985 F.2d 83
Prod.Liab.Rep. (CCH) P 13,474
Scott E. LEWIS and Victoria Lewis, Plaintiffs-Appellants,
BABCOCK INDUSTRIES, INC., McDonnell Douglas Corp., and
General Dynamics Corporation, Defendants-Appellees.
No. 538, Docket 92-7703.
United States Court of Appeals,
Argued Dec. 9, 1992.
Decided Feb. 4, 1993.
David Beekman, New York City (Milton G. Sincoff, Daniel M. Kolko, Kreindler & Kreindler, New York City, on the brief), for plaintiffs-appellants.
Louis R. Martinez, New York City (Haight, Gardner, Poor & Havens, New York City, on the brief), for defendant-appellee Babcock Industries, Inc.
Michael G. Biggers, New York City (Bryan Cave, New York City, on the brief), for defendant-appellee McDonnell Douglas Corp.
John F. Keating, New York City (Hill, Betts & Nash, New York City, on the brief), for defendant-appellee General Dynamics Corp.
Before: MESKILL, Chief Judge, OAKES and NEWMAN, Circuit Judges.
JON O. NEWMAN, Circuit Judge:
This appeal concerns the military contractor defense, see Boyle v. United Technologies Corp., 487 U.S. 500, 108 S.Ct. 2510, 101 L.Ed.2d 442 (1988). The precise issue is whether the defense protects a contractor from whom the Government purchased replacement parts with knowledge of a design defect even if the Government had not previously approved specifications for the product with knowledge of the defect during the design phase. This question arises on an appeal by Scott E. Lewis and Victoria Lewis from the June 10, 1992, judgment of the District Court for the Southern District of New York (John M. Cannella, Judge) dismissing their complaint on a motion for summary judgment. We uphold the military contractor defense under the circumstances of this case and therefore affirm.
On February 23, 1987, Scott Lewis was flying a United States Air Force F-111-F jet fighter over England. When the plane malfunctioned, he initiated ejection of the self-contained crew module. The crew module is designed so that once the pilot initiates ejection, all subsequent operations until landing occur automatically. After the module separates from the aircraft, a parachute system deploys, and large bags on the bottom of the module cushion the landing. In this case, however, the forward repositioning cable, one of two cables that connect the parachute to the module, severed, causing the module to land at an incorrect angle without adequate cushioning. Lewis sustained spinal injuries.
The Air Force concluded, after a post-accident investigation, that the breaks in the forward repositioning cable were caused by corrosion. The cable was made of high carbon steel, which is corrosive unless protected. The cable was protected by a polyolefin coating, which adequately protects the steel as long as the coating is not cut. During normal operations, the cable rests under a light metal cover in a channel in the center of the windshield. General Dynamics Corp. designed and manufactured the aircraft. It subcontracted for the design and manufacture of the crew module with McDonnell Douglas Corp., which in turn subcontracted with Babcock Industries, Inc. for the parachute cables.
During the design phase, the entire crew module was designated a "critical item" by the Air Force. Items were designated critical by virtue of their unusual complexity, developmental nature, extreme cost, or relevance to safety. Because of their importance, the Air Force subjected the design and development of "critical items" to extensive review. The forward repositioning cable itself, however, was a "non-critical item." The Government did not dictate the materials from which to make the cable, but only its required dimensions and strength. Early in the design process, the Government performed an extensive review of the entire crew module, as a critical item, during the so-called First Article Configuration Inspection ("the FACI") and during tests of the module. The parties dispute the level of attention paid by the Government to the forward repositioning cable, a non-critical item, during the FACI and whether the Government was aware of the cable's composition and susceptibility to corrosion at that time.
After General Dynamics delivered the F-111s to the Air Force, the Air Force made an initial redesign of the windshield of the F-111 for reasons not relevant to the pending appeal. The Air Force selected a different contractor for this redesign. The redesigned windshield was connected to the aircraft with clips located in the same channel as the forward repositioning cable. The Air Force eventually noticed corrosion of the cables in the aircraft with the new windshield design. After an investigation completed in 1980, the Air Force determined that the clips from the modified windshield were cutting the cable's protective coating, thus allowing moisture to reach the carbon steel cable and cause corrosion.
