An educational resource at the heart of criminological teaching, debate, and research


Flowers at the altar of profit and power Part 3: Was the disaster at Bhopal "unforeseeable"?

User Rating: 5 / 5

Star ActiveStar ActiveStar ActiveStar ActiveStar Active

Were there discrepancies in regard to the standards of worker health and safety and environmental protection between different Union Carbide Corporation (UCC) facilities – whether directly owned and controlled by UCC itself or by one or more of its subsidiary? If there were such discrepancies, is it reasonable to suggest that UCC and/or UCE and/or UCIL either knew about them or should have known about them? In particular, were there safer ways of making Sevin?

Bhopal and Institute


One way to explore this latter, crucial question is by establishing whether or not MIC and/or Sevin were produced in the 1980s more safely at other UCC facilities. The first and most obvious comparison is with the MIC/Sevin plant at Institute, West Virginia.

While large amounts of MIC were stored at both Bhopal and Institute, at the latter the storage tanks were much larger, there was an additional dedicated dump system with a capacity of 42,000 gallons and an additional and more powerful emergency back-up system. But there were differences of another kind. As we have seen, the emergency safety systems at Bhopal were either out of commission or turned off but none of them, if they had been switched on and working, would have had the capacity to fulfill their tasks—gas detection instrumentation was too insensitive, the refrigeration plant could not keep the MIC at a low enough temperature, the vent gas scrubber and flare tower were incapable of either dealing with the volume of escaped chemicals or with multi-phase rather than just gaseous   emissions. Finally, the Bhopal storage tanks were of a type unsuitable for Indian climatic conditions. In other words, there was a mismatch between the potential demands that might be made on the safety systems and their capacity to deal with these. This was a design flaw premised on a willingness to take unconscionable risks: the safety of India and Indians was taken much less seriously than that of America and Americans.

Given the degree of control exercised by UCC/UCE over UCIL (Union Carbide India Ltd.), there is every reason to believe that it not only knew of, but also allowed, or even approved of, the agreement signed in May 1983 between UCIL and the hourly-paid plant operatives and their union. Since reference was made to it earlier, here we wish to only focus on the following provisions:

"The selection, placement, distribution, transfer, promotion of personnel, fixing of working hours and laying down of working programmes,           planning and control of factory operations, introduction of new or improved production methods, expansion of production facilities, establishment of quality standards, determination and assignment of workload, evaluation and classification of jobs and establishment of production standards, maintenance of efficiency, maintenance of discipline in the factory ¼ are exclusively rights and responsibilities of the Management". (Cited in URG 1985a: 1‑2. For all references, see Pearce and Tombs Bibliography)

In practice this meant not merely a change but a reduction in standards, and that, even if changes violated SOPs, management had to be obeyed Thus, contra the claims of Anderson and others, in many ways the Bhopal plant was an inferior plant to the UCC plant at Institute, West Virginia.

Themistocles D’Silva, in his The Black Box of Bhopal: a Closer Look at the World's Deadliest Industrial Disaster, alleges that to compare Bhopal and Institute is inappropriate because while the "technology sold to UCIL included process improvements made over the first MIC plant at Institute, West Virginia", the original plant was "built in 1965-1966". The "MIC plant at Institute" was built much later, in 1978, and "is more streamlined and is a fully automated factory” (D’Silva 2006: 49). But, while accepting that some differences were due to the fact that these were based in part on fundamentally different designs, nevertheless some improvements, for example, enlarged MIC storage tanks, an additional dedicated dump system with a capacity of 42,000 gallons and an additional and more powerful emergency back-up system, could have been easily incorporated into the older design, particularly given that work on the Bhopal MIC and Sevin plants did not even start until 1979.

