Full opinion text
LAWRENCE, Chief Judge. INDEX PART ONE Page (1) OUTLINE OF CASE.................................... 264-265 (2) THE THIOKOL CONTRACT AND GOVERNMENT PROCUREMENT PROCEDURES......................... 265-268 (3) THE TRIP FLARE FACILITY BEFORE THE EXPLOSION...................................... 268-269 (4) ILLUMINANT PRODUCTION PROCESS.................. 271-272 (5) BLACK WEDNESDAY AT WOODBINE................... 272-273 (6) THE FACILITY: AFTER THE EXPLOSION (THE THIOKOL INVESTIGATION) ...................... 273-275 (7) DOD SAFETY STANDARDS ............................ 275-277 (8) THE CLASSIFYING OF ILLUMINANTS.................. 277-279 (9) DISSIDENT VOICES............... 279-280 (10) THE 1968 THIOKOL (WASATCH) TESTS................. 280-281 (11) THE ROLFE REPORT, BRITISH MINISTRY OF DEFENSE...................... 281 (12) CLASS 2 UPGRADED IN THE CASE OF VARIOUS PYROTECHNIC COMPOSITIONS ........................ 281-282 PART ONE Page (13) FAILURE TO NOTIFY THIOKOL OF THE DECISION TO UPGRADE DECISION.......................... 282-284 (14) CONTENTIONS OF PARTIES (A) Plaintiffs’..................................... 284 . (B) Defendant’s................................... 284 (15) THE LAW: A CONSPECTUS (a) Thiokol’s Status as an Independent Contractor and the Legal Consequences thereof in this Case.......... (b) The Discretionary Function Exception............... 285- 286 286- 288 (c) Liability of Employers Contracting with Independent Contractors for Inherently Dangerous Work.......... 288-289 (d) Proximate Cause............................... 290 PART TWO (16) WAS THE UNITED STATES NEGLIGENT?.......... 290-293 (17) WAS THE NEGLIGENCE OF THE GOVERNMENT A PROXIMATE CAUSE OF THE EXPLOSION?......... 293 (A) NEGLIGENT ACTS OF THIOKOL AS INTERVENING EFFICIENT CAUSE OF EXPLOSION............................... 293-297 (B) WOULD THE NECESSARY CHANGES IN SAFETY STANDARDS REQUIRED BY THE CLASS 7 DESIGNATION HAVE BEEN IMPLEMENTED BEFORE THE EXPLOSION IF THIOKOL HAD BEEN TIMELY NOTIFIED OF THE ARMY’S RECLASSIFICATION?.......................... 297-299 PART THREE FINDINGS OF FACT AND CONCLUSIONS OF LAW I. FINDINGS OF FACT............................ 299-301 II. CONCLUSIONS OF LAW ........................ 301-303 OPINION AND FINDINGS OF FACT AND CONCLUSIONS OF LAW PART ONE (1) OUTLINE OF CASE These Federal Tort Claim actions have resulted from the fire and explosion which occurred at the Woodbine, Georgia plant of Thiokol Chemical Corporation on February 3, 1971. Thiokol was engaged in the manufacture, under contract with the Army, of trip flares which were used by the military during the war in Viet Nam as an aid to troops subjected to attack at night. The flares are ignited by pulling the trip or cutting of the trip wire. A bright flame is emitted which lights up a considerable area. On February 3, 1971, around 60 employees of Thiokol were working in or near Building M-132 in which the flares were produced. At 10:53 A.M. a fire broke out at the “first fire” addition station in the facility. The loose illuminant material (magnesium and sodium nitrate) burns at a speed measured in milliseconds and reaches very high temperatures. The fire ran down the ignition pellet assembly line and eventually got into the cure room where 8,000 pounds of loose illuminants were being cured in trays. Also in the curing room were 56,322 candles containing approximately 0.3 pounds of illuminant each; 18,472 ignition pellets, and 100 pounds of first fire and intermediate mix. An enormous pressure built up as the result of the deflagration of the illuminants. The fire culminated in an explosion in the cure room that destroyed the building. Twenty-nine employees lost their lives. More than fifty other employees were injured. The amounts sought by the plaintiffs aggregate $717,526,391. The action against the United States by the injured employees and representatives of deceased employees is brought under the Federal Tort Claims Act, 28 U.S.C. § 1346(b) and §§ 2680 et seq. In many instances Thiokol Chemical Corporation (hereafter Thiokol) was named as a party defendant along with the United States. The Company was covered by workmen’s compensation insurance and its carrier paid the many claims in accordance with the Georgia Act. That statute provides the exclusive remedy against an employer of an employee injured in the course of his employment. The Negro plaintiffs challenged the constitutionality of the Workmen’s Compensation Act on the ground that the average-weekly-earnings basis of computing compensation payments discriminates against low income employees in violation of the equal protection clause of the Fourteenth Amendment. That contention was denied on December 21, 1973. Thiokol was dismissed as a defendant. See Massey et al. v. Thiokol Chemical Corporation and United States, 368 F.Supp. 668 (S.D., Ga.). After extended discovery, the case was tried on the liability issue alone at a trial which lasted approximately three weeks during June, July and August, 1974. The record in this case is over 7,500 pages long. It includes pleadings, interrogatories and answers, depositions, reports, photographs, manuals, drawings, treatises, films and many other materials. There is no transcript of the testimony with the exception of that of one witness. Apparently a transcript of the trial was not ordered by counsel. However, most of the witnesses had given lengthy depositions. They were introduced in evidence and are before the Court as a part of the record. The tapes of the testimony of other witnesses are available. I have listened to relevant parts of the recordings of what witnesses, previously not deposed, testified to at the trial. Unfortunately, the testimony of one of the Government witnesses, Dr. Francis R. Taylor, was not recorded. The failure was apparently due to malfunction of the machine or to some error by the Reporter who is now deceased. He took no stenographic notes of the evidence. The pretrial depositions do not vary in any material way from what is found in the depositions. As I recall, no effort was made by counsel for either side to question such witnesses as to any disparity. (2) THE THIOKOL CONTRACT AND GOVERNMENT PROCUREMENT PROCEDURES (A) Early in the 1960’s Thiokol established a plant at Woodbine, Georgia. One segment of its widespread business activities was known as the Aerospace Group. For some years the Company had specialized in advanced propulsion systems for missiles and space vehicles. It has engaged in the production of rocket components, flares and ordnance devices. At the time the Woodbine operation commenced the space program was concentrated on landing unmanned and manned space vehicles on the moon. Camden County was selected as the site of the Company’s plant for the manufacture of very large solid propellant motors. Woodbine was remote from any metropolitan area and was accessible to the water for shipment of engines and fuel to Cape Kennedy. In 1965 the National Space Program decided against the use of solid propellants and the Government withdrew financial assistance to Thiokol. The Company converted its Georgia facility to the production of various materials, including specialty chemical and ordnance systems for the military. On October 1, 1969, Thiokol entered into the contract with the United States through the Department of the Army to manufacture 754,000 M 49 Al surface trip flares. The contract was made pursuant to an Invitation for Bids sent by