Samsung Electronics Co., Ltd. v. Rambus, Inc.

Full title: SAMSUNG ELECTRONICS CO., LTD., Plaintiff, v. RAMBUS INC., Defendant

Court: United States District Court, E.D. Virginia

Date published: Jul 18, 2006

439 F. Supp. 2d 524 (E.D. Va. 2006)

Opinion

Civil Action No. 3:05cv406.

July 18, 2006.

Harris Dewey Butler, III, Charles Lewis Williams, James Curie Skilling, Butler Williams Skilling PC, Brian Charles Riopelle, McGuirewoods LLP, Richmond, VA, Ana Elena Kadala, David James Healey, Weil Gotshal Manges LLP, Houston, TX, Matthew D. Powers, Weil, Gotshal Manges, Redwood Shores, CA, for Plaintiff.

Gregory Paul Stone, Munger Tolles Olson LLP, Los Angeles, CA, Rowland Braxton Hill, IV, Craig Thomas Merritt, Michael W. Smith, Christian Barton LLP, Richmond, VA, for Defendant.

MEMORANDUM OPINION

PAYNE, District Judge.

Samsung Electronics Co., Ltd. ("Samsung") has moved for an award of attorney's fees against Rambus Inc. ("Rambus"), under 35 U.S.C. § 285 and the Court's inherent power. In deciding whether an award of attorney's fees is appropriate under § 285, it is necessary to determine whether this is an exceptional case. For the reasons set forth below, the Court finds that this is an exceptional case, but that an award of attorney's fees to Samsung is not appropriate under either § 285 or the Court's inherent power.

FACTUAL AND PROCEDURAL BACKGROUND

Samsung filed this action seeking a declaratory judgment, inter alia, that four patents held by Rambus are unenforceable by virtue of the doctrines of unclean hands, equitable estoppel, patent misuse, waiver, laches, and laches in the United States Patent and Trademark Office ("PTO"). The patents-in-suit were the same as the four patents-in-suit in Rambus, Inc. v. Infineon Technologies AG, No. CIV. A. 3:00cv524 (E.D. Va.) ("Rambus v. Infineon"): (1) U.S. Pat. No. 5,953,263 ("the '263 Patent"); (2) U.S. Pat. No. 5,954,804 ("the '804 Patent"); (3) U.S. Pat. No. 6,032,214 ("the '214 Patent"); and (4) U.S. Pat. No. 6,034,918 ("the '918 Patent"). Rambus asserted counterclaims against Samsung, alleging infringement of the '263 and the '918 patents.

A. Rambus v. Infineon Litigation

Rambus develops and licenses technology to companies that manufacture semi-conductor memory devices. Its patents are directed to various dynamic random access memory devices ("DRAMs"), Rambus DRAMs ("RDRAMs"), Synchronous Dynamic Random Access Memory ("SDRAM"), and Double Data Rate Synchronous Dynamic Random Access Memory ("DDR-SDRAM"). See Rambus, Inc. v. Infineon Tech. AG, 164 F. Supp. 2d 743, 747-48 (E.D. Va. 2001). Beginning in early 1998 and continuing through 1999 and 2000, Rambus developed, refined, and implemented a patent licensing and litigation strategy, which was aimed at several specifically identified DRAM manufacturers. Among the targeted DRAM manufacturers were Infineon, Samsung, and Hynix Semiconductor, Inc. ("Hynix").

Pursuant to that strategy, in June 2000, Rambus asserted, in this Court, patent infringement claims against Infineon with respect to the same four patents-in-suit that were at issue in Samsung's action for declaratory judgment. After extensive discovery and issuance of a claim construction opinion, there was a two week trial on Rambus' infringement claims, as well as Infineon's counterclaims. Ultimately, the judgment was appealed to the United States Court of Appeals for the Federal Circuit, which affirmed in part, reversed in part, and remanded for further proceedings. Additional discovery was conducted at that time and, during those proceedings, it was determined that spoliation of documents by Rambus warranted the piercing of Rambus' attorney-client privilege and work product protection.See Rambus, Inc. v. Infineon Tech. AG, 222 F.R.D. at 296-99. Subsequent discovery was permitted on the issue of spoliation and other issues.

In February 2005, a bench trial was held on Infineon's defense of unclean hands, which was based on Rambus' alleged spoliation of evidence and other litigation misconduct. Simultaneously, a corollary evidentiary proceeding was held with respect to spoliation of evidence, for which a sanction of dismissal was requested. At the conclusion of the trial of those issues, the Court ruled from the bench that Infineon had proven, by clear and convincing evidence, that Rambus was liable for unclean hands, thus barring Rambus from enforcing the four patents-in-suit. Additionally, the Court ruled that Infineon had proven, by clear and convincing evidence, that Rambus had spoliated evidence, for which dismissal was the appropriate sanction. Following that ruling, and before the Court issued findings of fact and conclusions of law, Rambus and Infineon settled the case.

B. Background And Procedural History Of This Action

Also pursuant to its licensing and litigation strategy, and while Rambus was prosecuting its actions against Infineon, Rambus entered license negotiations with Samsung. In October 2000, the parties entered into a license agreement that covered, inter alia, the patents-in-suit in Samsung's action for declaratory judgment. See Samsung Electronics Co., Ltd. v. Rambus Inc., 386 F. Supp. 2d 708, 712 (E.D. Va. 2005). Samsung and Rambus amended that license agreement in 2001 because of developments in the litigation between Rambus and Infineon. See id.

Samsung and Rambus began to renegotiate the terms of the license agreement in July 2004. As part of those negotiations, the parties discussed a so-called "Standstill Agreement" by which any litigation over the license agreement would be delayed for a year while negotiation continued. However, the negotiations did not go to the liking of Rambus. On June 6, 2005, when Samsung refused to accede to Rambus' demand for a contract provision that would allow Rambus to file litigation first in the venue of its choice, Rambus terminated the discussions respecting an extension of the license agreement and the license agreement itself. Simultaneously, Rambus filed a patent infringement action against Samsung in the United States District Court for the Northern District of California. See id., at 713-15. In that action, Rambus claimed that Samsung was infringing, inter alia, the '263 and '918 patents that were at issue in Rambus v. Infineon and in this action.

On June 7, 2005, one day after Rambus brought patent infringement claims against Samsung in the Northern District of California, Samsung filed this action for declaratory judgment, and filed its First Amended Complaint shortly thereafter. See id. at 712. Samsung's complaint and the amended complaint proceeded on the clearly articulated theory that the decision on the spoliation and unclean hands issues in Rambus v. Infineon barred any claim for infringement of the patents-in-suit. On July 12, 2005, Rambus counterclaimed alleging infringement of the '263 and '918 patents.

Pursuant to 28 U.S.C. § 1404(a), Rambus also moved to transfer this action to the Northern District of California so that Rambus could press the infringement claims in its chosen venue. At an evidentiary hearing on that motion, it was established that the General Counsel of Rambus had been directed by the company's management to avoid litigation in this district and to assure that Rambus controlled the selection of forum for any litigation between Samsung and Rambus. Samsung Electronics Co., Ltd. v. Rambus Inc., 836 F. Supp. 2d 703, 713 (E.D. Va. 2005). Indeed, it was for that reason that Rambus terminated the license renegotiation with Samsung and precipitously sued Samsung in the Northern District of California. Samsung Electronics, 386 F. Supp. 2d at 713, 723.

On August 5, 2006, Samsung moved for partial summary judgment on the issues of spoliation and unclean hands. Samsung argued that the Court's bench ruling in Rambus v. Infineon with respect to Rambus' spoliation and unclean hands should be given collateral estoppel effect and consequently that the four patents-in-suit were unenforceable. A briefing schedule was set for Samsung's motion for partial summary judgment, and argument was set for September 21, 2005.

C. Covenants Not To Sue

On September 6, 2005, six days before responding to Samsung's motion for partial summary judgment, Rambus filed an "unconditional" and "irrevocable" covenant not to assert patent infringement claims against Samsung with respect to the '804 and '214 patents ("First Covenant"). The First Covenant expressly extended to actions in the International Trade Commission as well. The scope of the First Covenant extended to "any and all methods, processes, and products made, used, offered for sale, sold, or imported by Samsung currently or at any time prior to the date of this covenant." However, the First Covenant did not extend to any other patents held by Rambus, related or unrelated, and Rambus expressly declined to concede the merits of Samsung's allegation that the '804 and '214 patents were unenforceable and invalid.

On September 12, 2005, Rambus filed its opposition to Samsung's motion for partial summary judgment on the theory of unclean hands based on spoliation. On September 13, 2005, Rambus and Samsung stipulated that the First Covenant "eliminates any need for declaratory relief that Samsung may have had with respect to the '804 Patent and the '214 Patent." Stipulation (Docket No. 42). Samsung, however, reserved its right to request that the Court declare the case exceptional and order Rambus to pay Samsung's attorney's fees under 35 U.S.C. § 285. Rambus expressly reserved the right to oppose such relief, and to argue that the First Covenant moots such relief. The stipulation also provided that Samsung's declaratory judgment action with respect to the '804 and '214 patents was to be dismissed without prejudice.

On September 14, 2005, Rambus' motion to transfer this action to the Northern District of California was denied. On the same date, counsel were ordered to confer about procedures to expedite the trial of this action and to report the results thereof to the Court on September 21, 2005. On September 20, 2005, Samsung filed its reply brief on its motion for partial summary judgment.

On September 21, 2005, the Court gave notice of its intent to take judicial notice of the record of the spoliation and unclean hands bench trial in Rambus v. Infineon. The hearing on Samsung's motion for partial summary judgment was rescheduled to September 28, 2005.

Also, on September 21, 2005, Mr. John Danforth, Rambus' General Counsel, signed a second covenant not to sue Samsung ("Second Covenant"), this time with respect to the '263 and '918 patents. The Second Covenant was filed with the Court on September 22, 2005. The language in the Second Covenant with respect to the covenant not to sue is identical to the language in the First Covenant. Rambus used the Second Covenant as a vehicle to withdraw its counterclaims in this action as well as its claims against Samsung in the Northern District of California, Rambus Inc. v. Samsung Electronics Co., Ltd., et al., Case No. C-05-02298-RMW (N.D. Cal.), which asserted infringement of the '263 and '918 patents. Rambus' counterclaims in this action were dismissed with prejudice by Order entered on September 28, 2005.

On September 28, 2005, the parties argued Samsung's motion for partial summary judgment. The motion was then submitted for decision.

Contemporaneous with the filing of the Second Covenant, on September 21, 2005, Rambus filed its motion to dismiss Samsung's declaratory judgment action for lack of subject matter jurisdiction under Fed.R.Civ.P. 12(b)(1). Rambus contended that the First and Second Covenants ("Rambus Covenants") moot any case or controversy, thereby depriving the Court of subject matter jurisdiction, including jurisdiction to rule on Samsung's request for attorney's fees.

Notwithstanding its argument on the latter point, on October 3, 2005, Rambus made a written offer to Samsung to pay reasonable attorney's fees incurred by Samsung in this action. In Rambus' view, that offer along with the Rambus Covenants "afford[ed] Samsung all of the relief to which it may otherwise be entitled in this action and therefore moots any further proceedings on the merits of any of Samsung's claims, including its allegations that this action qualifies as an exceptional case entitling Samsung to recover its reasonable attorney's fees."¹ However, in making its settlement offer, Rambus expressly declined to concede the merits of Samsung's allegations. Samsung argued that the Court retained jurisdiction not only to decide whether Samsung was entitled to payment of attorney's fees under 35 U.S.C. § 285 and the Court's inherent powers, but also to issue declaratory relief by ruling on Samsung's motion for partial summary judgment.

1 Reply In Support of Motion by Defendant Rambus Inc. To Dismiss Action (Docket No. 73), Ex. A (October 3, 2005 letter from Gregory Stone to David Healey).

On November 8, 2005, the declaratory judgment action was dismissed without prejudice as moot. The motion to dismiss Samsung's claim for attorney's fees was denied. See Samsung Elec. Co. v. Rambus Inc., 398 F. Supp. 2d 470 (E.D. Va. 2005). With respect to Samsung's claim for attorney's fees, the Court then set a schedule for submission of the Rambus v. Infineon record, for briefing the issue of attorney's fees, and for submission of proposed findings of fact and conclusions of law on the issue of exceptionality. A hearing was set for December 15, 2005.

D. Offer of Judgment

On November 29, 2005, Rambus made an offer of judgment to Samsung under Fed.R.Civ.P. 68 for the amount of Samsung's attorney's fees of $476,542.30, plus the full amount of any reasonable additional attorney's fees and costs incurred by Samsung in connection with this action after Samsung's November 22, 2005 filing, in which Samsung had specified the amount of attorney's fees it had incurred as of that date. Rambus argues that its offer of judgment with respect to attorney's fees moots the Court's jurisdiction to decide to impose sanctions under 35 U.S.C. § 285 and the Court's inherent power. For the reasons set forth in the accompanying Memorandum Opinion (Docket No. 133), the Court finds that Rambus' offer of judgment has not divested the Court of jurisdiction to determine whether sanctions are appropriate in this case.

Additionally, for the reasons set forth in the accompanying Memorandum Opinion (Docket No. 134), the Court finds that Samsung is a prevailing party under § 285. Thus, there remains for decision whether this is an exceptional case, and whether an award of attorney's fees under § 285 or the Court's inherent power is appropriate.

DISCUSSION

I. EXCEPTIONAL CASE

A. Standard

"The determination of whether a case is exceptional and, thus, eligible for an award of attorney fees under § 285 is a two-step process." Cybor Corp. v. FAS Technologies, Inc., 138 F.3d 1448, 1460 (Fed. Cir. 1998). It is first necessary to determine whether exceptionality has been proved by clear and convincing evidence. If the case is determined to be exceptional, then it is necessary to assess whether it is appropriate to impose the sanction of attorney's fees. See id.

An award of attorney's fees under § 285 serves two purposes. First, it "compensate[s] the prevailing party for its monetary outlays in the prosecution or defense of the suit," Central Soya Co. v. Geo. A. Hormel Co., 723 F.2d 1573, 1578 (Fed. Cir. 1983), "where it would be grossly unjust that the winner be left to bear the burden of his own counsel which prevailing litigants normally bear." Badalamenti v. Dunham's Inc., 896 F.2d 1359, 1364 (Fed. Cir. 1990) (quoting J.P. Stevens Co. v. Lex Tex Ltd., 822 F.2d 1047, 1052 (Fed. Cir. 1987)) (emphasis in original). Second, and of equal, if not greater, importance, the sanction serves to deter parties from bringing or prosecuting bad faith litigation, see Mathis v. Spears, 857 F.2d 749, 754 (Fed. Cir. 1988). The sanction thus protects litigants, the courts, and the judicial process from abuse.

"A case may be deemed exceptional when there has been somematerial inappropriate conduct related to the matter in litigation, such as willful infringement, fraud or inequitable conduct in procuring the patent, misconduct during litigation, vexatious or unjustified litigation, conduct that violates Fed.R.Civ.P. 11, or like infractions."Brooks Furniture Manf'g, Inc. v. Dutailier Int'l, Inc., 393 F.3d 1378, 1381 (Fed. Cir. 2005) (emphasis added). However, "[a]bsent misconduct in conduct of the litigation or in securing the patent, sanctions may be imposed against the patentee only if both (1) the litigation is brought in subjective bad faith, and (2) the litigation is objectively baseless." Id. (citing Professional Real Estate Investors v. Columbia Pictures Industries, 508 U.S. 49, 60-61 (1993) (defining "sham" litigation)). The Federal Circuit has declined to expand the scope of exceptional cases to include "a patentee's bad-faith business conduct toward an accused infringer prior to litigation."Forest Laboratories, Inc. v. Abbott Laboratories, 339 F.3d 1324, 1329 (Fed. Cir. 2003).

Samsung contends that this is an exceptional case for two reasons. First, Samsung argues that Rambus engaged in the spoliation of evidence as part of its plans for litigation against the DRAM industry, including Samsung. Indeed, Rambus was found to have done precisely that in Rambus v. Infineon. Based in part on that spoliation, Rambus was found in that case to have had unclean hands that barred enforcement against Infineon of the two patents-in-suit in Rambus' counterclaims.² For the reasons outlined in full in Part II, the Court finds that Rambus did in fact spoliate evidence relevant to the patent infringement claims in its counterclaims and at a time when Rambus anticipated litigation with DRAM manufacturers generally and Samsung specifically.

2 In Rambus v. Infineon, the grounds for the unclean hands defense also included the use of false testimony given by Rambus executives, abuse of the discovery process, and other discovery violations. That misconduct was not part of Samsung's unclean hands theory in this case. Rather, Samsung's theory was based only on Rambus' spoliation of evidence.

Second, Samsung contends that Rambus engaged in misconduct by gaming the system when it filed counterclaims alleging infringement of the patents-in-suit, enforcement of which had been barred by the findings of spoliation and unclean hands inRambus v. Infineon. According to Samsung's second theory of exceptionality, Rambus had no intention of ever litigating its counterclaims in this forum. Samsung charges that Rambus has attempted to game the court system by asserting the counterclaims merely to bolster its argument for transferring this action (including Samsung's action for declaratory judgment) to the Northern District of California, where Rambus had already asserted infringement claims with respect to the same two patents.³ After the motion to transfer was denied and the motion for partial summary judgment had been fully briefed, Rambus provided covenants not to sue with respect to the '263 and '918 patents and voluntarily dismissed its counterclaims with prejudice. Samsung contends that this was an attempt to manipulate the court system, thus providing an independent basis for treating this case as exceptional.

