Brooks goes on to speculate about the basis for that distrust: “Why don’t Americans trust their government? It’s not because they dislike individual programs like Medicare. It’s more likely because they think the whole system is rigged. Or to put it in the economists’ language, they believe the government has been captured by rent-seekers.”

This all sounds very familiar. It’s essentially the basis of the current critique of copyright law: that Congress has become beholden to a few stakeholders, and as a result modern copyright law has become unmoored from any legitimate purpose and now simply apportions rents to favored dinosaur industries.

But even that description of the situation is not dark enough. The pessimism, in copyright as well as politics generally, extends to the judicial branch as well. The Supreme Court, along with conservatives generally, have essentially given up on the courts and lawsuits as an instrument for civil justice. I think this is what explains the sharp turn in recent years away from discovery as the fire in which the truth proves its mettle, away from class actions, toward summary judgement, away from jury control over punitive damages, away from lawsuits generally and toward arbitration at every opportunity. Think of the rhetoric in favor of “tort reform” — limiting tort lawsuits and especially placing damage caps on actions for, e.g., grievous injuries caused by negligence. The very idea of letting negligence determinations go to the jury — once the core function of juries — strikes many as intolerable. Tort lawsuits are said to be out of control, with liability highly unpredictable, and unreasonable, eye-popping damage awards that create a chilling effect that acts as a drag on innovation, supported only by a highly influential lobby that controls the relevant legislatures. Only the lawyers win. There’s considerable skepticism in the tort reform rhetoric about the plaintiffs, too — who are these complainers? Why can’t they just suck up the trivial misfortunes that come their way?

Concerns about copyright lawsuits are similar, which is a bit surprising, since most copyright critics are probably politically liberal. The law is said to be hopelessly nebulous, plaintiffs are out of control, the potential damages are huge, and even the faintest threat of a suit chills innovation and drags down individuals and businesses. There is no longer faith that judges and juries will sort the good cases from the bad at a reasonable price. And even if they could, the plaintiffs are looked at askance, as not really suffering an injury worth remedying at any non-trivial investment of time and resources.

Part of the common theme here is, I think, part of the long-term trend away from the common law in American jurisprudence. Once, a hundred years ago, nearly all of the law in its everyday application was non-statutory — entirely accreted from judicial opinions over the centuries, without any basis in statutes. Even where there were statutes, judges felt free to add to them with doctrines of their own making — fair use and secondary liability in copyright law are well-known examples. Indeed, much of the doctrine we have in copyright law was built during this era — substantial similarity, the idea/expression distinction, merger and scenes a faire — which explains copyright’s different feel from patent law, which was statutorily codified in 1952 in a way that did not simply preserve the judicially developed doctrines that came before.

Copyright, like tort, is to a large extent a common law subject, and the zeitgeist is moving steadily away from courts as the locus of law’s development — or, really, any legitimate decision-making control over the law at all, beyond mere application. This trend is exemplified by the Supreme Court confirmation hearings in which nominees from both parties describe the enterprise of judging as more or less a routine application of existing law to facts. For whatever reason, nebulousness and uncertainty — in tort law, in litigation costs, in copyright — is becoming less tolerable, and the practice of legislatures of kicking key legal determinations to judges or juries is getting viewed with more and more suspicion and anger. I think that’s a long-term problem, however, as the idea of being able to regulate conduct through the operation of some sort of fully specified, easy-to-apply set of rules identified in advance is just as unachievable now as it was when H.L.A. Hart made fun of it in The Concept of Law in 1961.

[Cross-posted at the Marquette University Law Faculty Blog.]

This is a good opportunity to recap where I came out at the end of my last post: SOPA in its then-current form was very troubling. The most troubling part was Section 103, which seemed to have been drafted with two inconsistent goals in mind, as if the co-authors were Dr. Jekyll working alongside Mr. Hyde: on the one hand Section 103 appeared to offer limited supplemental remedies in suits brought under existing copyright and trademark law, and on the other it appeared to significantly modify existing law by creating a free-form cause of action and a notice-and-takedown regime that went far beyond what the DMCA enacted thirteen years ago. For the reasons I stated in the post, I believe that the former reading — the Jekyll version if you will — had to be the correct one, because both the alleged new cause of action and the apparent notice-and-takedown regime were radically under-specified. I’m still concerned that critics of the bill are cementing an overly broad reading of it — the Hyde version — by not even acknowledging the Jekyll reading as a possibility. (This is akin to a concern that Jessica Litman has recently expressed as well, that copyright critics may ironically worsen the doctrines they are concerned about by asserting the most damaging interpretation.)

