What management had given the people who would treat the injured, the Indian Council of Medical Research later wrote, was nothing usable about the chemical. MIC had been made at the Union Carbide factory in Bhopal since 1980. “No information whatsoever was ever provided before or after the Disaster by the Management regarding its toxicity, antidote, nor how to protect workers from being exposed.” The gas came into houses without warning. On the night of 2–3 December 1984 about 40 tons left tank 610. Pressure climbed between about 00:15 and 00:30. A siren around 00:30 alerted plant personnel. Water sprayed from about 01:00 did not reach the stack. The release stopped around 03:00. The report says water entering the tank triggered a runaway reaction, and that failed safety systems played a major role. The minimization mattered because the neighborhoods were already in the cloud, and the clinicians were not told what they were treating. A sabotage story has been told about that water. The court orders below do not adopt it as the cause.
What Left the Tank
The ICMR technical report on studies from 1985 to 1994 calls it the worst industrial accident, at a Bhopal pesticide plant it describes as owned by the Union Carbide Corporation. About 40 tons of MIC left tank 610 around midnight, temperature below 10 degrees Celsius. Of a population over 8 lakh, about 5 lakh were estimated affected. A conservative estimate was about 2,500 deaths. Three tanks, 610, 611, and 612, were meant to hold MIC near 0 degrees under nitrogen, with a scrubber that was not circulating caustic. What escaped was MIC plus other reaction products. The report says wet cloth and staying indoors could have reduced injury. That advice was not in residents’ hands. The siren it describes was for plant personnel.
Indian Council of Medical Research — technical report, health effects of the MIC leak, epidemiological studies 1985–1994: https://www.icmr.gov.in/icmrobject/custom_data/1720338129_bgdrc-technical_report.pdf
What the Supreme Court Counted
The Court’s orders do not agree with that 2,500 figure, and they do not agree with each other. The curative-petition order of 14 March 2023 says a horrendous tragedy occurred on the night of 2 and 3 December 1984 from deadly fumes at the factory owned and operated by Union Carbide India Limited. Union Carbide Corporation, a New York company, owned 50.9 percent of UCIL. The Court repeats its earlier label: “unparalleled in its magnitude and devastation.” The Union sued for about $3.3 billion. The 14 and 15 February 1989 settlement was $470 million, due by 31 March 1989, closing civil and criminal proceedings. The Court later recorded that the Union had told the High Court 2,660 persons died and 30,000 to 40,000 were seriously injured, and that the figures before the Supreme Court were about 3,000 fatal cases and about 30,000 grievous injuries verifiable from hospital records. The 2010 curative petition said those death figures were wrong and that the actual number of deaths was 5,295. A table in the 2023 order also lists 5,479 deaths as of 31 October 2022 beside the 5,295 figure. The order of 11 May 2011, in the criminal curative petitions, states the same night as a massive escape of lethal gas from the MIC storage tank at the UCIL plant, “causing the death of 5,295 people” and leaving 568,292 suffering injuries. A peer-reviewed review states that more than 40 tons of MIC leaked and immediately killed at least 3,800 people. The Department of Chemicals and Petrochemicals states the night of 2/3 December 1984 and the $470 million deposit in February 1989. It does not state a death toll. It states that original compensation of Rs. 1,549.33 crore had been disbursed to 573,959 claimants out of 574,394 by 31 July 2024. Claimants are not deaths. These doors disagree. This page does not pick a body count they do not share.
Supreme Court of India — curative petition order, Union of India v. Union Carbide Corporation, 14 March 2023: https://api.sci.gov.in/supremecourt/2010/60172/60172_2010_2_1501_42809_Judgement_14-Mar-2023.pdf
Supreme Court of India — order, CBI v. Keshub Mahindra, 11 May 2011: https://api.sci.gov.in/jonew/judis/37991.pdf
The Sabotage Claim Beside That Record
Edward Broughton’s 2005 review in Environmental Health says UCC shifted culpability to the Indian subsidiary and “fabricated scenarios involving sabotage by previously unknown Sikh extremist groups and disgruntled employees,” a theory he writes was impugned by independent sources. That is the reviewer’s verb, not a Supreme Court finding. The 14 March 2023 order identifies an escape of fumes from the UCIL factory and a settlement. It does not find sabotage. The 11 May 2011 order identifies lethal gas from the MIC tank, a CBI charge sheet filed 1 December 1987, and a Chief Judicial Magistrate judgment of 7 June 2010 convicting accused numbers 2 to 5, 7 to 9, and 12 under sections 304A, 336, 337, and 338 read with section 35 of the Penal Code, and sentencing them to two years. It records that accused No. 1, Warren Anderson, was absconding when charges were framed, and that the Court could not bring UCC or Union Carbide Eastern before it. Anderson is named as that accused. The order does not call the leak sabotage, and it does not call it murder: the 1996 judgment it refused to reopen had limited the charge, on the material then produced, to section 304A. The ICMR report’s own causal sentence is water in the tank plus failed safety systems, not a named saboteur. Sabotage remains a claim in the review. It is not the cause these Indian orders adopt.
National Library of Medicine — Edward Broughton, “The Bhopal disaster and its aftermath: a review,” Environmental Health, 2005: https://pmc.ncbi.nlm.nih.gov/articles/PMC1142333/
The Money, and What It Did Not Decide
The Department of Chemicals records the $470 million deposit in February 1989 and says original compensation began to be adjudicated in November 1992. A Supreme Court order of 3 October 1991 held that an earlier $5 million UCC interim deposit in the New York case, credited against any final judgment, did not stand outside the 1989 fund. The department’s later rupee figures are payments to claimants, including an ex-gratia death-category scale of Rs. 10 lakh after a June 2010 Group of Ministers recommendation. Those scales do not reconcile 2,500, 2,660, 3,000, 3,800, 5,295, and 5,479. The 14 March 2023 order dismissed the curative petitions and left the settlement closed. Neither order here holds that a saboteur opened the tank.
Department of Chemicals and Petrochemicals — Bhopal Gas Leak Disaster: https://chemicals.gov.in/bhopal-gas-leak-disaster
Supreme Court of India — order on the $5 million interim deposit, 3 October 1991: https://api.sci.gov.in/jonew/judis/12602.pdf
What the Record Settles
Documented: a methyl-isocyanate release from tank 610 at the Bhopal plant on the night of 2–3 December 1984; an ICMR finding that management supplied no toxicity, antidote, or protection information, and that the gas entered houses without warning; a $470 million settlement in February 1989; criminal convictions in 2010 under the negligence sections, with Anderson recorded as absconding; and death-and-injury figures that the doors do not share. Sabotage is a company claim described in a 2005 review. It is not a cause found by the Indian orders in this file. The Archive files the release, the warning gap, and the cases as documented / L2.