Picatinny Arsenal, Dover, New Jersey, the designated contracting office, to various proposed manufacturers. To it was attached a Technical Data Package containing the specifications for the item to be produced. Seventeen manufacturers submitted bids. Thiokol was the lowest bidder. The chain of command for procurement of the contract in question is as follows: The procurement in the present case originated with the Department of the Army and went down the chain of command to AMC and MUCOM. Picatinny Arsenal was the Procuring Contracting Office for privately-owned, privately-operated (POPO) plants. APSA was primarily responsible for procurement at Government-owned, contractor-operated (GOCO) facilities. DCASR is a joint service organization. It is set up as an entity within the DOD to administer military procurement contracts at the field level. Eugene C. Heckler was the Administrative Contracting Officer (ACO) in Atlanta. He was assisted by legal, safety and industrial specialists and quality assurance personnel. DCASR-Atlanta had authority over Government personnel stationed at the Woodbine plant. The head of such detailed personnel was Herbert Boyd, the Contracting Officer’s Representative (COR). Mr. Boyd’s function was one of quality assurance to endeavor to insure that the Government obtained a quality end product. C. F. Otken of the Picatinny Arsenal was the Procuring Contracting Officer (PCO). The division of function between the ACO and the PCO in this contract was as follows. The PCO was the ultimate contract administrator and made arrangements for production or engineering assistance from Picatinny Arsenal if Thiokol needed same. The ACO was responsible for on-site quality assurance, field contract administration, and safety. The safety involvement of the ACO was to make pre and post-award safety surveys. The latter consisted of annual safety surveys of the plant. Thiokol could call upon the ACO for safety advice or assistance if needed, and the ACO could call upon the PCO for safety assistance. If the Contracting Officer’s Representative (COR) observed a safety violation, he could request the contractor to remedy it immediately or advise the ACO. After the COR dealt directly with the contractor, he would advise the ACO in the event the contractor did not remedy the violation. The ACO could then deal directly with the contractor or have his safety personnel do so. If no remedy was achieved, the ACO would report the violation to PCO, who would intervene. The AGO could also order the quality assurance people out of the plant. This would have the effect of shutting down production. The PCO could default the contractor and terminate the contract. After it was determined that Thiokol was the low bidder for the trip flare contract, Mr. Otken sent a pre-award survey form to the commanding, officer of DCASR-Atlanta. The form requested that a pre-award survey of Thiokol be conducted to determine whether it could comply with the terms of the contract. A safety survey was performed on October 28, 1969, by Theron W. Driscoll, Chief, Office of Specialized Safety, DCASR-Atlanta. It indicated Thiokol would meet the safety requirements of the bid package. Building M-132, where the manufacturing was performed and in which the February 3, 1971, fire began, was empty at the time of the survey. However, DCASR accepted the contractor’s assurances that it was capable of performing the work. DCASR-Atlanta recommended approval of an award to Thiokol on November 5, 1969, and Mr. Otken certified its responsibility on November 25, 1969. The award was made and the contract was let under a document signed by Mr. Otken on December 12, 1969, effective as of December 2, 1969. On December 22, 1969, Thiokol requested its first delay in the contract time schedule relative to the date for submission of a first article for testing by the military. Thiokol’s original submission for first article approval was rejected and a request for further delay was made, this time in reference to the delivery schedule for the finished end-item flare. By August 12, 1970, production had begun on the M49A1 trip flare. This was ascertained by Stephen P. Carl, Office of Safety, DCASR-Atlanta, during a preaward survey in conjunction with a contract for manufacture of an 81 mm. illuminating round. He also indicated that Thiokol was producing the trip flare and CS-2 Riot Control Agent at the same time and that both contracts were scheduled for completion in October, 1970. (B) The Department of Defense “Contractors’ Safety Manual for Ammunition, Explosives and Related Dangerous Material” (October, 1968), PI. Ex. 2, is integrated with Thiokol’s contract. It provides: “103. Responsibilities a. The Procuring Contracting Officer (PCO) is responsible for: (1) Providing safety clauses and technical safety data in contracts for ammunition, explosives and related dangerous materials, using this manual, in whole or in part, as appropriate. (2) Seeking counsel or advise of the cognizant DoD Component safety staff to insure that adequate safety criteria are included in the contract. (3) Having pre-award safety surveys conducted to determine that the contractors can meet the applicable criteria contained herein. (4) Evaluating conditions found during pre-award surveys of contractors plants. When the applicable safety criteria cannot be met, and only upon authority of the cognizant DoD Component, amend the contract to reflect the authorized waivers or exemptions (see para 104). (5) Requiring the contractor to provide site and construction plans for review under conditions set forth in paragraph 105. (6) Taking appropriate action on safety matters reported by the ACO. b. The Administrative Contracting Officer (ACO) is responsible for: (1) Assuring that the contract contains safety clauses and technical safety data and that the contractor complies with same; reporting noncompliance to the PCO. (2) Seeking advice of the safety organization of the cognizant DoD Component on matters relating to safety. (3) Performing pre-award safety surveys when requested by the PCO. (4) Forwarding requests for waivers or exemptions in accordance with para 104, with appropriate recommendations.” The Manual states that “For contractor owned contractor operated plants, it is the responsibility of the Procuring Contracting Officer to determine the acceptable protection after considering the hazard involved and the risk that must be assumed.” The purpose of the DOD Manual is to prescribe “safe methods, practices, and standards for insuring continuity of production, safe-guarding personnel, and preventing property damage at contractor operated plants.” See Para. 103. The Contractors' Manual classifies manufacturing hazards. Class 2 is a fire hazard. See Para. 707. It embraces: “Items which burn vigorously with little or no possibility of extinguishment in storage situations are class 2. Explosions normally will be confined to pressure ruptures of containers and will not produce propagating shock waves or damaging blast overpressure beyond the magazine distances specified for this class. There may be a severe hazard of the spread of fire from tossing about of container materials, burning propellant, or incendiary materials. Toxic effects normally will not exist beyond the inhabited building distances specified for this class.” The most hazard is Class 7 (Mass Detonating Hazards). See Para. 709. Items in this classification are defined as “Those, most of