3 Rambus filed an action in the Northern District of California one day before Samsung filed this action. See Samsung Electronics Co., Ltd. v. Rambus Inc., 386 F. Supp. 2d 708 (E.D. Va. 2005). That conduct was part of a forum manipulation scheme in which Rambus engaged to avoid litigation against Samsung in this district. Id. at 713, 722-23.

B. Pre-Filing Spoliation As A Basis For Exceptionalty

The first question that must be addressed is whether the spoliation alleged by Samsung, which occurred several years before Rambus filed its counterclaims, constitutes misconduct in the conduct of the litigation within the meaning ofBrooks Furniture. According to Rambus, intentional spoliation of evidence that occurs before the commencement of litigation, which is marked by the filing of the relevant claim (here Rambus' counterclaim), does not constitute misconduct during the litigation, or, "misconduct in conduct of litigation," as the Federal Circuit put it in Forest Laboratories. Consequently, Rambus asserts that Samsung must satisfy the exceptional case standard for vexatious or unjustified litigation set forth by the Federal Circuit in Brooks Furniture. Thus, Rambus argues that Samsung must prove that Rambus filed its counterclaims in subjective bad faith and that the counterclaims were objectively baseless in order to demonstrate exceptionality under § 285. See Brooks Furniture, 393 F.3d at 1381. Given that the parties have framed the applicable legal standard so differently, the Court must first decide whether the intentional destruction of discoverable evidence at a time when the party anticipated litigation, or reasonably should have anticipated litigation, constitutes misconduct during litigation, notwithstanding the fact that no claim has been filed at the time of spoliation.

Rambus' argument, that pre-filing spoliation can only support an exceptional case finding if its counterclaims amounted to sham litigation, requires a tortured reading of Forest Labortories, the decision on which Rambus principally relies. In Forest Labortories, ONY, Inc. and Forest Laboratories, Inc. sought a declaratory judgment of noninfringement and invalidity of two patents, to which Abbott held an exclusive license. Abbott and the patent holder counterclaimed for infringement of the two patents. After trial, the district court awarded the plaintiffs a judgment of noninfringement as a matter of law and held that the defendants were equitably estopped from asserting infringement of the patents against the plaintiffs. The district court found that Abbott had, in bad faith, misled ONY and Forest to believe that it would not assert infringement of the two patents, by encouraging the continued development of the allegedly infringing product and by neglecting to inform the plaintiffs of possible infringement. Based on the conduct underlying the equitable estoppel finding and its determination that Abbott had pursued its counterclaims with reckless disregard for the facts, the district court found the case to be exceptional and awarded attorney's fees to the plaintiffs under § 285.

The Federal Circuit reversed the finding of exceptionality, holding that bad faith business conduct toward an accused infringer prior to litigation was not the sort of misconduct on which a court could ground an exceptional case finding. While the Federal Circuit acknowledged that such conduct might support a finding of equitable estoppel, it declined to extend "the consequences to include attorney fees under the rubric of exceptional case." Forest Laboratories, 339 F.3d at 1329. From that holding, Rambus attempts to extrapolate the general proposition that no misconduct that occurs prior to the filing of a claim, other than inequitable conduct in the procurement of a patent, can form the basis of an exceptional case finding. Rambus' broad reading of the holding in Forest Labortories strains the Federal Circuit's language far beyond its logical import.

In fact, the Federal Circuit has yet to rule on the question whether pre-filing spoliation of evidence in anticipation of litigation constitutes litigation misconduct that can form the basis for a finding of exceptionality. However, in Molins PLC v. Textron, Inc., 48 F.3d 1172 (Fed. Cir. 1995), the Federal Circuit confronted such a scenario, and did not overturn the district court's finding that pre-filing document destruction constituted litigation misconduct. InMolins, the district court held that the plaintiff's patents were unenforceable as a result of inequitable conduct before the Patent and Trademark Office. The court awarded attorney's fees under § 285, finding the case to be exceptional as a result of the plaintiff's inequitable conduct before the PTO and as a result of the plaintiff's litigation misconduct. With respect to litigation misconduct, the district court found that employees of the plaintiff had destroyed records pertaining to the prosecution of one of the patents after the plaintiff had begun contemplating litigation, thereby "precluding any potential defendant from conducting full and fair discovery." See id. at 1186. According to the plaintiff, the destroyed records were abandoned foreign patent files.

The Federal Circuit noted that "[c]ompanies are entitled to maintain file destruction programs without being found to have improperly destroyed evidence," but deferred to the district court's judgment as to whether the "destruction was part of an established and legitimate records disposal program." Id. District courts are "in the best position to monitor parties' litigation conduct," and are "the most injured by misconduct at the pretrial and trial stages." Id. (quoting Rolls Royce Ltd. v. GTE Valeron Corp., 800 F.2d 1101, 1111 (Fed. Cir. 1986)). TheMolins court found no clear error in the district court's findings of litigation misconduct, but because it had reversed some of the district court's findings with respect to inequitable conduct before the PTO, the Federal Circuit remanded the case to the district court to reassess the award of attorney's fees. The Federal Circuit directed the district court to reconsider its exceptional case finding and award of attorney's fees in light of these holdings.⁴

4 On remand, the parties apparently were able to reach a settlement and stipulated to dismissal with prejudice.

Given the Federal Circuit's remand and the subsequent settlement, Molins is certainly not controlling precedent. However, Rambus unduly minimizes the effect of Molins. Rambus contends that, at best, Molins supports the proposition that a district court may consider pre-filing spoliation in determining whether the patentee's infringement claims were brought in subjective bad faith and were objectively baseless. However, the Federal Circuit's decision cannot be read to foreclose a district court's reliance on pre-filing spoliation in considering misconduct during the litigation. Moreover, the court certainly did not hold that spoliation could be the basis for a finding of exceptionality only on a finding that the litigation was vexatious or unjustified. Indeed, the Federal Circuit's mandate to the district court expressed no concerns or hesitation with respect to grounding an exceptional case finding on a plaintiff's pre-filing destruction of evidence at a time when the plaintiff contemplated litigation. And, one might reasonably conclude from the Federal Circuit's discussion that pre-filing spoliation would constitute litigation misconduct that could support a finding of exceptionality.

In fact, Judge Nies dissented in part on the view that remand was unnecessary because the plaintiff's "destruction of documents, which outright precluded full and fair discovery," was itself the kind of action that warrants "an award of attorneys fees in the interests of justice and the fair allocation of the burdens of litigation." Id. at 1192 (Nies, J., dissenting in part). The majority simply considered remand to be the more prudent course in light of the fact that it had reversed one of the other grounds on which the district court had rested the exceptional case finding.

It is important to place the Federal Circuit's decision inBrooks Furniture in its proper context by recalling that theBrooks Furniture test was based on the Supreme Court's definition of sham litigation in Professional Real Estate Investors. It follows that the Federal Circuit's aim was to formulate a test to guide exceptional case determinations when the sole ground asserted for exceptionality is vexatious or unjustified litigation.⁵ It defies common sense, however, to extend the Federal Circuit's pronouncement with respect to vexatious or unjustified litigation in Brooks Furniture to pre-filing litigation misconduct. There is no logic to Rambus' proposed rule that sanctions may be imposed under § 285 for pre-filing litigation misconduct only if that misconduct meets the standard for sham litigation.

5 The Federal Circuit has emphasized repeatedly that "[a]lthough the trial judge may exercise his discretion to award attorney fees and costs because of inequitable conduct during prosecution of the patent or misconduct during litigation, attorney fees are not to be routinely assessed against a losing party in litigation[,] in order to avoid penalizing a party for merely defending or prosecuting a lawsuit." McNeil-PPC, Inc. v. L. Perrigo Co., 337 F.3d 1362, 1372 (Fed. Cir. 2003) (quotingRevlon, Inc. v. Carson Products Co., 803 F.2d 676 (Fed. Cir. 1986)). As the Federal Circuit has recognized, Congress had no intention of discouraging patent holders and licensees from vigorously enforcing their patent rights.

In reality, parties often contemplate and plan litigation long before the filing of a claim. As the district court found inMolins, pre-filing destruction of evidence can preclude any potential defendant from a full and fair opportunity to conduct discovery crucial to the defense of a patent infringement claim. To allow a party, which was anticipating litigation, to avoid sanctions under § 285 because it had the foresight to commit the spoliation of evidence before filing its infringement claim would be utterly irrational and would defeat the very purpose for which the judicial system condemns spoliation of evidence.

The judicial abhorrence of spoliation is based on the need to protect the integrity of the judicial system. As the Fourth Circuit has clearly explained, "[t]he policy underlying [the spoliation doctrine] is the need to preserve the integrity of the judicial process in order to retain [society's] confidence that the process works to uncover truth." Silvestri v. General Motors Corp., 217 F.3d 583, 590 (4th Cir. 2001). This is not mere hyperbole. The "[d]estruction of evidence undermines two important goals of the judicial system — truth and fairness." Lawrence B. Solam Stephen J. Marzen, Truth Uncertainty: Legal Control of Destruction of Evidence, 36 Emory L.J. 1085, 1138 (1987). As the authors so accurately observe:

Evidence destruction impedes the search for truth because it creates inaccuracy if the fact of destruction is unknown and uncertainty if the fact of destruction is revealed. Destruction of evidence is unfair because it potentially creates inequality of access to information.

Id. at 1138.

It is difficult to imagine conduct that is more worthy of being considered litigation misconduct or more worthy of sanction than spoliation of evidence in anticipation of litigation because that conduct frustrates, sometimes completely, the search for truth. And, it creates extra expense in the judicial process in the effort to uncover that which has been destroyed. The rule urged by Rambus — that the sanction of an attorney's fees award cannot be imposed if the spoliation precedes anticipated litigation — would rightly hold the judicial system up to scorn and ridicule.

Thus, the Court finds that where a patentee has destroyed evidence before the filing of an infringement claim, but at a time when the patentee anticipated, or reasonably should have anticipated, litigation, that spoliation constitutes misconduct during the litigation within the meaning of Brooks Furniture, which after all permits a finding of exceptionality "when there has been some material inappropriate conduct related to the matter in litigation." Brooks Furniture, 393 F.3d at 1381. It is nonsensical to assert that the fees incurred by Rambus' litigation targets in dealing with those issues cannot be awarded (if they are otherwise awardable) merely because Rambus destroyed the evidence before it engaged in litigation. Nothing in Brooks Furniture or Forest Laboratories suggests that the Federal Circuit has countenanced such a result. Thus, it is not appropriate to employ the standard for a finding of sham litigation where the ground asserted for the exceptional case finding is pre-filing spoliation in anticipation of litigation.

Consequently, it is necessary for the Court to examine whether the record supports a finding of spoliation against Rambus in regard to its anticipation of litigation with Samsung, and if so whether that spoliation makes this an exceptional case.

II. LITIGATION MISCONDUCT

A. The Record Respecting The Spoliation Issue

The issue respecting the alleged spoliation of evidence by Rambus has been fully developed in discovery in two cases:Rambus v. Infineon and Hynix Semiconductor, Inc. v. Rambus Inc., No. CC-00-20905, United States District Court for the Northern District of California ("Hynix v. Rambus"). In each case, there has been a complete trial involving scores of witnesses and hundreds of exhibits. Samsung eschewed further discovery on the issue whether Rambus engaged in the spoliation of evidence, choosing instead to rely on the record from the trial of the spoliation and unclean hands issues in Rambus v. Infineon. Rambus opposed consideration of that record evidence, and proposed some additional discovery. The proposal for additional discovery was rejected because the proposed discovery did not relate to whether Rambus had engaged in spoliation of evidence.

For example, much of the proposed discovery was directed to whether Samsung has committed intentional spoliation of evidence. Citing to the finding of spoliation entered against Samsung inMosaid Techs. Inc. v. Samsung Elecs. Co., 348 F. Supp. 2d 332, 338-339 (D.N.J. 2004), Rambus argues that Samsung destroyed technical emails on a rolling basis for three years with respect to the DRAMs and DRAM technology at issue in Rambus' infringement counterclaims. According to Rambus, if it had had the opportunity to take discovery on Samsung's email destruction, it could have proven Samsung's unclean hands. Rambus grounds this contention on the proposition that "unclean hands can be asserted in opposition to an equitable defense as well as being assertible as a defense to a claim for equitable relief." Scheiber v. Dolby Labs., Inc., 293 F.3d 1014, 1022 (7th Cir. 2002).

The glaring omission from Rambus' statement of the law is that "[i]f the plaintiff has unclean hands and seeks equitable relief, the defendant's own improper behavior serves as no bar to its equitable defenses."⁶ United Cities Gas Co. v. Brock Exploration Co., 995 F. Supp. 1284, 1296 n. 11 (D. Kan. 1998). See also Precision Instrument Mfg. Co. v. Automotive Maintenance Mach. Co., 324 U.S. 806, 814, 65 S.Ct. 993, 89 L.Ed. 1381 (1945) (holding that doctrine of unclean hands "is a self-imposed ordinance that closes the doors of a court of equity to one tainted with inequitableness or bad faith relative to the matter in which he seeks relief, however improper may have been the behavior of the defendant.");Mas v. Coca Cola Co., 163 F.2d 505, 510 (4th Cir. 1947) (citing Precision Instrument in rejecting plaintiff's contention that he was entitled to relief notwithstanding his fraudulent conduct because defendant was also guilty of fraud and unlawful conduct). Whether Samsung is guilty of spoliation or has unclean hands is entirely immaterial because it would not preclude Samsung from asserting its equitable defenses.

6 As counterclaim plaintiff, Rambus sought equitable relief in the form of a permanent injunction against Samsung under 35 U.S.C. § 283, enjoining Samsung from infringing Rambus' patents.

Rambus' other proposed discovery was equally irrelevant to the issue raised by Samsung: whether Rambus had engaged in spoliation of evidence. And, in any event, to the extent that knowledge of that topic does not reside within Rambus, there has been ample discovery on it.

To assure the just, speedy and inexpensive determination of the exceptional case aspect of this action, the Court incorporated the entire record of the unclean hands trial in Rambus v. Infineon (the testimony and exhibits offered by both sides) into this record.⁷ Also, because Rambus made arguments in this case respecting the issue of spoliation based on the record of the unclean hands hearing in Hynix v. Rambus, that record (transcript and exhibits), in its entirety, was incorporated into this record. The record in Hynix v. Rambus has been reviewed and considered in reaching the decisions reflected herein respecting the issue of spoliation.⁸ Rambus objected to consideration of the record of the unclean hands trial in Hynix v. Rambus, just as it objected to consideration of the record inRambus v. Infineon, but Samsung agreed that the record in that action should be incorporated and considered here.

7 Also included in this record were the objections to the admissibility of evidence made in Rambus v. Infineon and the rulings made thereon. Rambus reasserted here the objections that it previously had made in Rambus v. Infineon. The rulings on all exhibits used in this Memorandum Opinion are reflected in the Final Pretrial Order in Rambus v. Infineon (Docket No. 1067), the transcript of the final pretrial conference, and the trial record, all of which were incorporated as a part of this record.
The Court ordered briefing at one point as to whether judicial notice under Fed.R.Evid. 201 was the proper mechanism by which to incorporate the Rambus v. Infineon record. The Court determined that the better course was to have the proceedings from Rambus v. Infineon introduced into evidence, and directed Samsung to effectuate the introduction of the Rambus v. Infineon proceedings into this record. Nevertheless, the Fourth Circuit has held that it is appropriate for courts to take judicial notice of their own records, upon considerations of efficiency and justice, where the prior case is brought into the pleadings or where the two cases represent related litigation.See United States Fidelity and Guaranty Co. v. Lawrenson, 334 F.2d 464, 467 (4th Cir. 1964). This case is obviously related toRambus v. Infineon, and is brought into the pleadings, and thus the Court takes judicial notice of that record in addition to its physical incorporation into the record in this case.

8 In significant measure, the records in Rambus v. Infineon and Hynix v. Rambus contain the same information.

Rambus' objection is overruled because there was no conceivable or articulated basis for the objection by Rambus, which was a litigant in both cases. Samsung chose to rely solely on the Rambus v. Infineon record, but Rambus was given an opportunity to present additional evidence at the December 15, 2005 hearing on the exceptional case question. Rambus chose not to do so. This choice is perplexing in light of Rambus' repeated assertions that additional evidence uncovered since Rambus v. Infineon and subsequently introduced at the Hynix v. Rambus spoliation trial in October 2005 made that record the full and complete record on Rambus' alleged spoliation. In lieu of adducing additional evidence in support of its arguments, Rambus offered the Declaration of Steven M. Perry, which appears substantially to be a recap of the evidence presented in Hynix v. Rambus. At the Court's direction, Samsung supplemented the record in this case by incorporating the record from Hynix v. Rambus, ⁹ as Samsung has objected to the Declaration of Mr. Perry as inadmissible hearsay.

9 While Rambus' requests for additional discovery were entirely meritless, given that it has twice had full discovery and a full trial on the issue of spoliation and unclean hands, Rambus' discovery requests are mooted by the incorporation of theHynix v. Rambus record.

The findings of fact set forth below are based on Rambus' business documents, the testimony from Rambus v. Infineon, and to a more limited extent on the record created in Hynix v. Rambus. Where findings are based on the Hynix v. Rambus record, it has been so noted.

