Monday, June 05, 2006

Presumption of Irreparable Harm: Why?

A unique feature of copyright law (also found to varying degrees with patents and trademarks ), but one that in my opinion been misunderstood and misapplied is the judicially created principle of presuming irreparable harm automatically when the movant establishes a likelihood of success. The presumption is applied by all circuits, except the Fifth Circuit, which expressly does not apply the presumption. The Eleventh Circuit applies the presumption generally, but not where there is a bona fide fair use defense. It has not been approved by the District of Columbia Court of Appeals yet, although some trial judges in that circuit have applied it.

The presumption is of a relatively recent vintage, appearing as a discrete statement first in dictum in the American Metropolitan Enterprises of New York, Inc. v. Warner Bros. Records, 389 F.2d 903, 905 (2d Cir. 1968) and later in other circuits. The relevant passage from American Metropolitan is: “A copyright holder in the ordinary case may be presumed to suffer irreparable harm when his right to the exclusive use of the copyrighted material is invaded.” This dictum is all the more remarkable because the remedy at issue was strictly legal: the dispute was over mechanical compulsory license fees under the predecessor to current Section 115. Even more remarkable is the fact that the Second Circuit did not apply the presumption; instead, it affirmed the district court’s refusal to grant a preliminary injunction on the ground that an adequate legal remedy existed:

“[T]he usual rationale for granting a preliminary injunction upon the showing of a prima facie copyright infringement does not apply where the copyright covers musical compositions mechanically reproduced since the copyright holder does not have any exclusivity once he has permitted use. No irreparable injury may be presumed in such a situation. “

While this is correct regarding a compulsory license fact pattern, there was no supporting authority for the dictum that in the “ordinary case” irreparable harm may be presumed. The discussion immediately preceding the statement concerned a different proposition, i.e., a number of cases had held that “upon the showing of a prima facie case of infringement a copyright holder is entitled to a preliminary injunction without a detailed showing of irreparable harm.” The cases cited in support of this proposition state exactly that, but no more. Nor was there any guidance on what an “ordinary” copyright case is or how to apply the presumption: even in “ordinary cases” is the presumption a substitute for any obligation on plaintiff’s part to produce evidence of irreparable harm? What if defendant introduces some evidence showing there is no irreparable harm? Does the presumption vanish and does plaintiff then have to present evidence? A quick look at Federal Rule of Evidence 301 provides the answers.

“In all civil actions and proceedings not otherwise provided for by Act of Congress or by these rules, a presumption imposes on the party against whom it is directed the burden of going forward with evidence to rebut or meet the presumption, but does not shift to such party the burden of proof in the sense of nonpersuasion, which remains throughout the trial upon the party on whom it was originally cast.”

A presumption may be used, consistent with Rule 301, as a device to temporarily shift the burden of going forward with the contrary evidence. A presumption does not shift the ultimate burden of persuasion, which always remains on plaintiff. Thus, in the case of a motion for a preliminary injunction, the burden of establishing entitlement to that relief always remains on the movant regardless of whether a presumption may be applied for one or more factors, e.g., irreparable harm. A presumption merely affords the party for whose benefit it runs “the luxury of not having to produce specific evidence to establish the point at issue.” The luxury is short-lived and shallow, though: a presumption still involves an obligation to produce a minimum quantum of evidence sufficient to call into question the existence of the presumed fact, albeit “nothing more” When evidence contrary to the presumption is introduced, the presumption is not only rebutted, it vanishes. Or, as the Supreme Court put it:” The presumption, having fulfilled its role of forcing the defendant to come forward with some response, simply drops out of the picture.” At this point, plaintiff may still succeed (and perhaps quite easily in a given case) by presenting evidence of irreparable harm, but he is “now put to his factually-supported proof.”

There are two reasons why the presumption vanishes once a minimum quantum of evidence sufficient to call into question the existence of the presumed fact is introduced by the nonmoving party. First, a presumption is not evidence. The second reason is that Congress intended the presumption to vanish under the “bubble bursting” or “Thayer” theory. The legislative history on this point is definitive.