the entire quantity of which will explode virtually instantaneously when a small portion is subject to fire, to severe concussion of impact, to the impulse of an initiating agent, or to the effect of a considerable discharge of energy from without. Such an explosion normally will cause severe structural damage to adjacent objects and the direct propagation of the detonation to other separated explosives and ammunition placed sufficiently close to the initially exploding pile.” In the case of Class 2 hazards where over 20,000 pounds of the material are involved the site must be 190 feet from a public road. For Class 7 hazards there must be barricades and minimum distances between magazines depending on the net pounds of explosives involved. Among the provisions incorporated by reference in the contract is the following disclaimer of responsibility on the part of the Government for the safety of the public and its personnel: “(d) Neither the requirements of this clause nor any act or failure to act by the Government in surveillance or enforcement thereof shall affect or relieve the Contractor of responsibility for the safety of his personnel and his property and for the safety of the general public in connection with the performance of this contract, or impose or add to any liability of the Government for such safety. The Contractor is not entitled to rely on the requirements of this clause or on any Government surveillance or enforcement thereof, or lack thereof, or granting of any waiver of exemption in accordance with DOD 4145.26M in discharging the Contractor’s responsibility.” PI. Ex. 104, Armed Services Regulation, 7-104.79. (3) THE TRIP FLARE FACILITY BEFORE THE EXPLOSION Building M-132 was constructed in 1964 as part of Thiokol’s anticipated manufacture of rocket motors. It was a rectangular, one story structure, 36'8" X 134'8". Three outside walls were 8" concrete block. The north wall was a weak wall designed to serve as a gas relief in the event of fire. It would blow out an internal pressure of 0.5 psi. The building was considered a logical facility for production of trip flares. A cure room and storage room were added to M-132 in 1970. The loading area was enclosed. It was referred to as the east porch. M-132 was connected and formed a single complex with Buildings M-142 and M-143. They are connected by covered ramp-ways. The buildings in relation to each other, including the nearby maintenance building, are shown in the diagram below: The building was subdivided into eight bays each 12' X 15' with 12" reinforced concrete dividing walls on three sides; the fourth side opened into the north corridor of the building. The bays were in groups of two with a 5' hallway between each group. The hallways extended from the north corridor to the south corridor of the building. The bay walls were reinforced with # 4 bars on 12" centers in each direction and in each face. The roof of the basic building was constructed of cement asbestos sandwich panels (1%6" thick) covered by built-up roofing and supported by structural steel roof purlins. It was designed to blow off at 0.5 psi internal pressure. Thiokol Investigation Report, p. 9. The cure room was constructed in April, 1970, by building two new walls and using the existing walls of 8" concrete block. There were two doors. The dimensions of the room were 35'6" X 22'6" and approximately 10 feet in height-about 8,000 cubic feet. The sketch below shows Building M-132 and its interior layout: (4) ILLUMINANT PRODUCTION PROCESS The manufacture of trip flares took place mainly in Building M-132. The manufacturing operations consisted of the following separate processes: Illuminant Manufacture First Fire and Intermediate Mix Manufacture Illuminant Pellet Manufacture Ignition Pellet Manufacture Case Lining Candle Assembly Cover Pressing Assembly The production processes can be divided into two basic steps: (1) the manufacture of ignition and illuminant pellets; (2) the candle-loading process. The raw materials used in the manufacture of flares are ground sodium nitrate, magnesium powder and a binder consolidator (Laminae). Sodium nitrate is the principal oxidizer; magnesium the solid fuel. The loose materials burn at a speed measured in milliseconds. The mix which makes up a flare consists roughly of approximately 45% each of magnesium and sodium nitrate and 8% to 10% of “first fire” material composed of boron powder and barium chromate. First fire is not an illuminant; it is a pyrotechnic and provides the fire that ignites the illuminant more rapidly. Deposition of William L. Childers, p. 50. The raw materials used in the manufacture of the flares were delivered to Building M-132 and stored at the rear of Bay 101. The grinding process and mixing were conducted in that Bay. The material was mixed in 500 pound batches. The illuminant composition was then granulated or screened in Bay 106 and spread on aluminum trays to a thickness of about one inch. Each tray held about 12.5 pounds of illuminant. The trays were placed in racks and the loose composition transferred to the curing or “oven” room for a minimum period of 16 hours at a temperature of approximately 110 ° F. After cure, the illuminant was again granulated in Bay 106; collected in aluminum buckets, and transferred to fibre containers as required for “surge” or pelleting operations. It was placed in large plastic drums called lever packs. Following further granulation, the illuminant mixture was placed on trays and returned to the oven room for a second curing after which it was transferred to Bay 108. There it was pressed into pellets (200 per tray) and inspected. They were reloaded into the lever packs and stored in a corridor between Bays 106 and 107 pending removal to the head of the candlemaking line. The illuminant material was then taken to Bay 105 where it was stored until needed for dispensement at the “first fire” belt in front of Bays 103 and 104. After the dispensing procedure, the illuminant was loaded into dies which were moved by conveyor belt to the “first fire” addition station. The “first fire” raw material were delivered to the oven for storage and then transferred to the mixing area at Bay 101 where it was ground, weighed and mixed. After screening, it was moved into the cure room for drying and storage and thence taken to Bay 105 where it was stored in small quantities of 20 to 25 pounds. In quantities of one pound the first fire material was removed to the first fire addition station where it was added to the dies previously filled with the illuminant. After pressing, they were ejected and placed on trays containing 150 pellets per tray and returned to the oven room for storage. The illuminant casings were placed at the head of the candlemaking line. The illuminant loading assembly consists of three illuminant pellets and an ignition pellet bonded into a lined aluminum alloy case about 1.5" in diameter and 4.1" long. The total illuminant in each loading assembly is about 0.3 pound. The cover assembly is pressed on the top of the illuminant loading assembly. The illuminant loading assembly was commonly referred to as a “candle” and the unit with the cover assembly installed as a “flare”. The term “candle” is a trade name for the illuminant that has been consolidated and is ready to use as an “end item”. (5) BLACK WEDNESDAY AT WOODBINE On February 3, 1971, 60 to 64 employees began work on schedule at 8 A.M. at the flare facility in Building