Full opinion text
OPINION and ORDER KEENAN, District Judge: FACTUAL BACKGROUND On the night of December 2-3, 1984 the most tragic industrial disaster in history occurred in the city of Bhopal, state of Madhya Pradesh, Union of India. Located there was a chemical plant owned and operated by Union Carbide India Limited (“UCIL”). The plant, situated in the northern sector of the city, had numerous hutments adjacent to it on its southern side which were occupied by impoverished squatters. UCIL manufactured the pesticides Sevin and Temik at the Bhopal plant at the request of, and with the approval of, the Government of India. (Affidavit of John MacDonald (“MacDonald Aff.”) at 2). UCIL was incorporated under Indian law in 1934. 50.9% of its stock is owned by the defendant, Union Carbide Corporation, a New York corporation. (MacDonald Aff. at 1). Methyl isocyanate (MIC), a highly toxic gas, is an ingredient in the production of both Sevin and Temik. On the night of the tragedy MIC leaked from the plant in substantial quantities for reasons not yet determined. The prevailing winds on the early morning of December 3, 1984 were from Northwest to Southeast. They blew the deadly gas into the overpopulated hutments adjacent to the plant and into the most densely occupied parts of the city. The results were horrendous. Estimates of deaths directly attributable to the leak range as high as 2,100. No one is sure exactly how many perished. Over 200,000 people suffered injuries — some serious and permanent — some mild and temporary. Livestock were killed and crops damaged. Businesses were interrupted. On December 7, 1984 the first lawsuit was filed by American lawyers in the United States on behalf of thousands of Indians. Dawani et al. v. Union Carbide Corp., S.D.W.Va. (84-2479). Since then 144 additional actions have been commenced in federal courts in the United States. The actions have all been joined and assigned by the Judicial Panel on Multidistrict Litigation to the Southern District of New York by order of February 6, 1985, 601 F.Supp. 1035. The individual federal court complaints have been superseded by a consolidated complaint filed on June 28, 1985. The Indian Government on March 29, 1985 enacted legislation, the Bhopal Gas Leak Disaster (Processing of Claims) Act (21 of 1985) (“Bhopal Act”), providing that the Government of India has the exclusive right to represent Indian plaintiffs in India and elsewhere in connection with the tragedy. Pursuant to the Bhopal Act, the Union of India, on April 8, 1985, filed a complaint with this Court setting forth claims for relief similar to those in the consolidated complaint of June 28, 1985. By order of April 25, 1985 this Court established a Plaintiffs’ Executive Committee, comprised of F. Lee Bailey and Stanley M. Chesley, Esqs., who represented individual plaintiffs and Michael V. Ciresi, Esq., whose firm represents the Union of India. Jack S. Hoffinger, Esq., who represents individual plaintiffs, was appointed liaison counsel for the Plaintiffs’ Executive Committee. On September 24, 1985, pursuant to the Bhopal Act, the Central Government of India framed a “scheme” for the Registration and Processing of Claims arising out of the disaster. According to the Union of India’s counsel, over 487,000 claims have been filed in India pursuant to the “scheme.” There presently are 145 actions filed in the United States District Court for the Southern District of New York under the Judicial Panel for Multidistrict Litigation’s order of February 6, 1985, involving approximately 200,000 plaintiffs. Before this Court is a motion by the defendant Union Carbide Corporation (“Union Carbide”) to dismiss the consolidated action on the grounds of forum non conveniens. DISCUSSION The doctrine of forum non conveniens allows a court to decline jurisdiction, even when jurisdiction is authorized by a general venue statute. In support of its position that the consolidated action before the Court should be transferred to a more convenient forum within the Union of India pursuant to this doctrine, Union Carbide relies on the United States Supreme Court’s decisions in Gulf Oil Corp. v. Gilbert, 330 U.S. 501, 67 S.Ct. 839, 91 L.Ed. 1055 (1947) and Piper Aircraft Co. v. Reyno, 454 U.S. 235, 102 S.Ct. 252, 70 L.Ed.2d 419 (1981). The plaintiffs cite numerous other lower United States federal court cases in their briefs and seek to distinguish the Supreme Court’s decisions from this case. Of course, Gilbert and Piper are the touchstones in sorting out and examining the contentions of both sides to this motion on the various factors bearing on convenience. Piper teaches a straightforward formulation of the doctrine of forum non conveniens. A district court is advised to determine first whether the proposed alternative forum is “adequate.” This inquiry should proceed in the order followed below. Then, as a matter within its “sound discretion,” Piper at 257, 102 S.Ct. at 266, the district court should consider relevant public and private interest factors, and reasonably balance those factors, in order to determine whether dismissal is favored. This Court will approach the various concerns in the same direct manner in which Piper and Gilbert set them out. At this juncture, it would be appropriate to discuss the presumptions on a forum non conveniens motion. In Piper, the Court discussed its earlier finding in Koster v. Lumbermens Mutual Casualty Co., 330 U.S. 518, 67 S.Ct. 828, 91 L.Ed. 1067 (1947), which suggested that a plaintiff’s choice of forum was entitled to great deference when the forum chosen was the home of the plaintiff. This presumption was based on the fact that the choice of the home forum indicated a reasonable assumption that the choice was convenient. Koster at 524, 67 S.Ct. at 831. Conversely, the Piper Court found: When the plaintiff is foreign, however, this assumption is much less reasonable. Because the central purpose of any forum non conveniens inquiry is to ensure that the trial is convenient, a foreign plaintiff’s choice deserves less deference. Piper 454 U.S. at 256, 102 S.Ct. at 266 (footnote omitted). In the case now before the Court, in which the plaintiffs, including the Union of India, are foreign, and share a home forum which is not the instant forum, the assumption that this forum is convenient is not completely reasonable. The foreign plaintiffs’ choice of the United States forum “deserves less deference” than would be accorded a United States citizen’s choice. This Court will apply the presumption in favor of plaintiffs’ choice of forum with “less than maximum force.” Piper at 261, 102 S.Ct. at 268. See note 23 at 864, infra. 