The debate about the status of a presumption once rebutted may be traced to two conflicting schools. The first is that of James Bradley Thayer, whose 1898 treatise, A Preliminary Treatise on Evidence, took the view that a presumption drops out from the case once rebutted. This approach is also called the “Thayer” or “bursting bubble” theory. The second school is that of Professor Edmund Morgan, who in 1933 proposed that a rebutted presumption should remain in the case to be given whatever weight the trier of fact wishes to give it, although rebuttal would still shift the burden of production back to plaintiff. These two schools battled it out like the ancient Talmudic schools of Hillel and Shammai during the 13 years of study that went into the development of the Federal Rules of Evidence, culminating in their enactment by Congress in 1974.

Although the Supreme Court promulgated the rules on November 20, 1972, to take effect on July 1, 1973, questions were raised about the Court’s authority, and thus on February 5, 1973, after Congress blocked their implementation, Chief Justice Burger transmitted the proposed rules to Congress, which extensively revised them. The amendments made by Congress to Rule 301 are highly relevant. The Supreme Court’s version of Rule 301 provided that in all cases a presumption imposes on the nonmoving party the burden of proving that the nonexistence of the presumed fact is more probably than its existence. This approach is well beyond that advocated by Professor Morgan. Indeed, the Advisory Committee on Rules of Evidence of the Judicial Conference took the position that the Court went to far by permanently altering the burden of persuasion. The committee therefore proposed the approach advocated by Professor Morgan in the following language: “[A] presumption imposes on the party against whom it is directed the burden of going forward with the evidence, and, even though met with contradicting evidence, is sufficient proof of the fact presumed to be considered by the trier of fact.

The House passed the Advisory Committee version. In the Senate, the Advisory Committee and the Judicial Conference’s Standing Committee on Practice and Procedure agued for adoption of the House approach. The Senate Judiciary Committee rejected the proposal, describing the Advisory Committee/House version as “ill-advised.” The Senate’s rejection was based on its view that “[p]resumptions are not evidence, but ways of dealing with evidence,” and that treating presumptions as evidence requires deeming “as evidence” facts for which there is no evidence. The Committee therefore amended the proposed rule to its current form. The purpose of the amendment was “to make clear that while evidence of facts giving rise to a presumption shifts the burden of coming forward with evidence to rebut or meet the presumption, it does not shift the burden of persuasion on the existence of the presumed facts. The burden of persuasion remains on the party to whom it is allocated under the rule governing the allocation in the first instance.” As this passage makes clear, Rule 301 contemplates a situation where some actual evidence is initially introduced, which in turn gives rise to a legal presumption regarding a difference fact. The Senate’s explanation of Rule 301 does not contemplate a situation where no evidence is presented, yet a presumption still arises and for the obvious reason that a presumption is not evidence and therefore no legal conclusion can be dram from the utter absence of evidence.

Undaunted, the Advisory Committee and the Judicial Conference lobbied the conference committee, telling the committee that “basically the choice to be made in treating the effect of given presumptions lies between giving them only the effect of a “bursting bubble” and giving them the greater effect of imposing a burden of disproof once evidence has established the conditions that call the presumption into operation.” The conference committee recommended adoption of the Senate version, and that was the version that passed Congress and was enacted. Rule 301 thus adopts the Thayer/bursting bubble approach to presumptions.
The reason for the Senate (and ultimately Congress’s) rejection of the Morgan/Advisory Committee/House approach is easy to understand. That view correctly noted that rebuttal “does not eliminate from the case the evidence itself that gave rise to the presumptions.” And there should be no bar to such predicate evidence going to the trier of fact. But the Morgan/Advisory Committee advocates wanted more: once the predicate evidence was introduced but was rebutted, the presumption was nevertheless to remain, with its weight to be determined by the trier of fact.

A proper understanding of Rule 301 renders the per se presumption of irreparable harm in copyright cases seemingly utilized in the Ninth Circuit highly suspect since in copyright cases the presumption of irreparable harm is used in lieu of evidence and no as a gap-filler once predicate evidence is first established. In copyright cases, if there were evidence of irreparable harm, there would be no need for the presumption. Moreover, the presumption of irreparable harm in copyright arises from an unconventional source, a showing of likelihood of success on the merits. That showing, though, has nothing inherently to do with irreparable harm from an evidentiary standpoint. A plaintiff could have a virtual certainty of prevailing on the merits, but the loss be entirely compensable by monetary damages, as where plaintiff proves there is one-to-one displacement of sales, or where plaintiff elects statutory damages. Perhaps as important, the threshold showing for any grant of interim injunctive relief is, irreparable harm, not a likelihood of success on the merits (although such success is also required). To permit a showing of likelihood of success on the merits to establish irreparable harm is not only to collapse two factors into one; it collapses the dog into the tail.