B. Whether Rambus Engaged In Spoliation Of Evidence

1. General Background Facts

Rambus, which was founded in 1990, is a so-called "technology company." DTX 142. That is to say, Rambus develops and licenses technology to other companies. Rambus' income and profits come from royalties, paid by others, for the right to use Rambus' intellectual property.

In 1990, Rambus filed United States Patent Application Serial No. 07/510,898 ("`898 Application") with claims directed to a computer memory technology known as DRAM. The United States Patent and Trademark Office ("PTO") determined that the '898 Application purported to cover several independent inventions and therefore issued an eleven-way restriction requiring Rambus to elect one invention to pursue in its application. In response, Rambus filed numerous divisional and continuation applications, at least 31 of which have resulted in issued patents. The patents-in-suit are directed to the previously identified DRAM technology.

2. The Background Of The Spoliation

In August 2000, Rambus instituted a patent infringement action against Infineon. The four patents-in-suit included the '263 and '918 patents, which were the subject of Rambus' counterclaims in this action. As explained previously, there were two trials in Rambus v. Infineon. The issue of spoliation arose shortly before the beginning of the first trial and was first addressed in a post-trial opinion, wherein Rambus was "found to have committed various acts of litigation misconduct, including the intentional destruction of documents relevant to [Rambus v. Infineon]." Rambus, Inc. v. Infineon Techs. AG, 222 F.R.D. 280, 286 (E.D. Va. 2004) (citing Rambus, Inc. v. Infineon Techs. AG, 155 F. Supp. 2d 668, 680-83 (E.D. Va. 2001)). Further, it was held there that, while anticipating patent litigation, "Rambus implemented a 'document retention policy,' in part, for the purpose of getting rid of documents that might be harmful in litigation." Rambus, Inc. v. Infineon Techs. AG, 155 F. Supp. 2d at 682. Rambus did not appeal the findings that it intentionally had destroyed relevant documents in anticipation of the Rambus v. Infineon litigation. Therefore, the decision of the Federal Circuit in Rambus, Inc. v. Infineon Techs. AG, 318 F.3d 1081 (Fed. Cir. 2003) left those findings undisturbed and it was the law of the case in Rambus v. Infineon.

Following remand from the Federal Circuit, a status conference was convened for the purpose of planning for the second trial. In that conference, Infineon represented that, in litigation between Rambus and Micron Technologies, Inc. ("Micron"), Hynix Semiconductor, Inc. ("Hynix"), and the Federal Trade Commission ("FTC"),¹⁰ Rambus had produced a substantial quantity of documents that should have been produced before the first trial in Rambus v. Infineon, but were not. Rambus acknowledged that some of the documents produced in those proceedings should have been produced in Rambus v. Infineon, but asserted that others should not have been.

10 Those parallel litigations involve some of the patents-in-suit and many of the same issues that were presented in Rambus v. Infineon.

After further briefing and several hearings, an order was entered under the crime/fraud exception that pierced Rambus' claims of attorney-client privilege and work product protection based on the preliminary record respecting spoliation. See Rambus Inc. v. Infineon Tech. AG, 222 F.R.D. 280 (E.D. Va. 2004). That order was reviewed by the Federal Circuit on Rambus' petition for a writ of mandamus which was denied. Further discovery on the spoliation issue ensued.

The issue of spoliation was tried on remand in Rambus v. Infineon. Evidence of document destruction, as well as other evidence on the issue of unclean hands, was heard by the Court sitting without a jury for three and one-half days. The parties offered testimony (live and on videotape) of 25 witnesses and offered approximately 175 exhibits. In Hynix, the trial on spoliation and unclean hands consisted of eight days of evidentiary hearings and two days of argument. That record consists of the testimony of 20 witnesses and 127 exhibits.

3. Controlling Legal Principles: Spoliation

"Spoliation refers to the destruction or material alteration of evidence or to the failure to preserve property for another's use as evidence in pending or reasonably foreseeable litigation."Silvestri v. General Motors Corp., 271 F.3d 583, 590 (4th Cir. 2001). In ascertaining whether a party has spoliated evidence, the Court is governed by the law of the regional circuit, rather than the decisional law of the Federal Circuit, because spoliation is not a matter peculiar to patent law, and the redress of spoliation is a procedural matter. See Wang Labs, Inc. v. Applied Computer Sci., Inc., 958 F.2d 355, 357 (Fed. Cir. 1992); Silicon Image, Inc. v. Genesis Microchip, Inc., 271 F. Supp. 2d 840, 849 (E.D. Va. 2003). Additionally, spoliation issues are resolved under principles of federal law, rather than the law of any state. See Silvestri, 271 F.3d at 590. As the Fourth Circuit has noted, "spoliation is not a substantive claim or defense but a `rule of evidence,' and thus is `administered at the discretion of the trial court.'" Hodge v. Wal-Mart Stores, Inc., 360 F.3d 446, 450 (4th Cir. 2004) (quoting Vodusek v. Bayliner Marine Corp., 71 F.3d 148, 155 (4th Cir. 1995)).

"The right to impose sanctions for spoliation arises from a court's inherent power to control the judicial process and litigation, but the power is limited to that necessary to redress conduct `which abuses the judicial process.'" Silvestri, 271 F.3d at 590. The sanctions available to courts include adverse inferences, exclusion of evidence or expert testimony, and dismissal. As the Fourth Circuit explained in Silvestri,

The policy underlying this inherent power of the courts [to impose sanctions for spoliation] is the need to preserve the integrity of the judicial process in order to retain confidence that the process works to uncover the truth. `[B]ecause no one has an exclusive insight into the truth, the process depends on the adversarial presentation of evidence, precedent and custom, and argument to reasoned conclusions — all directed with unwavering effort to what, in good faith, it believes to be true on matters material to the disposition.' The courts must protect the integrity of the judicial process because `[a]s soon as the process falters — the people are then justified in abandoning support for the system.'

Id. (citing United States v. Shaffer Equip. Co., 11 F.3d 450, 457 (4th Cir. 1993)) (emphasis added). District courts have "broad discretion in choosing an appropriate sanction for spoliation," but "the applicable sanction should be molded to serve the prophylactic, punitive, and remedial rationales underlying the spoliation doctrine." Id. (quoting West v. Goodyear Tire Rubber Co., 167 F.3d 776, 779 (2nd Cir. 1999)).

(a) Factors In Determining Appropriate Sanction

Spoliation can occur when the destruction of evidence in anticipation of litigation is willful. However, it can also occur when the destruction is the result of inadvertent, albeit negligent, conduct. See Silvestri, 271 F.3d at 593-94. See also United States ex rel. Koch v. Koch Industries, Inc., 197 F.R.D. 488, 490 (N.D. Okla. 1999) ("Spoliation includes the intentional or negligent destruction or loss of tangible and relevant evidence which impairs a party's ability to prove or defend a claim."); Townsend v. American Insulated Panel Co., Inc., 174 F.R.D. 1, 4 (D. Mass. 1997) ("[S]poliation is the intentional, negligent, or malicious destruction of relevant evidence.").

In Trigon Ins. Co. v. United States, 204 F.R.D. 277, 284 (E.D. Va. 2001), this Court defined spoliation as a willful act, drawing on the Fourth Circuit's language in Vodusek. However, the Trigon decision was issued five days before the Fourth Circuit's decision in Silvestri. Silvestri teaches that both negligent and willful destruction can constitute spoliation and that the appropriate place to assess the effect of the spoliator's state of mind is in ascertaining an appropriate sanction, not in assessing whether spoliation has occurred. See Silvestri, 271 F.3d at 593-95. And while "a court must find some degree of fault to impose sanctions," id. at 590, the degree of fault need not rise to the level of willful spoliation. See also Residential Funding Corp. v. DeGeorge Financial Corp., 306 F.3d 99, 108 (2nd Cir. 2002) ("[T]he `culpable state of mind' factor is satisfied by a showing that the evidence was destroyed knowingly, even if without intent to breach a duty to preserve it, ornegligently."). For example, the Fourth Circuit has noted that, while the ultimate sanction of dismissal is generally reserved for instances of spoliation involving bad faith by the plaintiff, some instances of negligent spoliation will require dismissal because of the resulting prejudice to the defendant. See id. at 593.¹¹

11 Ultimately, however, the distinction between intentional and negligent spoliation is of little moment on the facts of this case. As set forth below, the Court finds that Rambus spoliated evidence but that it did so willfully and intentionally.

This segues to another important point. Unlike the equitable defense of unclean hands, prejudice to the adverse party is not an element that must be proven in order for a court to find that a party's destruction of evidence amounts to spoliation. Rather,Silvestri instructs that prejudice, like wilfulness, is a factor to be considered in deciding on an appropriate sanction.¹² Id. ("At bottom, to justify the harsh sanction of dismissal, the district court must consider both the spoliator's conduct and the prejudice caused and be able to conclude either (1) that the spoliator's conduct was so egregious as to amount to a forfeiture of his claim, or (2) that the effect of the spoliator's conduct was so prejudicial that it substantially denied the defendant the ability to defend the claim."). In Townsend, the court noted that it was appropriate to consider a number of factors in reaching an appropriate sanction: "(1) whether the defendant was prejudiced as a result of the destruction of the evidence; (2) whether the prejudice can be cured; (3) the practical importance of the evidence; (4) whether the plaintiff was in good faith or bad faith; and (5) the potential for abuse if the evidence is not excluded." Townsend, 174 F.R.D. at 4.

12 Indeed, if a party had to prove prejudice as an element of spoliation, an adverse inference instruction could hardly be deemed a sanction.

(b) Anticipation Of Litigation

Whether or not Rambus anticipated litigation with Samsung at the time of the alleged document destruction is a threshold issue in deciding the spoliation issue. "The duty to preserve material evidence arises not only during litigation but also extends to that period before the litigation when a party reasonably should know that the evidence may be relevant to anticipated litigation."Silvestri, 271 F.3d at 591. See also Kronisch v. United States, 150 F.3d 112, 126 (2nd Cir. 1998) ("This obligation to preserve evidence arises when the party has notice that the evidence is relevant to litigation — most commonly when suit has already been filed, providing the party responsible for the destruction with express notice, but also on occasion in other circumstances, as for example when a party should have known that the evidence may be relevant to future litigation.")). "Notice does not have to be of actual litigation, but can concern `potential' litigation. Otherwise, any person could shred documents to their heart's content before suit is brought without fear of sanction." Bayoil, S.A. v. Polembros Shipping Ltd., 196 F.R.D. 479, 483 (S.D. Tex. 2000) (citing ABC Home Health Serv., Inc. v. IBM Corp., 158 F.R.D. 180, 182 (S.D. Ga. 1994)).

In Kronisch, the district court rejected the defendants' claims that documents were destroyed to preserve the confidential identities of outside participants in a study sanctioned by the Central Intelligence Agency, to prevent incomplete documents from being misunderstood, and to prevent paper overflow. Instead, the district court found that defendants destroyed the documents out of fear of litigation at some point in the future. The district court held that the defendants had a duty to preserve the documents, even absent imminent litigation, and the Second Circuit affirmed.See also Byrnie v. Town of Cromwell, Bd. of Ed., 243 F.3d 93, 108 (2nd Cir. 2001) (citing Kronisch for proposition that "documents destroyed years before suit brought could reasonably be found to have been destroyed in anticipation of litigation where fear of potential future litigation plausibly motivated the spoliation.").

The determination of when a party anticipated litigation is necessarily a fact intensive inquiry, and a precise definition of when a party anticipates litigation is elusive. One helpful analytical tool is the more widely developed standard for anticipation of litigation under the work product doctrine. The work product doctrine provides that documents and tangible things prepared in anticipation of litigation or for trial, which are otherwise discoverable, are only discoverable upon a showing that the party seeking them has substantial need of them in preparation of its case and that the materials cannot be obtained by other means without undue hardship. See Fed.R.Civ.P. 26(b)(3); Hickman v. Taylor, 329 U.S. 495 (1947). The Fourth Circuit has acknowledged that "litigation is an ever-present possibility in American life." National Union Fire Ins. Co. v. Murray Sheet Metal Co., Inc., 967 F.2d 980, 984 (4th Cir. 1992). Thus, in order for a document to receive protection as attorney work product, it "must be prepared because of the prospect of litigation when the preparer faces an actual claim or a potential claim following an actual event or series of events that reasonably could result in litigation." Id.

The Second Circuit has since clarified that

[a]lthough the non-occurrence of the events giving rise to the anticipated litigation is a factor that can argue against application of the work product doctrine, especially when the expected litigation is merely a vague abstract possibility without precise form, there is no rule that bars application of work product protection to documents created prior to the event giving rise to litigation.

United States v. Adlman, 68 F.3d 1495, 1501 (2nd Cir. 1995). The court noted that "[i]n many instances, the expected litigation is quite concrete, notwithstanding that the events giving rise to it have not yet occurred." Id. For example, where "a party expects to be sued by a particular adverse claimant in the event it undertakes to trade under a disputed mark or publishes a book the copyright of which is contested," there is "no reason why work-product protection should not apply to preparatory litigation studies undertaken by that party (or its representatives) before it begins to trade under the contested mark or publishes the book." Id. The Fourth Circuit's real message in National Union Fire was "to stress the necessary causal relationship between the anticipated litigation and the creation of the document, rather than a requirement that the litigation-causing events have already occurred."¹³ Id.

13 An alternative characterization of this test is as follows:
a party anticipates litigation when (1) it subjectively believed that litigation was a real possibility and that belief was objectively reasonable and (2) the document was prepared or obtained because of the prospect of litigation. See United States ex rel. Fago v. M T Mortgage Corp., 235 F.R.D. 11, 16 (D.D.C. 2006).

While the work product definition of when a party anticipates litigation is not helpful with respect to negligent spoliation, by virtue of its causal element, it nevertheless provides a helpful guide when assessing intentional spoliation: at the very least, a party has anticipated litigation when it destroys relevant evidence because of the prospect of litigation on an actual or potential claim.

(c) Scope Of Duty To Preserve Evidence

Corporations are not obligated, "upon recognizing the threat of litigation," to "preserve every shred of paper, every e-mail or electronic document, and every backup tape." Zubulake v. UBS Warburg LLC, 220 F.R.D. 212, 217 (S.D.N.Y. 2003). Indeed, "[s]uch a rule would cripple large corporations." Id. Nevertheless, "[w]hile a litigant is under no duty to keep or retain every document in its possession . . ., it is under a duty to preserve what it knows, or reasonably should know, is relevant in the action, is reasonably calculated to lead to the discovery of admissible evidence, is reasonably likely to be requested during discovery, and/or is the subject of a pending discovery request."Wm. T. Thompson Co. v. General Nutrition Corp., Inc., 593 F. Supp. 1443, 1455 (C.D. Cal. 1984). "[A]nyone who anticipates being a party or is a party to a lawsuit must not destroy unique, relevant evidence that might be useful to an adversary."Zubulake, 220 F.R.D. at 217.

(d) Document Retention Policies

As the Supreme Court has noted, "`[d]ocument retention policies,' which are created in part to keep certain information from getting into the hands of others, including the Government, are common in business." Arthur Andersen LLP v. United States, 544 U.S. 696, 704 (2005). "It is, of course, not wrongful for a manager to instruct his employees to comply with a valid document retention policy under ordinary circumstances." Id. "In contrast, however, a document retention policy adopted or utilized to justify the destruction of relevant evidence is not a valid document retention policy," and "[i]t follows that implementing such a policy in advance of reasonably foreseeable litigation would not be proper and could constitute spoliation."Hynix Semiconductor, Inc. v. Rambus, Inc., 2006 WL 565893, *20 (N.D. Cal. 2006) (unpublished).

"Once a party reasonably anticipates litigation, it must suspend its routine document retention/destruction policy and put in place a `litigation hold' to ensure the preservation of relevant documents." Zubulake, 220 F.R.D. at 218. See also Broccoli v. Echostar Communications Corp., 229 F.R.D. 506, 510 (D. Md. 2005). "The obligation to retain discoverable materials is an affirmative one; it requires that the agency or corporate officers having notice of discovery obligations communicate those obligations to employees in possession of discoverable materials." National Ass'n of Radiation Survivors v. Turnage, 115 F.R.D. 543, 557-558 (N.D. Cal. 1987). It is no defense that particular employees were not on notice of the duty to preserve evidence or what kinds of evidence were material to the potential litigation. See Turnage, 115 F.R.D. at 557. "To hold otherwise would permit an agency, corporate officer, or legal department to shield itself from discovery obligations by keeping its employees ignorant." Id.

4. Spoliation: Findings Of Fact And Conclusions Of Law

Rambus does not contest that, beginning in early 1998, it developed a document retention policy. Nor does Rambus dispute the volume of documents that it destroyed in 1998, 1999, and 2000. The disputed spoliation issues are whether Rambus destroyed documents when it anticipated, or reasonably should have anticipated, litigation, and whether the destroyed documents were relevant to anticipated, or reasonably foreseeable, litigation. The findings of fact pertinent to each of those issues are set forth below.

(a) The Destruction Was Accomplished When Rambus Anticipated, Or Reasonably Should Have Anticipated, Litigation

i. From 1990 To 1998

Some events that occurred before the period at issue are helpful in understanding the anticipation of litigation issue. A brief look is thus in order.