The sabotage theory

This is a useful place to address the question of the ‘sabotage theory’, which played a key role in UCC’s ‘definitive version’ of the sequence of events that led to the leak at the plant. According to this ‘theory’, on the night of the incident a disgruntled employee who was not on duty removed a pressure gauge and then used a hose to put water into an MIC tank; his intention was to spoil a batch of chemicals rather than create a disaster. This version of events, circulated to the media and to UCC personnel, was most fully articulated when it formed the basis of a paper presented by ‘independent consultant’ Dr. Ashok Kalelkar at a London conference in May 1988 (Kalelkar 1988). Kalelkar had in fact been a member of the team organized by UCC in March 1985, which even then had mooted the possibility of sabotage, although a lack of evidence meant, it claimed, that "it was unable to develop this theory further at the time".

Yet this ‘definitive version’ was only the last in a series of such ‘theories’ involving alleged saboteurs. First, it had been claimed that the disaster itself was the result of the actions of careless or malicious employees who had placed a water line where a nitrogen line should have been used. The New York Times, on 26 March 1985, pointed out that neither an accidental nor a deliberate incorrect coupling were possible since the relevant nitrogen and water lines were of a different colour and the nozzles were of different sizes. That same day Union Carbide Chairman, Warren Anderson, had to withdraw the accusation at Congressional Hearings when he admitted that he had no evidence of sabotage. Then, between July 31, 1985 and January 3, 1986, UCC claimed that a group of Sikh extremists called the Black June Movement were responsible. But no such group was ever identified in any context other than allegedly putting up posters about Union Carbide; moreover, it was virtually impossible for anybody to actually plan a disaster of this kind. Not surprisingly, this claim was also quietly abandoned. In August 1986 a specific but unnamed employee was blamed — but it was not until May 1988 that all references to nitrogen lines were dropped and a pressure meter was mentioned.

Of course, Union Carbide and UCIL were hoping, first, to avoid vicarious liability on the ground that the employee would have been acting without authority and outside the course of his employment, and, secondly, to avoid liability under Rylands v. Fletcher, on the ground that an employee who comes onto his employers’ premises without authority and causes the escape of a dangerous thing is a ‘stranger’ for whose acts the occupier is not responsible. (Muchlinski 1987: 575).

There is no dispute between UCC and its opponents that a key link in the causal chain leading to the disaster was the introduction of 120 gallons or more water into MIC storage Tank E-610 and that this produced an exothermic chemical reaction which in turn initiated many more. The major dispute remains whether the water entered the tank accidentally, as a consequence of UCC/UCE and UCIL using under-trained and inexperienced plant operatives working with defective equipment in an understaffed facility, or, as UCC has always argued, whether the presence of water was the result of a deliberate act of sabotage. UCC’s critics have also suggested that impurities in the tank may have independently contributed to the genesis of exothermic chemical reactions even before water was added and, further, these and other impurities such as phosgene and cyanide may have been significant toxic components of the deadly gas cloud.

Perhaps the most elaborate development of, and defence of UCC’s arguments is to be found in D’Silva’s book, mentioned above. D’Silva worked as a research scientist in UCC’s Agricultural Chemicals Division for over 20 years. He was a member of its laboratory team that examined core samples taken from the residues left in the major source of the gas leak, Methyl Isocyanate (MIC) storage tank E-610, while another UCC team tried to determine which chemicals under what conditions could have produced this specific residue and the gas cloud. D’Silva was satisfied that his team identified all of the chemicals that were in the residue and that the other team had shown how, provided the temperature inside the tank never exceeded 275 degrees C, multiple interactions involving MIC, water and chloroform, and then further interactions involving both the original interactants and the newly created chemicals, could have produced a residue with this chemical composition.