Finally, compelling plaintiff to establish actual evidence of irreparable harm is consistent with the general rule that where facts relating to a disputed issue lie peculiarly within the knowledge of one party, placing the burden of proof on his or her adversary is unfair. Nowhere is this principle more appropriately applied than where on party claims irreparable harm. Such harm is entirely within the knowledge of the party asserting its existence.

In sum, the presumption of irreparable harm cannot be justified on a blanket basis, and if applied at all, is a mere burden-shifting device, not evidence. Once some contradictory evidence is introduced by the nonmoving party, the presumption vanishes. If plaintiff fails to introduce convincing evidence of actual irreparable harm, the request for injunctive relief must be denied.


Anonymous said...

While I don't disagree with your general analysis of the law with respect to the general evidentiary effects of a presumption, methinks you fly high and fast over the more fundamental issue: what does it mean to have an **exclusive** right?

The critical aspect of exclusivity is the ability to effectively exclude others -- not merely to collect monetary damages for violation of exclusive rights.

To hold, as some courts have, that the availablity of a damages remedy for subsequent infringements is an adequate remedy at law effectively results in a forced license -- especially in the context of a denial of permanent injunctive relief. In such instances, the copyright holder is required, notwithstanding his "exclusive" rights, to suffer the infringer's continued or future use of the copyrighted work, with the legal system providing only a forum in which to haggle over the price. Except for mechanical licenses, Congress has never approved the concept of such compulsory licenses.

It is, I believe, the recognition of this nature of an exclusive right that has led courts to "presume" irreparable harm, or in many cases to treat it as a given (i.e., an irrebuttable presumption). In other words, no independent evidence of irreparable harm is necessary because once infringement is established, because the threatened violation of an exclusive right is inherently irreparable.

Does this result in the second element collapsing into the first? Sure, and a number of courts have noted that such is the case. Is this inconsistent with the Copyright Act? Not if one takes the concept of exclusive rights seriously.

[BTW, re: DC Circuit and presumption of irreparable injury in copyright cases, see Walt Disney Co. v. Powell, 897 F.2d 565, 567 (D.C. Cir. 1990) -- at least in the context of permanent injunctions, the DC Circuit pretty clearly presumes irreparable injury once a threat of future infringement is shown, availability of monetary damages notwithstanding. Money quote from that case: where infringement is established at trial and there is a threat of continuing infringment, "the copyright holder is **entitled** to an injunction" (emphasis by the court).]


LKB in Houston

William Patry said...

The issue of denying injunctive relief came to the fore in a number of Supreme Court cases, Sony, where no justice thought that the courts had the power to order what was in effect a compulsory license, and in 2Live Crew, which following Judge Leval, took a more traditional equitable approach to simply denying requests in particular cases. But where injunctive relief is denied, the key question is, what happens when defendant continues the infringing activity? Is is compensable by monetary damagees and by what measurement?

Anonymous said...

Very interesting. In trademark law, I believe, the presumption is justified on the grounds that infringement always involves damage to reputation and monetary damages are always insufficient to repair reputation.

Perhaps, a similar justification applies in copyright insofar as most copyright claims include facts that might otherwise be a trademark claim.

If the infringer asserts that the copy comes from the author, then it falsely attributes distribution to the author. If the infringer denies recognition to the author entirely, then he passes off the work to his own credit, denying the reputational advantage to the author.

An infringer must do one or the other, and so an infringer must always be involved in reputational harm to the author.

In this light, the presumption is a species of moral right in the author. If the origin of the presumption rests on reputational harm, perhaps it should be attacked from a Dastar direction.

Anonymous said...

Prof Patry:

Respectfully, I must disagree with your analysis.

The fact that injunctive relief can, in appropriate cases, be denied (which Justice Thomas recently reconfirmed in the eBay decision) doesn't address the issue that I raise: the nature of an exclusive right *inherently* presents a situation of irreparable harm. Thus, the "presumption" of irreparable injury is not only appropriate but, in my opinion, should be irrebuttable once infringement is determined. In short, the "collapsing" of element 2 into element 1 is perfectly appropriate.