1. Preliminary Considerations. “At the outset of any forum non conveniens inquiry, the court must determine whether there exists an alternative forum.” Piper at 254, n. 22, 102 S.Ct. at 265, n. 22. The elements of that inquiry are set forth in Piper. First, the Court said, “[ojrdinarily, this requirement will be satisfied when the defendant is ‘amenable to process’ in the other jurisdiction.” Piper at 254, n. 22, 102 S.Ct. at 265, n. 22, quoting Gilbert 330 U.S. at 506-507, 67 S.Ct. at 842. Gilbert states that the doctrine of forum non conveniens “presupposes at least two forums in which the defendant is amenable to process.” Extending the limited inquiry of Gilbert, the Piper Court delved into the relevance of the substantive and procedural differences in law which would be applied in the event a case was transferred on the grounds of forum non conveniens. The Piper Court determined that it was theoretically inconsistent with the underlying doctrine of forum non conveniens, as well as grossly impractical, to consider the impact of the putative transferee forum’s law on the plaintiff in its decision on a forum non conveniens motion: “[I]f conclusive or substantial weight were given to the possibility of a change in law, the forum non conveniens doctrine would become virtually useless.” Piper 454 U.S. at 250, 102 S.Ct. at 263. The Court listed numerous practical considerations which led to its conclusion that an unfavorable change in law for plaintiff was not a relevant factor in the forum analysis. First, the Court observed that if the chance of a change in law were given substantial weight, choice of law questions would “become extremely important.” Piper at 251, 102 S.Ct. at 263. U.S. courts would “have to compare the rights, remedies, and procedures available” within the two proposed alternative forums, to determine whether a disadvantageous change in law would occur upon transfer. Id. Since “[t]he doctrine of forum non conveniens, however, is designed in part to help courts avoid conducting complex exercises in comparative law,” the change in law analysis would subvert the doctrine itself. Id. Thus, a court engaged in the inquiry regarding the existence and adequacy of an alternative forum should not hinge its decision on an unfavorable change in law. Another practical concern relating to the “change in law” inquiry was discussed by the Piper court. Based on the liberality of United States federal law as compared to much foreign law with respect to availability of strict liability for tort, malleable and diverse choice of law rules among the 50 states, availability of jury trials, contingent fee arrangements and extensive discovery provisions, the Court' observed that a change of forum might frequently involve an unfavorable change of law for foreign plaintiffs suing American defendants. Piper at 252, n. 18, 102 S.Ct. at 264, n. 18. Consequently, if the unfavorable change in law were a major factor in the analysis: [T]he American courts, which are already extremely attractive to foreign plaintiffs, would become even more attractive. The flow of litigation into the United States would increase and further congest already crowded courts. Piper at 252, 102 S.Ct. at 264 (footnotes omitted). At the point, however, where the possible change in law would provide “no remedy at all” to plaintiff, a court may conclude that no adequate alternative exists. As the Piper Court observed, it did not hold that: [T]he possibility of an unfavorable change in law should never be a relevant consideration in a forum non conveniens inquiry. Of course, if the remedy provided by the alternative forum is so clearly inadequate or unsatisfactory that it is no remedy at all, the unfavorable change in law may be given substantial weight; the district court may conclude that dismissal would not be in the interests of justice. Piper at 254, 102 S.Ct. at 265 (emphasis in original) (footnote omitted). Thus, while it is not a “major factor” in the analysis, a court must at least consider the effect on plaintiffs of a change in law upon transfer. To a great extent, the plaintiffs in this case argue that Indian courts do not offer an adequate forum for this litigation by virtue of the relative “procedural and discovery deficiencies [which] would thwart the victims’ quest for” justice. (Memorandum in Opposition by Plaintiffs’ Executive Committee (“Memo in Opp.”) at 2). The defendant disputes this contention. Plaintiffs’ preliminary concern, regarding defendant’s amenability to process in the alternative forum, is more than sufficiently met in the instant case. Union Carbide has unequivocally acknowledged that it is subject to the jurisdiction of the courts of India (Defendant’s Memorandum in Reply filed December 20, 1985 (“Reply Memo”) at 8); (oral argument January 3, 1986, transcript at 29, comment of Bud Holman, counsel for Union Carbide). Union Carbide is definitely amenable to process in India. Beyond this initial test, plaintiffs and amicus curiae argue that the Indian legal system is inadequate to handle the Bhopal litigation. In support of this position, plaintiffs have submitted the affidavit of Professor Marc S. Galanter of the University of Wisconsin Law School. Professor Galanter’s credentials are impressive; he was a Fulbright Scholar at the Faculty of Law of Delhi University and specializes in South Asian Studies at the University of Wisconsin Law School. He is not, however, admitted to practice in India and the Court views his opinions concerning the Indian legal system, its judiciary and bar as far less persuasive than those of N.A. Palkhivala and J.B. Dadachanji, each of whom has been admitted to practice in India for over 40 years. Both are Senior Advocates before the Supreme Court of India. Mr. Palkhivala served as Indian Ambassador to the United States from 1977 to 1979, and has represented the Indian government on three occasions before international tribunals. Although the outcome of this analysis, given the rule of Piper regarding change in law, seems self-evident, the Court will review plaintiffs’ argument on the inadequacy of the Indian forum out of deference to the plaintiffs. A. Innovation in the Indian Judicial System. Professor Galanter describes the Indian common law legal system, inherited from the British, in terms of its similarity to that of other common law systems. He compares the system favorably to that of the United States or Great Britain in terms of the appellate structure, the rule of stare decisis, the role of the judiciary as “guardian of [India’s] democratic structure and protector of citizens’ rights.” (Galanter Aff., at 6-12) before pointing to its ostensible deficiencies. According to Professor Galanter, India’s legal system “was imposed on it” during the period of colonial rule. (Galanter Aff. at 11). Galanter argues that “Indian legal institutions still reflect their colonial origins,” (Galanter Aff. at 12), in terms of the lack of broad-based legislative activity, inaccessibility of legal information and legal services, burdensome court filing fees and limited innovativeness with reference to legal practice and education. (Galanter Aff. at 12). On the question of innovativeness, Mr. Palkhivala responds with numerous examples of novel treatment of complex legal issues by the Indian Judiciary. In the words of the former ambassador of India to.the United States, “a legal system is not a structure of fossils but is a living organism which grows through the judicial process and statutory enactments.” (Palkhavala Aff. at 3). The examples cited by defendant’s experts suggest a developed and independent judiciary. Plaintiffs present no evidence to bolster their contention that the Indian legal system has not sufficiently emerged from its colonial heritage to display the innovativeness which the Bhopal litigation would demand. Their claim in this regard is not compelling. B. Endemic Delays in the Indian Legal System. Galanter discusses the problems of delay and backlog in Indian courts. Indeed, it appears that India has approximately one-tenth the number of judges, per citizen, as the United States, and that postponements and high caseloads are widespread. Galanter urges that the backlog is a result of Indian procedural law, which allows for adjournments