OPINION

REAL, District Judge.

I. BACKGROUND

Victims of torture, summary execution and disappearance filed suits for damages, in the form of a class action as well as individual direct actions, against the Estate of the former President of the Philippines, Ferdinand E. Marcos (MARCOS), for human rights violations. Specifically, the violations are alleged to have occurred during the period in which MARCOS, as President of the Philippines, declared martial law, from September 21, 1972 to February 25, 1986.

In 1986 MARCOS fled the Philippines and arrived in the State of Hawaii. MARCOS was a resident of Hawaii at the time he was served with the complaints that are the subject of this litigation but he died during the pendency of these actions. The Estate of Ferdinand E. Marcos (the ESTATE) has been substituted in MARCOS' place; his widow, Imelda Marcos, and his son, Ferdinand E. Marcos, Jr., have appeared before this Court as representatives of the ESTATE.

The action was tried in the three phases: (1) liability, (2) exemplary damages, and (3) compensatory damages, over a nine year period — from 1986 to 1995. In the compensatory damages phase, Phase III, this Court allowed the jury to consider the damages to a random sample of plaintiffs as representative of the injuries suffered by those in the three subclasses; i.e. (1) plaintiffs who were tortured; (2) the families of those individuals who were the subjects of summary execution; and (3) the families of those who disappeared as the result of the actions of MARCOS. Pragmatically, the jury could not hear testimony of nearly 10,000 plaintiffs in this action within any practicable and reasonable time, to do justice to the class members. The individual plaintiffs who opted out of the certified class action each presented his or her individual claim for compensatory damages to the jury in a separate part of the Trial.

This opinion addresses the compensatory damages phase of the trial. The Court deals here with the propriety of the use of inferential statistics to ascertain the damages suffered by each of the 9,541 class members.

II. MARCOS REGIME

MARCOS was elected President of the Philippines in 1965 and was re-elected in 1969. The Philippine Constitution of 1935, still in effect in 1972, was similar to the United States Constitution, in that it limited election of the President to two four-year terms. Thus, MARCOS would have had to leave the office of the Presidency by the end of 1973, but he did not.

On September 21, 1972 MARCOS imposed martial law on all of the Philippines through Proclamation 1081, which suspended the Constitution, in order to keep himself in office. The stated purpose for the imposition of martial law, as expressed in Proclamation 1081, was:

"to maintain law and order throughout the Philippines, prevent or suppress all forms of lawless violence as well as any act of insurrection or rebellion and to enforce obedience to all the laws and decrees, orders and regulations promulgated by me personally or upon my direction."

At the time martial law was declared, a Constitutional Convention, elected by the people, had been meeting and was near completion of proposed revisions to the 1935 Constitution. On orders from MARCOS, some delegates to the Convention were arrested and placed under detention while others went into hiding or left the country leaving the revisions uncompleted.

Without allowing for ratification of the new Constitution by a plebiscite, on January 17, 1973, MARCOS ordered ratification of a revised Constitution, tailor-made for his maintenance of power. With those actions MARCOS planted the seeds for what grew into a virtual dictatorship in the Philippines.

The new Constitution nullified the term limits for the President and provided that MARCOS could function as President, using his own judgment, for as long as necessary. Until he convened a new legislative body, MARCOS also had sole authority to rule in the Philippines.

Proclamation 1081 not only declared martial law, but also set the stage for what plaintiffs alleged, and the jury found, to be acts of torture, summary execution, disappearance, arbitrary detention, and numerous other atrocities for which the jury found MARCOS personally responsible.

MARCOS gradually increased his own power to such an extent that there were no limits to his orders of the human rights violations suffered by plaintiffs in this action. MARCOS promulgated General Order No. 1 which stated he was the Commander-in-Chief of the Armed Forces of the Philippines. The order also stated that MARCOS was to govern the nation and direct the operation of the entire Government, including all its agencies and instrumentalities. By General Orders 2 and 2-A, signed by MARCOS immediately after proclaiming martial law, MARCOS authorized the arrest, by the military, of a long list of dissidents. By General Order 3, MARCOS maintained, as captive, the executive and judicial branches of all political entities in the Philippines until otherwise ordered by himself personally.

Immediately after the declaration of martial law the issuance of General Orders 1, 2, 2A, 3 and 3A caused arrests of persons accused of subversion, apparently because of their real or apparent opposition to the MARCOS government. These arrests were made pursuant to orders issued by the Secretary of Defense Juan Ponce Enrile ("ENRILE"), or MARCOS himself.

