A disaster by design

Internal documents of Union Carbide Corporation, obtained for the first time, during a discovery procedure, point to the company's culpability in the Bhopal gas disaster.

THE seeds of the December 2-3, 1984 Bhopal gas disaster were sown 11 years earlier, on December 2, 1973.

On that day, three documents were presented to the management committee of Union Carbide Eastern Inc., (UCE), a subsidiary of Union Carbide Corporation (UCC), for its consideration: a two-page internal memorandum relating to a plan of UCC's another subsidiary, Union Carbide India Limited (UCIL), to begin manufacturing methyl isocyanate (MIC) at its Bhopal factory, supported by a four-page capital budget plan and a 44-page project proposal. Warren Anderson, former UCC Chairman, who for the past 11 years has been refusing to appear before the Bhopal District Court in connection with the criminal charges against him, and against whom the Indian government had shown inexplicable reluctance to initiate extradition proceedings as ordered by the court, was one of the seven management committee members who approved the MIC project. UCIL reported to UCE, headquartered in Hong Kong but incorporated in Delaware. UCE, in turn, reported to UCC, in Danbury.

In the internal memorandum, issued under the letter-head of UCE and signed by B.T. Burgoyne, it was made clear that "to the extent feasible, UCC would provide the necessary technology and process design and would review any technology and design developed outside UCC." Further, the memorandum said: "In addition to responsibility for these activities, UCC has also agreed to start up support and training outlined in the proposal." Claiming the support of UCC Worldwide Agricultural Policy Committee and UCE, the memorandum revealed that UCC's law, finance and environmental affairs departments had already reviewed the proposal. The memorandum and the relevant documents were brought to light, thanks to the discovery procedure in the ongoing class action suit filed against UCC and Warren Anderson by the survivors of the disaster, in the Federal Southern District Court of New York. Under the discovery procedure, the defendants are required to produce the documents on which they seek to rely, and each party is entitled to inspection of the documents annexed to or relied upon by the other.

To understand the significance of this memorandum, one needs to trace the history of the Bhopal factory. In 1966, UCC sent one of its technical representatives, Edward Munoz, to India to study the feasibility of establishing a pesticide-manufacturing facility. UCIL was in existence since 1934, duly registered as a public company.

Munoz concluded that such a project was feasible, and the proposal was reviewed and approved by the UCC management committee. Thereafter UCC assigned Munoz to UCIL with the responsibility to develop the project. In its application to the Government of India for a licence to manufacture MIC-based pesticides, submitted in 1966, UCIL stated that the proposed facility would manufacture the two principal raw chemical ingredients, alpha naphthol and MIC.

Subsequently, UCC began preparations to build the production facilities in Bhopal, based on the designs and experiences of its MIC facility in Institute, West Virginia, United States. UCC engineers in the U.S. began to design the Bhopal MIC plant at least one year prior to entering into two formal agreements with UCIL for design transfer and technical services in 1973. These agreements did not serve to limit UCC's involvement in the Bhopal plant since the entire operation was under its control pursuant to its majority ownership of UCIL and its standard corporate policies. They detailed the extensive undertaking and intent of UCC to carry out the Bhopal venture. The plant, owned by UCIL, consisted of five principal units set up and constructed from time to time. The unit manufacturing MIC began to function in February 1980.

In its written statement submitted to the Bhopal District Court on December 10, 1986, UCC claimed that it did not design or prescribe the procedures for the Bhopal plant. In an effort to absolve itself of any responsibility for the disaster, the UCC had stated: "The design was by UCIL in consultation with and by employing Indian engineering companies and firms, and the defendant (UCC) only supplied process design packages for certain portions of the plant. UCIL devised and prescribed its own operating procedures." The Indian government, representing the victims and the survivors, contested this claim and asked the court to disregard the legal persona of UCIL in the interest of justice, and to hold UCC liable. Regardless of the cause of the runaway reaction, which resulted in the gas leak, the Indian government maintained that UCC knew of the possibility of such an event, but that it failed to design the plant for such an eventuality. One of the engineers who designed the Bhopal plant, Gordon E. Rutzen, said that UCC's objective was to provide to UCIL a process design, which incorporated a number of safety features and elements, which when taken together would be able to cope with the situations that would conceivably occur.