M-132. On the previous day 6,700,000 ignition pellets and 6,000 candles had been produced and had been placed in the oven for storage. Eight mixtures of illuminant No. 2 had been produced and were placed on the racks and left in the cure room for the purpose of drying. Nothing unusual occurred on the morning of the 3rd until approximately 10:50 A.M. There had been small fires at the ejection station on two or three occasions during the two months preceding the catastrophe. The procedures at the first fire dispensing station were thus described by an employee who worked there: “Two operators worked at the first fire addition end of the ignition pellet line. One was referred to as the ‘first fire dispenser’ and the second as the ‘die turn-around operator’. As the dies charged with illuminant approached the station from the illuminant dispensing station, the die turn-around operator lifted the die from the belt and placed it on a plywood board. The board had been affixed to the conveyor frame over the belts to provide a stationary pad for adding the first fire and to prevent spilling material onto the belt. The first fire dispenser dipped first fire material from a conductive plastic container (capacity: 400 grams) with a small brass scoop and leveled the material in the scoop by drawing the top of the scoop across a plastic bar located over the plastic container. She then turned and dumped the scoopful of material into the die on top of the illuminant. “The die turn-around operator shook the die gently to level the material in the die and placed the die on the return belt of the conveyor for transport to the presses.” See Pl. Ex. 6, App. B-3. Carolyn Waye was the operator at the first fire addition station. Sandra Rawls who lost her life was the die “turn-around” operator on the assembly line. The investigation report made by the special investigators of the Alcohol, Tobacco and Firearms Division states that Sandra would remove the dies from the conveyor belt containing the illuminant material and place them on the plywood board above it. Mrs. Carolyn Waye would dip the first fire material and insert it into the candles. Sandra picked up a die, shook it to level the material, and placed the dies on the opposite belt. To quote from the report: “Carolyn Waye stated that she had just filled up her dipper and was turning toward the belt when she observed a spark come up out of the belt on her side of the wooden board. Mrs. Waye further related that Sandra Rawls was standing at her position at this time and that she (Waye) could not remember if Sandra Rawls had a die in her hand. Mrs. Waye related she and Sandra Rawls left the building through a door next to their station and that she heard three explosions; one that sounded like a shotgun blast, the second a louder explosion, and the final explosion being the biggest and loudest of all.” Thomas F. Aretz was working in the building as Process Inspector, Quality Control. He testified that the fire built up as it went along, “got bigger and burned harder”, feeding on the material and things that were in its path. He said that the fire travelled to the inspection table of the first fire pellets and the containers of material in its path and followed them through the corridor over to the oven. He did not see how the fire got into the cure room doors. Aretz said that the interval between when he first saw the fire at the ejection machine and the first small explosion was very short-two or three seconds. The interval between then and the explosion that blew him into the air was three or four minutes as best as he could recall. See Mr. Aretz’s statement to the Court in the course of the testimony of Dr. J. H. McLain, pp. 94-97, 101-102. At 10:30 operations on the pellet line were proceeding smoothly. What shortly followed is thus summarized in the report of the Investigation of the Fire and Explosion: “At approximately 10:53 a. m. a fire started in the north corridor of the building at the west end of the ignition pellet line. At this point, the die turn-around operator would take a die of alluminant from the conveyor belt and place it on a board shelf affixed to the conveyor frame over the belts. The first fire dispensing operator would then add a scoopful of first fire to the die. The operator at this station reported that just as she was adding the first fire material to a die that a fire seemed to come ‘right from the board itself’. The fire spread rapidly to dies at the first fire station and down the ignition pellet line. Personnel in the north side of the building immediately evacuated. One operator, who was in the mixer bay on the northwest corner of the building, went to Building M-142, pounded on the door hollering fire, and then ran to the woods north of M-142. “Personnel on the south side of the building became aware of the fire by the shouts and by observation of a very bright fire through the hallways leading to the north corridor. The fire appeared to come down the hallways to the south corridor, and the south side of the building was also evacuated. One employee turned to a fire hose to fight the fire, but then realized the fire was too large to fight and ran from the building. Personnel in the trailer were alerted to the fire by the shouts of the people evacuating the building and by observation of the fire through the trailer windows, and evacuation of the trailer was started. “A massive illuminant fire developed in the north side of the building and erupted from the north side before a heavy explosion. The development of this fire was accompanied by two dull conflagrations which were audible to people north of the building. After heavy fire erupted from the north side of the building, there was a heavy explosion which blew the building apart and knocked down most people within several hundred feet. A large brilliant fire ball accompanied or followed the explosion and witnesses reported ‘fire was everywhere’. Burning candles and debris rained down over a wide area. A rain of burning candles was reported over a wide area, including the area near the Maintenance Shop, more than 500' from M-132. The extreme flight of a burned candle was about 4200 feet. “It is apparent from the location where the dead and the seriously injured were located after the explosion, that all personnel had evacuated Building M-132 pri- or to the explosion and that most of the people were a substantial distance from the building at that time.” (6) THE FACILITY: AFTER THE EXPLOSION THE THIOKOL INVESTIGATION The explosion all but obliterated M-132. “Only the concrete cubicles in Building M-132 and some parts of the exterior wall which were shielded by the concrete bays of these cubicles remained standing. The other walls were sheared at floor level and broken into fragments. Fragments, candles, and structural roof members from the cure room were thrown considerable distances. ... The rear wall of Bay 106 (12" reinforced concrete); the bay nearest the cure room, was shattered and forced inward, and the rear of Bay 105, the companion bay to Bay 106, showed significant cracking. The building floor in the area of the cure room showed a large dished depression about 7" deep and two smaller depressions. There was no evidence of any other explosion in the building.” There was damage to nearby structures. The east end of Building M-142 was collapsed. Building M-143 showed effects of the blast wave. Building M-131 which was 380 feet from M-132 was