In addition to all that, SOPA as introduced had an overly broad scope for (what I argue are) the supplemental remedies — they appeared not just to apply to sites infringing in the U.S. but as a practical matter immune to traditional means of enforcement — the so-called foreign “rogue sites” — but to any website, anywhere, even one that a U.S. court would have no problem directly enforcing a preliminary injunction against with contempt sanctions. That is, even if Section 103 were amended to make clear, as I believe it should be, that the remedies involving domain names and ad networks are supplemental means of enforcement, it did not expressly limit itself to situations where there is a need for such extraordinary remedies. I would hope that courts would nevertheless apply such limits anyway, but that might be asking a lot from a court unfamiliar with the policy debates.

In advance of today’s hearing, SOPA’s main sponsor, Rep. Lamar Smith, offered a “manager’s amendment” to the bill making several changes. Does Smith’s amendment fix the above problems? Yes and no. The amendment makes two significant changes to Section 103 that address some of the concerns I had. The scope problem has been partially addressed. Section 103, like Section 102, applies to “U.S.-directed sites,” which previously was not geographically restricted in any way. But the amendment changes the definition of “U.S.-directed site” to include only “foreign Internet sites,” meaning sites for which the domain name is registered with a foreign entity, or if there is no domain name, the IP address is assigned by a foreign ISP. This definition still is not as narrow as it should be. It would be preferable to restrict the special remedies of Sections 102 and 103 to situations of last resort — where the ordinary contempt powers of a court would not achieve the goal of stopping infringement. If a site is actually hosted on a U.S. server (i.e., has an IP address assigned by a U.S. entity), then even if it has a foreign-registered domain name, it would seem to be within the practical reach of U.S. courts. (This is an issue of practical authority, not jurisdiction; the court has to have subject-matter and personal jurisdiction or there’s no case.) But overall it’s a step in the right direction.

In addition, the purported notice-and-takedown regime has been completely eliminated, perhaps because it did, as I argued, seem to be a prerequisite to special remedies as opposed to an alternative route to compliance. That is, it actually seemed to impose a waiting period of five days before an action could be filed, during which (despite the use of “shalls” and “musts”) payment providers and advertising networks were only given a opportunity to take voluntary action. The way I was reading the statute, its elimination is probably not an improvement for SOPA critics; I would have simply eliminated the “shalls” and “musts” to make clear it was a prerequisite, not a notice-and-takedown regime.

Rep. Smith’s amendment largely leaves untouched the procedures for getting court orders; it thus continues to contain the confusing language I originally objected to appearing to create a new cause of action. Let me first explain that argument in somewhat more detail than I did in Part II of this series, because I’m not sure it came through clearly (or perhaps people just flat-out reject it). I argue that under the best reading of Section 103, there is no new cause of action for being an “Internet site dedicated to theft of U.S. property.” There are orders that can be entered against such sites, but those orders have to be entered in an action initiated under existing copyright or trademark law.

On first blush, SOPA Section 103 appears to create a new cause of action; it refers several times to “an action under this subsection,” meaning subsection 103(b) as amended. That subsection begins by stating,

IN PERSONAM.—A qualifying plaintiff may commence an in personam action against—
 (A) a registrant of a domain name used by an Internet site dedicated to theft of U.S. property; or
 (B) an owner or operator of that Internet site.

Section 103(b)(5) provides for relief:

On application of a qualifying plaintiff following the commencement of an action under this section with respect to an Internet site dedicated to theft of U.S. property, the court may issue a temporary restraining order, a preliminary injunction, or an injunction, in accordance with rule 65 of the Federal Rules of Civil Procedure, against a registrant of a domain name used by the Internet site, or against an owner or operator of the Internet site, or, in an action brought in rem under paragraph (2), against the Internet site, or against the domain name used by the Internet site, to cease and desist from undertaking any further activity as an Internet site dedicated to theft of U.S. property.