The import of the 1973 memorandum is that irrespective of the contents of the two agreements it signed with UCIL in 1973, UCC intended to review any technology and design, even if it was developed outside UCC. Not surprisingly, under the 1973 design transfer agreement, UCC provided the design for the entire MIC unit at the Bhopal plant. It included specifications for an emergency relief system, including a vent gas scrubber and a flare tower. The Union Carbide design personnel also decided to store the ultrahazardous MIC in large quantities despite the existence of alternative, safer methods of production. In the preparation of the design reports, Union Carbide engineers had calculated the potential entry of water into the MIC tanks which could result in a runaway reaction and leak of the toxic material, and compiled safety considerations reports. In other words, UCC monitored and approved the detailed design and construction work at the Bhopal plant.

Having had a clear responsibility, did UCC intend to do justice to it? The 1973 capital budget plan brings out the startling revelation that UCC ordered under-investment in the highly dangerous sevin project, in order to sidestep Indian regulations requiring a dilution of foreign equity, which seemed to threaten UCC's control of UCIL.

In January 1974, the Foreign Exchange Regulation Act (FERA) came into force. UCIL was one of the companies affected since it had a foreign equity ownership of 60 per cent. UCC had to dilute its equity to 50.9 per cent during 1977-78, principally to meet the government's conditions attached to the permission to construct the MIC project. Indeed, UCC was exempted from the FERA regulation that required the dilution of foreign equity to 40 per cent and was allowed to maintain majority ownership and thus control of UCIL primarily because the government trusted its claims of exclusive expertise regarding the MIC process.

It was UCC's policy to sell its interest in any subsidiary if, at any time, it lost control over its operations. In order to maintain its control, UCC had developed detailed policies in recognition of the political reality that many nations, such as India, would attempt to exercise some degree of regulation to protect their sovereign interests. In one internal publication, entitled `Legal control of a 50-50 Joint Venture Affiliate', UCC detailed a number of devices or expedients that would allow it to gain control of its affiliates.

In its 1973 plan, UCC hints at one such device or expedient: besides the sevin project, UCC had two other projects in mind. "Both these projects will be the subject of separate capital budget proposals, but the financial arrangements will be affected by Government of India legislation requiring a dilution of foreign-held equity whenever new capital expenditures are made. Since our total equity dilution is critical to our strategy in India, it is important to review the financial plan for all three projects," UCC's finance plan for the sevin project said. (Sevin is the commercial trademark name given by UCC to the pesticide that they were making.) UCC held 60 per cent stakes in UCIL, which was required by the government to raise 25 per cent of the estimated cost of expansion through the issue of new equity to local shareholders and the Indian public.

UCC proposed to make a single equity issue for all three projects and to maintain a minimum of 53.5 per cent ownership of the company. In order to accomplish this, UCC claimed that it would have to negotiate with the Government of India to reduce the amount of investment for purposes of the 25 per cent new equity from approximately $28 million (Rs.215 million) to $20.6 million (Rs.163 million). "The negotiated amounts will be mainly on the sevin project," the plan revealed.

UCC's specific objective, the plan said, was not to accept any conditions that would reduce its equity below 51 per cent. Obviously, the sevin project had to compromise on many safety standards, because of the limited funds available.

For example, UCIL stored liquid MIC in huge tanks, far in excess of what would have been permitted in its parent plant in the U.S.

MIC is a dangerously volatile chemical and these tanks were supposed to be kept cooled to 0 degree Celsius. However, the refrigeration system had been switched off in June 1984, probably to save the cost of freon gas. If the estimate of the cost of refrigeration as given in the 1973 project proposal is any indication, UCC hoped to cut costs substantially by switching off the refrigeration system. Sevin production at the plant had stopped in October 1984, following huge accumulation of stocks.

UCC, in its March 1985 investigative report on the disaster, admitted: "The temperature of MIC in tank 610 (where the gas leak happened) before the incident was at 15-20 degree Centigrade as compared to the requirement of about 0 degree Centigrade refrigeration. The lower temperature would have retarded the reaction rather considerably and extended the time available for corrective action. The refrigeration system provided to cool the MIC in the storage tank had been made non-operational in June 1984." UCC, it appears, did not go into the reasons for switching off the refrigeration system, for obvious reasons.