The Air Force decided to replace the forward repositioning cables on all F-111s. The cable in the aircraft involved in Lewis's accident was replaced on January 21, 1983, with a cable purchased from Babcock. The Air Force also changed its maintenance manual to warn its personnel to be careful not to cut the cable's coating during maintenance operations. The Air Force did not change the windshield design, nor did it change the cable composition from that of the original Babcock cable. At least by this time, the Air Force was aware of the cable's susceptibility to corrosion if its coating was cut. After the crash of Lewis's plane, the Air Force performed more tests. Based on these tests, the Air Force decided to continue use of the Babcock cable and to make a second redesign of the windshield to avoid its cutting of the cable's coating.
Lewis ejected from a plane equipped with a replaced cable and a windshield constructed according to the first redesign. The cable broke in two places, only one of which was in the portion of the cable resting in the channel where the windshield clips were located. The parties dispute whether the other break was caused by corrosion or whether it was the result of the added stress created as the portion of the cable originally located in the channel was breaking. Lewis and his wife, claiming derivatively, sued General Dynamics, McDonnell Douglas, and Babcock (collectively "appellees") alleging negligence in their design of the cable because the high carbon steel chosen was susceptible to corrosion from moisture and the polyolefin protective coating was easily cut, thereby limiting its effectiveness. Their complaint also contained claims based on theories of strict liability and breach of warranty. Judge Cannella granted appellees' motion for summary judgment based on the military contractor defense of Boyle.
I. The military contractor defense
In Boyle, the Supreme Court held that the federal common law provides certain military contractors with a defense against suits based on state tort law. The defense is applicable only to those contractors for whom the duties of care imposed by state tort law conflict with the duties imposed by their federal contract. Under this defense, conflicting state law is displaced "when (1) the United States approved reasonably precise specifications; (2) the equipment conformed to those specifications; and (3) the supplier warned the United States about the dangers in the use of the equipment that were known to the supplier but not to the United States." Boyle, 487 U.S. at 512, 108 S.Ct. at 2518. As we explained in our first extended discussion of Boyle:
The first element ensures that the design specification at issue actually was considered by a government official--or, in other words, that a government official had made the type of policy decision considered a discretionary function under the [Federal Tort Claims Act]. The second element ensures that the supplier actually acted upon the Government's discretion. The final element encourages the contractor to provide all information at its disposal to the government official and in so doing facilitates a fully informed decision by the government official.
In re Joint Eastern and Southern District New York Asbestos Litigation, 897 F.2d 626, 629 (2d Cir.1990) ("Grispo ").
As to the first requirement, the District Court found that during the FACI, the Government examined and approved every component of the module, including the cable, and that the Government conducted extensive testing of the crew escape module, including the cable. The District Court found especially persuasive the fact that the Air Force continued to use the cable after it found out that it was susceptible to corrosion when it was cut. In effect, the District Court held that approval may be established by the Government's continued use of a product after it learns of the defect. The District Court found that appellees met Boyle's second requirement because the Government had inspected and accepted the aircraft when it was first delivered and had also inspected and accepted the replacement cables produced by Babcock. Finally, the Court held that appellees met Boyle's third requirement because the Air Force, from its experience with the aircraft, had greater knowledge than appellees of the corrosion problem.
II. Does Boyle have prerequisites?
Lewis urges us to adopt an interpretation of Boyle under which the three-part test outlined above is not applied unless two prerequisites are established: a significant conflict between the requirements under the federal contract and state tort law, and the Government's exercise of its discretion in accepting a safety risk. In rejecting this argument, the District Court concluded that it makes more sense to consider whether a significant conflict exists in determining the first element of the Boyle test, approval of reasonably precise specifications, rather than as a prerequisite, because "[b]y satisfying this element, the contractor establishes that a conflict exists between state law which imposes liability and the duty imposed by the Government contract." In considering the first element, the District Court also examined whether the Government exercised discretion or merely "rubber stamped" the design.