The arrest orders were means for detention of each of the representatives of the plaintiff class as well as each of the individual plaintiffs. During those detentions the plaintiffs experienced human rights violations including, but not limited to the following:

1. Beatings while blindfolded by punching, kicking and hitting with the butts of rifles;

2. The "telephone" where a detainee's ears were clapped simultaneously, producing a ringing sound in the head;

3. Insertion of bullets between the fingers of a detainee and squeezing the hand;

4. The "wet submarine", where a detainee's head was submerged in a toilet bowl full of excrement;

5. The "water cure", where a cloth was placed over the detainee's mouth and nose, and water poured over it producing a drowning sensation;

6. The "dry submarine", where a plastic bag was placed over the detainee's head producing suffocation;

7. Use of a detainee's hands for putting out lighted cigarettes;

8. Use of flat-irons on the soles of a detainee's feet;

9. Forcing a detainee while wet and naked to sit before an air conditioner often while sitting on a block of ice;

10. Injection of a clear substance into the body a detainee believed to be truth serum;

11. Stripping, sexually molesting and raping female detainees; one male plaintiff testified he was threatened with rape;

12. Electric shock where one electrode is attached to the genitals of males or the breast of females and another electrode to some other part of the body, usually a finger, and electrical energy produced from a military field telephone is sent through the body;

13. Russian roulette; and

14. Solitary confinement while handcuffed or tied to a bed.

All of these forms of torture were used during "tactical interrogation", attempting to elicit information from detainees concerning opposition to the MARCOS government. The more the detainees resisted, whether purposefully or out of lack of knowledge, the more serious the torture used.

Eventually, MARCOS, his family and others loyal to him fled to Hawaii in February of 1986. One month later, a number of lawsuits were filed, including those that are the subject of this case.

III. CLASS ACTION

On September 22, 1992, in the liability phase of the trial, the jury found defendants liable to 10,059 plaintiffs, for the acts of torture, summary execution and disappearance. On February 23, 1994 the jury awarded plaintiffs $1.2 billion in exemplary damages.

In the compensatory damages phase, the class action plaintiffs presented their case to the jury by using damages sustained by a random sample of plaintiffs as representative of damages suffered by the entire class. After reviewing the deposition of 137 claimants and hearing the live testimony of several class members who could come to Court, the Special Master presented a report to the jury recommending the damages suffered by the 137 claimants, to give the jury a statistically valid representation of damages suffered by the entire class. On January 20, 1995, the jury reconvened and after hearing several representatives of the class and the testimony of the Special Master found the defendant liable to the class for over $766 million in compensatory damages, with individual plaintiff's awards ranging from $150,000 to $700,000.

The Court held that damages of 137 of the claimants, presented to the jury in the form of a report presented by the Special Master, was representative of damages sustained by the entire class, and introduction of such report did not offend due process. Furthermore, the fact that defendants did not have the opportunity to cross-examine all class plaintiffs, because only the testimony of 137 claimants was presented in the report, did not violate defendant's right to a jury trial under the Seventh Amendment of the United States Constitution.

This opinion will address judgment as to the class plaintiffs only.

IV. DISCUSSION

A. Issues Presented

All threshold issues in this case have been previously resolved by the Ninth Circuit. At this time there are two issues before this Court. The primary question is whether the use by this Court of a random sample of plaintiffs, as representative of the injuries suffered by others in the class, violates defendant's due process rights. The second question is whether use of the random sample violates the defendant's Seventh Amendment right to a jury trial.

B. Random Sampling

1. Introduction

The ESTATE asserts random sampling is inappropriate for this case, and each claim should be individually tried. This Court holds otherwise. The use of aggregate procedures, with the help of an expert in the field of inferential statistics, for the purpose of determining class compensatory damages is proper.

James Dannemiller, an expert in the field of inferential statistics and survey sampling for twenty five years, assisted in this case. He has testified as an expert in those areas in both state and federal courts. Mr. Dannemiller formulated a plan so that only 137 randomly selected claims, of the 9,541 claims found to be valid, would have to be examined in order to achieve a 95% statistical confidence level that all claims would fall within the ambit of the 137 randomly selected claims.

Mr. Dannemiller testified that inferential statistics is a recognized science which uses mathematical equations to infer the probability of events occurring or not occurring. One branch of that science is the sampling theory, which deals with the selection of sample sizes sufficient to produce results that can be applied to a larger population from which the sample was selected with a specified probability of error. The formula Dannemiller used in this case is a well-known statistical tool that is found in Leslie Kish, Survey Sampling 53 (New York, John Wiley and Sons 1962) (KISH FORMULA).

