Until the U.S. Court of Appeals for the Federal Circuit issued its en banc decision in Knorr-Bremse Systeme Fuer Nutzfahrzeuge Gmbh v. Dana Corp., Nos. 01-1357, 02-1221 and 02-1256 (Fed. Cir. Sept. 13, 2004), if an opinion was not produced in litigation, an inference could be drawn either that no opinion was obtained or that the opinion was unfavorable to the accused infringer.1

However, on Sept. 13, 2004, the Federal Circuit changed this law, holding that "no adverse inference that an opinion of counsel was or would have been unfavorable flows from an alleged infringer's failure to obtain or produce an exculpatory opinion of counsel."2

The court reaffirmed that the determination of whether an accused infringer has willfully infringed a patent must be made based on a consideration of the totality of the circumstances, and the court cited several cases listing factors to be considered in a totality of the circumstances analysis.3

The Federal Circuit also considered whether a substantial defense to infringement should be sufficient to defeat liability for willful infringement even if no legal advice has been secured. The court concluded that a substantial defense is one factor to be considered with others under the totality of the circumstances analysis, but the court rejected giving this factor per se treatment, preferring to having it be given different weight in different cases.4

The Knorr-Bremse decision has substantially improved the way that willful infringement allegations will be handled in patent litigation. This article discusses why this change in the law concerning treatment of adverse inferences was needed, some of the benefits that will occur as a result of this change and some practical advice for accused infringers in light of this holding.

Why a Change in the Law Was Needed
A change in the treatment of adverse inferences was needed because the attorney-client privilege and work product doctrine are central to the relationship between an attorney and client and to the conduct of litigation. Clients feel free to candidly and honestly discuss their legal situations with their attorneys only if they are assured that their communications will remain privileged and nondisclosed. Similarly, attorneys feel free to candidly and honestly give their clients advice that discusses specifics, as opposed to being limited to broad generalizations, only if they are assured that their communications will remain privileged and nondisclosed.

Society should encourage, and in fact before Knorr-Bremse in every type of litigation other than that involving patent infringement society has encouraged, such candid and honest communication from the client to the lawyer and such candid, honest and specific advice from the lawyer to the client.

Before Knorr-Bremse, it was only in patent litigation that clients were compelled to disclose otherwise privileged communications. "The attorney-client privilege is the oldest of the privileges for confidential communications known to the common law. … The privilege recognizes that sound legal advice or advocacy serves public ends and that such advice or advocacy depends upon the lawyer's being fully informed by the client."5

Similarly, the work product doctrine long has been upheld in the common law and in the Federal Rules of Civil Procedure (see Rule 26[b][3]). As stated in the seminal work product doctrine case, Hickmon v. Taylor, "it is essential that a lawyer work with a certain degree of privacy, free from unnecessary intrusion by opposing parties and their counsel."6

It is in the interest of the patent system and of patentees to have potential infringers seek advice from their attorneys as to whether they are infringing a valid patent. However, if the clients are discouraged from seeking such advice or communicating honestly or if the lawyers are discouraged from discussing specifics and raising issues as to how the patent can be designed around, the result will be more, not less, patent infringement.

Another reason that a change in the treatment of adverse inferences was needed is that the Federal Circuit has suggested in its decisions that the problems caused by pressuring an accused infringer to produce its opinion of counsel may be ameliorated through the use of bifurcation.7 However, bifurcation did not cure this problem.

If an adverse inference was to be drawn, in order to avoid prejudice from this inference during the determination of liability, both discovery and trial on willfulness should be bifurcated. However, trial courts have been reluctant to stay discovery and to hold a separate trial on willfulness, fearing that if they bifurcate the willfulness issue, there will be delay, inefficiency and worse yet, they may have to empanel a second jury.8

If an accused infringer was going to be, in essence, forced to disclose its opinion of counsel, the question of willfulness should have been decided in a bifurcated proceeding. The focus should not have been on whether bifurcation produces a more efficient proceeding (it will be more efficient if no liability is found in the first phase of the case, but in many cases, it will not be more efficient).

Rather, the focus should have been on whether bifurcation was needed as a matter of fairness in order to ameliorate the prejudice caused by pressuring the defendant to disclose its opinion of counsel.

However, many courts were not doing this. As a result of the Knorr-Bremse holding, the disclosure of an accused infringer's opinion of counsel is no longer being compelled, and thus, the need for bifurcation to ameliorate any prejudice ought to be lessened.

