involved; any special military expertise in the matter inherent in the decision; the extent to which judicial intervention would interfere with the military function; and whether judicial intervention would unduly impact on any special requirements of the military community.

Putting aside for the moment the force of the precedents of the Court of Claims cited in the court's opinion, and applying the above analysis, Dr. Fisher's case falls safely on the side of judicial reviewability. First and most important is the fact that the issue before us is not one of military governance and authority going to the question of war fighting; the issue is an administrative one, the outcome of which results only in a grant or denial of disbursements from the Treasury.

Looking then to the suggested factors to be balanced, the right asserted by Dr. Fisher is one to well-established pension benefits for those who suffer duty-related injury, and the denial of which would be a clear and specific financial injury. From the military side, the discretionary decision whether Dr. Fisher should be compensated for an alleged injury does not seem to be particularly a military discretion as such; its judicial review would not seem to intrude to any great extent on the needed prerogatives of the President in conducting military affairs. Nor is the expertise required to decide the case especially that of the military; to the extent courts are called upon to review medical decisions of all sorts to determine whether there has been an abuse of discretion, this one is not that different. Further, a decision in favor of Dr. Fisher does not interfere with a particularly military function — the same problem of eligibility for disability compensation arises in the civilian context as well, and since the issue is not whether Dr. Fisher should be returned to military duty but how if at all he should be compensated for past duty, any special requirements of the military community would not seem to be significantly implicated.

In my view, the Government's reliance on almost century-old cases, and on sweeping pronouncements from an earlier era in administrative law, is less persuasive than it needs to be in order for the Government's argument of absolute non-reviewability to prevail in this case. At the same time, the Government is correct that it is anomalous for courts to continue to review on the merits military disability decisions when the same issues arising in the federal civil service are by court rule and statute not so reviewable. If the policy of the United States is to make agency disability decisions for federal employees subject only to the quite limited review provided by the Lindahl standard, there seems no compelling reason not to apply that same standard to the same issues arising when the agency is a part of the military establishment. In providing uniformity of civilian/military treatment, we would also more properly reflect the deference owed to the Executive in the area of military governance.

The initial choice of the standard of review in military disability cases was made by the judiciary, and presumably an appropriate body of the judiciary could change it. It would certainly be within the authority of Congress to apply the same statutory review standard to both situations. Given the state of the law, however, this is not a matter that a panel of this court has the power to correct.