This research along with detective work involving additional non-laboratory, ad hoc, experiments and data gathered from interviews provided the basis for three confident judgments. The first judgment was that the UCC teams had demonstrated that water could not have accidentally entered Tank E-610. The second was that they could show how it had been added deliberately and even identify those responsible - they claimed that during a shift change a local pressure indicator was removed from Tank E-610 and water was introduced into the tank. D’Silva is not trained as a detective and is hence is hardly qualified to claim a detective’s authority: he is not skilled in investigation, in interrogation, in assessing the veracity and utility of what ‘persons of interest’ say or do. His own judgments of the credibility of witnesses is compromised by his identification with UCC and UCIL and his trust in their executives; for example, he takes at face value statements by two of the eight UCIL executives subsequently convicted in June 2010 guilty of causing “death by neglect.” He tries to discredit UCC’s critics. He dismisses the evidence of Edward Munoz, on the grounds that he was bitter over losing his position with UCC in 1978 and for meretriciously accepting a consultancy fee from the plaintiff’s lawyers while acting as a witness for them - but he does not apply the same standards to himself and his friendly witnesses since he finds it quite acceptable that their testimony occurred when their salaries were paid by UCC/UCIL. He is also dismissive of another witness on the grounds that he was a disgruntled worker and an active trade unionist: contributory evidence it seems to his being the saboteur in question (D’Silva et al. 1986; D’Silva 2006). Since none of this evidence has been subjected to the rigours of the legal process, it is mere speculation and ultimately of little interest. Of much more significance, however, precisely because it is within the realm of D’Silva’s expertise, is his third judgment, which is that while there is no evidence that phosgene or hydrogen cyanide escaped into the air:

"The materials that could have escaped out of the stack along with MIC would …include carbon dioxide, and varying amounts of ammonia, dimethylamine, trimethylamine, chloroform, dichloromethane, and hydrogen chloride. All these materials, being heavier than air, would form a low–lying cloud in the vicinity of the plant and would probably not drift very far. The varied and unexplained medical conditions reported for the victims in different locations may be the result of exposure to any one or a combination of these materials".   (Emphasis added. D’Silva 2006: 110)

We remain unconvinced that this is the only possible scenario. For example, there is good evidence that the temperature inside Tank E-120 would have at times exceeded 420o C (Ball 2010), leading to a decomposition of the MIC into the toxic chemicals cyanide (HCN) and Carbon Monoxide (CO) (Blake and Ijadi-Maghsoodi 1982) and the decomposition of other chemicals thereby eradicating any trace of them. Nevertheless, we believe that D’Silva has demonstrated that interactions between MIC, water and chloroform could have produced the chemicals his team identified in the residue of Tank E-610 and those he identifies as escaping in a gaseous form. But this is an astonishing admission for it underscores that the MIC producing process at Bhopal would have been extraordinarily dangerous even if the plant had been in excellent condition.


This remarkable state of affairs generates three crucial, quite specific questions. First, why was it possible to remove a pressure dial by hand when this was connected to such a toxic and volatile chemical? Second, why was there such inadequate security to protect the plant from foolish or mischievous strangers? Thirdly, why was there water in the area? As a leading specialist on safety in chemical plants has written, if water is not there ‘it cannot leak in, no matter how many valves leak or how many errors are made’ (Kletz 1988: 86). This discussion makes it clear that Bhopal was a less safe plant than that at Institute.

This point being made, it needs emphasizing that, if not as dangerous as the Bhopal plant, that at Institute was still a dangerous place to be employed or to have as a neighbour. In the immediate aftermath of the disaster, UCC claimed that at the Institute plant there had only been 28 MIC accidental releases in the previous five years, yet within days of this statement it changed its story and disclosed that there had been 61 methyl isocyanate (MIC), 107 phosgene, and 22 mixed MIC-phosgene releases in that period. UCC subsequently admitted that it had been less than honest about its workplace safety performance too and it was fined for misleading OSHA inspectors (Jones 1988: 163‑186; Pearce 1990).