Benefits of the Knorr-Bremse Decision
Now that the Federal Circuit has ruled that an adverse inference may not be drawn from the absence of a produced opinion of counsel or from no opinion of counsel ever being obtained, patent litigation and accused infringers will experience three benefits.

The first benefit is that the patentee will no longer have an unfair advantage in meeting its burden to prove willfulness. Willful infringement is a question of fact, including elements of intent, reasonableness and belief, and because it is a punitive finding that can increase damages, it must be proved by the patentee by clear and convincing evidence.9 A patentee that is able to prove willfulness can be rewarded not only with up to treble damages, but also with attorney fees if the case is found to be exceptional.

However, when an affirmative duty to seek an opinion of counsel was imposed on an accused infringer and the accused infringer suffered an adverse inference if it failed to disclose the opinion of counsel, the burden of proof of willfulness was effectively shifted. Now that the adverse inference will not be drawn, the burden will not be shifted.

The second benefit that patent litigation and accused infringers will experience now that the threat of an adverse inference has been removed is that the expensive and wasteful ancillary litigation over the issue of the scope of waiver should be needed less often. However, where it does occur, the uncertainty as to how much has been waived will remain until the Federal Circuit addresses this issue.

When an opinion of counsel is produced, inevitably questions are raised as to what is the scope of waiver of documents and communications beyond the opinion itself. There can be no question that communications between the attorney and client concerning the legal opinion being relied upon are waived (i.e., a narrow waiver of privilege). But the tougher question is whether the patentee can obtain all documents reviewed by or prepared for legal counsel, including all drafts, notes, and opinion and mental impression work product, even when such documents were not communicated to the client (i.e., a broad waiver of privilege).

Litigating this dispute as to the scope of waiver can be expensive. In addition, the lack of certainty as to where the scope of waiver will be drawn makes the already difficult choice of whether to produce an opinion of counsel that much more difficult. The amicus curiae brief filed by the Federal Circuit Bar Association in the Knorr-Bremse case10 does an excellent job of demonstrating how divergent rulings have been, even sometimes in the same district, in terms of the scope of waiver found:

• Opinion counsel's work-product, if not communicated to the client, need not be produced. Thorn EMI N. Am. [Inc.] v. Micron Tech. Inc., 837 F. Supp. 616, 622 (D. Del. 1993).
• Opinion counsel's work-product that formed the factual basis of the opinion must be provided even if not disclosed to the client. D.O.T. Connectors Inc. v. J.B. Nottingham & Co., 2001 WL 34104928 (N.D. Fla. Jan. 22, 2001).
• All of opinion counsel's work-product must be produced whether or not disclosed to client. Mosel Vitelic Corp. v. Micron Tech. Inc., 162 F. Supp. 2d 307, 312-[13] (D. Del. 2000).
• Trial counsel's work-product must be produced if it contradicts or casts doubt on opinion counsel's opinions. Thermos Co. v. Starbucks Corp., 1998 WL 299469 (N.D. Ill. Nov. 4, 1998).
• Post-complaint work-product need not be produced. Micron Separations Inc. v. Pall Corp., 159 F.R.D. 361, 363-64 (D. Mass. 1995).
• There is no bright line immunity for post-complaint work-product. Chiron Corp. v. Genentech Inc., 179 F. Supp. 2d 1182, 1188 (E.D. Cal. 2001).
• When opinion counsel becomes trial counsel, all post-complaint legal advice is discoverable. Novartis Pharms. Corp. v. Eon Labs. Mfg. Inc., 206 F.R.D. 396, 399 (D. Del. 2002). But see Bristol-Myers Squibb Co. v. Rhone-Poulenc Rorer Inc., 52 U.S.P.Q.2d (BNA) 1908, 1909 (S.D.N.Y. 1999) (although certain post-complaint legal advice was ordered to be produced to determine whether it contradicted the earlier opinion of counsel, the defendant did not waive its trial counsel's trial strategy or work product).

Thus, there is tremendous uncertainty as to where the line will be drawn in any particular case as to the scope of the waiver, and the consequences of this determination can be great. The accused infringer may not be concerned about producing its opinion of counsel, and even all factual bases for that opinion of counsel, but it may be very concerned that advice from its trial counsel may need to be disclosed.