Although, at the end of the analysis, federal common law provides the contractor with a defense only if there is both (a) a conflict between the requirements of state tort law and those of the Government contract and (b) an exercise of Government discretion in considering the design feature in question, we agree with the District Court's approach that these inquiries are part of the Boyle test and not prerequisites to it. The purpose of the first part of the test, approval of reasonably precise specifications, is to determine whether a conflict with state law exists at all. As we indicated in Grispo, a federally imposed contract requirement displaces only a parallel state law requirement. 897 F.2d at 630. If the Government did not approve reasonably precise specifications for the design feature in question, there is no conflict with state law.1 On the other hand, if the Government did approve such specifications, conflicting state tort law is preempted. Thus, answering the question whether the Government approved reasonably precise specifications for the design feature in question necessarily answers the question whether the federal contract conflicts with state law.
We also disagree with Lewis's proposition that the Government's exercise of discretion in accepting a safety risk should be considered a prerequisite to applying the three-part Boyle test. In its search for a limiting principle for the military contractor defense, the Supreme Court rejected the Feres doctrine, see Feres v. United States, 340 U.S. 135, 71 S.Ct. 153, 95 L.Ed. 152 (1950), which states that the Federal Tort Claims Act ("the FTCA") does not cover injuries to military personnel during military service, and instead chose the discretionary function exception to the FTCA, see 28 U.S.C. § 2680(a) (1988). See Boyle, 487 U.S. at 509-12, 108 S.Ct. at 2517-18. The Supreme Court indicated that the first two parts of the Boyle test implement this limitation by "assur[ing] that the suit is within the area where the policy of the 'discretionary function' would be frustrated-- i.e., they assure that the design feature in question was considered by a Government officer, and not merely by the contractor itself." Id. at 512, 108 S.Ct. at 2518. Thus, in determining whether the Government approved reasonably precise specifications, a District Court necessarily inquires into whether the Government adequately exercised its discretion and "thereby limit[ed] the contractor's ability to accommodate safety in a different fashion." Grispo, 897 F.2d at 632. The purpose of this inquiry is to exclude from the defense those cases where the Government merely "rubber stamps" a design, see Trevino v. General Dynamics Corp., 865 F.2d 1474, 1480, 1486 (5th Cir.), cert. denied, 493 U.S. 935, 110 S.Ct. 327, 107 L.Ed.2d 317 (1989), or where the Government merely orders a product from stock without a "significant interest" in the alleged design defect, see Boyle, 487 U.S. at 509, 108 S.Ct. at 2517.
III. At what point in time should the Boyle test be applied?
The parties' dispute turns in large part on the time when the Boyle test is to be applied. Lewis asserts that a contractor can meet the requirements of Boyle, if at all, only during the design phase and argues that with reference to that time, summary judgment is not appropriate in this case. Appellees respond that they can satisfy the requirements of Boyle through Government approval that occurs at any time prior to the accident. We understand the District Court's opinion to adopt the appellees' approach to the time frame. Judge Cannella found the first element satisfied by events that happened at the design stage and by the Air Force's responses to problems of which it learned during its experience with the F-111 after the aircraft were delivered. He found the second element satisfied based on the Government's acceptance, after review, of the aircraft and the replacement cables. He found the third element established because the Air Force knew of the cable's susceptibility to corrosion long before the accident occurred based on its analysis of the corrosion problem that led it to replace all the cables in the early 1980s. At least for the third requirement, the District Court did not determine whether appellees, during the design phase, adequately informed the Government of problems in the cable's design of which their knowledge was superior to the Government's. The Court made this determination only at the post-design stage.
The Fourth Circuit has twice dealt with situations presenting the question of when the Boyle test should be applied. In the first case, Dowd v. Textron, Inc., 792 F.2d 409 (4th Cir.1986), cert. denied, 487 U.S. 1233, 108 S.Ct. 2897, 101 L.Ed.2d 930 (1988), two servicemen were killed when the helicopter they were flying crashed. Their widows brought suit against the manufacturer. They alleged that the rotor system was defectively designed permitting, under certain circumstances, the rotor to sever the mast to which it was attached. See id. at 410. The rotor system had been used in previous versions of the helicopter. See id. at 410-11. At least forty-six incidents had occurred prior to the accident at issue in which the rotor severed the mast of helicopters using the same rotor system. The Army studied the problem, recommended educating the pilots about the risk, and decided to study the problem further. The manufacturer also performed several studies and recommended three variations to the Government. Although the Army chose to modify the rotor system on previous versions of the helicopter, it decided to continue to use the original system on a new version of the helicopter then being designed. This new version was the one involved in the crash. See id. at 411.