Mr. Dannemiller testified under the KISH FORMULA, 137 randomly selected valid claims examined from a larger population of 9,541 validly submitted claims by class members would produce a 95% confidence level. The Court then considered the details of deposing 137 randomly selected claimants.

This Court appointed a Special Master, to facilitate the taking of depositions of 137 randomly selected plaintiffs. The Special Master's appointment had a three-fold purpose: first, he supervised the taking of the 137 depositions in the Philippines; second, he served as a court-appointed expert on damages, under Federal Rule of Evidence 706, to review the deposition transcripts along with the claim forms; finally, he made recommendations on compensatory damages for the 137 claimants as well as the remaining class members to the jury. The Special Master's 182 page findings and recommendations, and the six page addendum thereto, are attached hereto as Appendix A and Appendix B, respectively.

The depositions which the Special Master oversaw were noticed and taken in accordance with the Federal Rules of Civil Procedure. Although having notice of the depositions of the 137 class member sample and the names of the individual class members, the ESTATE chose not to participate and did not appear at any of the depositions, which were taken during October and November of 1994. Nor did the ESTATE choose to depose any of the 9,541 class members to test the procedure employed by the Court, or to acquire evidence to refute the fairness to the defendant of this random selection process using inferential statistical methodology.

The Special Master was directed by this Court to review the depositions for the following three elements: (1) whether the abuse claimed fell within one of the three definitions, with which the Court charged the jury at the liability phase of the trial; (2) whether the Philippine military or paramilitary was involved in such abuse; and (3) whether the abuse occurred during the period of September 1972 through February 1986. The claims of all the class members were filed with the Court and examined by the Special Master. Each claim was made under oath. After considering the deposition of the 137 claimants and the claims filed by each of the class members, the Special Master prepared the attached report.

Of the 137 randomly sampled claims, 67 were torture victims, 52 were execution victims and 18 were disappearance victims. Based upon the depositions of each of the 137 randomly selected class member's claims and review of all the claims of the remaining class members, the Special Master recommended damages under Philippine, International, and American law, for each of the three categories of claims. During the Special Master's testimony, the Court advised the jury that they, in determining damages, could accept, modify or reject the recommendations of the Special Master. The jury was also instructed that they could, independently, on the basis of the depositions of the 137 randomly chosen class members, make their own judgment as to the individual damages of the 137 claimants and the aggregate damages suffered by the class. Copies of the Special Master's and Court-Appointed Expert's Report and Addendum thereto were supplied to each member of the jury. After five days of deliberations, the jury returned a verdict of over $766 million, approximately $1 million less than the Special Master had recommended.

In his report and testimony, the Special Master made damage determinations for torture victims by ranking each claim from 1-5, with 5 representing the worst abuses and suffering. The torture claims were evaluated based upon Judge Real's decision in Trajano v. Imee Marcos-Manotoc, aff'd, In re: Estate of Ferdinand E. Marcos Litigation, 978 F.2d 493 (9th Cir. 1992), cert. denied, ___ U.S. ___, 113 S.Ct. 2960, 125 L.Ed.2d 661 (1993), as part of this matter, and the following considerations: (1) physical torture, including what methods were used and/or abuses were suffered; (2) mental abuse, including fright and anguish; (3) amount of time torture lasted; (4) length of detention, if any; (5) physical and/or mental injuries; (6) victim's age; and (7) actual losses, including medical bills. Although each claim of torture could have been but were not totally unique, as the Court Appointed Expert on damages, the Special Master, was able to determine that there were sufficient similarities within a rating category to recommend a standard damage amount to each victim within that grouping.

For summary execution and disappearance claims, whether there was any torture prior to a victim's death or disappearance weighed into the damages recommended for these two categories. Applying Philippine law, loss of earnings in torture claims was also factored into each of the Special Master's recommendations. After the aggregate lost earnings were computed, the result was converted into American dollars by dividing the figure by twenty-four, which was the approximate exchange rate, as of December 1994, between U.S. dollars and Philippine pesos.

Because there were discrepancies between some transcripts stating income earned in gross, some in net, and some giving no amount, serving as the Court Appointed Expert on damages, the Special Master, recommended that it was necessary to place a cap upon lost earnings; $120,000 was the maximum a claimant could receive. When a witness did not state the amount of income earned by a summary execution or disappearance victim, an average for the victim's occupation was utilized. For example, when the victim was a farmer, the average earnings for one harvesting the same crop on the same amount of land was used. If a person stated the victim's income in terms of perharvest, and if there were three harvests per year, for example, then that victim's earnings would be multiplied by three. For any victim who did not work, there was no award given for lost earnings.