in mid-hearing, and for multiple interlocutory and final appeals. Numerous appeals and “[cjonsiderable delay [are] caused by the tendency of courts to avoid the decision of all the matters in issue in a suit, on the ground that the suit could be disposed of on a preliminary point.” (Galanter Aff. at 17; 18-20, 21, quoting Indian Law Commission, 54th Report (1973) pp. 12-13). This Court acknowledges that delays and backlog exist in Indian courts, but United States courts are subject to delays and backlog, too. See Remarks of Honorable Warren E. Burger, Chief Justice, Supreme Court of the United States, 100 F.R.D. 499, 534 (1983). However, as Mr. Palkhivala states, while delays in the Indian legal system are a fact of judicial life in the proposed alternative forum, there is no reason to assume that the Bhopal litigation will be treated in ordinary fashion. The Bhopal tragedy has already been approached with imagination in India. Demonstrating the creativity and flexibility of the Indian system, the Parliament of India has passed the Bhopal Act in order to deal with the cases arising from the sad events of December 3, 1984. The Bhopal Act permits the cases to be treated “speedily, effectively, equitably and to the best advantage of the claimants.” (Palkhivala Aff. at 11). Mr. Dadachanji refers to another Indian case which arose from a gas leak in New Delhi. The Chief Justice and another Justice of the Supreme Court of India ordered the presiding court to expedite adjudication of claims. MC Mehta v. Union of India. (Dadachanji Aff. at 11 and Annexure A thereto). In another instance, the Indian Supreme Court directed the High Court to hear a given matter on a daily basis, and set a deadline for delivering judgment (Dadachanji Aff. at 11 and Annexure B thereto). Other means of coping with delay are appointment of special tribunals by the Government of India (Dadachanji Aff. at 12 and Annexure C thereto), and assignment of daily hearing duties to a single special judge, otherwise unburdened, to hear a special matter. (Dadachanji Aff. at 11). This Court is persuaded, by the example of the Bhopal Act itself and other cases where special measures to expedite were taken by the Indian judiciary, that the most significant, urgent and extensive litigation ever to arise from a single event could be handled through special judicial accommodation in India, if required. C. Procedural and Practical Capacity of Indian Courts. Plaintiffs contend that the Indian legal system lacks the wherewithal to allow it “to deal effectively and expeditiously” with the issues raised in this lawsuit. (Memo in Opp. p. 53). Plaintiffs urge that Indian practitioners emphasize oral skills rather than written briefs. They allegedly lack specialization, practical investigative techniques and coordination into partnerships. These factors, it is argued, limit the Indian bar’s ability to handle the Bhopal litigation. As Mr. Dadachanji indicates, Indian lawyers have competently dealt with complex technology transfers, suggesting capability within the technological and scientific areas of legal practice, if not “specialization.” (Dadachanji Aff. at 8). Moreover, Indian attorneys use experts, when necessary. As to investigative ability, Mr. Dadachanji persuasively points out that the Central Bureau of Investigation (“CBI”) of the Union of India is well equipped to handle factual inquiry, as is the Commission of Enquiry constituted by the state of Madhya Pradesh. (Dadachanji Aff. at 8). While Indian attorneys may not customarily join into large law firms, and as Mr. Palkhivala states, are limited by present Indian law to partnerships of no more than twenty, this alone or even in concert with other factors does not establish the inadequacy of the Indian legal system. (Palkhivala Aff. at 8). There is no reason the Indian legislature could not provide for the expansion of law-firms, if such a choice is required. In any event, this Court is not convinced that the size of a law firm has that much to do with the quality of legal service provided. Many small firms in this country perform work at least on a par with the largest firms. Bigger is not necessarily better. Moreover, since the Union of India purports to represent all the claimants, it is likely that if the case were transferred to India, the Attorney General or Solicitor General of India and the Advocate General of Madhya Pradesh, with attendant staffs, would represent the claimants. The Indian bar appears more than capable of shouldering the litigation if it should be transferred to India. (Palkhivala Aff. at 9). Next, plaintiffs and Professor Galanter argue that the substantive tort law of India is not sufficiently developed to accommodate the Bhopal claims. Plaintiffs trace the lack of sophistication in Indian tort law to the presence of court fees for litigants as inhibiting the filing of civil suits. Though the filing fees may have had historical significance, they are irrelevant here. Professor Galanter acknowledges that court fees may be waived for “poor parties or for specific classes of litigants.” (Galanter Aff. at 28). In fact, filing fees have been waived for claimants in India in the Bhopal litigation already begun there. Professor Galanter asserts that India lacks codified tort law, has little reported case law in the tort field to serve as precedent, and has no tort law relating to disputes arising out of complex product or design liability. (Galanter Aff. at 30-36). As an illustration of the paucity of Indian tort law, Professor Galanter states that a search through the All-India Reports for the span from 1914 to 1965 revealed only 613 tort cases reported. (Galanter Aff. at 32). Mr. Dadachanji responds that tort law is sparsely reported in India due to frequent settlement of such cases, lack of appeal to higher courts, and the publication of tort cases in specialized journals other than the All-India Reports. (Dadachanji Aff. at 16-17; Palkhivala Aff. at 10). In addition, tort law has been codified in numerous Indian statutes. (Dadachanji Aff. at 16-17). As Professor Galanter himself states, “the major categories of tort, their elements, the [theories] of liability, defenses, respondeat superior, the theories of damages — are all familiar.” (Galanter Aff. at 37). What is different, Galanter asserts, is the complete absence of tort law relating to high technology or complex manufacturing processes. This is of no moment with respect to the adequacy of the Indian courts. With the groundwork of tort doctrine adopted from the common law and the precedential weight awarded British cases, as well as Indian ones, it is obvious that a well-developed base of tort doctrine exists to provide a guide to Indian courts presiding over the Bhopal litigation. In any event, much tort law applied in American cases involving complex technology has its source in legal principles first enunciated in Victorian England. See, e.g., Rylands v. Fletcher, 1868, L.R. 3 H.L. 330. As Mr. Palkhivala stated in his affidavit: The plant itself was the product of highly complex technology, but complexity of the technology cannot be equated with complexity of legal issues. The principles of liability and damages involved in the Bhopal cases are all well established in India. The complexity is not in the nature or determination of legal issues but in the application of the law to the events which took place in Bhopal. Well settled law is to be applied to an unusual