In 1990, the year of its founding, Rambus patented its RDRAM technology and began licensing it. DTX 142. By 1992, Rambus was of the view that its technology was "covered by 18 [United States] patents, with over 300 claims." In its business plan for the five year period 1992-1997, Rambus expressed the opinion that:

Because Rambus Technology represents such an innovative and unique way to provide high bandwidth logic-to-memory interconnect, the patents are extensive and fundamental. It is Rambus' opinion that the patents will be issued largely as filed and that companies will not be able to develop Rambus-compatible technology or Rambus-like technology without infringing on multiple fundamental claims of the patents.

DTX 142, § 2.1.2 (emphasis added).

It was also disclosed in the company's 1992 business plan, that Rambus had come to learn that a Joint Electron Device Engineering Council ("JEDEC") Committee, an industry standard-setting committee, had been working on specifications for a standard for SDRAMs. At the time, Rambus believed that JEDEC would establish a standard at the end of 1992, at the earliest. Rambus also was of the view that SDRAMs infringed some of Rambus' patents and that it could file for additional patents that covered features of the SDRAMs made according to the forthcoming JEDEC standard. DTX 142. It was also Rambus' view that, when that was accomplished, the company would be "in position to request patent licensing (fees and royalties) from any manufacturer of [SDRAMs]." Id. The company's articulated "action plan [was] to determine exact claims and file the additional claims by the end of Q3/92. Then to advise [SDRAM] manufacturers in Q4/92." Id.

In late December 1991, Rambus began attending JEDEC meetings, and officially joined JEDEC as a member in February 1992. Rambus Inc. v. Infineon Techs. AG, 318 F.3d 1081, 1085 (Fed. Cir. 2003). Beginning in 1992, Richard Crisp and Billy Garrett attended the JEDEC meetings on behalf of Rambus. Crisp reported regularly to the Rambus executive staff about the technology being considered for the JEDEC standard and the progress of the standard-setting process. As envisioned by the 1992 business plan, Crisp also discussed with Rambus' patent lawyers how to amend or develop claims that would bring within the reach of Rambus' patents products that were made in accord with the forthcoming JEDEC standard. In other words, as the Federal Circuit put it, "Rambus thought it could cover the SDRAM standard and tried to do so while a member of an open-standard setting committee." Rambus, Inc. v. Infineon Techs. AG, 318 F.3d at 1104.

As early as the spring of 1992, the company's patent counsel, Lester Vincent, a partner in the Blakely Sokoloff Taylor Zafman firm, advised Richard Crisp and Allen Roberts, who was then in charge of obtaining Rambus' patents, that the company could encounter a defense of equitable estoppel if Rambus, by attending JEDEC meetings, created the impression that Rambus would not enforce its patents. DTX 1535; DTX 1523. These warnings were repeated by Vincent and Anthony Diepenbrock, then Rambus' in-house patent counsel, in 1993 and 1995. By January 1996, Rambus had become sufficiently concerned that its participation in JEDEC could adversely affect its patent enforcement strategy that it determined to attend no more JEDEC meetings. DTX 4169. This decision was made by Rambus' CEO, Geoff Tate, and other executive staff. Rambus withdrew from JEDEC in June 1996. Rambus Inc. v. Infineon Techs. AG, 318 F.3d at 1085.

However, throughout the time that Crisp and Garrett attended JEDEC meetings, they, but mostly Crisp, communicated with the numerous Rambus executives, largely by way of email, about the technological developments at JEDEC and the progress of the JEDEC SDRAM standard. Crisp and Garrett also accumulated a substantial number of documents from JEDEC, such as agendas, meeting notes, and informational presentations made by other DRAM manufacturers which were JEDEC members.

Also, by the late 1990s, Rambus had come to the view that DRAM manufacturers were using its technology and patented inventions to make infringing products. And, by then, Rambus had come to the view that steps must be taken to have the DRAM industry take seriously Rambus' intellectual property rights. To that end, in October 1997, Rambus hired Joel Karp, formerly employed by Intel and a consultant to Samsung, to implement its plan to secure royalties from DRAM manufacturers whose products Rambus considered to infringe its patents and trench on its technology.

ii. 1998

Not long after Karp's arrival, he became head of the intellectual property department at Rambus, and he began to take steps to formulate a patent licensing and litigation strategy. In DTX 4071, apparently prepared in the fall of 1999, Rambus recounted the company's "Top Level Key Results for 1998." Section 18 of the document identified, under the heading, "Develop and enforce IP," as a 1998 key result:

C. Get all infringers to license our IP with royalties 203a than RDRAM (if it is a broad license) OR sue.

Id. Karp testified that, as head of the intellectual property department, he provided the input for that section of the 1998 key results document.

To help accomplish these corporate objectives, Karp, in January 1998, telephoned Diane Savage, a partner at the law firm Cooley Godward. Karp told Savage that "he was working at Rambus, and that he was looking for some litigation — somebody to provide him with litigation assistance." DTX 9008. In response to that request, Savage asked Dan Johnson, a litigation partner at Cooley Godward, to meet with Karp. Id.

On February 12, 1998, Karp met with three lawyers from Cooley Godward. Karp recorded the topics that were discussed and the action items agreed on at that meeting, and he memorialized them in a document entitled:

COOLEY GODWARD MEETING 2/12/98 Licensing/Litigation Strategy

DTX 3681.

According to that document, Karp and the lawyers "[r]eviewed a license program document" as to which there were "minor changes." As part of that notation, Karp reported that "[r]oyalty rates will probably push us into litigation quickly." As a consequence, Karp noted that Rambus needed to make itself "battle ready" by gathering critical documents to put into a searchable, coded electronic database and that the company needed a "document retention policy."¹⁴ Karp's notes then reflect that it would be necessary to clean out all attorney notes from patent prosecution files so that they would conform with the official files. Further along, Karp reported that Rambus:

14 The precise entry says:

Make ourselves battle ready. Start gathering critical documents in company so we can start putting together an electronic database. Gather all documents into searchable, coded database. Need company policy on document retention policy.

[N]eed[ed] to litigate against someone to establish [a] royalty rate and [to] have [a] court declare patent valid.

Karp also reported that Johnson identified a suit for breach of contract as an alternative to a patent infringement action because it was easier to prove breach of contract than to prove patent infringement. Consequently, according to Karp, "[t]hey [Cooley Godward] will review Micron, Fujitsu, and Samsung and Hyundai contracts and formulate litigation strategy driven by results of the analysis — breach — scope of license, NDA or patent infringement" (emphasis added). The notes also reflect that experts needed to be selected "in advance."

The companies mentioned during the discussion on February 12, 1998 (Micron, Fujitsu, Samsung and Hyundai (now Hynix)) were major DRAM manufacturers. And, some of them had RDRAM licenses or other agreements with Rambus that were to be assessed as predicates for the breach of contract litigation alternative.

After the February 12 meeting, Karp and Johnson prepared a document dated February 23, 1998 and entitled "PROPOSED STRATEGY FOR RAMBUS." DTX 3678. One aspect of the strategy was a "tiered litigation strategy." The strategy included three scenarios, the second of which involved Rambus filing a patent infringement suit. Interlineated in hand at the end of the document, in Karp's writing, are the notations "Document retention policy — patent attorney files."

One week later, on March 2, 1998, Karp made a presentation entitled "Licensing and Litigation Strategy" to Rambus' board of directors. DTX 3680. The board was told:

If licensing discussions do not result in resolution, tiered litigation strategy kicks in.

Another slide outlined options:

"Option 1: Breach of Contract Remedy"

"Option 2: Patent Infringement Suit."

The document continues:

If the infringing party is unlicensed, patent suit can be brought in venue of our choice

• ITC

• Southern California

• Eastern District of Virginia (the rocket docket)

The Licensing and Litigation Strategy presentation to the board also explained that Rambus could sue an alleged infringing DRAM manufacturer for breach of contract if the manufacturer had an existing license for compatible products (RDRAMs), but that a patent infringement suit was the only option if the infringing party did not have such a license. Of course, the alleged "infringing" parties included manufacturers of SDRAMs and DDR-SDRAMs.

The Licensing and Litigation Strategy then identified, for the Rambus board, a hierarchy of "companies chosen." First, were "licensees which currently have released non-compatible product;" second, "present licensees which are currently well along with alternate development;" and, third, "non-licensees with released product covered by Rambus IP."

The Licensing and Litigation Strategy then outlined to the board a timetable which began with obtaining samples of alleged infringing products, followed by reverse engineering and claim charts, then sending a letter and holding a meeting, and then scheduling a second meeting. The final step was designated as "commence legal action," which was forecast to occur four to six months after obtaining samples of the infringing product.

The Licensing and Litigation Strategy presentation concluded by identifying for the board several "NEAR TERM ACTIONS." Those actions, in order, were:

• Need to create document retention policy

• Need to prepare discovery database

• Need to organize prosecuting attorney's files for issued patents]

The record thus establishes that, by February and March of 1998, Rambus had determined that a number of DRAM manufacturers, some of which held Rambus licenses for compatible products (or RDRAMs), and others of which did not, were making incompatible products using Rambus technology. Rambus intended to do what was necessary to secure royalties for the use of its inventions. It retained counsel to provide advice on how to achieve that end, and, from the beginning, the means to that end included both licensing and litigation.¹⁵

15 Sometime in late February or early March 1998, Johnson left Cooley Godward and became a partner at Fenwick West. However, Johnson continued to advise Rambus and, to some extent, so did Cooley Godward.

Although, at the time, the final decision to litigate had not been made, and although achieving a license with a DRAM manufacturer would eliminate the need to litigate with that manufacturer, the prospect of litigation certainly was not, as Rambus would have it, a vague or far-off possibility. To the contrary, even in the early spring of 1998, Rambus considered litigation to be a realistic alternative.

The litigation alternative was sufficiently concrete, even by then, that the company had identified potential litigation targets by industry type (DRAM manufacturers) and by name, and it had identified potential venues in which to prosecute the litigation and the theories of liability (breach of contract or patent infringement) to assert in it. In fact, Rambus had identified, as components of aggressive readiness, the need to prepare a discovery database and the need to select experts. And, as the Rambus board was told, the litigation strategy involved the near term actions of establishing a document retention policy and cleansing the patent prosecution files.

On March 19, 1998, two weeks after Karp made that presentation to the board, Savage, in response to a telephone request, sent to Karp a document entitled "Document Retention Policy Guidance." DTX 3676.¹⁶ The document was a general outline of document retention practices.

16 A slightly different version of the document was sent to Karp on March 20, 1998. DTX 3700. Another iteration of that document is shown as dispatched to Karp on April 27, 1998. DTX 3683. It appears that, by this date, Johnson had left Cooley Godward.

The company's business plans for its intellectual property activity in 1998 were outlined in a document entitled "IP TOP LEVEL GOALS Q298 FINAL CONFIDENTIAL."¹⁷ This business planning document contains four topic headings:

17 The document is undated, but its text shows that it was prepared before April 1, the first day of the second quarter of 1998.

1. Patent Activity;

2. Infringement Activity;

3. IP Protection Activity; and

4. Litigation Activity.¹⁸

The fourth IP Top Level Goal was outlined under the heading "Litigation Activity." The following entry appears as item B in that category:

"Propose policy for document retention — Done"

See Rambus Inc. v. Infineon, 222 F.R.D. 280 (E.D. Va. 2004).

18 Most, if not all, of the twelve items under Patent Activity, Infringement Activity, and IP Protection Activity are of the sort which reasonably could be expected to generate documents, some of which likely would be relevant to patent litigation involving the patents-in-suit. And, some of those activities (e.g., procuring suspect devices, reverse engineering them, and preparing claim charts) surely would generate documents that would be relevant in patent litigation.

Using the guidance document sent by Savage, Karp prepared the Rambus document retention policy. And, on May 14, 1998, Karp in a memorandum bearing the subject line "Back Up Strategy/Document Retention Policy," announced that Rambus was "on the eve of `starting the task of implementing a company-wide retention policy.'"

The company's planning documents for the third quarter (July-September) of 1998 include a series of documents entitled "IPQ3'98 Goals." Each of these documents has four major category headings: (1) Patent Activity; (2) Infringement Activity; (3) IP Litigation Activity; and (4) Negotiation Activity. One of those "IPQ3'98 Goals" documents reflects, under the heading, "IP Litigation Activity," as Goal A, the implementation "of the document retention plan" and, as Goal B, "[s]taff training event for document retention program with outside counsel."

Before the document retention policy was implemented Karp reviewed it with Rambus' senior management, including the company's CEO, Geoff Tate, Allen Roberts, the Vice President of Engineering, and other appropriate vice presidents. DTX 9009. There is evidence that the document retention policy was adopted for some of the legitimate reasons that warrant adoption of such policies. It is also quite clear that the Rambus policy was adopted to rid the company of documents that would be discoverable in litigation. That is evident from the testimony of several Rambus executives. For example, Anthony Diepenbrock, formerly the in-house patent lawyer at Rambus, testified that one reason for implementing a document retention policy was concern about documents being discoverable in litigation. Diepenbrock recalled that a particular focus of that concern was email communications. DTX 9012. Allen Roberts testified that one of the reasons given by Karp for purging files was that they were discoverable in litigation. DTX 9016. Roberts himself sent an email on March 16, 1998, in which he noted "there is a growing worry about the e-mail backups as being discoverable information" and proposing a systematic regular deletion of email backup tapes. DTX 5185.

This record linking concern over litigation discovery with the document destruction plan is consistent with the presentation that Karp and Johnson gave to Rambus' managerial staff on July 22, 1998. One of the slides used in that presentation, DTX 3686, identifies a list of discoverable documents including email messages, files stored on individual computers, corporate databases, backup tapes, system records and logs, and computers. That presentation specified that special care should be taken with email and electronic documents because, inter alia, email communications "are generally less formal and thoughtful than written correspondence" and, therefore, "candid comments" are made that "can have a significant impact on the outcome of a case." Johnson explained that emails should be treated like written documents for discovery or destruction purposes. The presentation begins and ends with two interesting notations: "Before Litigation" a "Document Retention/Destruction Policy" and the "Eve of Litigation" "The Need for an Effective Document Retention Policy."

The targeting of discoverable documents is evident also in presentations given by Karp to the staff at divisional meetings within the company. DTX 4024. For example, Karp's presentation begins with the admonition "Email is Discoverable In Litigation Or Pursuant To A Subpoena."

In the middle of 1998, Karp arranged to retain, as outside counsel for Rambus, Neil Steinberg, a lawyer with whom Karp had a previous relationship in connection with litigation between Samsung and Texas Instruments.¹⁹ At the time he began to represent Rambus in mid-1998, Steinberg says that his responsibility was "licensing and preparation for litigation, and, of course, prosecution . . . related to Rambus technology and Rambus patents." DTX 9007. According to Steinberg, the "licensing and preparation for litigation" to which he referred was related to Rambus' patents that were thought to relate to third parties who used RDRAMs and third parties who used SDRAMs and DDR-SDRAMs. Id. ²⁰ Also, in the summer of 1998, Rambus was putting the finishing touches on the document retention program and beginning to implement it, with Shred Day occurring on September 3, 1998. DTX 4026. On that day, burlap bags were handed out to every Rambus employee and documents were placed in the bags and delivered to an outside contractor for shredding which incidentally occurred on site. On Shred Day and the next day, Rambus destroyed the equivalent of 291 boxes (or 757,531 pages) of documents.²¹

19 Steinberg was in-house counsel for Samsung and Karp was an expert witness for Samsung in that litigation.

20 After the attorney-client and work product privileges were pierced, and documents were produced disclosing the nature and extent of Rambus' document destruction, Steinberg changed his testimony and insisted that there was no anticipated litigation until almost immediately before Rambus sued Hitachi in January 2000 and that there was no anticipation of litigation after the Hitachi litigation settled on June 23, 2000 until almost immediately before Rambus initiated the action against Infineon on August 8, 2000. The Court believes the testimony given by Steinberg in DTX 9007 in January 2001 before the attorney-client privileges were pierced and the documents respecting spoliation were uncovered. The Court does not believe, and thus rejects, the testimony of Steinberg given in DTX 9020 and October 2004 in the Infineon action (DTX 9021) wherein Steinberg asserts that litigation was not anticipated until just before Rambus sued Hitachi and Infineon, respectively. As to the latter testimonies, the Court finds that Steinberg was not telling the truth.

21 Even before Shred Day 1998, Rambus had sent 1,268 computer tapes to be degaussed on or before July 1, 1998. HTX 107; HTX 157; HTX 29. This magnetic sanitizing made the information on the tapes irrecoverable. Robert Kramer testified in Hynix v. Rambus that it was his understanding the tapes were erased pursuant to Rambus' document retention policy. Rambus made sure to save backup tapes that contained information on technical projects, but everything else was sent for erasure. 6 Trial Tr. 1105:17-1106:25, 1113:11-14. The testimony in Hynix v. Rambus also establishes that at some point before these computer tapes were destroyed, Karp accessed one of the tapes in order to retrieve a document that would help prove the date of conception for Rambus' patents. Karp saved that one tape, and the rest (1,268) were destroyed. 2 Trial Tr. 224:4-8, 225:3-226:25.

Approximately one month later, on October 1, 1998, Steinberg gave a presentation to Rambus executives which, on Rambus' privilege log, is identified as "Patent Litigation Strategy Update." This presentation helps to explain why Rambus did not actually implement its Licensing/Litigation Strategy until late 1999. To understand the situation in the fall of 1998, it is well to recall that, although Rambus had filed the '898 application in 1990, its progress in securing patents was slowed because the PTO required Rambus to reconfigure its applications and that led to the necessity to file continuation and divisional applications. Also, in late 1991, Rambus joined JEDEC and, as reflected in the company's 1992 business plan, learned that JEDEC was formulating a standard applicable to SDRAMs. The process of developing the standard was a slow one. As Rambus attended the meetings, its representatives, Crisp and Garrett, obtained information that was used to improve existing applications or to file new ones in a deliberate effort to cover the evolving SDRAM standard.