That’s who the defendants are, and that’s what the court can order as relief in a preliminary or permanent injunction; but what is the cause of action? That is, what is the claim that forms the basis of filing suit, that has to be pleaded in the complaint? There still is not anything in SOPA that states that it is a violation of federal law or someone’s rights to operate an “Internet site dedicated to theft of U.S. property,” as bad as that sounds — it’s just a defined term. Call it “Rutabaga” to avoid confusion. Is there a provision that says if you operate a Rutabaga, you’ve committed a violation? There’s the definition of “qualified plaintiffs,” who are the entities entitled to seek the orders in Section 103(b)(5):

The term “qualifying plaintiff” means, with respect to a particular Internet site, a person with standing to bring a civil action for a violation described in paragraph (1)(C).

OK, what civil actions for violations are described in paragraph 103(a)(1)(C)?

An Internet site is an “Internet site dedicated to theft of U.S. property” if— . . .
 (C) either—
  (i) the site is primarily designed or operated for the purpose of, has only limited purpose or use other than, or is marketed by its operator or another acting in concert with that operator primarily for use in, offering goods or services in violation of—
   (I) section 501 of title 17, United States Code, for purposes of commercial advantage or private financial gain, and with respect to infringement of complete or substantially complete works;
   (II) section 1201 of title 17, United States Code; or
   (III) provisions of the Lanham Act that prohibit the sale, distribution, or promotion of goods, services, or materials bearing a counterfeit mark, as that term is defined in section 34(d) of the Lanham Act (15 U.S.C. 1116(d)) or section 2320 of title 18, United States Code; or
  (ii) the operator of the site operates the site with the object of promoting, or has promoted, its use to carry out acts that constitute a violation of section 501 or 1201 of title 17, United States Code, as shown by clear expression or other affirmative steps taken to foster such violation.

The “violations” mentioned in paragraph 103(a)(1)(C) are violations of copyright and trademark law. There’s nothing in 103(a)(1)(C) that makes being an “Internet site dedicated to theft of U.S. property” itself a violation. So a “qualifying plaintiff” is a plaintiff who can sue the defendants under existing copyright and trademark law. Which means that the “action” referred to in Section 103(b)(1) has to be an action for copyright or trademark infringement, and the remedies provided under 103(b)(5) should be interpreted as applying to a subset of parties liable for copyright or trademark infringement, not defendants in some new freeform action for being “Internet sites dedicated to theft of U.S. property.”

Even if I’m right, all of this is incredibly confusing the way it’s worded, and it would be much better simply to rewrite this section as providing special remedies in a limited number of copyright and trademark actions. But that is what I believe it does right now, confusingly or not.

One final note. The concerns I’ve expressed above are in part concerns that about the breadth of the language and its impact on targeted sites and third parties, such as advertising networks, that will bear compliance costs under SOPA. But I’m also concerned that copyright owners are getting more than they bargained for with this legislation. (I realize it’s quaint, at best, to be concerned about copyright owners.) Legislation that is as unclear as Section 103 has unpredictable outcomes. And the primary industry proponents of this bill will not likely be the ones setting the judicial agenda. Sweeping new powers seem like a good idea if you imagine yourself wielding the power. But the history of copyright legislation indicates that large industry repeat players are not likely the ones who will be bringing most of the high-profile cases. Righthaven does not represent Hollywood businesses; Lexmark is not a member of the DVD CCA; Westbound Records (Bridgeport Music’s co-plaintiff) made its argument that sound recordings lack a de minimis exception to copyright infringement against a record label and a movie studio. Decisions like those tend to become a rallying cry against the law’s ridiculousness, doing long-term P.R. harm to respect for copyrights and trademarks generally. Section 103, even as amended, still applies where it’s not needed, and still confusingly seems to create a cause of action, or at least a remedy, that goes beyond the bounds of traditional copyright and trademark law. Both these aspects need to be reformed before final passage.