structurally damaged. About 30,000 flares in it were apparently ignited by burning candles thrown from M-132. See PI. Ex. 6, Thiokol Investigation Report, p. 20. The physical damage was estimated at as high as $1,000,000. According to Dr. Joseph H. McLain, an expert witness for plaintiffs and an acknowledged authority in the field of pyrotechnics, “What happened was this material in the trays was ignited went from burning to explosion, in the time of milliseconds, and exploded in the particular confined room was enough and destroyed the building, created depressions in the floor, and demolished walls and threw projectiles.” Testimony of Dr. McLain, p. 45. Dr. Melvin Cook who is an expert of acknowledged repute in the field of explosives was employed by Thiokol to inspect the site and to give his opinion as to the cause of the explosion. He did not testify as a witness in the case. However, a statement by him at a meeting of the A.T. and F. investigators on February 23, 1971, is part of the record. Dr. Cook concluded that “the fire had originated at the first fire station, spreading to nearby containers of ignition pellets, which by their nature of ignition, flew down the corridor to a rack of granulated illuminant, causing it to ignite, and that this heat either ignited additional material in the corridor between 106 and 107 or through its own heat, was sufficient to enter the oven at the end of the corridor. Dr. Cook stated that when this entrance was made, the amount of illuminant material was sufficient to explode, and that this material had reached a ‘critical mass’ state, and that it deflagrated at such speed, as to cause the destruction.” See PI. Ex. 101. The Thiokol investigators stated in their report concerning the fire and explosion: “Examination of the depressions in the floor of the cure room show that the major force of the explosion was at the location of the trayed illuminant.” The investigating team concluded that “the probable cause of the accident was the ignition of first fire powder by friction or impact caused by an operator at the first fire addition station on the ignition pellet line.” Post-explosion tests of the first material showed its high sensitivity to friction. It may be “ignited readily by the normal forces and velocities which may be exerted by an operator during handling of dies or striking two objects together.” Thiokol Investigation Report, App., C-l. Possible sources of ignition by friction included: “1. Bumping or dropping a die, thereby igniting first fire contaminant in the die or on the follower plate. 2. Striking the scoop on a die. 3. Bumping a die against another die or other metal object, such as the conveyor frame, igniting first fire contaminant on the outside of the die or on the frame. 4. ' Sliding, bumping, or dropping a die on the board over the conveyor on which the dies were placed for loading, thereby igniting first fire material spilled on the board.” PI. Ex. 6, p. 3. The operation at Woodbine was a “high risk” one. Deposition of Theron Driscoll, Office of Special Safety, Atlanta, p. 8. Dr. Joseph H. McLain testified as to the high sensitivity of the material in its loose state to friction and impact. There were five fires in Building M-132 prior to February 3, 1970. At least two occurred at the ejection station. According to McLain, where this type of illuminant is being handled there is “a finite probability” that you are going to have fires. He mentioned “human error, human frailty, people do drop things, and people do get misalignments in the pressing....” McLain testimony, pp. 50-51. The report by Thiokol was issued on July 30, 1971. See Pl. Ex. 6. The investigation team was composed of Thiokol personnel, two technical consultants and one investigative consultant. They were assisted by the Alcohol, Tobacco and Firearms Division of the United States Treasury Department and by the Safety Research Center of the United States Bureau of Mines. An ATF investigative report was separately made and filed. See Pl. Ex. 101. The Treasury Department also submitted its own report. Apparently, the Department of the Army declined to participate in the investigation. The Thiokol Investigating Team listed the following major contributing factors to the explosion. (PI. Ex. 6, Report, p. 4): “1. Inadequate hazard classification of the illuminant material. The hazard classification (Class 2 Fire Hazard) ascribed to the materials and the information relative to the hazards of the materials were inadequate to characterize the dangers in manufacturing operations. The explosion, not fire, was responsible for the severity of the accident. 2. Inadequate separation of materials. The relative locations of materials resulted in fusing through the building to the cure room. 3. Adequacy of the fire protection svstem. The inability of the fire protection system to contain the initial fire to a small area of the building probably resulted from the failure of the operators to pull the manual deluge valve as they exited the building and the slow response time of the automatic activation devices relative to the rate of spread of the fire.” The Report recommended that the government thoroughly evaluate, or fund contractors to evaluate, the hazards associated with pyrotechnic materials at all stages of processing and provide this information to all bidders on programs that require the processing or use of such materials. (7) DOD SAFETY STANDARDS The Contractors’ Safety Manual (PI. Ex. 2) is incorporated in the Thiokol contract by reference. The Contractor is required to comply with it. Contract Addendum; ASPR 7-104.79, Pl. Ex. 104. The purpose of the Manual was to prescribe “safe methods, practices, and standards for insuring continuity of production, safeguarding personnel . . .. ” Para. 100. The Contracting Officer (PCO) and the Administrative Contracting Officer (ACO) had the responsibility of providing “safety clauses and technical safety data in contracts for ammunition, explosives and related dangerous materials”. Those officials were also charged with seeking “counsel and advice of the cognizant DOD Component safety staff [Picatinny] to insure that safety criteria is included in the contract”. Para. 103 a, b. The Manual (para. 701 c) provides: “For contractor owned contractor operated plants, it is the responsibility of the Procuring Contracting Officer to determine the acceptable protection after considering the hazard involved and the risk that must be assumed.” The Contractor was under the duty of developing and implementing a safety program to assure the safety of its personnel, property and the general public with respect to the performance of work pursuant to the terms of the contract. Safety Manual, Para. 103 d(4). See also depositions of Herschel Q. Holley, pp. 80-82; Charles F. Otken, Jr., pp. 70-71. “The cardinal principle to be observed in any location or operation involving explosives, ammunition, severe fire hazards or toxic materials is to limit the exposure of minimum number of personnel, for a minimum time, to a minimum amount of the hazardous material consistent with safe and efficient operations.” Para. 603 b(1). According to an ASPR Regulation (7-104.79, Pl. Ex. 104), no act or failure to act by the Government in surveillance or enforcement of the Manual “shall affect or relieve the Contractor of responsibility for the safety of his personnel and his property and for the safety of the general public....” However, both