In the best case scenario from the accused infringer's point of view, it may be able to limit the scope of the waiver, but usually such success is achieved only after significant attorney fees have been incurred litigating this issue, perhaps in discovery and depositions concerning this issue, and in motion practice on this issue.

Therefore, in light of the Knorr-Bremse decision, while fewer accused infringers may choose to produce their opinions of counsel, for those who do choose to do so, the problem of different courts (and sometimes judges in the same district) applying different standards to the scope of waiver issue will continue.

Because this is either an issue unique to patent cases or one that bears an essential relationship to matters committed exclusively to the Federal Circuit's exclusive control,11 the Federal Circuit should set a uniform standard for the scope of waiver for all patent infringement cases.

The third benefit that patent litigation and accused infringers will experience now that the threat of an adverse inference has been removed is that accused infringers may be able to choose to have their opinion counsel serve as litigation counsel if they so desire. It is often the case that a company will ask a patent attorney with whom it has worked closely to provide an opinion of counsel when it learns of the existence of a patent. Such counsel may be very familiar with the client's business, products and services and may be able to prepare such an opinion very cost effectively.

Under the law before the Knorr-Bremse decision, it was likely that the client would produce its opinion of counsel in order to avoid an adverse inference, and there was a reasonable chance that the attorney who authored the opinion would need to testify at trial. Accordingly, such an attorney,12 and in some jurisdictions, the attorneys' entire firm,13 could be disqualified from representing the client at trial.

For a small company, the effect of preventing opinion counsel from being trial counsel can be devastating. The economic impact of paying for a second firm (or at least different attorneys at the first firm) to come up to speed can be substantial. A defendant should be free to choose the counsel it desires to represent it in litigation. Opinion counsel may still be disqualified from being trial counsel if the company discloses its opinion of counsel, but now an accused infringer has more control over the decision as to whether or not to disclose its opinion of counsel.

Practical Advice for Accused Infringers in Light of the Knorr-Bremse Decision
In light of the Knorr-Bremse decision, a company that might be accused of infringing a patent and that has knowledge of the patent must decide whether to obtain an opinion of counsel before beginning any infringing conduct. Because an accused infringer must still show that it acted with due care and because having a substantial defense is likely to become an important factor in the totality of circumstances analysis (notwithstanding the Federal Circuit's rejection of giving it per se status), an accused infringer would be well served to still obtain an opinion of counsel.

In fact, the opinion of counsel obtained post-Knorr-Bremse may be more valuable to a company than prior exculpatory opinions could be. Now, a company may feel free to communicate honestly and candidly with its attorney about its situation and the lawyer may feel free to communicate more honestly, candidly, and specifically about the company's situation, including exploring ways to design around the patent.

However, before this candid and honest communication occurs, the lawyer and the client must fully consider the ramifications of not producing the opinion of counsel that will be prepared. A conclusion should be reached as to whether the opinion will be withheld or whether it will be produced because this could have ramifications on the information supplied by the client and the advice given by the attorney.

Another reason to obtain an opinion of counsel is that it might supply a substantial defense. As a practical matter, a company may wish to seek a substantial defense before beginning its allegedly infringing activity, and may in fact only begin such activity if such a defense is found. This is true because the company may not want to risk that no substantial defense will be found during subsequent litigation once the company has committed resources to the allegedly infringing conduct and has damages exposure.

Unfortunately, the Knorr-Bremse decision does not cure all problems raised by willfulness allegations. If a decision is made to produce the opinion of counsel, the problem remains that where the court will draw the line on the scope of the waiver is uncertain. Accordingly, trial counsel must carefully consider whether an opinion should be disclosed in any particular case. In order to analyze this issue, an assessment of the factors used in the totality of circumstances analysis should be made.

Notes

1 See, e.g., Kloster Speedsteel AB v. Crucible Inc., 793 F.2d 1565, 1580 (Fed. Cir. 1986), cert. denied, 479 U.S. 1034 (1987); Fromson v. Western Litho Plate & Supply Co., 853 F.2d 1568, 1572-73 (Fed. Cir. 1988).