Moreover, as a result of trade union pressure, UCC had already adopted in its chemical facility in Béziers, France, a safer just-in-time production process that could use small amounts of MIC, stored in sealed and sturdy stainless steel drums each containing only 213 kg of MIC, all of which was immediately consumed in the production of Carbaryl, thus eliminating the need to store MIC. It is clear then that: Bhopal (owned by UCIL, owned and controlled by UCE, itself, owned and controlled by UCC) was less safe than Institute (owned and controlled by UCC) which was less safe than Béziers (also controlled by UCC). Thus, there were known discrepancies in worker health and safety and environmental protection standards between different UCC facilities – and that, in part, these discrepancies can be explained through the prism of power relations within the respective plants. Although, all was not well in France either. However good the Bèziers production process, safety was undermined in another part of the commodity chain since piece work pressures at the Marseilles dock reduced the possibility of taking the necessary care and time required to safely handle the barrels of MIC safely.

Now, D’Silva acknowledges that once water entered the MIC tank and set in motion an exothermic chemical reaction "the Bhopal tragedy … was aggravated by management miscalculations" (D’Silva 2006: 137), which included ‘[t]he needless storage of large quantities of [MIC] in very large containers for inordinately long periods as well as insufficient caution in design", and he concludes that “the combination of conditions for the accident were inherent and extant” (D’Silva 2006: 242). If these were ‘inherent and extant’, any number of unexpected events could have caused a runaway exothermic reaction that the safety systems in place could not have contained. Thus the more fundamental issue is the ease with which a disaster of almost unprecedented scale could occur.

So, even if we accept that the incident was caused by sabotage, this only serves to demonstrate how unsafe the actually plant was. That is, any sabotage theory itself only serves to underline UCC, UCE and UCIL’s responsibility for the gas leak and its consequences. We do not accept the ‘sabotage theory’ but this is a very secondary question. How the water entered the tank is of much less significance than the fact that it could do so, that the “accident” could escalate into a massive gas leak, and that this then gassed a nearby population of 200,000 people never having been told of the dangers to which they were being exposed and for whom no evacuation plan had been designed, no medical services equipped. On these matters, even this flawed book makes available new information and, in fact, provides evidence that inadvertently supports the interpretations developed by Sarangi, Bee and other radicals.



In December 1986, UCC sold its worldwide Agricultural Products business to the French chemical and pharmaceuticals company, Rhone Poulenc, for $585 million. The deal included Rhone-Poulenc's purchase of Union Carbide's chemical plant in Institute, West Virginia. In, 2002, two years after the creation of Aventis merger of Rhone-Poulenc and AgrEvo to become Aventis, the new company sold the facility to Bayer Crop Science.   In 2009 there was an explosion during start-up of the Methomyl unit. A report for the House Energy and Commerce Committee found that the explosion “came dangerously close” to igniting several tons of MIC stored nearby. The explosion and resulting fire were caused by a runaway reaction. The CSB reported critical safety features had to have been overridden to reach the required temperature in the unit that exploded.


(For a contextualization of this event see Mac Sheoin 2010 and for an extraordinary and horrifying video see the US Chemical Safety Board website

Frank Pearce, Professor of Sociology, Queen's University, Canada, and Steve TombsProfessor of Sociology, Liverpool John Moores University.

For Part 1, click on Flowers at the altar of profit: the continuing disaster at Bhopal.

For Part 2, click on Flowers at the Altar of Profit Part 2: Explaining the disaster at Bhopal

For info on the Pearce and Tombs column here on CrimeTalk: Crimes of the Powerful and Insurgent Resistance

For Part 4, click on Flowers at the Altar of Profit and Power Part 4: the Bhopal "Settlement

For Part 5, click on Bhopal: criminal, immoral or the cost of business as usual?

Click on References for the extensive bibliography on the Bhopal disaster used in this article and the series to follow. The series makes use of ch. 6 of Frank Pearce and Steve Tombs' Toxic Capitalism: Corporate Crime in the Chemical Industry, 1998, Ashgate: Aldershot; paperback version, 1999, Canadian Scholars Press, Toronto. See also: Tombs, S. and Whyte, D. (2007) Safety Crimes. Cullompton: Willan. Support CrimeTalk by buying these books through the CrimeTalk Shop.

A FREE e-copy of Toxic Capitalism can be downloaded at Frank Pearce's website here.

Text Size