occurrence. (Palkhivala Aff. at 7). Plaintiffs next assert that India lacks certain procedural devices which are essential to the adjudication of complex cases, the absence of which prevent India from providing an adequate alternative forum. They urge that Indian pre-trial discovery is inadequate and that therefore India is an inadequate alternative forum. Professor Galanter states that the only forms of discovery available in India are written interrogatories, inspection of documents, and requests for admissions. Parties alone are subject to discovery. Third-party witnesses need not submit to discovery. Discovery may be directed to admissible evidence only, not material likely to lead to relevant or admissible material, as in the courts of the United States. Parties are not compelled to provide what will be actual proof at trial as part of discovery. These limits on discovery are adopted from the British system. Similar discovery tools are used in Great Britain today. This Court finds that their application would perhaps, however, limit the victims’ access to sources of proof. Therefore, pursuant to its equitable powers, the Court directs that the defendant consent to submit to the broad discovery afforded by the United States Federal Rules of Civil Procedure if or when an Indian court sits in judgment or presides over pretrial proceedings in the Bhopal litigation. Any dismissal of the action now before this Court is thus conditioned on defendant’s consent to submit to discovery on the American model, even after transfer to another jurisdiction. The ostensible lack of devices for third-party impleader or for organizing complex cases under the law of the state of Madhya Pradesh are two other procedural deficiencies which plaintiffs assert preclude a finding that India offers an adequate alternative forum. Assuming for the moment that, upon appropriate transfer, the Bhopal litigation would be adjudicated by the local district court in Bhopal, and that the law of Madhya Pradesh would be applied, this Court is still not moved by plaintiffs’ argument regarding impleader or complex litigation. Although no specific provision in the Indian Code of Civil Procedure permits the impleading of third-parties from whom contribution is sought, other provisions in the Code do provide for impleader. As both parties to this motion state, Order 1, Rule 10(2) of the Indian Code of Civil Procedure “allows the court to add additional parties if the presence of those parties is ‘necessary in order to enable the Court effectively and completely to adjudicate upon and settle all questions involved in the suit.’ ” (Galanter Aff. at 60; Dadachanji Aff. at 18). Professor Galanter posits that a joint tortfeasor would not be considered a necessary party, and would not be joined. Defendant’s expert, conversely, asserts that a party can be added to prevent multiplicity of suits and conflicts of decisions. Thus, Mr. Dadachanji argues, defendants would be able to seek contribution from third-parties if joinder would prevent repetitive litigation or inconsistency. Moreover, the broad provision of inherent powers to aid the ends of justice, as codified at Section 151 of the Indian Code of Civil Procedure would prevent an ultimate miscarriage of justice in the area of impleader. (Dadachanji Aff. at 19). The absence of procedures or mechanisms within the Indian judiciary to handle complex litigation is presented as support for plaintiffs’ position regarding the nonexistence of an adequate alternative forum. Professor Galanter asserts, for example, that Indian judges do not promote settlements. The point is wholly irrelevant to the question of whether an adequate alternative forum exists. In any event, this Court has labored hard and long to promote settlement between the parties for over a year, to no avail. It would appear that settlement, although desirable for many reasons, including conservation of attorneys’ fees and costs of litigation, preservation of judicial resources, and speed of resolution, is unlikely regardless of the level of activism of the presiding judge. Plaintiffs’ next contention is that since no class action procedure exists in India expeditious litigation of the Bhopal suits would be impossible. As with all of plaintiffs’ other arguments, this purported deficiency does not constitute “no remedy” at all. Professor Galanter himself acknowledges that Order 1, Rule 8 of the Indian Code of Civil Procedure provides a mechanism for “representative” suits, “where there are numerous persons having the same interest in one suit.” (Galanter Aff. at 54). Even if the current state of Indian law regarding “representative” suits involves application of the mechanism to preexisting groups such as religious sects or associations, there is no reason to conclude that the Indian legislature, capable of enacting the Bhopal Act, would not see its way to enacting a specific law for class actions. In addition, it does not appear on the face of Order 1, Rule 8 that the “representative” suit is expressly limited to preexisting groups. The Indian district court could adopt the rule for use in a newly created class of injured, whose members all have “the same interest” in establishing the liability of the defendant. An Indian court has law available to create a representative class, or perhaps a few different representative classes. The “scheme” for registration and processing of claims, see supra, at 4, could perform the task of evaluating the specific amounts of claims. Moreover, Mr. Dadachanji gives at least three examples where Indian courts have consolidated suits pursuant to their inherent power under Section 151 of the Indian Code of Civil Procedure. In at least one case, such consolidation allegedly occurred without consent of the parties. (Dadachanji Aff. at 9). The absence of a rule for class actions which is identical to the American rule does not lead to the conclusion that India is not an adequate alternative forum. Final points regarding the asserted inadequacies of Indian procedure involve unavailability of juries or contingent fee arrangements in India. Plaintiffs do not press these arguments, but Mr. Palkhivala touches upon them. They are easily disposed of. The absence of juries in civil cases is a feature of many civil law jurisdictions, and of the United Kingdom. Piper at 252, n. 18, 102 S.Ct. at 264, n. 18 and citations therein. Furthermore, contingency fees are not found in most foreign jurisdictions. Piper at 252, n. 18, 102 S.Ct. at 264, n. 18. In any event, the lack of contingency fees is not an insurmountable barrier to filing claims in India, as demonstrated by the fact that more than 4,000 suits have been filed by victims of the Bhopal gas leak in India, already. According to Mr. Palkhivala, moreover, well-known lawyers have been known to serve clients without charging any fees. (Palkhivala Aff. at 8). Plaintiffs’ final contention as to the inadequacy of the Indian forum is that a judgment rendered by an Indian court cannot be enforced in the United States without resort to further extensive litigation. Conversely, plaintiffs assert, Indian law provides res judicata effect to foreign judgments, and precludes plaintiffs from bringing a suit on the same cause of action in India. (Galanter Aff. at 63-65). Mr. Dadachanji disputes this description of the Indian law of res judicata. He asserts that the pendency, or even final disposition, of an action in a foreign court does not prevent