For computing the total amount of damages for summary execution and disappearance victims, depending on the individual facts, there were different variables which went into the equation: (1) torture prior to death or disappearance; (2) the actual killing or disappearance; (3) the victim's family's mental anguish; and (4) lost earnings, calculated in the above described manner.

2. Precedent

The utilization of random sampling was fully examined in Cimino v. Raymark Industries, Inc., 751 F. Supp. 649 (E.D.Tex. 1990), an asbestos class action. There, Chief Judge Robert M. Parker of the Eastern District of Texas allowed the use of inferential statistics in determining the amount of damages to award each class plaintiff. The 2,298 class members were divided into five categories based on the plaintiff's individual claims. The Court then selected a random sampling from each disease category created by the asbestos and the damages attributed to each sample category was then presented to the jury. Each plaintiff who was a member of the random sampling was awarded his individual verdict and the average verdict for each disease category constituted the damage awarded for each non-sampled class member.

As in Cimino, here class members were divided into three categories — torture, summary execution and disappearance — based upon a plaintiff's claims. The court next selected a random sampling from the population of 9,541 plaintiffs. Each plaintiff who was in the random sampling category and was found to have a valid claim was awarded his or her individual verdict. The average verdict for each category was the amount awarded to the class members who were not in the sample class.

3. Due Process

In a case such as this one, where there are 9,541 class members, most of whom live in other areas of the world, a balancing of interests must occur to obtain justice to the parties. A due process analysis must weigh defendant's claim to the right to trial in each individual case against judicial economy and manageability by use of a valid statistical procedure. The Court in Cimino, rejected defendants argument that they were entitled to a one-on-one trial for each of the 2,298 cases. The Court held due process is not necessarily limited to the traditional sense as argued by defendants, "but should also encompass the impact on plaintiffs and even the obvious societal interests involved." Id. at 666. The Court in that case was concerned that a one-on-one trial for each case, assuming the Court could close thirty cases a month, would take six and one-half years.

This Court was moved by the same concerns as Chief Judge Parker in Cimino. Here, individual trials for each of the 9,541 plaintiffs would take decades. Most of that time would be wasted since the nature of the injuries would be similar, if not identical, the testimony would be largely duplicative. Utilizing the procedure employed by the Court the injuries could be accurately categorized, and the source of the injuries would be identical.

This Court believes individual testimony from each of the plaintiffs, i.e., testimony of all 9,541 plaintiffs, could well have been repetitive. Although the facts in this case are not identical to those in Cimino, the damages here are much more objective, i.e., the torture, summary execution and disappearance come from the same source with the same objective. Inferential statistics with random sampling produces an acceptable due process solution to the troublesome area of mass tort litigation.

The issue remains whether this Court's use of inferential statistics in using aggregate procedures, denied defendant's their constitutional due process right to a one-on-one trial. This Court believes, "the aggregate trial is, in some vital respects, superior to the individual trial" and does not violate the substantive or procedural due process rights of either the plaintiffs or the defendant.

This Court finds persuasive the analysis of Professors Saks and Blanck in their discussion that aggregate trials do not violate due process. In Mathews v. Eldridge, 424 U.S. 319, 96 S.Ct. 893, 47 L.Ed.2d 18 (1976) the Supreme Court identified three factors to be balanced by the judge in determining what kind of process is due. The three factors to be balanced are: (1) the private interest affected, (2) the risk of erroneous deprivation of the interest through the procedures used, and (3) the government's interest, including all fiscal and administrative burdens that the additional procedure would require.

The use of random samples in aggregate procedures meets every requirement of the Mathews standard. Under the first prong of Mathews, since liability was established in Phase I of the trial, the private interest affected is the harm done to plaintiffs by MARCOS and what damages defendant would be required to pay plaintiffs for their injuries. It is probable the judgment against the ESTATE, had this Court allowed one-on-one trials, would be significantly more than the judgment the jury returned in the aggregate procedure. Nor were amounts awarded nearly as high as the amounts awarded in other "bipolar" human rights litigation. See, e.g., Filartiga v. Pena-Irala, 577 F. Supp. 860 (E.D.N.Y. 1984) (over $10 million for a single summary execution); see Estate of Ferdinand E. Marcos Litigation, 978 F.2d 493 (9th Cir. 1992) (over $4 million for a single summary execution), aff'g Trajano v. Imee Marcos-Manotoc. It is therefore evident that one-on-one trials would produce substantially higher verdicts than those returned in the aggregate.