Upon consideration of the remaining factors (especially public interest and balancing of the harms), a court could still determine that injunctive relief is unwarranted in an appropriate case. This fact, however, does not mean that irreparable injury should not be taken as a given once infringement is established, or that the copyright owner ought to be forced to present evidence of irreparable injury beyond the facts that (1) he has an exclusive right (i.e., a valid copyright that the defendants' activity violates), (2) the defendants' future activities threaten further infringement, and (3) the copyright owner does not consent to the defendants' continued violations.

I also disagree that as to "what happens when the defendant continues the infringing activity," the "key question" ought to be "whether and how much compensation ought to be awarded." As I noted in my original post, this analysis ignores the copyright holders' right to exclude -- especially in the permanent injunctive context. My point is that reducing the irreparable injury inquiry to compensibility implies that the courts can or should just force a license on the copyright owner. As you correctly note, the Supremes in Sony reaffirmed that courts cannot do this.

Here's another complication to reducing the inquiry to the compensibility of the injury, especially in the permanent injunction context . . . .

If we assume that there has been adjudication of infringement (e.g., a posttrial request for a permanent injunction), there is no question that the defendant actually knows that his continued activity will be an act of copyright infringement; indeed, that fact is res judicata. Thus there will be little question that future infringement will be willful.

If we also assume that the retail value of the copies that will be made or distributed in the future will exceed $1,000 (which is a reasonable assumption, given that virtually no one is going to actually try a copyright case involving smaller amounts), then the defendant's threatened future activity will facially satisfy all the elements of a 506(a) criminal offense.

Should a court, in deciding a request for permanent injunctive relief, simply ignore that the continuation of the infringing activity will be a criminal offense? (This was not an issue in eBay, because there is no analog to 17 USC 506 in the Patent Act.) The fact that a wrong may be compensible by civil money damages never provides a defense to criminal behavior. And, I submit, preventing future criminal behavior is fit grist for the injunction mill.

As always, I look forward to your comments.

LKB in Houston

Anonymous said...

I blogged about the overlap between this post and the eBay decision over at

The conclusion:

"As Patry and the eBay court both say, it may be that almost all infringement cases do indeed merit injunctions. (A concurring opinion in eBay emphasizes that viewpoint.) But the move away from even analyzing the question, toward an automatic presumption, sweeps too broadly. Sometimes the unauthorized use of content just means rightsholders lose potential licensing fees, and that’s it. Why is it insufficient in such cases to award monetary damages keyed to the licensing fee that would have been charged? Note that, in situations where such damages would be too low, a plaintiff also has the option to elect statutory damages instead under section 504(c) of the Copyright Act."

Anonymous said...

Pensans, I feel that you present a false dichotomy. Couldn't an alleged infringing work include an attribution such as, "this work includes samples from" or even, "the following passage is from" for a direct quotation?

Also, I understand that when you state, "monetary damages are always insufficient to repair reputation", you may very well be quoting established jurisprudence. It seems to challenge common sense, however, that courts may assign monetary value to human life but not to reputation.

Anonymous said...

I don't know if this is true anymore, but it certainly always used to be the case that the largest volume of copyright infringement cases filed in Federal Court involved cases brought by performing rights societies.

A PRO in the United States and in many other countries is required to issue a license to all comers. In the instance where the user doesn't obtain a license or rejects the offer of a license from a PRO, the PRO can sue for copyright infringement and seek damages as well as injunctive relief.

The damages available to the PRO, thanks to statutory damages, are far, far in excess of the amount that a reasonable license would have been for the use of the works and the PRO has the capability to aggregate its infringement claims brought in the name of its members such that the damages can reach astronomical sums. We are now seeing the record companies doing the same in cases such as the ones brought against Napster and XM Satellite Radio.

The infringements by the user, if they continue, are not extinguished in any way by a prior damage award. Under the copyright act, the infringing conduct continues to hold its status as infringing and new claims can be brought each time a new performance of the music occurs.

Why would you want an injunction in these circumstances? Well, contempt is a powerful threat and a useful tool likely to be much more streamlined than bringing another infringement action. Injunctions can bring the court in as an active supervisor of continuing conduct. And a prior record of contempt in addition to an earlier damage award makes a better argument for willfulness in the next go round and even a useful convincer in trying to get a criminal case going.