Further, Steinberg ascertained that some of the pending applications might not be effective in covering SDRAM products to be made in compliance with JEDEC standard. Therefore, he took over the patent prosecution of a number of patents in 1998 with a view to strengthening the Rambus portfolio as to SDRAM and DDR-DRAM products.

Steinberg's strategy update for October 1998 is in several pages. On the page entitled "Strategy Update 1098-1," Steinberg advised the Rambus executives:

DO NOT ROCK THE DIRECT BOAT

• We should not assert patents against Direct partners until ramp reaches a point of no return (TBD)

• Probably not until Q1/00

DTX 3687. Testimony established that the term "Direct" means the RDRAM and thus, in effect, Steinberg advised the Rambus executives not to assert Rambus' patents against DRAM manufacturers who had licenses for the RDRAM technology until those manufacturers had "reached a point of no return." In other words, he advised that Rambus should forestall an assertion of its patent rights against DRAM manufacturers with whom it had licenses until those manufacturers had become committed to use of RDRAM technology. At the same time, Steinberg posed a significant question:

However, Big Question Is — WHAT'S THE RUSH?

• What is compelling business reason? I can't think of any

* * *

• Let's not snatch defeat from the jaws of victory

The next slide, Strategy Update, 10/98-2, specifies that Rambus should give top priority to strengthening its portfolio by filing continuation cases based on the '898 filing in 1990, the objective of which was to "cover SDRAM, DDR, SLDRAM, any and all forms of synchronous memory (static and dynamic)." To that end, a new series of filings were recommended. Meanwhile, Rambus was proceeding with reverse engineering efforts and Steinberg advised that the company should " Continue In Stealth Mode During '99."

The following slide, Strategy Update, 10/98-3, suggested to the Rambus board a way to implement the point made on slide 2. Specifically, all prosecutions based on the '898 filing were then to be given to Steinberg who would add eight to twelve new continuation cases and continue with the five then being prosecuted by the Blakely Sokoloff law firm. It was Steinberg's expectation that all of the cases would issue within 12 to 18 months from filing. Thus, he suggested that the strategic portfolio of Rambus patents would be ready to present to the industry during calendar year 2000. He advised also that all reverse engineering should be completed in advance of that time.

The purpose of strengthening the strategic portfolios was also outlined by Steinberg. In his view, it "Should Result In Quick Settlements By Several Companies."

Thus, the record is that, in October 1998, one month after implementing the document destruction plan that was part of its Licensing and Litigation Strategy, Rambus decided to slow down its schedule slightly. It did so for two reasons. One was to allow its adversaries to get locked into a position with production of products that used Rambus technology. The other reason was to allow Rambus to strengthen its patent portfolio. But, both reasons were in pursuit of the company's Licensing and Litigation Strategy. And, a key part of that strategy was to get rid of discoverable documents. The first purge had occurred by the time the schedule was slowed slightly.

iii. 1999

After the documents were shredded in September 1998 and the company's schedule was adjusted in October 1998, Karp set about to further refine the enforcement scenario for 1999. To that end, by email entitled "Enforcement Scenario For 1999," Karp sent to Rambus' chairman, president and vice president of engineering, and other officers a scenario that he developed based on "inputs from the 11/30 [1998] management staff meeting." Karp obviously understood the significance of this document because he instructed the recipients not to make copies of it and to return all copies to him after a discussion on the following Monday. DTX 3691.

The attachment is titled "Patent Enforcement 1999." According to Karp and Johnson, this document was prepared by Karp with input from Johnson, whose practice specialty is litigation. The subtitle of the attachment is "NUCLEAR WINTER SCENARIO."

The first part of the document refers to a breakdown in the relationship between Rambus and Intel, Rambus' largest licensee, and to the possibility that the relationship between the two might come to an end. The complete breakdown of the Intel/Rambus relationship was thought to be unlikely. Thus, the discussion as to litigation with Intel is described as hypothetical.

The second part of the document, however, discusses "COMPLAINTS AGAINST DRAM COMPANIES." The discussion, which is not described as hypothetical, articulated "potential causes of action against DRAM companies" to include patent infringement, breach of contract, other common law claims (fraud and business disparagement or defamation), and antitrust actions. The document also identified specific litigation targets and articulated the strategy that those targets should be sued in separate suits and in separate, specifically identified fora, so as to minimize the ability of the targets to cooperate with each other in defending the case. Five candidates were identified. They were: Fujitsu (or Mitsubishi), Hitachi, Hyundai (now Hynix), NEC, and Siemens (now Infineon).

In a section entitled "NEGOTIATION TACTICS," the enforcement plan for 1999 suggested meetings at which the targets would be shown how they infringe by using claim charts. The proposed royalties would be high (5% to 10%), if settlement was not thought desirable in a particular case, or low (1% to 2%), if settlement was desired by Rambus. However, Karp concluded by stating:

My recommendation is that we should not be too concerned with settlement at this point and should push for very high rates. We would give them a very short time to accept our royalty terms by setting a short expiration date of the offer to settle. During the meeting we would make it very clear to them that protracted negotiations are not possible and that we are ready to file suit.

DTX 3691, p. 3 (emphasis added). That, of course, was the approach actually taken with both Hitachi and Infineon. Specifically, Rambus started the process with Hitachi on October 22, 1999, DTX 5380, and, when Hitachi did not agree in short order, Rambus filed suit on January 18, 2000. Rambus first asserted its claims against Infineon on June 23, 2000, and, when a license did not materialize quickly, Rambus filed an action against Infineon on August 8, 2000.

Karp's statement in DTX 3691 is but one reason why the Court considers incredible his testimony in both the Infineon andHynix unclean hands trials that, in 1998 and 1999, Karp and Rambus regarded litigation as unlikely because the goal was to reach settlement. This document, like other contemporaneous business records, substantially disproves that and kindred assertions.

In April 1999, Rambus, through Karp, instructed its outside patent counsel, Lester Vincent of the Blakely Sokoloff firm, to comply with the Rambus document retention policy and to "clean out all the Rambus files that had issued." DTX 3710. According to Vincent, by April 19, 1999, he had followed Karp's April 5th request and cleaned 11 of 49 issued patent files at Blakely Sokoloff in accord with the Rambus document retention policy. At that time, Karp asked Vincent to speed the process up and a secretary was assigned full time to file clearance. DTX 3737; DTX 9007.²²

22 Citing the testimony of Johnson, and deposition testimony of individuals from IBM, as well as Deipenbrock, Rambus' former in-house counsel, and Vincent, Rambus contends that, once a patent is issued, it is standard practice to strip the patent prosecution file of all documentation other than that which is in the public record. The Court is doubtful that there is such a practice, but, even if there is such a practice, that kind of destruction cannot take place by outside lawyers when the client is anticipating, or reasonably should anticipate, litigation.

In June 1999, Steinberg made a presentation to Rambus executives. DTX 3689. According to a Rambus privilege log, the topics were intellectual property and litigation strategy. In a section of the presentation entitled "KR99.5 Update for IP," the executives were given a status report.²³

23 "KR" stands for "key results" and "99.5" stands for "half way through 1999."

Under the category "Current: IP ACQUISITION AND PROTECTION," Steinberg advised that more than ten continuation cases based on the 1990 filing (the '898 application) with claims directed to SDRAM, DRDRAM, and SLR-DRAM had been filed. Steinberg further announced as an objective, the commencement of license negotiation with at least one company with the purpose of starting the "clock for calculation of damages by Q4/99-Q1/00." Steinberg then proposed amended goals, the second of which was to begin license negotiations with one company, thereby starting the clock for calculation of damages by Q4/99, and then to begin license negotiations with two additional companies during Q1/00. Finally, he advised that the amended goal would include the choosing of "one company to litigate with during Q1/00" and to "[c]ommence litigation during Q2/00 upon ex/board approval."

Steinberg also informed the Rambus executives of the selected "SDRAM Targets" for whom infringement cases would be prepared in "Q4'99." Those targets included Hitachi and Infineon, against whom Rambus actually filed patent litigation actions in January 1999 and August 2000, respectively. It was the stated objective to select the first target by "early Q4'99" and the second and third target by "mid-Q4'99." The factors to be considered in selecting the target included the opponent's "Licensing and Litigation Capabilities," and the "Economic Impact of Licensing and Litigation" in the United States, Europe, and Korea.

Thus, by June 1999, the scheduled slowdown proposed by Steinberg in October 1998 went by the wayside and planning for litigation resumed apace. Indeed, in June 1999, Karp and Steinberg were preparing Rambus' intellectual property department's third quarter goals. A first cut of those goals was prepared on June 27, 1999. DTX 4067. The third section of the "IPQ3'99 Goals" was addressed to the company's "Licensing/Litigation Readiness." The goals included the preparation of licensing positions against three manufacturers (item 3D), the preparation of a litigation strategy against one of the three manufacturers thusly identified (item 3E), to beready for litigation with 30 days notice (item 3F), and to "organize 1999 shredding party at Rambus" (item 3G) (emphasis added).

To achieve item 3G of Rambus' 1999 Licensing/Litigation Readiness goals, a second shredding event occurred on August 26, 1999. DTX 4068; DTX 3759. On this occasion, the company's chief executive officer, Geoff Tate, advised that he was aware of, but would not be attending, the shredding party. DTX 3759.²⁴ On Shred Day 1999 (August 26, 1999), Rambus destroyed 188 boxes (or 487,688 pages) of documents.

24 The '263 patent issued on September 14, 1999, and the '918 patent issued on March 27, 2000.

On September 24, 1999, Karp and others made an "IP Strategy" presentation to Rambus executives, addressing the question whether there was life at Rambus if Intel should pursue other technology. DTX 3698. The presentation goes on to explain that the "[b]est route to IP credibility is through victory over a major DRAM manufacturer." Slide 4 advises that

Even if we gain some initial settlements, we will have to ultimately pursue remedies in court

Companies like Micron will fight us tooth and nail and will never settle

The next slide explains why DRAM manufacturers are the most attractive litigation target: direct infringers, good litigation story, larger number of patents-in-suit, and limited exposure to counterclaims. DTX 3698.

On October 14, 1999, the board of directors was given an update respecting the targets for anticipated litigation. DTX 3675. In particular, they were shown a detailed matrix that Steinberg, in continued pursuit of the company's Licensing and Litigation Strategy that had begun in 1998, had created for selecting litigation targets. The matrix for legal considerations included such topics as the adversary's experience in battle, the litigation story, Rambus' exposure to counterclaims, and venue flexibility. There was also a matrix for business considerations such as the size of the recovery, the ability to settle, the significance of the products of the adversary, and the ability to drive competitor's standards.

The matrix for each area, legal and business, reflected the company's goals of confirming and establishing respect for "Rambus IP." This, of course, was one of the reasons for hiring Karp and, for developing the company's Licensing and Litigation Strategy beginning in early 1998.

The board was informed that, of 13 potential litigation adversaries, those achieving the highest score on the legal matrix were, in order, Hitachi, NEC, Samsung, Fujitsu, and Toshiba, while those achieving the highest score on the business matrix were, in order, Micron, Hitachi, AD, Hyundai (now Hynix), and Samsung. The overall combined weighting (60% for the legal factors, 40% for the business factors) produced the following overall rankings, in order: Hitachi, Samsung, Hyundai (now Hynix), NEC, and Micron. The board was also informed of a timeline for negotiations which culminated in the filing of a complaint in Delaware.

The "#1 Target Recommendation" was Hitachi, which was to be approached during the fourth quarter for settlement discussions. If, after six weeks, a license had not been achieved, the objective was to file suit in Delaware "ASAP." That strategy and those tactics are precisely those proposed in Karp's November 1998 paper entitled "Enforcement Scenario for 1999," DTX 3691, p. 3, "NEGOTIATION TACTICS," which, of course, is simply a more refined articulation of the Licensing and Litigation Strategy developed in February 1998.

The board was also briefed on "VENUE SELECTION" and the importance of selecting a proper venue in which to conduct the anticipated litigation. The presentation disclosed that:

3 Venues have been considered

• Eastern District VA (rocket docket)

• Delaware

• Northern CA

Each venue was evaluated and Delaware was considered to be the "most attractive" forum.

By letter of October 22, 1999, DTX 5380, Rambus implemented its plan against Hitachi. That, of course, is in the fourth quarter of 1999.

In November 1999, Rambus retained Gray Cary Ware Freidenrich ("Gray Cary") to represent it in the patent infringement action to be filed in the District of Delaware (a previously selected venue) against Hitachi. Also, in late 1999 or early 2000, Rambus selected Howrey Simon to represent it in a case against Hitachi in the International Trade Commission.

iv. 2000

In a document entitled "IP Update 1/18/00," DTX 3673, the Rambus board of directors was informed that the action against Hitachi would be filed in the District of Delaware as of 3:30 p.m. eastern standard time that day. A copy of the cover page of the complaint for patent infringement was attached and the board was told that the company expected to expand the Delaware litigation to include 10 U.S. patents and then to expand the litigation to Germany based upon the European counterpart of the original Horowitz disclosure. The action was eventually transferred to the Northern District of California.

Lester Vincent, acting on his own, stopped "cleaning" the patent prosecution files in his office when he learned that Hitachi had been sued. No one at Rambus gave him that instruction. On June 22, 2000, Vincent learned, by email from Karp, that the Hitachi litigation had settled. The next day, June 23, 2000, Vincent resumed cleaning out his Rambus prosecution files in accord with the previous instructions from Karp to act expeditiously in completing that task.

Also, on June 23, 2000, Rambus first asserted the '804 patent against Infineon by way of a letter from Steinberg to Infineon's vice president, Dr. Andreas von Zitzewitz. On that day, Vincent cleaned out his '804 prosecution file. DTX 3784.

After June 23, 2000, Infineon and Rambus engaged in negotiation and discussions respecting a possible license by Infineon and a possible means of settling the patent infringement claim made by Rambus. Once again, as envisioned by Rambus' Licensing and Litigation Strategy, when a settlement did not occur in short order, Rambus filed an action against Infineon on August 8, 2000.

On July 17, 2000, approximately one month after having first asserted the '804 patent against Infineon and six months after suing Hitachi and approximately three weeks before filing the action against Infineon, Steinberg issued a memorandum to all Rambus executives that was entitled "Reminder of Document Destruction Policy re: Contracts." DTX 3700. Steinberg instructed the Rambus executives to destroy draft contracts and materials used during negotiations that are not part of the final contract. According to Steinberg, that pertained to "all licenses." At the time that Steinberg issued the July 17 destruction order, Rambus had negotiated license contracts with at least 14 different DRAM manufacturers, at least seven of which were on the litigation target list.²⁵

25 Documents respecting these contracts and negotiations could be relevant or lead to the discovery of admissible evidence respecting the issue of damages and reasonable royalties as well as other reasonable licensing terms in any litigation, including the one initiated three weeks later by Rambus against Infineon.

In December 2000, some three and a half months after filing the action against Infineon, a company-wide document destruction plan was conducted in conjunction with a transfer of offices from one location to another. At that time, during the pendency of Rambus v. Infineon and in the midst of discovery in that case, Rambus destroyed 575 boxes (or 1,495,575 pages) of documents.

iv. Other Findings Of Fact

The foregoing facts, which are established by clear and convincing evidence, permit the Court to make other factual findings, also shown by clear and convincing evidence. The foregoing facts are summarized briefly below, and the permissible inferences and additional findings follow.

In February 1998, Rambus consulted Cooley Godward with an eye toward developing a licensing and litigation strategy to enforce its intellectual property. Following these consultations, Karp, who was the head of Rambus' IP department, realized that the royalty rates that Rambus would seek to extract would push the company into litigation quickly. Rambus' board of directors were informed of this likelihood in March 1998. Indeed, the board was also informed of an already extant tiered litigation strategy, which involved filing a suit for breach of contract or patent infringement, or both, in specified venues. The board presentation defined near term actions that were necessary to ready the company for litigation. One of the identified near term actions was the development and implementation of a document retention policy. Another was the cleaning out of Rambus' patent prosecution files, so that they would conform to the official files maintained by the PTO, commonly referred to as "file wrappers."

Rambus relies on the testimony of Karp and Johnson in support of the notion that Rambus' discussions with Cooley Godward in February 1998 and the presentation given to the board in March 1998 were focused on licensing, rather than litigation. Rambus also relies on the testimony of Johnson and Karp to advance the related notion that, in 1998, litigation was a distant prospect, which would only ensue in the event that Rambus' licensing efforts were not successful. Johnson and Karp go so far as to assert that litigation was not even in contemplation or seriously considered in the first quarter of 1998 or as of the time (September 1998) when the first wholesale destruction of documents took place.

The testimony of Karp and Johnson to that effect is rejected as not credible. Johnson's testimony is flatly at odds with that of Diane Savage, who at the time was Johnson's law partner, and who the Court accepts as a credible witness because her demeanor was convincing and she has nothing at issue in this dispute. Additionally, her testimony is consistent with Rambus' contemporaneous business records and Cooley Godward's billing records. Savage testified that Karp told her that he needed "litigation assistance," and, for that reason, she put him in contact with Johnson, a litigator. Licensing lawyers were involved in the early meetings as well, but that is unremarkable given that Rambus' licensing strategy was also under review. Karp's notes of the February 12 meeting, the Licensing and Litigation Strategy document which Karp and Johnson prepared together, and Karp's presentation to the board of directors refute the testimony given by Johnson and Karp that the focus of Rambus' efforts at this stage was confined to licensing and that litigation was merely an ephemeral possibility.