actually and contractually the Government prescribed basic safety standards and practices and saw to it that the contractor complied. In that respect as well as assurance of the quality of the end product the United States possessed and exercised the right to oversee the operation. The Contractors’ Safety Manual classifies the hazards of explosives and similar dangerous material. As noted earlier, the lowest rating is Class 2. It is a fire hazard only. It covers items that burn vigorously with little or no possibility of extinguishment in storage situation. Explosions of such material, according to the DOD Manual, normally will be confined to pressure ruptures of containers and will not produce shock waves or damaging blast overpressure. Class 7 is the highest hazard classification. It covered Mass Detonating Hazards. The materials are defined as those that will explode virtually instantaneously when a small portion is subject to fire, severe concussion of impact, or impulse of an initiating agent. The explosion of items in Class 7 normally will cause severe structural damage to adjacent objects. Clearly the degree of hazard under the Army’s safety standards related to detonability of material rather than to danger from fire or deflagration of an item. The standards established by the Manual were geared to the concept of the hazard by detonation as opposed to danger by fire. The A.T. and F. officials who investigated the Thiokol incident reported that “the production of the military trip flare presented a very distinct hazard to both personnel and property ... [T]he hazard of production was identified as that of fire, not explosion.” Pl. Ex. 101. The safety regulations were largely oriented to that concept. Whether a material was classified as Class 7 or as 2 depended on what is called the “card gap test”. It is explained in Explosives Hazard Classification Procedures, 1967, Pl. Ex. 5 and in J. D. Settles, “Deficiencies in the Testing and Classification of Dangerous Materials”, supra, (see footnote 9) pp. 201-202. See also testimony of Joseph H. McLain, pp. 22, 26. The test consists of placing the candidate material in a tube above (or below) a steel witness plate which bears “witness” to the detonative effect. The plate is %th of an inch thick. The sample under test is ignited by a blasting cap at the top of the tube. Lucite cards are interposed in a gap between the booster and the material being tested. The witness plate rests on a wooden stand, 6 inches above the ground. “Detonation is indicated when a clean hole is cut in the witness plate.” Then and then only is the material regarded as Class 7 under the definition. Since illuminants do not fulfill that requirement for detonation, they are a Class 2 fire hazard only. The safety standards established by the Manual are “the Quantity-Distance” (Q-D) requirements. If Class 2 items weigh between 20,000 and 30,000 pounds, they should not be located less than 235 feet from “inhabited buildings”, public railroads or highways. The distances prescribed in the instance of Class 7 items (Mass-Detonating Hazards) depends upon whether or not there were barricades. When such detonable materials weigh from 30,000 to 35,000 pounds, the distance requirement was 1,310 feet in the case of a barricade and 2,410 feet in the absence of one. A strange thing about the Quantity-Distance standard as applied in the present case is that the employees working in M-132 were not in an “inhabited” building. They were working where the manufacturing process took place which was the point from which the distance to “inhabited buildings” was to be measured. M-132 was an operational building; it was not “inhabited” so as to come within the distances prescribed in the Q-D tables. The nearest “inhabited” building was E-103 which was 1,420 feet from M-132. Deposition of Harry L. Padgett, Thiokol Safety Director, p. 53. The Army’s hazard precautions were seemingly not designed to protect employees of the contractor working in an operational building-the persons most endangered by an explosion during the manufacturing process. As the Government argued in its brief on the motion for summary judgment (p. 76): “[T]he classifications are established for ‘Q-D’ measurements, or for establishing minimum distances between the explosives and other facilities and buildings. They do not apply within a building itself. Thus, the classifications were not relevant to the plaintiffs but only, at best, to workers in other buildings. Plaintiffs mistakenly assume a higher classification would have protected them when the building within which they worked was irrelevant to the classification scheme.” (8) THE CLASSIFYING OF ILLUMINANTS (A) The Government’s contention that the trayed, untamped illuminants were properly Class 2 rather than Class 7 material requires it to argue that depressions in the concrete floor of the cure room (6 inches in depth at one point) were not craters. No crater, no detonation; no detonation, no Class 7 item, according to defendant. It strikes this Court that we are allowing technicalities to obscure the real problem. The issue is not whether there was a detonation. It is whether the explosion in the cure room on February 3, 1971, and the resulting injuries and loss of life was proximately attributable to any negligence by the defendant. According to Walter G. Queen, Chief of the Safety Office of the AMC, deflagration of illuminants under confinement may lead to an explosion. He testified that if enough heat is applied to the mix, rapidly enough and there is sufficient confinement, there is reason to believe that “you can get certainly a violent explosion. Perhaps not a detonation in the sense of TNT or some other materials but something that would look factually like this, in any event.” Dep., pp. 38, 60. The General Manager of the Longhorn Division of Thiokol, Herschel Q. Holley, testified that Class 2 material (illuminants) would not themselves explode but vigorously burn and can produce an explosive-like rupture such as that of the cure room at Woodbine. Dep., pp. 271, 106. The Safety Manual, 605b(4), in dealing with firefighting, notes that Class 2 ammunition may explode “if material is confined”. The Government conceded at the trial that “there was material in the cure room that either exploded or caused an explosion in the cure room”. Government counsel’s statement reported in the testimony of Dr. McLain at p. 42. It is admitted by defendant that there is a risk of explosion from Class 2 materials. Brief on Government’s motion for summary judgment, p. 26. Theron Driscoll, Safety Engineer of DCASR at Atlanta, was aware, prior to 1971, that flash fires involving pyrotechnic materials had resulted in explosions. He was also cognizant that under conditions of confinement Class 2 illuminants could explode. Dep., pp. 85-86; 123, 132. The record in this case includes several reports as to such incidents that were made to the appropriate Army command between 1954 and 1967. See PI. Exhibits 84 through 88. (B) Prior to November 3, 1967, loose illuminants in process were classified as Class 7 items. However, the following year certain illuminants were downgraded to Class 2. In 1970 the Thiokol (Woodbine) flares were not classified or treated as Class 7 items by contract or otherwise. An “Alert” which is part of the contract deals with “Transportation of Explosives” and classifies the material as Class 2. However, for the purpose of the Quantity-Distance requirements the material is designated as Class 2 and the contracting parties treated it as such. At the time of the explosion on February 3, 1971, illuminants were classified as Class 2, according to the Procurement Contract Officer at Picatinny. The ascending scale of classification used by the Army indicates the dangers or safety factors involved. Holley dep., pp. 68-69, 117. This Court cannot agree with the contention of the Chief of the Munitions Engineering Division at Picatinny that Thiokol and other contracting manufacturers make the classification of illuminants. Deposition of Seymour Fleishnick, pp. 25, 30-33, 41, 44-45, 48. Even less do I agree with the Government’s argument that the Hazardous Item Contracts regulation of April 15, 1969 prohibits the Procuring Contracting Officer from specifying hazard classifications except in respect to the end item or product. AMC Regulation 385-17, Para. 5c.