plaintiffs from suing in India upon the original cause of action. Plaintiffs would not be limited, Mr. Dadanchanji argues, to an Indian action to enforce the foreign judgment. (Dadachanji Aff. at 19-20). In addition, he states that an Indian court, before ordering that a foreign judgment be given effect, would seek to establish whether the foreign court had failed to apply Indian law, or misapplied Indian law. (Dadachanji Aff. at 20). The possibility of non-enforcement of a foreign judgment by courts of either country leads this Court to conclude that the issue must be addressed at this time. Since it is defendant Union Carbide which, perhaps ironically, argues for the sophistication of the Indian legal system in seeking a dismissal on grounds of forum non conveniens, and plaintiffs, including the Indian Government, which state a strong preference for the American legal system, it would appear that both parties have indicated a willingness to abide by a judgment of the foreign nation whose forum each seeks to visit. Thus, this Court conditions the grant of a dismissal on forum non conveniens grounds on Union Carbide’s agreement to be bound by the judgment of its preferred tribunal, located in India, and to satisfy any judgment rendered by the Indian court, and affirmed on appeal in India. Absent such consent to abide by and to “make good” on a foreign judgment, without challenge except for concerns relating to minimal due process, the motion to dismiss now under consideration will not be granted. The preference of both parties to play ball on a distant field will be taken to its limit, with each party being ordered to be bound by the decision of the respective foreign referees. To sum up the discussion to this point, the Court determines that the Indian legal system provides an adequate alternative forum for the Bhopal litigation. Far from exhibiting a tendency to be so “inadequate or unsatisfactory” as to provide “no remedy at all,” the courts of India appear to be well up to the task of handling this case. Any unfavorable change in law for plaintiffs which might be suffered upon transfer to the Indian courts, will, by the rule of Piper, not be given “substantial weight.” Differences between the two legal systems, even if they inure to plaintiffs’ detriment, do not suggest that India is not an adequate alternative forum. As Mr. Palkhivala asserts with some dignity, “[wjhile it is true to say that the Indian system today is different in some respects from the American system, it is wholly untrue to say that it is deficient or inadequate. Difference is not to be equated with deficiency.” (Palkhivala Aff. at 4). Piper at 254,102 S.Ct. at 265. The inquiry now turns to a weighing of the public and private interest factors. 2. Private Interest Concerns. The Gilbert Court set forth a list of considerations which affect the interests of the specific litigants to an action, and which should be weighed in making a forum non conveniens determination. The so-called private interest factors, along with public interest factors discussed below, were not intended to be rigidly applied. As the Court stated in Piper, “[E]ach case turns on its facts.” If central emphasis were placed on any one factor, the forum non conveniens doctrine would lose much of the flexibility that makes it so valuable. Piper at 249-50, 102 S.Ct. at 263. Recognizing that “[particularly with respect to the question of relative ease of access to sources of proof,” “the private interests point in both directions,” the Supreme Court nevertheless upheld a district court’s decision to dismiss a case in favor of the relative convenience of a forum in Scotland. Piper at 257, 102 S.Ct. at 267. By contrast, this Court finds that the private interests point strongly one way. As in Piper, it appears that the burdensome effect of a trial in this forum supports a finding that the private interest factors in this case weigh strongly in favor of dismissal. A. Sources of Proof. The first example of a private interest consideration discussed in Gilbert is “relative ease of access to sources of proof.” As stated, the analysis of this issue must hinge on the facts. Limited discovery on the issue of forum non conveniens has taken place, pursuant to the Court’s order of August 14, 1985. The Court can therefore proceed to discuss this question. Union Carbide argues that virtually all of the evidence which will be relevant at a trial in this case is located in India. Union Carbide’s position is that almost all records relating to liability, and without exception, all records relevant to damages, are to be found in and around Bhopal. On the liability question Union Carbide asserts that the Bhopal plant was managed and operated entirely by Indian nationals, who were employed by UCIL. (Affidavit of Warren J. Woomer, formerly Works Manager of the Bhopal plant (“Woomer Aff.”) at 2). Defendant asserts that the Bhopal plant is part of UCIL’s Agricultural Products Division, which has been a separate division of UCIL for at least 15 years, and that the plant had “limited contact” with UCIL’s Bombay headquarters, and almost no contact with the United States. (Woomer Aff. at 4, 32). Woomer claims to have been the last American employed by UCIL. He departed from Bhopal in 1982. (Woomer Aff. at 2). Woomer describes the structure and organization of the Bhopal facility at the time of the accident. The plant had seven operating units, each headed by a manager or department head, each an Indian national. The managers or department heads each reported either directly to the plant’s General Works Manager, or to one of three Assistant Works Managers. (Woomer Aff. at 6). Each of these is also an Indian national. Three of the operating units which at this very early stage of inquiry into liability appear to have been potentially involved in the MIC leak are the Carbon Monoxide, MIC/Phosgene and Carbamoylation units. (Woomer Aff. at 7-10). The Carbon Monoxide and MIC/Phosgene units together employed 63 employees, all Indian nationals. (Woomer Aff. at 9). The Carbamoylation unit employed 99 Indian nationals. (Woomer Aff. at 10). Mr. Woomer states that an inquiry into the cause of the accident would require interviews with at least those employees who were on duty at the Bhopal facility “immediately prior or after the accident;” Mr. Woomer asserts that there are 193 employees, all Indians, who must be interviewed. (Woomer Aff. at 58). In addition to the seven operating units, the Bhopal plant contained seven functional departments which serviced operations. The seven heads of the units reported within the plant much as the department heads did. The maintenance unit was apparently subdivided into departments including Instrumentation, Mechanical Maintenance, both part of the Agricultural Chemical Maintenance unit, which employed 171 people in total, and Plant Engineering and Formulation Maintenance, which employed 46 people. (Woomer Aff. at 11-12). In addition, the Utilities and Electrical department employed 195 people. (Woomer Aff. at 13). According to Mr. Woomer, the various maintenance organizations performed repairs on equipment, provided engineering support, fabricated certain equipment, salvaged other portions, and controlled utilities, temperatures and pressures throughout the plant. (Woomer Aff. at 11-14). Moreover, according to Mr. Woomer, these UCIL departments also kept daily, weekly and monthly records of plant operations, many of which