One does not make oneself "battle ready," nor establish a discovery database, nor identify potential venues, nor identify specific litigation targets, nor select experts, to address a vaguely anticipated litigation potentiality. The efforts of Karp and Johnson to suggest that those terms mean something else amount to nothing more than after-the-fact spin.

Johnson's testimonial demeanor, when addressing the issue of anticipation of litigation, was that of an advocate for Rambus, rather than that of a professional who was disinterested in the outcome. His bias is understandable, to some extent, because he was defending advice that he had given. However, Johnson's bias can hardly be thought to aid the credibility of testimony, especially when it is inconsistent with Rambus' contemporaneous business records. Johnson's defensive and adversarial manner might be attributed in part to the fact that his advice to Rambus was incomplete. In Johnson's presentations to Rambus, he failed to explain that spoliation could occur not only on the "eve of litigation," as he put it, but also if documents were destroyed when the company anticipated, or reasonably should have anticipated, litigation. That is a serious omission given that the test for spoliation universally includes this component. Nor did Johnson's presentation, or anything else he claims to have said, explain that, in order to have an effective suspension of the document destruction plan during litigation, employees must be specifically instructed respecting what documents are relevant to the litigation (and thus cannot be destroyed) and what documents are not relevant (and thus can be destroyed). In sum, the fact that Johnson's testimony, in part, is offered in defense of his less than complete advice is another reason not to credit his testimony respecting the issue of whether and when Rambus anticipated litigation. When that is considered in perspective of the conflict between Johnson's testimony and that of Diane Savage and the Rambus business records, Johnson's testimony on the topic of Rambus' anticipation of litigation is rejected as lacking credibility.

Karp's testimony is likewise incredible on that topic. It too is thoroughly at odds with the corporate business records and with Karp's contemporaneous writings on the subject. Indeed, Karp sent an email discussing Rambus' document retention policy to other Rambus executives, on May 14, 1998, referring to Johnson as "Rambus' litigation counsel." DTX 3697. Karp cannot now credibly be heard to say that Johnson's advice on document retention was sought without regard to the benefit it would have in future litigation. Furthermore, Karp's testimonial demeanor when answering questions on the topic of when Rambus anticipated litigation with DRAM manufacturers was evasive and unconvincing.

By the summer of 1998, Rambus had retained Neil Steinberg as outside counsel. One of Steinberg's responsibilities was to prepare for litigation related to Rambus technology and Rambus patents against third parties using SDRAMs and DDR-DRAMs, as well as the RDRAM. This significantly undercuts Rambus' position that it did not anticipate litigation with DRAM manufacturers until much later, and further demonstrates the unreliability of the testimony given by Karp and Johnson on this point.

The presentations made by Johnson and Karp in July and August 1998 provide further support for the factual conclusion that the September 1998 document destruction was executed in anticipation of litigation. In those presentations, Rambus' managerial executives and staff were informed of the document retention policy. The content of the presentations makes clear that a principal animating factor for the adoption of the document retention policy was the apprehension of forthcoming litigation with DRAM manufacturers and the discoverability of Rambus' files in that litigation. Several of Rambus' top executives confirmed that they were told that eliminating discoverable documents (paper and electronic) was a principal impetus for the adoption of the document retention plan. Rambus' internal communications confirm the concern over the discoverability of email communication, which according to Johnson is a particularly troublesome area because of the candid views often expressed in this casual form of communication.

In October 1998, one month after implementing its document destruction plan on Shred Day 1998, and at Steinberg's urging, Rambus decided to slow down its schedule for implementing the Licensing and Litigation Strategy. The purpose of this deceleration was to allow Rambus to secure a strategic and tactical advantage over the targeted DRAM manufacturers. By slowing the pace of its Licensing and Litigation Strategy, which Rambus knew and expected to lead to litigation with DRAM manufacturers, Rambus sought to strengthen its patent portfolio through the filing of continuation cases and to allow sufficient time for DRAM manufacturers to fully commit to RDRAM technology. The slow down also allowed Rambus to take advantage of the information it had obtained from JEDEC's efforts to set an industry standard on SDRAM.

In November 1998, with Johnson's assistance, Karp prepared a memorandum that outlined Rambus' patent enforcement plans for 1999. In that memorandum, Rambus took the view that it possessed three enforceable patents: the '327 patent covering DDR (and dual edged clocking), the '481 patent covering DDR (with PLL circuitry), and the '580 patent covering the DDR and PC100 (access time register). These patents were issued respectively in 1996, 1997, and November 1998. At that point, Rambus had already prepared claim charts relative to Micron and Fujitsu devices under the '580 patent. This sort of activity, which must have been accomplished before the date of the correspondence, is typically done in connection with the anticipation of litigation.

The patent enforcement strategy for 1999 (prepared in November 1998) provided for a short period for negotiating a license and settlement, to be immediately followed by the initiation of litigation. So, even if one were inclined to disregard the clear evidence of anticipated litigation before September 1998 (which the Court is not inclined to do), the record clearly shows that, by November 1998, the enforcement scenario for 1999 anticipated litigation with DRAM manufacturers. Indeed, Rambus followed this strategy to the letter with both Hitachi and Infineon in late 1999 and 2000. In that regard, it is important to note that Karp, one of the authors of the 1999 enforcement strategy, said at the time the strategy was articulated in 1998 that litigation (not settlement) was the likely consequence of the company's strategy.

In April 1999, Neil Steinberg was hired on as in-house patent counsel. At a deposition given in Rambus v. Infineon, he was asked what the scope of his duties and responsibilities were when he came on board. His response was:

I was handling prosecution matters, patent prosecution matters. I was handling licensing matters. I was handling — we were preparing for litigation.

DTX 9007.²⁶ That, of course, was what he had been doing since mid-1998 when he was retained as outside counsel.

26 Steinberg later testified to the contrary when, after the piercing orders, he realized it was necessary to move the anticipation of litigation to a much later time. In so doing, he testified falsely.

As far back as February 1998, Rambus had been advised or had concluded that it needed to clean out its patent prosecution files, whether in-house or outside, to assure that they conformed with the official files. To that end, in an April 5, 1999 meeting with Lester Vincent, Rambus' outside counsel, Karp instructed Vincent to comply with that policy and to achieve that result. Karp met with Vincent again on April 19, 1999, and Vincent reported that he had accomplished approximately one-fourth of the task. Karp pressed Vincent to increase the pace and to finish the job. Consequently, a secretary was assigned full time to the purging project. Clearly then, the Rambus document destruction program was back in full swing by April 1999.

Karp and Steinberg headed the intellectual property department at Rambus. Their goal for the third quarter of 1999 was to have a litigation strategy prepared against one of the three manufacturers as to whom licensing negotiations would be initiated, and to be ready for litigation within 30 days notice. In other words, the deceleration of the litigation strategy called for in October 1998 was over by June 1999. By October 14, 1999, Rambus had prepared a detailed litigation matrix to help select appropriate targets. It then launched its program against Hitachi in October 1999, with whom it filed suit in January 2000.

Thus, although in the fall of 1998, Steinberg slowed the progress toward litigation, the mobilization for litigation was back in full operation by April 1999 and certainly no later than June 1999. The documentary record for 1999 confirms that, although the litigation anticipated by Rambus as of 1998 had to be delayed, the company nonetheless anticipated that litigation would occur against specifically identified DRAM manufacturers in specifically identified fora by the end of 1999 or early 2000. It is of little moment, in assessing whether and when Rambus anticipated litigation, that the plans for litigation and the anticipation of litigation demonstrated by the documents as early as 1998 were slowed down by the deliberate conduct of Rambus in order to allow its identified litigation targets to become so entrenched in their positions that, for economic and business reasons, the targets might be more inclined to settle than to litigate.

Thus, the record clearly and convincingly establishes that, by the time it implemented its document destruction policy on Shred Day in September 1998, Rambus held the view that it would soon be in litigation with manufacturers of RDRAMS, SDRAMs and DDR-DRAMs over patents which Rambus thought covered the products being manufactured by those manufacturers. This is evident from the official business records of Rambus, which reveal that by early 1998 Rambus had developed a Licensing and Litigation Strategy. The licensing component of the strategy called for such high royalties that the company expected to be in litigation soon after seeking those royalties. Indeed, the company expected that litigation would be necessary to establish both its intellectual property rights and its royalty rates. Rambus' official business records demonstrate that the document retention program was an integral component of this litigation strategy.

By the time Rambus implemented its document destruction policy on Shred Day in September 1998, it had identified the most likely and attractive litigation targets, and had settled on a number of possible legal theories to press against specific targets, depending upon whether the target was already a licensee. Rambus had also begun its reverse engineering efforts and planned to develop claim charts shortly thereafter. Additionally, Rambus had selected three fora it thought would be most advantageous to it. Among the factors that Rambus considered in selecting these fora was the average time to trial, as Rambus believed it could gain a tactical advantage over DRAM manufacturer targets by developing an electronic document database in advance. All of these efforts on the part of Rambus were components of its overarching Licensing and Litigation Strategy, as manifested by the contemporaneous business records.²⁷

27 Karp testified that the document destruction program was motivated by fear of subpoenas from third-parties which might need Rambus documents in cases to which Rambus was not a party. There is no evidence to support the existence of such an apprehension and Karp's testimony on the subject is regarded as incredible.

By mid-1999, Rambus once again pressed forward with the implementation of its Licensing and Litigation Strategy. The company's business documents show that the destruction of documents on Shred Day in August 1999 was an integral part of that strategy. Just as it had in 1998, Rambus continued to anticipate that licensing would not succeed because of the high royalty rates that would be proposed, and thus Rambus anticipated litigation throughout 1999. In fact, the company used 1999 to continue refining its approach to selecting litigation targets and the most advantageous fora in which to conduct litigation against those targets.

Finally, it is well to remember that, in December 2000, while discovery in Rambus v. Infineon was underway, another mass document purge took place at Rambus. That destruction took place while litigation that had been anticipated as of 1998 (DTX 3691) was actually in progress.

The record proves that Rambus engaged in pervasive document destruction in 1998 and 1999 while it anticipated litigation, or reasonably should have anticipated litigation, and in 2000 while it was actually engaged in litigation.

It is also appropriate to assess the record respecting when Rambus anticipated litigation with Samsung. Rambus executed a patent license agreement with Samsung in October 2000. Under that agreement, Samsung was permitted to use certain Rambus technology, including that which is the subject of the two patents-in-suit in Rambus' counterclaims, in the manufacture of DRAM and other products. Under the original Samsung/Rambus license executed in October 2000, Samsung was obligated to pay a running royalty that was a percentage of sales revenue for SDRAM, DDR-DRAM and controller products made by Samsung. In June 2005, after Samsung and Rambus failed to reach an agreement on a renewal of the license or a Standstill Agreement, Rambus filed suit in the Northern District of California. Samsung then filed an action for declaratory judgment on the following day in this Court.

While Rambus might plausibly argue that it did not anticipate litigation with Samsung from October 2000 until some point in the months before June 2005, Rambus cannot muster a colorable argument that it did not anticipate litigation with Samsung before the execution of the license agreement. Samsung was one of the DRAM manufacturers targeted by Rambus' matrices. By late 1999, Samsung was considered by Rambus to be the second most attractive litigation target. So while Rambus arguably did not anticipate litigation with Samsung when it destroyed documents in December 2000, it most certainly did anticipate litigation with Samsung on both Shred Day 1998 and Shred Day 1999. To conclude otherwise is to ignore the logical, common sense inferences to be drawn from the specific text of Rambus' business records, which simply pile one on top of the other, to lead to that conclusion.

(b) Were Relevant Documents Destroyed?

The factual record on this point begins with the finding made at the conclusion of the first trial in Rambus v. Infineon. At that time, the Court held that Rambus had destroyed documents relevant to litigation. Rambus, Inc. v. Infineon Techs. AG, 155 F. Supp. 2d at 682. That finding, of course, was made on a record that is not as complete as the one which was developed subsequently in Rambus v. Infineon and in Hynix v. Rambus. However, the subsequent record confirms, clearly and convincingly, this facet of the spoliation test.

Additionally, the record clearly establishes that Rambus' document retention policy actually targeted discoverable documents, including email messages, files on individual computers, network servers or floppy disks, corporate databases, backup tapes, system records and logs, and computers and disks. DTX 3686. This reference was included in the presentation made by Johnson to the Rambus managers. Likewise, Karp's presentation focused on information that was discoverable in litigation, emails in particular.

A number of Rambus' witnesses testified that the reason for adopting the document retention policy was to get rid of discoverable documents. Of particular concern were emails which were described as potentially quite harmful in litigation. Accordingly, email backup tapes were eliminated and employees were told to purge their own individual email files unless it was necessary to keep an email for some purpose and then it should be saved to a particular file or reduced to hard copy. DTX 5185; DTX 9016.²⁸

28 In the Hynix litigation, Rambus discovered some of the back up tapes and some of them were restored to readability. However, the backup tapes did not include files on UNIX and Macintosh workstations during the period from 1998 to 2000. 9 Trial Tr. 1574:9-22.

Discoverability is a concept that exists only in the context of litigation. To target, and then to destroy, documents that are thought to be discoverable is to identify documents that would be relevant or would lead to the discovery of admissible evidence in litigation. For Rambus, that meant documents that were relevant in patent litigation because that is where Rambus expected litigation to occur. There can be no doubt that Rambus destroyed discoverable documents pursuant to its document retention policy.

The document retention policy also provided for the destruction of documents, drafts and materials used during negotiations (including notes of meetings and telephone conversations, presentation slides, memoranda, and checklists). Steinberg reemphasized the need to destroy contract-related documents as late as July 17, 2000. Those kinds of documents have particular relevance in the damages aspect of patent litigation. They can also lead to the discovery of admissible evidence.

Rambus instructed its outside patent counsel, Lester Vincent, to purge the patent prosecution files so as to make them conform to the file wrapper and thereby to eliminate much information that typically is useful in patent litigation in addressing validity and infringement issues as well as the conduct of the applicant before the PTO. The prosecution files that were moved in-house when Steinberg joined Rambus were likewise cleansed and with the same results. Documents of the sort destroyed by Vincent quite clearly are relevant to patent litigation. At a minimum, they are reasonably calculated to lead to the discovery of admissible evidence.

Rambus also destroyed email archives and other electronic files. From 1990 until shortly before Karp arrived on the scene in October 1997, the company had used Macintosh computers. By the time Karp arrived, most of the Macintosh computers had been eliminated. However, there were Macintosh backup tapes that had been saved from the early days forward (e.g. 1990 to 1997). Many of the Macintosh backup tapes were destroyed in the implementation of the document retention policy. By virtue of the document retention policy, the backup tapes for emails were destroyed every three months from its inception in September 1998 forward. That some of those tapes were stumbled upon during the Hynix v. Rambus litigation does not negate that others were destroyed.

The record also establishes that prior art documents were discarded by Rambus. Steinberg admitted that, in 1999, he threw away prior art documents on which he had made substantive notations.

Of course, Rambus' litigation adversaries cannot, and cannot be expected to, demonstrate with certainty the content of destroyed documents. On this record, and under the applicable law, that responsibility is on the shoulders of Rambus. For example, inAnderson v. Cryovac, Inc., 862 F.2d 910 (1st Cir. 1998), the court explained that:

[i]n the case of intentional misconduct, as where concealment was knowing and purposeful, it seems fair to presume that the suppressed evidence would have damaged the non-disclosing party. (citations omitted) It seems equally logical that where discovery material is deliberately suppressed, its absence can be presumed to have inhibited the unearthing of further admissible evidence adverse to the withholder, that is, to have substantially interfered with the aggrieved party's trial preparation.

Id. at 925 (emphasis added). The court went on to hold that "[t]he presumption, if it arises, should be a rebuttable one. It may be refuted by clear and convincing evidence demonstrating that the withheld material was in fact inconsequential." Id.

The court explained the need for this stringent standard with the following observation:

We are keenly aware of the stringency of the standard, yet we believe it to be an appropriate antidote for deliberate misconduct. A party who is guilty of, say, intentionally shredding documents in order to stymie the opposition, should not easily be able to excuse the misconduct by claiming that the vanished documents were of minimal import. Without the imposition of a heavy burden such as the `clear and convincing' standard, spoliators would almost certainly benefit by having destroyed the documents, since the opposing party could probably muster little evidence concerning the value of papers it never saw. As between guilty and innocent parties, the difficulties created by the absence of evidence should fall squarely upon the former.

Id. at 925 (emphasis added).

Further, from a description of many of the destroyed documents, it is reasonable to infer that many were relevant to the patent litigation planned by Rambus as part of its litigation strategy.Telectron, Inc. v. Overhead Door Corp., 116 F.R.D. 107, 133 (S.D. Fla. 1987). And, as the court held in Telectron:

while it is now impossible to determine precisely what or how many documents were destroyed, the bad-faith destruction of a relevant document, by itself, `gives rise to a strong inference that production of the document would have been unfavorable to the party responsible for its destruction.' Coates v. Johnson Johnson, 756 F.2d 524, 551 (7th Cir. 1985) (citations omitted); see also Nation-Wide Check Corp. v. Forest Hills Distributors, 692 F.2d 214, 217 (1st Cir. 1982); National Association of Radiation Survivors v. Turnage, 115 F.R.D. 543 (N.D. Calif. 1987); Wm. T. Thompson Co. v. General Nutrition Corp., 593 F. Supp. 1443, 1455 (C.D. Calif. 1984).