The second prong of the Mathews test involves the erroneous deprivation of defendants rights through the procedures used, i.e., aggregated trials. With aggregate trials, the only issue decided is damages; liability is determined subject to objections and appeals; and liability is tried before the jury. Thus there is little risk of erroneous deprivation of defendants interest through the procedures used. In this case there can be no erroneous deprivation of the ESTATE's interest, which is to pay as little in damages as possible. There is no proof that the ESTATE would pay less had damages been determined on a bipolar basis. In fact, as stated above it appears that had such a procedure been utilized, each claim would have brought a higher judgment against the ESTATE.

Finally, aggregate trials are consistent with the third prong of the Mathews test. Clearly it cannot be questioned that a one-on-one trial is more burdensome for the Court than an aggregate trial. The costs involved in conducting bipolar trials with 9,541 plaintiffs in this case would substantially surpass the costs of an aggregate trial which lasted only about one and one-half weeks. The judicial and administrative time and costs of holding bipolar trials would also have been virtually, if not absolutely, prohibitive. Lastly, because class members are mostly impecunious, the cost of bringing them to the forum or even taking their depositions would have prevented their claims from ever being determined. Moreover the whole jurisprudence of class action treatment of numerous claims supports the conclusion that the ESTATE has suffered no due process violation.

4. Seventh Amendment

The issue here is whether the use of random samples, in an aggregate trial, violates the Seventh Amendment right to a jury trial. The Seventh Amendment "was designed to preserve the basic institution of jury trial in only its most fundamental elements, not the great mass of procedural forms and details, varying even then so widely among common-law jurisdictions." The Seventh Amendment provides no formula for the procedures to be used in a trial by jury. Rather, it is the rules of evidence and procedure that impact jury trials. Pragmatic application of these rules, consistent with justice, is all that is necessary for the presentation of the facts necessary for a jury determination. To claim otherwise certainly raises form over substance to a new level in today's jurisprudential world.

Here, the jury did determine the facts of the case, as the substance of the action was presented to the jury. There would be no benefit to either side in having the entire class testify given the repetition in the claims. Rule 23 of the Federal Rule of Civil Procedure does not mandate the presence of each member of the class. Therefore, by choosing a random sample of 137 claimants in an aggregate trial, neither side was deprived of even the form of their right to a jury trial.

In recent years, both complexity of cases and the concern of the length of trials have been the bases upon which several courts have refused jury trial demands. This Court did not go that far. Defendant was given its day in court with the jury, by procedures facilitating the presentation of evidence by use of random sampling in an aggregate damage trial.

C. Federal Common Law

Because plaintiffs in this action are citizens of the Philippines who are complaining of human rights abuses which occurred in that country, this case arises under two statutes — the Alien Tort Statute, 28 U.S.C. § 1350, and the newly enacted Tort Victim Protection Act ("TVPA") of 1991 (codified in the note to 28 U.S.C. § 1350). The Alien Tort Statute does not address damages; the TVPA only provides that an individual who abuses others under color of law of a foreign country is subject to liability for damages. Virtually all of the nations of the world, including the United States and the Philippines, are in agreement that human rights victims should have enforceable rights to fair and adequate compensation. Therefore, the issue of damages is one of federal law.

Because Congress in the TVPA offered no methodology as to how damages should be determined, federal courts are free to and should create federal common law to provide justice for any injury contemplated by the Alien Tort Statute and the TVPA or treaties dealing with the protection of human rights. See Textile Workers Union of America v. Lincoln Mills of Alabama, 353 U.S. 448, 457, 77 S.Ct. 912, 918, 1 L.Ed.2d 972 (1957) ("Some [problems] will lack express statutory sanction but will be solved by looking at the policy of the legislation and fashioning a remedy that will effectuate that policy."); Banco Nacional de Cuba v. Sabbatino, 376 U.S. 398, 425, 84 S.Ct. 923, 939, 11 L.Ed.2d 804 (1964) (". . . an issue concerned with a basic choice regarding the competence and function of the Judiciary and the National Executive in ordering our relationships with other members of the international community must be treated exclusively as an aspect of federal law."); National Audubon Society v. Dept. of Water, 858 F.2d 1409 (9th Cir. 1988); First National City Bank v. Banco Para, 462 U.S. 611, 103 S.Ct. 2591, 77 L.Ed.2d 46 (1983).

Therefore, the aggregation of compensatory damage claims is appropriate under federal common law for human rights claims.

V. CONCLUSION

The use of an aggregate procedure for determining compensatory damages, under the procedures followed in this litigation, was neither a violation of the parties' due process rights nor their right to a jury trial under the Seventh Amendment. The aggregation of compensatory damage claims vindicates important federal and international policies, permits justice to be done without unduly clogging the court system, and was shown to be fair to the defendant.

Judgment shall be entered for plaintiffs.