(1)(d), Gov’t. Ex. 80. The contention is based on the responsibility assigned to AMC and the Procuring Contracting Officer in the matter of “Specification of hazard classifications of end items and pertinent safety standards in contract solicitations and contract documents.” That the Regulation did not withdraw the authority of AMC in respect to classifying in-process materials is amply evidenced by the conduct of the Army. Reference was made on page 20 to the sensitivity of loose illuminant compositions and their susceptibility to ignition by friction or impact. Evidently that is what occurred at Woodbine and it was not the first time. In the weeks prior to February 3, 1971 five minor fires occurred in Building M-132. Two of them took place at the first fire addition station where the fire on February 3, 1971 originated. At least two of these incidents were reported to Picatinny, one by telephone on January 4, 1971, and another by letter on January 20, 1971. The Thiokol Safety Board met on January 8, 1971. Its subsequent report listed as the cause several unsafe practices. Officials at Picatinny were aware of the frequence in the industry of small fires in processing of illuminant materials. Otken dep., pp. 63-64. See also Gov’t. Ex. 84 which reports tests following a fire on an illuminant production line at the Navy’s Ammunition Depot at Crane, Indiana on January 4, 1968. The Navy’s tests looked to better methods of putting out such fires rather than preventing them. (9) DISSIDENT VOICES (A) Among the witnesses at the trial was Dr. Joseph H. McLain who appeared as an expert on behalf of the plaintiffs. He is President of Washington College at Chestertown, Maryland. Dr. McLain received his doctorate in chemistry and has had wide experience in the manufacture of pyrotechnics and explosives. During World War II he was stationed for a time at Edgewood Arsenal in Maryland in the capacity of Chief of the Smoke Branch, pyrotechnic division. After the war, Dr. McLain was a consultant at Edgewood. He has engaged in the business of manufacturing illuminants. In the summer of 1968 and for several years thereafter he conducted a week-long seminar at the Franklin Institute in Philadelphia. It was attended by personnel of various Government agencies, including high ranking military officers. McLain strongly opposed the idea prevailing in high circles that pyrotechnics will not per se explode. It had become with him, he said, “almost a labor of love, evangelical zeal . . . to try to eradicate this superstitious belief that these things will not detonate”. Testimony, p. 20. In his lecture on safety at Franklin Institute he took particular exception to the classification given such pyrotechnics by the Army in its technical bulletin, TB 700-2. He thought such materials should have a classification of their own on the basis of output and sensitivity. Testimony of McLain, p. 54. “I would like to propose that pyrotechnics be given a position of their own, separate and distinct from explosives ... based upon definitive testing and experience, and just as there are class A, B, and C explosives there should be some similar classification for pyrotechnics.” McLain, “Pyrotechnics and Solid State Chemistry”, 17-1 (1973), PI. Ex. 50. Dr. McLain is particularly critical of the military’s test for determining whether a substance is an explosive or detonable. The card gap test “miserably” fails, he said, “to establish a hazard classification.” Testimony, p. 26. In the lectures referred to he discussed the question: “What is the nature of protection that must be given to personnel? Should it protect against shock wave, over pressure and sympathetic detonation between buildings in the case of a detonable material, or should it protect against fragmentations, radiant heat, smoke inhalation and/or suffocation as is the case with most pyrotechnics?” According to Dr. McLain, “A medium to high rate explosion which can occur from a pyrotechnic mix kills and damages with fire ball and fragments. Barricades unless of proper height and design only serve to increase the distance of throw of fragments and burning debris.” In summarizing, he said: “[Ejxplosives and pyrotechnics differ widely in methods of protection, characteristics of output and initiation. Industry and Government must realize these basic differences and take action to make a more realistic code for the protection of life and property.” Pl. Ex. 50. (B) J. E. Settles was Manager of Safety, Chemical Propulsion Division of Hercules, Inc. at Wilmington, Delaware. Reference has been made above to his paper (published in 1968) entitled “Deficiencies in the Testing and Classification of Dangerous Materials”. Pl. Ex. 47. His views correspond with those of Dr. McLain as to the inadequacy of the “card gap test”. They are in accord as to the defects of the DOD classification system as applied to illuminants based on the detonation theory and fully agree concerning the danger of explosion of Class 2 items. In his article Mr. Settles attacked “the general acceptance of the philosophy that all reactions other than the high-velocity detonation are fire hazards only.” Analyzing 81 accidents that had occurred in the munitions industry since 1959, he reached the conclusion that of 78 fatalities all but one resulted from non-detonating forces. Mr. Settles found serious inconsistency in accepting a “fire hazard only” in respect to reactions of such violence and destructive energy as high-rate, medium-rate and low-rate “explosions”. In his opinion, a “searching reappraisal of past and present practices” was needed in the industry. (10) THE 1968 THIOKOL (WASATCH) TESTS Illuminant hazard tests were conducted in 1968 by the Wasatch Division of Thiokol at the Army Air Force installation at Brigham City, Utah. The purpose was to determine what results could be expected should Thiolite B-7 ignite during any of the various stages of illuminant processing. See Pl. Ex. 10 and 11. The tests which were filmed in color are in evidence as Pl. Ex. 12 and 13. The films were shown at the trial. This Court has looked at them again. They dramatically illustrate the conclusion in Thiokol’s Wasatch Report (PI. Ex. 11, p. 3): “B-7 illuminant in the untamped, unsecured condition when ignited would explode violently. In open containers the material, whether ignited on the surface or below the surface, would instantly cast the burning mass up out of the container from which point the mass would further explode in all directions.” It was further determined that