Id.

The record shows deliberate destruction of documents designed by Rambus to rid itself of documents discoverable in litigation, and documents of the type that likely would be relevant in litigation. Rambus has failed to establish that the destroyed documents were of minimal or little import.

Rambus has not tried to prove what kind of documents were among the 1,054 boxes (2.7 million pages) it destroyed or the computer tapes it degaussed. Rather, it argues that it produced many relevant documents; and, it did. But, where, as here, the destruction was part of a litigation strategy, the volume of documents destroyed was so great, the destruction targeted documents that were discoverable in litigation, and the destruction is shown to have reached so many types of likely relevant documents, that alone is inadequate to discharge Rambus' burden.

Rambus argues that the document destruction program was designed to save documents relevant to litigation and that the "litigation holds" had that result. Each of these points is addressed in turn.

The first argument has its genesis in Karp's presentation to the Rambus staff in August 1998. It is true that Karp's presentation slides had a notation at the foot of each page, in bold letters, articulating "LOOK FOR THINGS TO KEEP." DTX 4024. According to Johnson, that instruction is confusing and frustrates the purpose of a legitimate document retention policy. Rambus points to this language as evincing an intent not to destroy documents relevant to litigation.²⁹ In any event, wholly apart from the fact that the instruction does not even mention litigation, the general precatory language does not identify what documents would be relevant in litigation, and thus does not support Rambus' contention. Unless employees are instructed as to what to save, the general "look for things to keep" means nothing.³⁰

29 That argument, if accepted, supports the view that the document retention policy was adopted when Rambus was anticipating litigation; otherwise, there would be no need to mention retaining documents relevant to litigation.

30 Rambus' own management characterized the policy as the "`scorched earth' theory of document retention." HTX 330. In dealings with Intel, Rambus itself was unable to locate many important documents that had been swept up in the wake of the document retention policy and destroyed.

The second argument is directed to take advantage of the principle that a document retention program must be suspended when litigation is, or reasonably should be, anticipated, so as to remove from the reach of destruction documents relevant to the anticipated litigation. Rambus asserts that, at two different times, it complied with that requirement. First, Rambus relies on the testimony of Neil Steinberg that, when litigation was anticipated in late 1999, he instructed a number of people from Rambus to retain documents "relevant to litigation." Having found that Steinberg has testified falsely on important matters, the Court will not credit his testimony on that point. Furthermore, Steinberg's testimony is substantially undercut by the evidence that shows that, in July 2000, he actually instructed Rambus executives to destroy contract drafts and documents related to the license negotiations. This, according to Steinberg, was accomplished pursuant to the company's "Document Destruction Policy." Taking into account that Steinberg is not a credible witness and that he actually acted contrary to the instructions he purportedly gave, the Court finds Steinberg's testimony on the point to be unbelievable and further finds that Steinberg gave no instructions to anyone at Rambus to retain documents "relevant to litigation."

Rambus also has offered testimony from a lawyer at Gray Cary that when he, and others in his firm, interviewed Rambus witnesses (approximately 40 in number), they usually gave a standard admonition not to destroy documents "relevant to litigation." And, a few Rambus witnesses said that they recalled instructions to that effect. Those instructions were said to have been given in late 1999 or early 2000, depending upon when the interviews occurred. There is evidence that Cecilia Gonzalez, of Howrey Simon, gave similar instructions to Steinberg.

Assuming that those instructions were given to some employees, Rambus has offered no proof to show that the recipients were informed what documents were relevant to the litigation. That is an especially significant failure of proof when one considers that the document retention policy was widely broadcast within the company as necessary to get rid of documents that were discoverable in litigation. Further, in the runup to the extensive destruction of documents in December 2000 (575 boxes, amounting to almost 1.5 million pages), the record discloses that no specific instructions respecting retaining documents relevant to litigation were issued. This is confirmed by Melinda Kauffman, DTX 9017, who said that specific instructions related to pending litigation were not given and instead employees were given a general reminder about the document retention policy. Also, Rambus' Rule 30(b)(6) witness, Robert Kramer, Director of Litigation at Rambus, testified that he was unaware of any instructions other than to follow the document retention policy.

Moreover, there is evidence that Rambus employees and agents did not follow the instructions. For example, Karp, the author of the policy, destroyed files in June and July 2000. Steinberg instructed employees to destroy relevant documents in July 2000. Vincent (who also had been interviewed by Gray, Cary) destroyed relevant files after June 23, 2000. And, of course, a very large quantity of documents were destroyed in December 2000 while the action in Rambus v. Infineon was underway.

Additionally, it is important to note that on Shred Day 1998 Rambus destroyed 291 boxes, on Shred Day 1999 the company destroyed 188 boxes, and in December 2000, it destroyed 575 boxes, a total of 1,054 boxes. Assuming approximately 2,600 pages of documents per box (which is the average number of pages in the boxes of documents that Rambus did produce), it appears that from September 1998 through December 2000, Rambus destroyed 2.7 million pages of documents.³¹ Of course, the volume of destroyed documents alone does not establish their relevance to the litigation. But, in view of the other evidence, it is probative of that point.

31 By contrast Rambus produced 138,000 pages in 59 boxes before the first trial in Rambus v. Infineon and, on remand, it produced 105,000 pages of documents in 38 boxes.

Considering the lack of specificity in defining what documents would be relevant to litigation, considering that the document retention policy focused on destruction of documents discoverable in litigation, considering that Rambus kept no records of the kinds of documents that were destroyed, considering the volume of documents destroyed, and considering the extent and kind of evidence destroyed after the vague "hold" instruction was given, the naked instruction not to destroy relevant documents is not sufficient to overcome the presumption. Rambus has not satisfied its burden.

5. Additional Conclusions of Law: Spoliation

Rambus defends the spoliation charge by arguing that its document retention policy is like many others, was adopted largely in the form recommended by counsel, and that, if its program constitutes spoliation, most, if not all, of corporate America is in jeopardy. For several reasons, Rambus' argument misses the mark.

First, the allegation of spoliation is not directed to the constituent elements of Rambus' policy. Nor is it asserted that the adoption and implementation of document retention policies are per se acts of spoliation. Indeed, many corporations have such policies. There is no need to canvass and recite the case law finding such policies to be generally permissible. And, since there has been no per se attack on the Rambus policy, there is no need to survey the policy to assess how it measures to standard.

Second, there is no dispute that counsel were involved in advising as to the content of the policy and in helping to present the policy to employees. That, however, does not change the fact that Rambus implemented the policy when it anticipated, or reasonably should have anticipated, litigation. Thus, the involvement of Johnson and Steinberg, as shown by the record, does not militate against a finding of spoliation.

Contrary to Rambus' third argument, neither corporations nor individuals are at risk of a finding of spoliation merely because they adopt or implement a proper document retention policy. However, any company that implements a document retention policy during or in anticipation of litigation, and destroys documents relevant to the actual or anticipated litigation, will face and lose a spoliation charge. But, that is as it should be.

In any event, the law recognizes that document retention policies and actual or anticipated litigation can coexist. For example, a company can modify its policy to preserve documents reasonably thought relevant to the actual or anticipated litigation. To accomplish that, however, the company must inform its officers and employees of the actual or anticipated litigation, and identify for them the kinds of documents that are thought to be relevant to it.³² Other mechanisms, such as collecting the relevant documents and segregating them, may accomplish the same result. These points are by way of illustration and are not exclusive.

32 If management does not desire to disclose to employees the fact of anticipated litigation, it can tell employees to preserve documents meeting certain specific descriptions, and/or then collect and save the specific documents.

It is not sufficient, however, for a company merely to tell employees to "save relevant documents," without defining what documents are relevant. If the testimony from Gray Cary is accepted, that is all that was done in late 1999 or in 2000. As explained above, this sort of token effort will hardly ever suffice, and it will always fall short when the document retention program had an articulated principal focus of ridding the company of documents that are discoverable in litigation.

Nor can a company make a document retention program an integral part of its litigation strategy and, pursuant thereto, target for destruction documents that are discoverable in litigation. Rambus has cited no decision that supports such an approach. The Court has found none. And, if the result of finding spoliation on this record is to deter others from such conduct under like circumstances, then that is desirable.

Generally, whether spoliation has occurred is measured, asSilvestri teaches, by looking at the totality of the circumstances. A legitimately adopted and implemented policy will pass muster. The conduct of Rambus does not for the reasons outlined above. Fortunately, that conduct appears to be sui generis. None like it is reported in the periodicals, the trade press, or the decisional law. That further underscores the point that corporate America has little to fear from a decision holding Rambus accountable for spoliation of evidence.

Lastly, Rambus points to the decision in Hynix Semiconductor, Inc. v. Rambus Inc., 2006 WL 565893 (N.D. Cal. 2006), wherein it was held, as part of a decision on the issue of unclean hands, that Rambus did not engage in the spoliation of evidence. The record of that proceeding and the resulting decision have been studied carefully. For the reasons that follow, the decision is not persuasive.

First, it appears that the decision gives considerable credence to the slant put on the Rambus business records by both Karp and Johnson. As explained above, the testimony of Karp and Johnson³³ is pointedly at odds with the company's contemporaneous business records and the law firm's billing records. In fact, the contemporaneous business records convincingly disprove the version of events sponsored in the testimony of Karp and Johnson. Also, as explained previously, Karp's testimony is at odds with the text of documents that he authored in which he spelled out his own views on the likelihood of litigation. And, the testimony of Karp and Johnson is at odds with the testimony of Johnson's partner, Diane Savage, about why Karp contacted Cooley Godward and why Johnson was assigned to respond to Karp's request for litigation assistance. Furthermore, for reasons set forth in the findings of fact, the testimony of Karp and Johnson is not credible on the issue of anticipation of litigation, and their efforts to show that Rambus did not anticipate litigation in 1998 and 1999 are simply unconvincing in the face of such a strong documentary record to the contrary.

33 Karp and Johnson both gave essentially the same testimony in Rambus v. Infineon and Hynix v. Rambus.

Second, the Hynix decision appears to be significantly influenced by the view that "[t]he evidence here does not support the conclusion that Rambus intentionally designed its Document Retention Policy to get rid of particular damaging documents."Hynix, 2006 WL 565893, *25. The record, it is respectfully submitted, shows quite clearly that Rambus acted intentionally to rid its files of discoverable documents because of the damage that such documents could do in litigation. Rambus may not have identified specific documents as damaging and then destroyed those specifically identified documents, but Rambus engaged in equally pernicious conduct by designating for destruction nearly all discoverable documents at a time when it anticipated litigation.

Documents are not discoverable unless they either contain evidence that supports a claim or defense or they are reasonably calculated to lead to the discovery of admissible evidence. Fed.R.Civ.P. 26(a) and (b). The decision to use the description "discoverable" to mark documents for destruction was not made by accident. It was a description chosen by Johnson, a litigation specialist, and Karp, who talked extensively with Johnson and who by virtue of previous employment was intimately familiar with litigation in the patent arena.

Third, and relatedly, the Hynix decision places significance in the conclusion that Rambus' document retention plan was "content neutral." Hynix, 2006 WL 565893, *24. With respect, that conclusion does not square with the evidence that the program was implemented principally to rid the company of discoverable documents at a time when it anticipated litigation with DRAM manufacturers.

Fourth, the decision in Hynix appears to have been influenced by a peculiarity of California work product law. According to the Hynix decision, the undisputed fact that documents dated in 1998 and 1999 were claimed inRambus v. Infineon to be protected by the work product doctrine is irrelevant, despite the fact that work product protection is available only when the document was produced in anticipation of litigation. According to the Hynix decision, that fact "does not support the conclusion that litigation was anticipated because the [privilege] log was prepared by California lawyers," and "California law differs from federal law in that it protects a lawyer's work product prepared `in a nonligitation capacity.'" Id.

Assuming this to be a proper statement of the California work product doctrine, the California work product doctrine is entirely irrelevant. Federal law governs claims of work product protection and attorney-client privilege in federal cases. See Fed.R.Civ.P. 26(b)(3); Fed.R.Evid. 501. If there were any doubt on that point, it would be dispelled by the fact that Rambus' patent infringement claims are federal question claims. Rule 26(b)(3) could hardly be more clear in requiring the anticipation of litigation with respect to the work product doctrine:

a party may obtain discovery of documents and tangible things otherwise discoverable under subdivision (b)(1) of this rule and prepared in anticipation of litigation or for trial by or for another party or by or for that other party's representative (including the other party's attorney, consultant, surety, indemnitor, insurer, or agent) only upon a showing that the party seeking discovery has substantial need of the materials in the preparation of the party's case and that the party is unable without undue hardship to obtain the substantial equivalent of the materials by other means (emphasis added).

See also In re Echostar Communications Corp., 448 F.3d 1294, 1301 (Fed. Cir. 2006) (applying Rule 26(b)(3) in patent infringement action).

Rambus has continually emphasized that it hired respected counsel, and it can only be assumed that even Rambus' California-based lawyers were aware of this most basic Federal Rule of Civil Procedure and the fact that they were litigating a federal question case in federal court.

The Court has not relied here on the fact that Rambus labeled a number of the relevant business documents as work product in determining that Rambus anticipated litigation with DRAM manufacturers when it destroyed relevant documents, but it is nonetheless a permissible inference.³⁴ However, the rejection of the argument in the Hynix decision is another reason that the decision is not persuasive.

34 Furthermore, even the cited California case afforded work product protection only to the mental processes of lawyers in a nonlitigation capacity. Many of the 1998-1999 documents for which the protection was claimed were those authored only by Karp who is not a lawyer. Of course, if there was anticipated litigation and Karp was working with a lawyer as to it, Karp's work can get protection. And, in Rambus v. Infineon, Rambus sought that protection for Karp's work and it was given. See Rambus Inc. v. Infineon, 222 F.R.D. 280 (E.D. Va. 2004), the sealed version of which is Docket No. 715 in Rambus v. Infineon (a copy of which shall be placed under seal in this record).

Fifth, the decision in Hynix sought to distinguish the controlling decision on spoliation in this circuit, Silverstri v. General Motors Corp., by observing that the spoliating plaintiff in Silvestri knew he was going to make a claim, had hired experts, and had retained counsel. Hynix, 2006 WL 565893, *21. These indicia were sufficient to show anticipation of the litigation that was filed some three years later. By contrast, said the Hynix decision, in 1998 and 1999, "the path to Rambus planned litigation was neither clear nor immediate." Id. at 22.

This distinction is more ephemeral than real. To begin, as explained above, Rambus well knew by 1998 and in 1999 that it was going to engage in patent litigation against specific DRAM manufacturers in specific fora. And, Rambus had set a timetable for reverse engineering as set forth in many of its quarterly goal setting documents. Thus, the real distinction is that Rambus had not, in 1998 or until late 1999, hired counsel to prosecute the litigation it planned to bring against specific DRAM manufacturers. Relatedly, theHynix decision fixed the point at which "litigation became probable" in late 1999 when Rambus undertook to hire counsel to file an action against Hitachi. Id. at 24.³⁵ At least in this circuit, the point at which litigation becomes probable does not necessarily correspond with when a party anticipated, or reasonably should have anticipated, litigation. Nor did the Second Circuit adopt such a rule in Kronisch, the case cited as controlling in the Hynix decision.

35 The decision acknowledges that "hiring of litigation counsel or actually filing suit is certainly not necessary to demonstrate that a company contemplates litigation." Id.

In the Hynix decision, the court suggests that litigation was not "probable" because a number of contingencies had yet to occur:

(1) the direct RDRAM ramp had to be sufficiently developed so as not to jeopardize RDRAM production; (2) Rambus's patents covering non-RDRAM technology had to issue; (3) product samples from potentially infringing DRAM manufacturers had to be available in the market; (4) the non-compatible products had to be reverse engineered and claim charts made showing coverage of the actual products; (5) Rambus's board had to approve commencement of negotiations with a DRAM manufacturer; and (6) the targeted DRAM manufacturer had to reject Rambus's licensing terms.

Id. at *22. To suggest that all of these contingencies had to occur before Rambus can be deemed to have anticipated litigation is to ignore the reality that Rambus actually expected and anticipated that all of these contingencies would come to pass and lead Rambus into litigation with DRAM manufacturers.³⁶ Rambus developed its document retention policy expressly for the purpose of preparing for the coming litigation. As explained above, it is not unusual that the anticipation of litigation is quite concrete, notwithstanding that the events giving rise to the litigation have not yet occurred. Adlman, 68 F.3d at 1501. That is exactly the situation presented here.

36 Interestingly, it does not appear that all of these contingencies had come to pass at the point in September 1999 when the Hynix court found that litigation became probable.See id. at 23.

Sixth, the Hynix decision seems to be influenced significantly by evidence that Johnson informed Rambus that it would cost $1.5-3.0 million to prepare for litigation, but Rambus never budgeted such an amount specifically for litigation. The record is not well-developed respecting Rambus' budget process but it does show that the budget about which Karp was testifying extended from the fall of 1999 through the second quarter of 2000. Moreover, the record, taken as a whole, shows that Rambus reasonably expected litigation at the time of the budget presentation to which Karp referred. And, thus, the fact that there was not a line item for litigation in that specific presentation is of little consequence in assessing anticipation of litigation.