2 Knorr-Bremse, slip op. at 3 (emphasis supplied).

3 Knorr-Bremse, slip op. at 6, citing, Rolls-Royce Ltd. v. GTE Valeron Corp., 800 F.2d 1101, 1110 (Fed. Cir. 1986), and Read Corp. v. Portec Inc., 970 F.2d 816, 826-27 (Fed. Cir. 1992) (factors include whether the infringer deliberately copied the ideas or designs of another; whether the infringer, after knowledge of the patent, investigated the scope of the patent and formed a good-faith belief that it was invalid or that it was not infringed; the infringer's behavior as a party to the litigation; the infringer's size and financial condition; the closeness of the case; the duration of the infringer's misconduct; the remedial action taken by the infringer; the infringer's motivation for harm; and whether the infringer attempted to conceal its misconduct).

4 Knorr-Bremse, slip op. at 15.

5 Upjohn Co. v. United States, 449 U.S. 383, 389 (1981).

6 Hickman v. Taylor, 329 U.S. 495, 510-11 (1946).

7 See, e.g., Quantum Corp. v. Tandon Corp., 940 F.2d 642, 644 (Fed. Cir. 1991) (court should give serious consideration to a separate trial on willfulness where the defendant is confronted with the dilemma between waiving the privilege to avoid a willfulness finding, thereby prejudicing itself on the question of liability, and maintaining the privilege, thereby risking a finding of willfulness); Fromson, 853 F.2d at 1572 (noting a separate trial on willfulness "may be useful in meeting the attorney-client privilege problem").

8 See, e.g., General Signal Corp. v. Applied Materials Inc., 46 U.S.P.Q.2d 1160 (D. Del. 1997) (rejecting bifurcation of liability and willfulness issues and stating that evidentiary rulings and instructions can adequately prevent jury confusion even though the opinion of counsel stated that some of defendant's products literally infringe the relevant patents); Calmar Inc. v. Emson Research Inc., 850 F. Supp. 861, 866 (C.D. Cal. 1994) (rejecting bifurcation because it is not the most efficient method of handling the case, it would complicate the case and would cause needless delay while acknowledging that "certain discovery problems may arise with respect to Emson's attorney opinions"); Keyes Fibre Co. v. Packaging Corp. of America, 763 F. Supp. 374, 375-76 (N.D. Ill. 1991) (rejecting bifurcation based on considerations of convenience, expedition and lack of complexity and stating that there is no threat of prejudice from evidence being applied to the wrong issue because the case will be tried by the court without a jury).

9 Pall Corp. v. Micron Separations Inc., 66 F.3d 1211, 1221 (Fed. Cir. 1995), cert. denied, 520 U.S. 1115 (1997).

10 Brief of amicus curiae, the Federal Circuit Bar Association, in response to the court's order inviting briefs at 2-4, Knorr-Bremse.

11 Madey v. Duke Univ., 307 F.3d 1351, 1358 (Fed. Cir. 2002), cert. denied, __ U.S. __, 123 S. Ct. 2639 (2003).

12 See Rohm & Haas Co. v. Lonza Inc., 1999 WL 718114, at *1 (E.D. Pa. 1999) (attorneys who provided the opinion of counsel relied upon by the defendant are disqualified from serving as trial counsel whether or not willfulness is bifurcated); Model Rules of Professional Conduct R. 3.7(a) (June 2003) ("A lawyer shall not act as advocate at a trial in which the lawyer is likely to be a necessary witness unless: (1) the testimony relates to an uncontested issue; (2) the testimony relates to the nature and value of legal services rendered in the case; or (3) disqualification of the lawyer would work substantial hardship on the client.") The Model Rules of Professional Conduct, unlike the predecessor Model Code of Professional Responsibility, do not automatically extend the lawyer-witness prohibition to the partners and associates of a testifying lawyer. See Rule 3.7(b).

13 The Model Code of Professional Responsibility provides for broader disqualification of a law firm. See, e.g., DR 5-101(B) of Ohio's Code of Professional Responsibility:

A lawyer shall not accept employment in contemplated or pending litigation if the lawyer knows or it is obvious that the lawyer or a lawyer in the firm ought to be called as a witness, except that the lawyer may undertake the employment and the lawyer or a lawyer in the firm may testify: (1) If the testimony will relate solely to an uncontested matter. (2) If the testimony will relate solely to a matter of formality and there is no reason to believe that substantial evidence will be offered in opposition to the testimony. (3) If the testimony will relate solely to the nature and value of legal services rendered in the case by the lawyer or the firm to the client. (4) As to any matter, if refusal would work a substantial hardship on the client because of the distinctive value of the lawyer or the firm as counsel in the particular case. Ohio Rev. Code Ann. § DR 5-101(B)(1-4) (West 2004).