the ignition of untamped illuminant produced a definite shock wave of undefined magnitude as the result of the rapid deflagration or a true detonation. Among the conclusions reached as a result of the tests were the following: “An ignition of untamped illuminant results in an extremely rapid deflagration that produces a shock wave of devastating proportions.” “A large quantity of untamped, uncured B-7 illuminant in a container, surface ignited, will begin to burn rapidly in approximately % seconds. The rate of the flame front growth will then become extreme and a shock wave will result.” The recommendations following the tests stated in part: “It is recommended that the frangible roof concept be utilized for facilities in which large quantities of untamped illuminant will be processed.” “Exposure of personnel to untamped illuminant should be kept to an absolute minimum. Operations should be accomplished remotely to the greatest extent practicable.” “Extreme precautions must be taken in process and facilities design to prevent an incident.” “Facilities and equipment should be designed as though any incident involving personnel in the area not protected by properly designed barricades will be fatal.” There is no evidence that the Thiokol reports were made known to the Air Force but it was apparently aware of certain changes that were implemented by the tests. Dep. of William L. Childress, p. 73; and his supplementary letter of March 11, 1974. See PI. Ex. 15. (ID THE ROLFE REPORT, BRITISH MINISTRY OF DEFENSE In 1969 or about that period the British Ministry of Defense in order “to assist safety considerations allied to processing and storage”, conducted experiments relating to the explosive effects of a variety of pyrotechnic compositions. See P. J. Rolfe, “The explosive effects of pyrotechnic compositions: Assessment of mass explosion hazards from some pyrotechnic compositions.” Royal Armament Research and Development Establishment Memorandum 57/69 (December, 1969), Pl. Ex. 35. The tests included flare compositions of which magnesium and sodium nitrate were constituents. Ignition occurred in less than 200 milliseconds and the materials were consumed in 1 or 2 seconds. Deflagration resulted in considerable gaseous residues. The Rolfe Report concluded that the product explodes in confinement even in the instance of one pound in a cartridge. For a summary of the Report see Dr. McLain’s testimony, p. 219. (12) CLASS 2 UPGRADED AS TO VARIOUS PYROTECHNIC COMPOSITIONS It is clear from the above that leaven was at work in 1968-1970 in the reevaluation of hazards and classification of illuminant compositions. It is less clear to what extent these developments and the researches in pyrotechnics influenced Picatinny in that period. At any rate, in November, 1970 illuminant material was reclassified and became a Class 7 item. Prior thereto, tests had been conducted by AMC personnel at Picatinny Arsenal and by General Electric Company at the NASA facility in Mississippi on the subject of hazard classification and evaluation of selected pyrotechnic compositions. In October, 1970 there went up through channels to the Commanding General, U.S. Army Materiel Command, from Munitions Command a report and recommendations dated October 29,1970. PI. Ex. 43A. The communication stated in part: “8. It is recommended that pyrotechnic compositions no longer be tested in accordance with the procedures given in Army Bulletin TB 700-2, Chapter 3. It is further recommended that an interim hazard classification of Class 7 be assigned to loose pyrotechnic compositions during handling, storage and transportation, until more meaningful tests can be adopted. “9. Picatinny Arsenal in its hazard classification program is developing a series of tests which will determine the deflagration to detonation characteristics of pyrotechnic compositions. These tests include rates of detonation, confined ignition temperature and pressure-time behavior under heavy confinement. Those compositions which appear to detonate, low to high order, will be tested for TNT equivalency for potential damage assessment. It is believed that the data obtained from these tests will give more meaningful criteria for hazard classification of pyrotechnics.” Accompanying the report was a communication to the General from Walter G. Queen, Chief Research & Labs, Safety Officer. It stated: “Pending completion of the Picatinny Arsenal program for development tests to determine the hazard classification of pyrotechnic compositions, loose pyrotechnic compositions should be considered hazard class 7.” Mr. Queen testified that it had been agreed that they should “take a second look to see if Class 2 really did define this well”. While the evidence was not conclusive in connection with the pyrotechnic classification procedures, to be on the conservative side, “we did agree with them that we should, in fact, retract our classification that was established, I believe, in ’68 and we so advised MUCOM.” Queen dep., p. 32. It appears that on November 16, 1970, AMC approved the recommendation to upgrade illuminants to Class 7. At this point something went skew in army bureaucratic circles. For some reason the approval was not disseminated. It was not until March 25, 1971, after the explosion, that Thiokol was notified of the action taken by AMC in February that all pyrotechnic compositions, in storage or process, “must be considered as Explosive Hazard Class 7, even though they have been tested here and determined by TB 700-2 to be Class 2 explosive materials.” Pl. Ex. 15. One month before, on February 25, 1971, the commanding officer at Picatinny had informed the General Manager of the Georgia Division of Thiokol: “As a precautionary measure, and until such time as further evaluations or new test criteria are established, it is recommended that the unconsolidated magnesium-sodium nitrate illuminant composition be reclassified back as Class 7 material.” Pl. Ex. 30. The Government says that this was not an order, only a recommendation, a fact exemplifying the lack of control of the United States over the matter. Brief on motion for summary judgment, p. 73. Defendant argues that it had no responsibility to disseminate the classification change. It says that it was without authority under the Thiokol contract to fix an in-process classification and that the Company had control over the classification, not the United States. See defendant’s brief on motion for summary judgment, pp. 73, 76. Mr. Fleishnick testified in answer to the question whether or not Picatinny classified the illuminant and designated such classification, as follows: “We did not . . . the company does that based on their background and their experience... . They make that classification based on the information that they have in many documents relating to pyrotechnics.” Dep., pp. 30, 41-42, 44. I cannot agree. The Government witness Queen testified that it is the function of the Army Materiel Command to make hazard classifications of potentially dangerous munitions, pyrotechnics and explosives. Dep., pp. 17, 26. Mr. Queen stated that the general effect of the decision made in November, 1970, to up-grade illuminants was to return to the Class 7 designation that prevailed in 1967. Dep., pp. 34-35. The Safety Engineer DCASR-Atlanta testified that the new Class 7 designation existed by virtue of the authority of Picatinny Arsenal “because they have authority to put a classification on