But at some point don't we need to add up all the weapons and all the intimidations arrayed for the protection of the exclusivity that LKB in Houston cherishes? Damages beyond your wildest economic expectations, attorney's fees, the right to prosecute sequentially and interminably to the same result, the right to shut down your opponent's business altogether, the right to get your very own Federal District Judge running your opponent’s business for you and the right to send the infringer to jail all for the protection of [fill it in with something extraordinary] or [fill it in with something completely useless].

Then comes the suggestion that we pile on some more and create presumptions for injunctive relief so strong that they become self-executing or that they effectively distract the court from engaging in a fair analysis and balancing as required by the jurisprudence of sitting in equity to issue an equitable remedy!

The notion that a copyright owner enjoys "a right to exclude" is a fascinating topic on its own and maybe Prof. Patry will amuse us in the future with a riff on it. For the time being I would say it is intellectually unfair to hold to that concept as an absolute in a world of collective licensing, statutory licensing, compulsory licensing and de facto an de jure requirements for non-exclusive and non-discriminatory licensing. And, it is important sometimes to see the forest not the trees. Copyright owners are well and adequately armed. They don’t need any more extraordinary deference than they already richly enjoy absent some compelling need unserved by the existing gargantuan tome of the copyright law.

Anonymous said...

My reading of the cases applying the presumption of irreparable harm upon a finding of likelihood of success on the merits is not at odds with with your reading of FRE 301.

There's a lot of caselaw to the effect that the "burden of proof" can refer to either the burden of production or the burden of persuasion, and that they are often lumped together because they are usually assigned to the same party. But there are instances where the difficulty of proof or access to the pertinent evidence will shift the burden of production to the party who does not have the ultimate burden of proof. The classic example is a defense of insanity in a murder case. The state has the burden of persuasion beyond a reasonable doubt on the elements of murder, but does not have to present evidence negating insanity until and unless the defendant meets a burden of production that raises the issue.

In copyright cases, the likelihood of prevailing on the infringement claim doesn't really give rise to a "per se presumption of irreparable harm" -- it gives rise to a per se presumption of harm: loss of the exclusive right. The burden of production shifts to the defendant to adduce evidence that the harm is reparable because the plaintiff would otherwise have the impossible task of proving a negative -- that the harm is not reparable.

Evidence that the harm is measurable, that the loss of the exclusive right is compensable, "bursts the bubble" by demonstrating a genuine issue of material fact that must be resolved as a matter of fact; but that doesn't mean that the ultimate burden of persuasion doesn't remain with the plaintiff.

In American Metropolitan, the harm is still presumed, but there is no occasion for shifting the burden of production on whether it is irreparable because it doesn't depend upon evidence of showing a genuine issue of material fact. The availability of an adequate remedy is a matter of law, not an issue of fact. If the Eleventh Circuit doesn't shift the burden where there is a viable fair use defense, that would seem to go the adequacy of its showing on the likelihood of success on the merits, and whether there is any harm at all.

In the "ordinary case," it is fair to shift the burden to the defendant to introduce evidence that there is a genuine issue of fact as to irreparable harm because the defendant has unique access to the evidence necessary to determine whether the damages can be calculated and satisfied. In that way, I think most of the cases can be reconciled with the "bubble-bursting" approach dictated by FRE 301.

However, it's also important to bear in mind that the shifting burden of proof, whether it's the burden of production or the burden of persuasion, is not on an element of the offense of copyright infringement. It is in aid of equitable determination of the need for preliminary relief. As an exercise in equity, it also takes into account the comparative harm to the defendant and the public interest. Once the plaintiff has demonstrated a likelihood of satisfying its own unambiguous burden of proof, i.e. burden of persuasion, on the elements of the offense, including damages. Once the plaintiff has satisfied the court in equity that it is harmed by the the defendants unlawful conduct, equity might fairly look to the defendant for proof that, in equity, its unlawful conduct should be allowed to continue.

John Noble

William Patry said...


Sorry for the delay in responding. You seem to assume the question I raised, which is that infringement of intellectual property inherently results in irreparable harm. I don't dispute that in particular cases that is true, but what I do question is the formulation of a general presumption that applies in all cases and that serves to seriously tip the scale against defendants. The roots the presumption are shallow and I think cannot support its current form, especially in the Ninth Circuit where it is all but irrebutable.