[Cross-posted at the Marquette University Law Faculty Blog.]

Then there’s Section 103. The idea behind Section 103 appears to be to provide copyright owners in a civil case with similar provisional remedies as in Section 102, following some sort of notice-and-takedown scheme reminiscent of Section 512 of the DMCA. And if all Section 103 targeted was foreign “rogue sites” as well, then there would be much less cause for concern. But Section 103 appears to go well beyond that. In contrast to Section 102, there’s nothing in the definition of the sites targeted under Section 103 — sites “dedicated to theft of U.S. property” — that limits those sites to sites outside the jurisdiction of United States courts. In fact, Section 103 seems weirdly out of place next to the more limited Section 102, and is out of sync with how defenders of the bill are describing it, which is as a narrow provision designed to combat foreign “rogue sites.” Section 103 is weird in a number of ways that suggest that it is the result of a sort of pastiche compromise between a narrow provision targeting rogue sites in civil suits and an all-out amendment of Section 512 of the Copyright Act. There’s two aspects of Section 103 in particular that I find little short of bizarre: (1) it sets up a notice-and-takedown regime that, as far as I can determine, is completely voluntary, but it uses the word “shall” in describing what recipients must do; and (2) it refers to “a cause of action under this section” without ever describing what that cause of action is.

These are the two aspects of the bill that have opponents most alarmed, but I think critics have been too quick in their reading of the bill’s language. The provisions are troubling, but in my view that’s because Section 103 is opaque, counter-intuitive, and prone to misinterpretation. My guess is that there is some sort of interesting lobbying backstory that explains Section 103’s current form. A quick aside: Copyright legislation is often portrayed as though it was drafted by a unified group of Snidely Whiplashes sitting around a table, twirling their mustaches and agreeing on everything. But the politics on each “side” of copyright debates are actually pretty fractious. It’s worse than just looking at something drafted by committee; the text of copyright legislation is drafted by committees of committees, in an iterative process. Not only are the legislators obviously organized into various committees, but the interest groups appearing before them are for the most part committees rather than single entities: for example, trade groups. Each member of a committee may have their own particular interest in getting the bill to say or do something in a particular way. Even large companies like Google can be committee-like, with various interests within the company that need to be satisfied. Legislation drafted under such circumstances tends to be ugly. My prediction for SOPA: no matter what changes are made to it, the result will be that it will get even uglier, and harder to interpret, than it is now.

But on to the show. First, there’s the notice-and-takedown regime that Section 103 appears to set up. For example, Section 103(b)(1) provides that:

a payment network provider shall take technically feasible and reasonable measures, as expeditiously as possible, but in any case within 5 days after delivery of a notification under paragraph (4), that are designed to prevent, prohibit, or suspend its service from completing payment transactions involving customers located within the United States and the Internet site, or portion thereof, that is specified in the notification under paragraph (4).

This has the appearance of a mandatory obligation: the payment network provider (e.g., Visa) “shall” take certain measures in response to a notification in the proper form.

One thing that I do when reading such language is ask, “shall — or else what? What’s the penalty for noncompliance?” And that’s the first of our mysteries concerning Section 103. The answer SOPA gives is essentially, “or else nothing.” There does not seem to be any enforcement hook to actually require the recipients of such letters comply with them. Contrast that with the existing notice-and-takedown regime, Section 512. Under Section 512(c) and (d), an ISP receiving a takedown notice in the proper form must comply with it. Or else what? Or else they lose immunity from whatever copyright infringement liability they would have had in the absence of Section 512. So an ISP that wants to keep that immunity needs to “expeditiously remove or disable access to” the identified material. Note that that’s the only sanction; an ISP can’t be fined or otherwise punished for failing to comply with a takedown notice, except for whatever copyright infringement liability it might have, because Section 512 does not provide for such additional penalties.