More importantly, one does not have to budget for litigation to anticipate it. As explained fully above, Rambus clearly anticipated litigation from early 1998 forward. Quite clearly, in June 1999, Steinberg and Karp set goals for the third quarter of 1999, one of which was to be ready for litigation on 30 days notice. And, Rambus was at the time engaged in reverse engineering and preparing claim charts. Furthermore, one of Steinberg's goals was to start the damage clock running not later than the fourth quarter of 1999. Of course, the concept of damages is intimately linked with litigation. To anticipate starting the damages clock is to anticipate litigation.

For those reasons and in perspective of the rest of the record, the absence of a budget figure for litigation in the presentation to which Karp referred, does not support the conclusion that Rambus did not anticipate litigation in 1998 or in June 1999, the date of the budget presentation about which Karp testified.

In a post-hearing motion, Hynix sought reconsideration of the unclean hands decision in Hynix on the ground that the court had adopted a "reasonably probable" standard when it cited with approval the American Bar Association's Civil Discovery Standards. The post-hearing motion of Hynix was denied when the court explained that "reasonably probable" equated with "reasonably foreseeable."³⁷ And, in the Hynix decision, the court held that the word "probable" in the ABA Standard means "that litigation is more than a possibility." Id. at *21.

37 Hynix Semiconductor, Inc. v. Rambus Inc., No. C-00-20905RMW, Order On Motion For New Trial Or Permission To Appeal.

Thus, the Hynix decision fixed the point at which litigation became more than a possibility as late 1999 when Rambus undertook to retain counsel to prosecute its case against Hitachi. With respect, that does not square with the record created by a great number of Rambus' contemporaneous business records and the testimony given by Steinberg to the effect that in 1998 and mid-1999 his job was, in part, to prepare for litigation. Those records and the testimony are outlined fully above and will not be repeated here. As explained above, the record, taken as a whole, rather clearly shows that litigation "became more than a possibility" in 1998. In fact, patent litigation was by then a foregone conclusion. As seen by Karp, Rambus' Vice President of Intellectual Property, litigation was the preferred, if not essential, course to setting a reasonable royalty for Rambus' patent licenses.

In concluding otherwise, the Hynix decision also relied significantly on the testimony of Geoff Tate. See id. at *23. Then Rambus' Chief Executive Officer, Mr. Tate has been determined to have given false sworn testimony, and hence not to be a credible witness. Rambus Inc. v. Infineon Techs. AG, 155 F. Supp. 2d 661, 681-82 (E.D. Va. 2001).

For the foregoing reasons, the decision in Hynix does not aid Rambus.

And, for the reasons set forth above, the record shows clearly and convincingly that in 1998, 1999 and 2000 Rambus engaged in the spoliation of evidence.

5. Exceptional Case

Here, Rambus planned for litigation beginning in 1998 and throughout 1999. As part of that plan, it identified and selected litigation targets — the DRAM industry generally and Samsung specifically. It identified the most desirable fora for litigation as part of the plan. And, as part of the plan, Rambus established and implemented a pervasive document destruction program. Rambus' destruction plan called for the destruction of discoverable documents. Thus, its program cannot be characterized as content neutral or innocent. Rambus' licensing and litigation strategy has generated a substantial amount of litigation and a large part of that litigation has been devoted to uncovering, and dealing with, the facts surrounding, and the consequences of, a most far-reaching document destruction plan.

For the foregoing reasons, the Court finds that Rambus' spoliation of evidence relevant to the '263 and '918 patents, while in anticipation of litigation with DRAM manufacturers such as Samsung, renders this case exceptional.³⁸

38 The key determination on this point is whether Rambus' litigation misconduct makes this case exceptional. Whether Rambus' spoliation would also have supported a finding of unclean hands that would bar enforcement of the patents is irrelevant.

III. VEXATIOUS OR UNJUSTIFIED LITIGATION

The second ground asserted by Samsung for a finding of exceptional case essentially amounts to a claim of vexatious or unjustified litigation. Samsung argues that Rambus asserted the counterclaims in bad faith, given that the Court already had found the relevant patents unenforceable in Rambus v. Infineon by virtue of the unclean hands finding. Samsung charges that the counterclaims were asserted merely to bolster the motion to transfer. Further, Samsung asserts that Rambus' tendering of the covenants not to sue, rather than confronting the motion for partial summary judgment, is proof that Rambus never had any intention to litigate the counterclaims in this Court. This, of course, is the paradigm case to which the Brooks Furniture test applies. In order to ground an exceptional case finding on this charge of vexatious litigation, the Court must find that Rambus brought its infringement counterclaims in subjective bad faith and that they were objectively baseless.

It is settled that, by June 2005, Rambus' management had decided that patent litigation with Samsung must not be allowed to occur in the Eastern District of Virginia. It is certainly logical to infer that the decision issued from the bench inRambus v. Infineon was a motivating factor behind that decision. While Rambus had initially preferred to litigate in multiple districts, it is plausible that Rambus had come to regret that approach and, by June 2005, preferred that all of its patent litigation should take place, if possible, in its home forum, the Northern District of California. That is especially so considering that it was involved in other litigation there.

When Rambus was confronted with the action filed in this district by Samsung, it had to decide, inter alia, whether to assert counterclaims. Rambus explains its decision to file counterclaims as the product of unsettled law with respect to whether its infringement claims on the patents-in-suit were compulsory counterclaims. For that reason, says Rambus, it asserted the counterclaims on the '263 and '918 patents out of an abundance of caution in order to preserve its infringement claims. To the contrary, however, the Federal Rules of Civil Procedure settled that issue decades ago. The notion that Rambus' counterclaims were compelled is without merit.

The general rule with respect to compulsory counterclaims is that

A pleading shall state as a counterclaim any claim which at the time of serving the pleading the pleader has against any opposing party, if it arises out of the transaction or occurrence that is the subject matter of the opposing party's claim and does not require for its adjudication the presence of third parties of whom the court cannot acquire jurisdiction.

Fed.R.Civ.P. 13(a). See also St. Paul Fire and Marine Ins. Co. v. Seafare Corp., 831 F.2d 57, 58 (4th Cir. 1987). However, one of the specifically enumerated exceptions to the compulsory counterclaim rule is that a party need not assert a counterclaim if "at the time the action was commenced the claim was the subject of another pending action." Id. Rambus filed patent infringement claims against Samsung in the Northern District of California on June 6, 2005, which included the '263 and '918 patents. Samsung filed its action for declaratory judgment in this Court on June 7, 2005. While Rambus' infringement claims with respect to the '263 and '918 patents had not been the subject of a pending action for very long when Samsung filed its action for declaratory judgment, the action was nevertheless pending and the infringement claims as to the '263 and the '918 patents were among the subjects of that action. Given the settled law on this point, the Court cannot accept Rambus' reliance on that theory to explain its motivations for asserting the counterclaims.

Absent any other explanation of its decision to assert patent infringement counterclaims, and given the established record that Rambus sought to avoid litigation in the Eastern District of Virginia at almost any cost, it is a permissible inference that Rambus filed the counterclaims in order to bolster its motion for transfer of venue. Indeed, the counterclaims and the motion to transfer venue were both filed on July 12, 2005, and Rambus concedes that the two were filed in direct connection with one another. The inference is further justified by the fact that, after Rambus' motion to transfer venue was denied, Rambus voluntarily dismissed the counterclaims with prejudice and gave Samsung covenants not to sue. Clearly, preservation of its infringement claims on the '263 and '918 patents was not nearly so important to Rambus as circumventing litigation in the Eastern District of Virginia.

However, Rambus also asserts that it was warranted in asserting the counterclaims, notwithstanding the bench ruling that the patents-in-suit were unforceable against Infineon for unclean hands and spoliation. On that score, Rambus is correct. When Rambus filed its answer and counterclaims on July 12, 2005, the court in Hynix Semiconductor, Inc. v. Rambus, Inc., Case No. CV0020905RMW (N.D. Cal.) already had ruled on April 22, 2005 that this Court's bench ruling in Rambus v. Infineon could not be given collateral estoppel effect on the issue of unclean hands because the Court never issued findings of fact or conclusions of law as a result of the parties' settlement. Further, Rambus was actively litigating the issues of spoliation and unclean hands in the Northern District of California, and a trial date on those issues had been set. Moreover, the course of this litigation can hardly be characterized as vexatious in light of the fact that, shortly after Samsung filed this action, Rambus provided covenants not to sue on the '263 and '918 patents and voluntarily dismissed its counterclaims with prejudice.

For the foregoing reasons, it cannot be said that Rambus asserted its counterclaims in subjective bad faith, and thus Rambus' assertion of its counterclaims cannot support an exceptional case finding. That being the case, there is no need to assess whether the counterclaims were objectively baseless.See Brooks Furniture, 393 F.3d at 1381 ("Since we conclude that the first requirement (subjective bad faith) is not satisfied here, we need not decide whether the second (objectively baseless) standard was met."). IV. WHETHER RAMBUS' PRE-FILING SPOLIATION WARRANTS AN AWARD OF ATTORNEY'S FEES UNDER 35 U.S.C. § 285 OR THE COURT'S INHERENT POWER

After having found this to be an exceptional case based on Rambus' spoliation of evidence, the determination as to whether an award of attorney's fees is appropriate under § 285 is committed to the Court's discretion. Having presided over the case, "the trial judge can best weigh the relevant considerations, such as the closeness of the case, the tactics of counsel, the flagrant or good faith character of the parties' conduct, and any other factors contributing to imposition of punitive sanctions or to fair allocation of the burdens of litigation." Perricone v. Medicis Pharmaceutical Corp., 432 F.3d 1368, 1380-1381 (Fed. Cir. 2005). See also J.P. Stevens Co., Inc. v. Lex Tex Ltd., Inc., 822 F.2d 1047, 1051 (Fed. Cir. 1987). "The court's choice of discretionary ruling should be in furtherance of the policies of the laws that are being enforced, as informed by the court's familiarity with the matter in litigation and the interest of justice." S.C. Johnson Son, Inc. v. Carter-Wallace, Inc., 781 F.2d 198, 201 (Fed. Cir. 1986) (citing Yarway Corp. v. Eur-Control USA, Inc., 775 F.2d 268, 277 (Fed. Cir. 1985)). "[A]ttorney fees are not to be routinely assessed against a losing party in litigation[,] in order to avoid penalizing a party for merely defending or prosecuting a lawsuit, and are awarded to avoid a gross injustice." Knorr-Bremse Systeme Fuer Nutzfahrzeuge GmbH v. Dana Corp., 372 F. Supp. 2d 833, 851 (E.D. Va. 2005) (quotingRevlon, 803 F.2d at 679).

Whether to exercise the discretion to award attorney's fees begins with the recollection that Samsung, not Rambus, initiated this action. Not long thereafter, Rambus voluntarily dismissed its counterclaims with prejudice and signed covenants not to sue with respect to the patents-in-suit. Rambus terminated its claims at a sufficiently early stage in the litigation that it would not be grossly unjust for Samsung to bear the burden of its own litigation costs. And, considering that Rambus promptly issued covenants not to sue on the four patents-in-suit after Samsung filed this action and dismissed its counterclaims with prejudice early on, this case is unique in that an award of attorney's fees would not serve as deterrence to the litigation misconduct at issue.³⁹

39 Of course, that would not be the case if Rambus had persisted in the pursuit of its counterclaims or further defended Samsung's declaratory judgment action, thereby making it necessary to further litigate the spoliation of evidence.

Moreover, in exceptional cases, the aggrieved party is entitled to an award of "the portion of its attorney fees which related to the vexatious litigation strategy and other misconduct."Beckman Instruments, Inc. v. LKB Produkter AB, 892 F.2d 1547, 1553 (Fed. Cir. 1989) (emphasis added). "The determination of the amount of the award remains within the discretion of the trial court, since it is the trial judge who is in the best position to know how severely [the offending party's] misconduct has affected the litigation." Id. Thus, where litigation misconduct forms the basis for a finding that there is an exceptional case under § 285, the record must establish a causal nexus between the fees claimed and the misconduct.

The record here is insufficient to establish a causal nexus between the misconduct found (pre-filing spoliation of evidence) and the fees sought by Samsung. In that regard, it is necessary to remember that the spoliation of documents was the basis for Samsung's request for a declaration that Rambus' unclean hands barred enforcement of the patents-in-suit. That misconduct, although it was the basis for the requested declaration, did not cause Samsung to incur the fees for which it now seeks an award. Instead, the misconduct merely afforded Samsung a ground upon which to seek the declaration, a course on which it embarked voluntarily.

Nor was it necessary for Samsung to prove the spoliation as the basis for an unclean hands defense in opposition to Rambus' counterclaims because Rambus dismissed those with prejudice. Thus, Samsung cannot be heard to assert that the fees which it now seeks were causally connected to the spoliation as a predicate for an unclean hands defense to Rambus' counterclaims. For these reasons, Samsung has not established that the requested attorney's fees are related to the misconduct on which the Court has found the case exceptional, and thus an award of attorney's fees is not appropriate.

Samsung also seeks an award of attorney's fees under the Court's inherent power to sanction the prosecution of bad faith litigation and litigation misconduct. "When rules alone do not provide courts with sufficient authority to protect their integrity and prevent abuses of the judicial process, the inherent power fills the gap." Shepherd v. American Broadcasting Cos., Inc., 62 F.3d 1469, 1474 (D.C. Cir. 1995). "The inherent power encompasses the power to sanction attorney or party misconduct," and includes the power to assess attorney's fees "when a party has acted in bad faith, vexatiously, wantonly, or for oppressive reasons." Chambers v. NASCO, Inc., 501 U.S. 32, 45-46 (1991) (quoting Alyeska Pipeline Service Co. v. Wilderness Society, 421 U.S. 240, 259 (1975)) (omitting internal quotation marks). In Chambers, the Supreme Court held that

if a court finds that fraud has been practiced upon it, or that the very temple of justice has been defiled, it may assess attorney's fees against the responsible party, as it may when a party shows bad faith by delaying or disrupting the litigation or by hampering enforcement of a court order. The imposition of sanctions in this instance transcends a court's equitable power concerning relations between the parties and reaches a court's inherent power to police itself, thus serving the dual purpose of vindicating judicial authority without resort to the more drastic sanctions available for contempt of court and making the prevailing party whole for expenses caused by his opponent's obstinacy.

Id. at 46 (omitting internal citations and quotation marks). Of course, inherent powers must be exercised with restraint and discretion. See id. at 44.

With respect to when it is appropriate to invoke the inherent power, the Supreme Court has advised that the sanctioning scheme found in various statutes and rules has not displaced courts' inherent power. See id. at 46. However, "when there is bad-faith conduct in the course of litigation that could be adequately sanctioned under [a statute or] the Rules, the court ordinarily should rely on the [statute or] Rules rather than the inherent power." Id. at 50. However, "if in the informed discretion of the court, neither the statute nor the Rules are up to the task, the court may safely rely on its inherent power." Id. As the Federal Circuit has characterized the Supreme Court's admonition, "courts should only resort to further sanctions when misconduct remains unremedied by those initial tools," i.e. "where the rules or statutes do not reach the acts which degrade the judicial system." Amsted Industries Inc. v. Buckeye Steel Castings Co., 23 F.3d 374, 379 (Fed. Cir. 1994) (quoting Chambers, 501 U.S. at 41-42, 111 S.Ct. 2123).

Given that the Court found this case to be exceptional, it cannot be said that § 285 was inadequate to reach Rambus' spoliation. Rambus' litigation misconduct was not left unremedied by § 285. Rather, the Court simply exercised its discretion in finding that an award of attorney's fees on that basis was not appropriate. Thus, with respect to spoliation, it is unnecessary further to consider whether an award of attorney's fees under the inherent powers is necessary to sanction Rambus' spoliation.

However, the Court rejected a finding of exceptionality based on Rambus' alleged "gaming of the system," and thus it is necessary to consider whether it is appropriate to impose, under the Court's inherent power, sanctions for the asserted vexation or the bringing the counterclaims in bad faith. For the reasons set forth previously, the Court finds that Rambus did not assert its counterclaims in bad faith or for the purpose of vexation. Nor, given the posture of the Hynix v. Rambus case in the Northern District of California at the time the counterclaims were filed can it be said that the counterclaims were objectively baseless. Under the facts of this case, it cannot be said that the presentation of the counterclaims was vexatious or in bad faith, notwithstanding that Rambus spoliated evidence in anticipation of litigation.⁴⁰

40 Samsung's claim for attorney's fees, whether under § 285 or the Court's inherent power, is not directed toward fees incurred in addressing Rambus' defense of Samsung's declaratory judgment complaint.

CONCLUSION

Considering the record as a whole, and for the foregoing reasons, the Court finds that this is not an appropriate case in which to award attorney's fees to Samsung. Hence, SAMSUNG'S MOTION FOR FINDING THAT SAMSUNG IS A PREVAILING PARTY AND THE AWARD OF REASONABLE ATTORNEY'S FEES (Docket No. 87) and SAMSUNG'S MOTION TO FIND THIS AN EXCEPTIONAL CASE AND FOR ATTORNEY FEES UNDER 35 U.S.C. § 285 (Docket No. 89) will be denied.

The Clerk is directed to send a copy of this Memorandum Opinion to all counsel of record.

It is so ORDERED.