The Court rejected the plaintiffs' argument that since the manufacturer originally designed the rotor system without Army participation, the Government had not approved reasonably precise specifications. It reasoned that "[t]he length and breadth of the Army's experience with the 540 rotor system--and its decision to continue using it--amply establish government approval of the alleged design defects." Id. at 412 (emphasis added). The Army required the manufacturer to use the original rotor system in the new version of the helicopter, and the manufacturer could not modify the design without Government approval. See id. The Court was also concerned that if the manufacturer's negligence at the initial design stage created permanent future liability, the manufacturer's incentive to propose safety modifications for its product would be diminished. See id.
In the second case, Ramey v. Martin-Baker Aircraft Co., 874 F.2d 946 (4th Cir.1989), the ejection seat of an F-18 fighter aircraft triggered during its removal and injured a mechanic. The mechanic, alleging that the manufacturer defectively designed the component and failed to apprise the Navy of the danger the seat posed to mechanics, brought suit against the manufacturer. Adopting the language in Dowd quoted above, the Court found both Boyle 's first prong and third prong met because the "Navy was well aware of the seat's harmful propensities" before the accident occurred through its experience with the aircraft. Id. at 951. This awareness was evidenced by internal documents indicating the Navy's concern that the method of removing the ejection seat, prescribed in the maintenance manuals, was deficient. See id.2 The Court found that the Navy's continued use of the component after it had experience with the potential design problems constituted approval of reasonably precise specifications, see id. at 950, and that the manufacturer could not have informed the Navy of anything the Navy did not already know, see id. at 951.
Our case presents a factual scenario that is less compelling for the application of the military contractor defense at the post-design phase than Dowd, but more compelling than Ramey. In Dowd, the Army was in the process of ordering a new helicopter that would modify certain aspects of the previous version of the helicopter. The Army's experience with the rotor system, a design that it had not originally approved, came from its experience with the prior versions of the helicopter. In addition, to aid the Army's evaluation of the alleged design defect, the manufacturer presented alternative designs. In effect, the decision in Dowd to continue use of the rotor system was made at the design stage of the new version of the helicopter and was informed by alternative designs. In our case, by contrast, the Air Force was evaluating an alleged design defect in an aircraft that had already been produced, and alternative designs for the cable were not evaluated until after Lewis's accident.
In Ramey, when the accident occurred, the Navy was continuing to use the deficient maintenance manual while apparently in the process of deciding how to deal with the risk created by the alleged design defect. See 874 F.2d at 951 n. 9. In our case, on the other hand, the Government had completed its investigation, and based on its conclusions, ordered the same part to use as a replacement for the corroded cables. In essence, the Government decided that changes in the maintenance manual and the replacement of existing cables were an adequate response to the corrosion problem.