were purportedly seized by the CBI and selected for copying by CBI immediately after the accident. The records and reports of the various maintenance units would likely be relevant to the question of liability at trial. Of the additional functional units, it is possible that Quality Control, with 54 employees, Purchasing, with 53, or Stores may have been directly involved in the disaster by virtue of their participation in analyzing plant output, procuring raw materials for the chemical processes of the plant, and maintaining spare parts and certain chemicals. (Woomer Aff. at 14-19). Thus, the records and reports of these three departments may be necessary to an investigation of liability. While examination of members of the Works Office department and Industrial Relations department would likely be less directly useful, information regarding plant budgets and employee histories might be of relevance. Of great importance are the records and reports of the Safety/Medical department, which was responsible for daily auditing of safety performance in all departments, training and testing on safety rules, maintaining safely statistics and planning and implementing safety drills. (Woomer Aff. at 22-23). The 31 Indian employees of this department worked with the Central Safety Committee of the plant, whose members were drawn from plant management, and the Departmental Safety Committees. Operating units were required to monitor plant safety mechanisms weekly, and to keep monthly checklists. (Holman Aff. # 2 at 9). The Central Safety Committee met monthly, as did the Departmental Safety Committees. (Woomer Aff. at 39). The MIC Unit held monthly safety committee meetings, for example, and issued monthly reports. (Woomer Aff. at 41). Quarterly “Measures of Performance” reviews also covered safety issues, and were required of each operating unit. (Woomer Aff. at 40). Certainly, interviews of the plant personnel involved in safety reports and audits would be particularly relevant to the investigation of the disaster. Plaintiffs refer to three occasions upon which Union Carbide, not UCIL, employees conducted safety audits at the Bhopal plant. As defendant correctly argues, these three events constitute a very small fraction of the thousands of safety audits conducted at the Bhopal facility. The three audits, moreover, were conducted in 1979, the fall of 1980 and in May of 1982, many years prior to the accident which is the subject of this lawsuit. (Plaintiffs’ Memo in Opp. at 25). Two accidents which occurred previously at the Bhopal plant might also be of relevance to the liability inquiry in this litigation. On December 24, 1981, a phosgene gas leak killed a UCIL maintenance worker. Reports of the fatality were sent to Union Carbide management in the United States. (Woomer Deposition, Exs. 30 and 31). Plaintiffs assert that the accident report called for increased training in Bhopal by United States employees of Union Carbide’s Institute, West Virginia, plant. Defendant states that the responsibility for remedying problems in the Bhopal plant rested with the plant itself, and that Union Carbide did not make any recommendations, and was involved only to the extent of receiving a copy of the report which called for its involvement in further training. (Woomer Aff. at 41). The second accident at Bhopal prior to the disaster of December, 1984 took place on February 9, 1982, when a pump seal, perhaps improperly used, failed. (Memo in Opp. at 24; Woomer Aff. at 41). Many employees were injured, and at least 25 were hospitalized. Plaintiffs discuss the fact that Robert Oldford, president of Union Carbide Agricultural Products Company (“UCAPC”) a wholly-owned subsidiary of Union Carbide headquartered in the United States, was in Bhopal at the time of the February 1982 leak. (Memo in Opp. at 24). Union Carbide asserts that Mr. Old-ford was visiting UCIL’s Research and Development Centre, located several miles from the Bhopal plant for an unrelated purpose, and was only coincidentally in Bhopal when the leak occurred. To the extent that this presence in India in 1982 has any significance, Mr. Oldford, and any other United States employees of Union Carbide who conducted safety audits in Bhopal or were present when accidents occurred there, may be flown to Bhopal for testimony or discovery. In addition to safety data, two other types of proof may be relevant to a trial of this case on the merits. Information regarding plant design, commissioning and start-up may bear upon the liability question. Information pertinent to employee training should also have significance. Leaving aside the question of whether the Government of India or UCIL chose the site and product of the Bhopal plant, the Court will evaluate the facts which bear on the issue of relevant records. The findings below concern the location of proof only, and bear solely upon the forum non conveniens motion. The Court expressly declines to make findings as to actual liability at this stage of the litigation. Plaintiffs and defendant agree that in 1973 Union Carbide entered into two agreements with UCIL which were entitled “Design Transfer Agreement” and “Technical Service Agreement.” According to plaintiffs, Union Carbide, pursuant to the Design Transfer Agreement, provided a process design to UCIL, the “detailing [of which] was undertaken in India.” (Memo in Opp. at 17). The process design package consisted of the basic plan of the factory, which was to be fleshed out in the detailing phase. Plaintiffs state that at least nine Union Carbide technicians travelled to India to monitor the progress of the project. Union Carbide also allegedly assigned a “key engineer,” John Couvaras, to serve as UCIL Bhopal project manager. Mr. Couvaras allegedly “assumed responsibility for virtually every aspect of the detailing of the process design,” and approved detail reports of “not only UCIL but also independent contractors, including Humphreys & Glasgow Consultants Private Ltd. and Power Gas Limited” of Bombay, India. (Memo in Opp. at 17-20). Plaintiffs also claim that “[n]o change of any substance was made from Union Carbide’s design during the detailing phase.” Plaintiffs note that only “one portion” of the process design work provided to UCIL by Union Carbide was not used. (Memo in Opp. at 20). In effect, plaintiffs seek to establish that Union Carbide was the creator of the design used in the Bhopal plant, and directed UCIL’s relatively minor detailing program. They urge that for the most part relevant proof on this point is located in the United States. Defendant seeks to refute this contention, with notable success. Turning first to the affidavit of Robert C. Brown, who describes himself as “chief negotiator for Union Carbide Corporation in connection with the two agreements it entered into with ... UCIL in November, 1973,” the Court is struck by the assertion that the two agreements were negotiated at “arms-length” pursuant to Union Carbide corporate policy, and that the Union of India mandated that the Government retain “specific control over the terms of any agreements UCIL made with foreign companies such as Union Carbide Corporation.” (Brown Aff. at 3-4). Mr. Brown alleges that the Letter of Intent issued by the Union of India in March 1972, pursuant to which construction and design of the plant were allowed to ensue provided, inter alia, that: (2) [F]oreign collaboration and import of equipment be settled to the satisfaction