Part I: Why Williams Isn’t the Answer

Sorry, but I can’t let go of the Thomas-Rasset affair quite yet. It is not my habit to criticize court opinions. They are what they are, and the important thing for us lawyers is to understand them. But in this case, I’d like to criticize the Eighth Circuit’s decision in Thomas-Rasset to move toward a better way of thinking about statutory damages in copyright cases.

In between the snarks, the main narrative of my last two posts about Thomas-Rasset was that:

1. The Eighth Circuit had to choose between two different U.S. Supreme Court decisions:

a. Williams, which was directly on point but very old; and

b.  Gore, which was not on point but relatively recent.

2. The Eighth Circuit rationally chose to follow Williams.

3. That choice dictated the result: affirmation of the statutory damages of $9250 per song.

4. Although that choice was defensible, the result felt unjust because it punished the wrongdoer far more than her net worth.

I want to question assumption 1a: that Williams was directly on point. It isn’t. That doesn’t necessarily change the Eighth Circuit’s result, but it does show that there’s much less Supreme Court guidance on the constitutionality of statutory damages than we thought we had, and that the Eighth Circuit’s uncritical reliance on Williams was misplaced.

Once Upon a Time: St. Louis, Iron Mnt. & S. Ry. Co. v. Williams

Long ago, Arkansas used to have a statute that regulated how much railroad companies could charge passengers. Violators were subject to a civil penalty of between $50 and $300 per violation—which was a lot of money back then—plus attorney’s fees. In 1915, Williams and some other plaintiffs sued when they found out they had been overcharged by a mere 66 cents, although, that, too, was a lot of money for an ordinary consumer back then. The trial court awarded the plaintiffs $75 each, plus attorney’s fees, which was on the low end of the range.

One might wonder at this stage: why was Arkansas implementing price controls? Arkansas evidently thought passengers were vulnerable to being gouged by the railways. This may not sound very free-market-oriented, but it’s likely that Arkansas’ citizens were essentially facing monopolies, at a time when anti-trust had not yet taken off. Price controls were the most obvious way of dealing with the problem.

The railway appealed, arguing that the penalty violated due process. No, wait, that’s not QUITE correct. The railway argued that the “provision for the penalty was repugnant to the due process of law.” So, the railway wasn’t arguing that $75 was too much. It was arguing that the $50 to $300 range set by the statute was too much. As the Supreme Court framed the issue:

The ultimate question is whether a penalty of not less than fifty dollars and not more than three hundred dollars for the offense in question can be said to bring the provision prescribing it into conflict with the due process of law clause of the Fourteenth Amendment.

Id. at 66.

This wasn’t the question before the Eighth Circuit in Thomas-Rasset. In that case, the question wasn’t that there was necessarily anything wrong with the $750 to $150,000 range provided in the Copyright Act*, but that the amount that the jury decided on—$9250—was unconstitutionally too high. You can tell from the opinion that Thomas-Rasset made this distinction explicitly, but the Eighth Circuit either didn’t pick up on that, or chose to elide it. More on that later.

* Actually, in a classic case of overreaching, Thomas-Rasset did argue, in the alternative, that the whole statutory scheme was unconstitutional as applied to her because even $750 would be too much.

The Three-Ringed Government

This distinction sounds theoretical, but it might matter a lot because it mirrors a fundamental distinction between two branches of government. Go back to basic Civics (or, if you’re a lawyer, to Con Law I) and separation of powers. What was at issue in Williams was what the legislative branch had done: it had set a certain range of penalties for a certain bad act. What was at issue in Thomas-Rasset is what the judicial branch had done: picking the appropriate figure within that range.

In passing and enforcement of laws, legislatures and courts play very different roles. Legislatures, obviously, pass statutes. They turn policy into law. Laws are by their nature general in application.* Courts, by contrast, apply the laws to a specific set of facts. To the extent the legislature left any such room, courts (and this includes both judges and juries) are to apply their sense of justice, fairness and common-sense. Surely, when Congress sets a range of penalties without guidelines, it’s not telling courts to just go nuts, right? Congress expects courts to do what courts do: mete out justice firmly but fairly.

* Putting aside so-call “private acts.” 

Here, the statute in question is § 504(c) of the Copyright Act, wherein Congress set a range of $750 to $150,000 (for willful infringement) per work infringed. Congress gave no guidelines for determining the amount, leaving courts with wide, wide latitude in setting damages.

The big question is whether the Eighth Circuit applied Williams generally to the range (i.e., to § 504(c)) or specifically to the award of $9250 per song. If the former, the Eighth Circuit is essentially saying that, so long as the jury stays within the $750 to $150,000 range (for willful infringement), courts may not intervene, no matter how nutty the actual award is. If the latter, the Eighth Circuit is simply saying that $9250 per song isn’t so nutty under the circumstances to require intervention (but not really shedding much light as to why).

When the Eighth Circuit was confronted by this issue, it essentially changed the subject:

Thomas-Rasset contends that the range of statutory damages established by § 504(c) reflects only a congressional judgment “at a very general level,” but that courts have authority to declare it “severe and oppressive” and “wholly disproportioned” in particular cases. The district court similarly emphasized that Thomas-Rasset was “not a business acting for profit, but rather an individual consumer illegally seeking free access to music for her own use.” By its terms, however, the statute plainly encompasses infringers who act without a profit motive, and the statute already provides for a broad range of damages that allows courts and juries to calibrate the award based on the nature of the violation.