Consideration of these Fourth Circuit cases leads us to agree with the District Court and appellees that a contractor may satisfy the Boyle test after the design stage has been completed, at least in the situation presented here. Although the Government was not in the process of designing a new aircraft as in Dowd, the Government reordered the same cable to replace existing ones. It was one of these replacement cables that failed in the accident at issue. This reorder occurred after the Government had completed its investigation into the problem and while it knew that the Babcock cable was susceptible to corrosion if the protective coating was cut. We hold that when the Government reordered the specific Babcock cable, with knowledge of its alleged design defect, the Government approved reasonably precise specifications for that product such that the manufacturer qualifies for the military contractor defense for any defects in the design of that product.3
The Air Force exercised its discretion in deciding that changes in the maintenance manual along with replacement of the existing cables were a sufficient response to the corrosion problem. After Lewis's accident, faced with alternative designs for the cable, the Air Force chose to redesign the windshield a second time and to continue use of the Babcock cable. Whether or not these decisions were ill-advised, or should have been made earlier, it is not our role to second-guess the Air Force's judgment. And since the Government specifically requested the cable that failed despite awareness of its alleged design defect, thus precluding characterization of the cable as a stock item that happens to have a design defect, we also cannot second-guess the contractor's decision regarding the materials to be used for the cable. Based on the reorder, the contractor can claim: " 'The Government made me do it.' " Grispo, 897 F.2d at 632. The imposition of liability under state law would constitute a significant conflict with the Government's decision to order replacement cables. Furthermore, if liability were imposed on the contractor based on its previous actions, the contractor would seek to raise the price of the replacement cables to cover its anticipated liability from the damage that might be caused by their failure. Such reaction would frustrate the policy underlying the FTCA's discretionary function exception by placing the cost of the Government's discretionary decisions on the Government itself when it contracts for a product. See Boyle, 487 U.S. at 511-12, 108 S.Ct. at 2518.
The remaining two parts of Boyle were also appropriately determined in appellees' favor on motion for summary judgment. Appellees have met the second part of the Boyle test, conformity with reasonably precise specifications, because when the Government reordered the cable, it received exactly what it sought--a cable that had certain dimension and strength characteristics, made of carbon steel coated by polyolefin. The Government also approved the replacement cable shipments it received.
Moreover, since we are evaluating the Boyle test at the time the Government reordered the cable, it is clear that the third requirement is also met. There is no requirement that appellees inform the Air Force of dangers already known to the Air Force. See Boyle, 487 U.S. at 512, 108 S.Ct. at 2518. Because of its experience with the aircraft, the Air Force had greater knowledge of the problem than appellees had. See Ramey, 874 F.2d at 951; Harduvel v. General Dynamics Corp., 878 F.2d 1311, 1321-22 (11th Cir.1989), cert. denied, 494 U.S. 1030, 110 S.Ct. 1479, 108 L.Ed.2d 615 (1990). As early as 1980, the Government determined that the cable materials were appropriate as long as care was taken not to cut the coating. Lewis argues that although the Government was aware of a problem, they were misled by appellees that the cause of the problem was the windshield redesign rather than the easy penetrability of the coating. We find Lewis's distinction unpersuasive. The Government knew that the coating could be cut in other ways; in its 1980 report, the Government characterized the windshield clips as the "primary cause of damage." Joint Appendix at 867. This conclusion also renders the parties' dispute over the cause of the other break irrelevant to our disposition. Even if the other break were caused by corrosion, the Government's decision to reorder the cable in light of its knowledge that the coating was susceptible to cutting in several ways also serves to immunize appellees for accidents caused by cuts from sources other than the initial windshield redesign.4
Because the District Court correctly granted appellees' motion for summary judgment under the military contractor defense based on the Government's actions and knowledge in the post-design stage, we affirm.
This proposition was illustrated by the Supreme Court in Boyle when it noted that there would be no conflict between a federal contract for the purchase of an air conditioner that specified only the cooling capacity, leaving the precise design to the contractor, and state law requiring a certain safety feature. See Boyle, 487 U.S. at 509, 108 S.Ct. at 2517. Since a manufacturer could comply with both the federal contract and state tort law, the Government had not approved reasonably precise specifications for the safety feature in question
The Court alternatively ruled that the Navy's participation in the design amounted "to more than a rubber stamping," and thus was sufficient to qualify as approval of reasonably precise specifications. Id. at 950
We do not decide whether the contractor can invoke the military contractor defense where the Government merely tolerates a defect through continued use of a product in the face of knowledge of a design defect acquired after the design stage ended. The District Court may have relied primarily on the Government's continued use of the cables, though it did refer to replacement
We need not decide whether the initial windshield redesign was, as a matter of law, an intervening cause of the cable's failure such that appellees would be relieved of liability
Furthermore, Lewis has not presented a claim or a sufficient factual basis supporting a claim that the defense should not apply because the appellees' actions at the design stage so constricted the Government's discretion when it reordered the cable that the Government was precluded from making the same decision (to reject or demand modification) that it would have made if presented with the issue at the design stage.