of the Government. Mr. Brown claims, on personal information, that UCIL told him that Union Carbide would not be allowed to be involved in the Bhopal project beyond the provision of process design packages. (Brown Aff. at 5). The Design Transfer Agreement indicates that Union Carbide’s duty under the Agreement was to provide process design packages, and that UCIL, not Union Carbide, would be responsible to “detail design, erect and commission the plant.” (Defendant’s Ex. 4, § 4.1). Union Carbide, accordingly, issued limiting warranties with respect to the design packages, detailing of which it would not be involved with. (Brown Aff. at 7, Ex. 4, §§ 4.1, 12.3). The nature of UCIL’s detail design work is discussed in the affidavit of Ranjit K. Dutta, who has held various positions at UCIL and UCAPC. From 1973 through 1976, Mr. Dutta was employed as General Manager of the Agricultural Products Division of UCIL. (Dutta Aff. at 2). Mr. Dutta asserts that the Bhopal facility was built by UCIL over the eight years from 1972 to 1980. (Dutta Aff. at 8). He asserts that Union Carbide’s role in the project was “narrow”, and limited to providing “certain process design packages for certain parts of the plant.” (Dutta Aff. at 9). He continues, stating: Once it did that, it had no further design or engineering role, and that: [T]he process design packages which Union Carbide Corporation provided are nothing more than summary design starting points____ They set forth only the general parameters____ A plant cannot be constructed from a process design package. The detail design comprises approximately 80 percent of the sum of the man hours involved in the design of any project and transposes the general process design parameters into an actual design which can be used for purchasing equipment and actual construction. (Dutta Aff. at 9-12). (emphasis omitted). According to Mr. Dutta, during the five years between the date upon which Union Carbide submitted process designs, and the date upon which the plant started-up, there were only four visits to Bhopal by Union Carbide process design engineers. (Dutta Aff. at 14). In contrast, he asserts that ten to fifteen UCIL engineers, working primarily out of Bombay, were involved in design detailing. (Dutta Aff. at 16). These UCIL engineers oversaw the 55 to 60 Indian engineers employed by the Bombay engineering firm which performed the detail design work. This firm, Humphreys and Glasgow, submitted designs and drawings to the UCIL engineers for approval. Corrected drawings were returned by UCIL to Humphreys and Glasgow for changes, and sent back to UCIL for final approval. (Dutta Aff. at 19-24). Mr. Dutta alleges that “at no time were Union Carbide Corporation engineering personnel from the United States involved in approving the detail design or drawings prepared upon which construction was based. Nor did they receive notices of changes made.” (Dutta Aff. at 24). Mr. Dutta expressly states that the MIC storage tank and monitoring instrumentation were fabricated or supplied by two named Indian sub-contractors. The vent gas scrubber is alleged to have been fabricated in the Bhopal plant shop. (Dutta Aff. at 25). Of the 12,000 pages of documents purportedly seized by the CBI regarding design and construction'of the Bhopal plant, an asserted 2,000 are design reports of Humphreys and Glasgow, UCIL or other contractors. Defendant claims that blueprints and calculations comprise another 1,700 pages of documents held by the CBI. Five thousand pages of contractors’ files, including specifications and contracts are asserted to be in India. In addition, Union Carbide claims that blueprints and diagrams may not reflect final design changes as incorporated into the actual plant, and that the detail design engineers’ testimony will be needed to determine the configuration of the actual plant. (Holman Aff. # 2 at 15-16). One final point bearing on the information regarding liability is contained in the affidavit of Edward Munoz, at a relevant time the General Manager of UCIL’s Agricultural Products Division. He later acted as Managing Director of UCIL. Mr. Munoz has submitted an affidavit in which he states that Union Carbide decided to store MIC in large quantities at the Bhopal plant, despite Mr. Munoz’ warnings that MIC should be stored only in small amounts because of safety. (Memo in Opp. at 15-16; Munoz Aff.). Mr. Dutta, for defendant, asserts that there was never any issue of token storage of MIC at Bhopal, as Mr. Munoz states, and that there is no truth to Mr. Munoz’ assertion that he was involved in the storage issue. (Dutta Aff. at 30). The Court cannot make any determination as to the conflicting affidavits before it. This question, which involves credibility concerns, is left for later in the litigation. To the extent that this particular matter bears upon the relative ease of access to sources of proof, Mr. Munoz and Mr. Dutta both may be called to testify at trial or discovery. Mr. Dutta’s home is in Bhopal. (Dutta Aff. at 1). The Court is not aware of the whereabouts of Mr. Munoz at this time. Either of the two could travel to either alternative forum. In addition to design and safety records, material regarding training of Bhopal personnel is likely to be relevant to the question of liability. Plaintiffs state that Warren Woomer supervised the training of UCIL personnel at Union Carbide’s Institute, West Virginia plant. According to plaintiffs, 40 UCIL employees were transported to Institute’s MIC facility for lengthy training. (Memo in Opp. at 22). Mr. Woomer states in reply that the 40 employees thus trained represented a fraction of the over 1,000 employees who were trained exclusively in Bhopal. (Woomer Aff. at 43). In addition, Mr. Woomer asserts that the training at Institute was pursuant to an arms-length agreement, that UCIL selected the parties to be trained, and that UCIL paid Union Carbide for the training. (Woomer Aff. at 43). Moreover, Mr. Woomer’s description of the training provided at Bhopal suggests that each of the plant’s employees had lengthy cumulative training, of which the Institute training was but a very small portion. (Woomer Aff. at 46). Personnel records, in any event, are located in Bhopal. (Holman Aff. # 2 at 4). The briefs and affidavits contain considerable discussion on the matter of commissioning and start-up of the Bhopal plant. The Court need not resolve the question of who was responsible for these aspects of plant operation. However, the Court determines that the manual regarding start-up was prepared by Indian nationals employed by UCIL. (Woomer Aff. at 48). In the aggregate, it appears to the Court that most of the documentary evidence concerning design, training, safety and start-up, in other words, matters bearing on liability, is to be found in India. Much of the material may be held by the Indian CBI. Material located in this country, such as process design packages and training records of the 40 UCIL employees trained at Institute, constitutes a smaller portion of the bulk of the pertinent data than that found in India. Moreover, while records in this country are in English, a language understood in the courts of India, certain of the records in India are in Hindi or other Indian languages, as well as in English. (Holman Aff. #2 at 12). The Indian language documents would have to be translated to be of use in the United States. The reverse is not true. It is evident to the Court that