More important, the 1973 project proposal reveals that UCC had realised that the technology to be adopted at the Bhopal plant was not a proven one, and yet it opted to take the risk in view of the desired long-term objectives of minimum capital and foreign exchange expenditures. "The comparative risk of poor performance and of consequent need for further investment to correct it is considerably higher in the UCIL operation than it would be, had proven technology been followed throughout," the proposal admitted. The MIC-to-sevin process, as developed by UCC, had had only a limited trial run, the proposal cautioned. It envisaged that there would be interruptions in operations and delays in reaching capacity or product quality, which might have been avoided by the adoption of proven technology. In particular, the proposal underlined that the particular combination of materials to be disposed of was new and would, therefore, create difficulty.

These 1973 documents have only proved what the subsequent events revealed; that contrary to its claims in Indian courts, UCC continued to exercise control over the Bhopal plant, its design and its functioning. The discovery process has yielded additional material in the form of telex messages tht were exchanged between UCC and UCIL officials before and in the immediate aftermath of the disaster. These messages show that UCC had to help UCIL by sending experts from the U.S. to dispose of the remaining MIC gas after the disaster and for safe conversion of the material. Before the disaster, UCC even considered dismantling and shipping the plant to overseas locations to mitigate probable losses, probably because it felt that the plant was unviable and unsafe. It showed that UCC knew or could have known the full dangers inherent in the Bhopal plant.

The discovery procedure has also led to the unravelling of UCC's in-house findings about the contamination and massive pollution of groundwater near its abandoned factory in Bhopal. The key document, entitled the `Presence of Toxic Ingredients in Soil/Water Samples inside Plant Premises', gives a frightening insight into the cover-up perpetuated by UCC. It said: "The seriousness of the issue needs no elaboration. Samples drawn in June-July 1989 from land-fill areas and effluent treatment pits inside the plant were sent to R&D (research and development). They consisted of nine soil/solid samples and eight liquid samples. The solid samples had organic contamination varying from 10 per cent to 100 per cent and contained known ingredients like naphthol and naphthalene in substantial quantities. Majority of the liquid samples contained naphthol and/or sevin in quantities far more than permitted by ISI for onland disposal. All samples caused 100 per cent mortality to fish in toxicity assessment studies and were to be diluted severalfold to render them suitable for survival of fish." This internal finding stands in contrast to what UCC had been claiming about the safety of the local drinking wells. UCC had used the results of the National Environmental Engineering Research Institute (NEERI) study to assert the absence of any danger of contamination of water in the area.

A Greenpeace (an international environmental non-governmental organisation) report published in 1999 found severe contamination of the factory site, surrounding land and groundwater. In some places, the levels of mercury were six million times higher than expected. Drinking water wells near the factory used by local people were heavily polluted with chemicals known to produce cancers and genetic defects. In 2001, a study conducted by the Fact Finding Mission on Bhopal set up in 1998 by experts from different fields, concerned with various aspects of the disaster found lead, mercury and organochlorines in the breast milk of nursing mothers in the affected area.

On the eve of the recent anniversary of the disaster, Greenpeace activists from abroad attempted to clean up the contaminated site, invited eviction and arrest by the police. However, this seems to have forced Madhya Pradesh Chief Minister Digivijay Singh to announce that his government will petition the Supreme Court to force Dow Chemical (the current owner of UCC) to pay for the clean-up of the site. According to independent estimates, this could cost the company up to $500 million. The costs of failure to ensure corporate accountability would certainly be greater than this amount.

The survivors' organisations have handed over the new documents to the Central Bureau of Investigations (CBI), which is now pursuing the extradition of Warren Anderson to India, following repeated orders to do so from the Bhopal District Court. The Indian government had all along assumed, on the basis of flawed legal advice given to it, that the evidence on record was not sufficient to meet the `probable cause standard' applicable in the U.S. in extradition matters. (A U.S. court will not agree to extradite an accused person to another country unless there is prima facie evidence directly linking him to the cause of the disaster Frontline, January 18, 2002.) It may now have to reconsider its stand in the light of these "discoveries".