Do you see how Thomas-Rasset’s point about a court’s ability to declare a specific award “severe and oppressive” turned into a discussion about how to treat non-commercial infringers? But Thomas-Rasset’s crucial question is left unanswered: having held that a “prescribed penalty” is constitutional, may a court nevertheless hold that a particular application of the “prescribed penalty” is unconstitutional, at least where the “prescribed penalty” is described as a range of damages?

The General Specific

It’s actually hard to tell from the opinion. On the one hand, the Eighth Circuit characterized Williams (incorrectly) as holding “that damages awarded pursuant to a statute violate due process only if they are ‘so severe and oppressive as to be wholly disproportioned to the offense and obviously unreasonable.’” That’s a correct reading of Williams only if you substitute “prescribed damages for violation of a statute” for “damages awarded pursuant to a statute.” This characterization implies that Williams may be applied to a specific award. Elsewhere, the Eighth Circuit purported to apply the “Williams standard” to the actual award of $9250 per song.

On the other hand, the Eighth Circuit discusses Congress’ “wide latitude of discretion” (also a quote from Williams) and Congress’ reasons for setting the statutory range (to combat piracy and promote creation). That sounds like it’s addressing the range of penalties, rather than the specific award itself. On balance, I’d say that the Eight Circuit believes that Williams is applicable to a specific award, but that it gives legislatures so much discretion that in setting the range that it doesn’t much of a difference.

The Eighth Circuit takes the trial court to task for substituting what it regarded as an arbitrary award—$2250 per work, or treble the minimum award—for the jury’s award. Why, indeed, should treble the minimum award be the constitutionally maximum award under these circumstances? Thomas-Rasset was a mere consumer with no real commercial ambitions, but she was also a spoliator and a liar and was engaged in a practice that was as widespread as it was difficult to detect. Evidence suggested strongly that, more than your average file-sharer, she knew exactly what she was doing and that it was wrong.

That, alas, doesn’t answer the question of why $9250 is constitutional—or even whether that’s a relevant question to ask. The Eighth Circuit does argue that, if an award for “aggravated … willful infringement” at “near the bottom of the statutory range is unconstitutional…, it would be the rare case of noncommercial infringement to which the statute could be applied.”* But even that argument was made in reference to the statute and not to the award.**

* As an aside: this sentence makes roughly zero sense to me. Is the court suggesting that, if $9250 for a song is too much, then there won’t be enough room between, say, $2250 and $750 for less egregious noncommercial infringements? If so, so what? From an ordinary person’s point of view, there’s a huge difference between $9250 and $2250, and a huge difference between $2250 and $750. Or is the court saying that some works will be so lacking in intrinsic value that even $750 would be unconstitutional? But isn’t one of the points of § 504(c) that, no matter how lame your work is, you always get at least $750 for it, and no matter how innocent the infringement, you’re on the hook for $250 (which is the absolute minimum if the infringement was truly “innocent”)? Besides Thomas-Rasset and her supporters, does anyone really have a problem with that?

** To explain further: the court was responding to an argument from the U.S. Solicitor General that one shouldn’t read the trial court’s decision to reduce the award as a constitutional attack on the statute itself. The trial court, after all, purports to be following Williams. The Eighth Circuit’s point is that a constitutional rule that $2250 was the maximum for non-commercial infringement, even where the conduct is “aggravated” and “willful,” such a rule would wreck the statutory scheme.

Missing: An Analytical Framework

Part of the problem is that Williams doesn’t really provide a framework for looking at specific awards (because that question wasn’t posed in Williams). All Williams tells us to do is look at the range and decide whether Congress had a rational basis for it.* What that question boils down to is whether the lower end of the range is acceptable in the least egregious case you could think of, and (less important) whether the upper end is acceptable in the worst case you could think of. Since that question is so abstract, the answer is almost always going to be yes.

* The Court obviously didn’t use the phrase “rational basis,” but that’s pretty much what the Court was getting at.

Courts don’t like to make constitutional decisions in the abstract. In fact, they’re not really supposed to. In a sense, that’s what the Williams court did: it didn’t look at the $75 award, but at “the overcharge possible in any instance” (emphases mine). It’s not until you confront an actual award set against its actual circumstances—i.e., in Constitution-speak, have a “Case or Controversy”—that you can start to perceive than an award is, somehow, out of whack. And not just out of whack, but unconstitutionally out of whack.

Another problem is that, when you try to apply Williams to a specific award, its analysis breaks down. The Williams Court’s analysis is pretty thin, but it does suggest a couple of factors to look at: (1) the public interest at stake and (2) the “numberless opportunities for committing the offense.”* The first factor is only applicable to looking at the statute as a whole (and even then it’s pretty circular**). The second factor does have specific application to file-sharing, but it’s also one of the main reasons for having statutory damages in the first place. Besides, that can’t be the only relevant factor, surely?

* There’s a third factor: “the need for securing uniform adherence to established passenger rates.” But this seems like a restatement or subset of the first factor. Anyway, the Eighth Circuit all but ignores it, so we will, too. 

** It’s circular because you’re trying to see if the legislature exceeded its power, but your main source for “the public interest” is the legislature. Of the three branches of government, it’s the legislative branch that expresses the public interest. Where else are we to get evidence of the “public interest”? That’s why the inquiry in Williams boils down to a rational-basis test.

Where does this leave us? Williams did not address the question Thomas-Rasset put squarely before the Eighth Circuit, and the Eighth Circuit’s treatment of Williams is actually pretty ambiguous. There are still two fundamental questions for us to answer:

1. As a threshold matter, may judges ever interfere with a jury’s award of statutory damages for copyright infringement?

2. If so, when is this permissible, and how does the judge determine the reduced award?

I will actually suggest answers to those questions in my next post or two.

Thanks for reading!