records concerning the design, manufacture and operation of the Bhopal plant are relatively more accessible in India than in the United States, and that fewer translation problems would face an Indian court than an American court. Since Union Carbide has been directed to submit to discovery in India pursuant to the liberal grant of the American Federal Rules of Civil Procedure, and this opinion is conditioned upon such submission, any records sought by plaintiffs must be made available to them in India. The private interest factor of relative ease of access to sources of proof bearing on liability favors dismissal of the consolidated case. The Indian Government is asserted to have been involved in safety, licensing and other matters relating to liability. Records relating thereto are located in India, as are the records seized by the CBI. Although plaintiffs state that all such records could and would be made available to this Court, it would be easier to review them in India. Transmittal and translation problems would thereby be avoided. B. Access to Witnesses. Gilbert teaches a second important consideration under the heading of private interests, the “availability of compulsory process for attendance of willing, and the cost of obtaining attendance of unwilling, witnesses.” Gilbert, 330 U.S. at 508, 67 S.Ct. at 843. As discussed in detail above, most witnesses whose testimony would relate to questions of causation and liability are in India. Engineers from UCIL and Humphreys and Glasgow and other subcontractors, of whom there are hundreds, are located in India. Shift employees from the possibly malfunctioning units, safety monitoring personnel, those responsible for training, safety auditing, procurement, compliance with regulations and other operations might be required to testify. More than likely, many of these potential witnesses do not speak English, and would require translators. Many of the witnesses are not parties to this litigation. Therefore, as the Court of Appeals for the Second Circuit has stated in the context of a forum non conveniens motion: In fact, the plaintiffs’ cases on liability will depend in large measure upon the knowledge and activities of such witnesses as the employees of [companies] who are not parties to this litigation, but who directly participated in the events which gave rise to it. The United States District Court in New York, however, has no power to subpoena any of these witnesses. It is unlikely that many would be willing to travel to New York to testify; and the cost, in any event, would be prohibitively great. Fitzgerald v. Texaco, 521 F.2d 448, 451-52 (2d Cir.1975), cert. denied, 423 U.S. 1052, 96 S.Ct. 781, 46 L.Ed.2d 641 (1976) (footnote omitted). In contrast, the relatively few witnesses who reside in the United States are primarily employed by Union Carbide. As employees of a party they would probably be subject to the subpoena power of Indian courts. Transportation costs would also be lower, since fewer people would have to make the journey to testify. The presence of the Indian Government in this action is also of critical importance on this motion. Plaintiffs assert that “all necessary officials and employees of the Central Government will voluntarily comply with requests to attend trial.” (Memo in Opp. at 70; Answer to No. 124 of Defendant’s First Requests for Admission, Exhibit 55). This statement does not provide for attendance by officials of Madhya Pradesh or the Bhopal municipality, whom Union Carbide indicates might be impleaded as third-party defendants. As witnesses only, these officials would not be subject to this Court’s subpoena power. As third-party defendants, they might be immune from suit in the United States by the terms of the Foreign Sovereign Immunities Act, 28 U.S.C. § 1602 et seq. State and city officials might also lack sufficient contacts with this district to allow this Court to exercise personal jurisdiction over them. While Union Carbide might be deprived of testimony of witnesses or even potential third-parties if this action were to proceed in this forum, no such problem would exist if litigation went forward in India. The unavailability of compulsory process for Indian non-party witnesses, of whom there are many, such as would ensure their presence at a trial in this country, the high cost of transporting the large number of Indian nationals to the United States, as well as the need to translate their testimony should they appear, all support the argument favoring dismissal of this action on forum non conveniens grounds. The private interest concerns regarding witnesses emphasize the logic of defendant’s position. Relatively fewer witnesses reside in the United States than in India. Almost all of the witnesses located in this country are employees of defendant, and would be subject to compulsory process in India as a result. Transportation costs for the relative few would not compare to the alternate costs of transporting hundreds of Indian witnesses. Since English is widely spoken in India, less translation would be required for foreign witnesses in India than in the converse situation. Should this case be tried in India, fewer obstacles to calling state and local officials as witnesses or parties would face the defendant. The Court determines that this private interest factor weighs in favor of dismissal. C. Possibility of View. The third private interest factor articulated in Gilbert is the ease of arranging for a view of the premises around which the litigation centers. Plaintiffs assert that the notion that a jury view of the plant and environs is necessary is “simply preposterous.” (Memo in Opp. at 71). Plaintiffs note that a viewing of the premises is rarely conducted in products liability cases, since videotapes, pictures, diagrams, schematics and models are more instructive than an actual view. (Memo in Opp. at 71). A viewing of the plant and hutments would probably not be of utmost importance in determining liability, and this consideration is not afforded great weight on this motion. However, the instant case is not identical to the product design defect case cited by plaintiffs, in which a district court judge determined that “the present appearance of the defendants’ facilities may or may not be relevant to production which occurred” in the period in which the allegedly violative manufacture occurred. Hodson v. A.H. Robins Co., Inc., 528 F.Supp. 809, 822 (E.D.Va.1981), aff'd, 715 F.2d 142 (4th Cir. 1983). In the instant case, the site of the accident was sealed after the leak, and the present condition of the plant might be relevant to a finding of liability. A viewing may not be necessary, but conceivably could be called for later in the litigation. An Indian court is in a far better position than this Court to direct and supervise such a viewing should one ever be required. This consideration, though minor, also weighs in favor of dismissal. In summary, then, the private interest factors weigh greatly in favor of dismissal on grounds of forum non conveniens. Since the “balance is strongly in favor of the defendant” and foreign plaintiffs’ choice of a foreign forum is given less than maximum deference, the Court determines that dismissal is favored at this point in the inquiry. Gilbert 330 U.S. at 508, 67 S.Ct. at 843. 3. Public Interest Concerns. The Gilbert Court articulated certain factors which affected the interests of non-parties to a litigation to be considered