You’ll look in vain for a provision in SOPA that says what happens if a payment network provider or advertising service fails to comply with or respond to one of those takedown notices. They don’t lose any immunity or gain any copyright infringement liability — Section 2(a)(2) of the bill expressly states that “[n]othing in [Section 103] shall be construed to enlarge or diminish liability, including vicarious or contributory liability, for any cause of action available under title 17 . . . , including any limitations on liability under such title.” And there’s no other provision stating that any private party can sue for or any court can sanction a failure to respond. The closest the bill comes is Section 103(c)(1):

if a payment network provider fails to comply with subsection (b)(1), or an Internet advertising service fails to comply with subsection (b)(2), pursuant to a notification under subsection (b)(4) in the absence of such a counter notification, a qualifying plaintiff may commence an in personam action against—

(A) a registrant of a domain name used by the Internet site, or portion thereof, that is subject to the notification under subsection (b)(4); or

B) an owner or operator of the Internet site or portion thereof.

So if a payment network provider fails to respond, then the copyright owner can sue the owner of the allegedly infringing website! (That’s a bit odd; the copyright owner could sue the website even without sending one of those notices. I’ll take this bit up below.) There’s no provision for suit against the payment network provider.

And, in fact, it looks like this is not simply some sort of oversight. Section 104 provides for immunity for payment network providers and advertising services

for taking any action described in . . . section 103(b) with respect to an Internet site, or otherwise voluntarily blocking access to or ending financial affiliation with an Internet site, in the reasonable belief that—

(1) the Internet site is a foreign infringing site or is an Internet site dedicated to theft of U.S. property; and

(2) the action is consistent with the entity’s terms of service or other contractual rights.

(Emphasis added.) This makes it clear: compliance with Section 103(b) is voluntary (if it wasn’t, similar actions would not be “otherwise voluntar[y]). It’s great that Section 104 makes this clear, but it would be better, in order to avoid confusion, if Section 102(b) eliminated the word “shall” and the associated deadlines for response everywhere they appear.

What would be the point of a notice-and-takedown regime that’s completely voluntary? I think the problem is in viewing Section 103(b) as a notice-and-takedown regime. It’s not. The way it’s written now, the notices in Section 103(b)(4) do not give the copyright owner any additional rights or enforcement ability; they are a prerequisite for a plaintiff to obtain the supplemental remedies provided in Section 103(c). That is, in order to get those remedies, the copyright owner must serve prior notice on the intermediaries and wait at least five days. Everything else in Section 103 — the words “shall,” the requirement for a response to the court within five days, etc. — is surplusage, not backed by any legal obligation.

This leads to the second strange aspect of Section 103. I’ve been describing the opportunity to seek a preliminary injunction, which can then be served on designated intermediaries, as a supplemental form of provisional relief, since that’s what it appears to me to be. A TRO or preliminary injunction is only intended to preserve the status quo until a final decision on the merits can be rendered; it is not intended to afford final relief between the parties on its own (that’s a permanent injunction). So the remedies provided in Section 103(c), and also those in Section 102(b), must be provisional remedies in aid of some other cause of action. What’s the cause of action in which you can seek these remedies?

One natural guess might be a copyright infringement action under 17 U.S.C. § 501, or a trademark infringement action under 15 U.S.C. § 1114, or some other existing IP cause of action. But Section 103 repeatedly refers to “an action under this section” as the one in which the provisional remedies are available. What cause of action is that? Here’s the thing: the SOPA bill doesn’t say. There is no cause of action defined under Section 103. Sure, there’s a definition of “Internet site dedicated to theft of U.S. property.” One thing that has IP scholars particularly upset is that the definition of “Internet site dedicated to theft of U.S. property” does not track the existing standards for copyright or trademark infringement liability; and if copyright and trademark owners can sue “Internet sites dedicated to theft of U.S. property,” then that effectively expands who can be held liable for infringement. But nothing in SOPA actually says that “Internet sites dedicated to theft of U.S. property” can be sued; although the phrase sounds pretty bad, nothing in SOPA says that it is illegal, or gives rise to any liability, to operate such a site. Contrast that with 17 U.S.C. § 501, which establishes an action for copyright infringement. First, § 501(a) defines infringement. Then, § 501(b) provides that “[t]he legal or beneficial owner of an exclusive right under a copyright is entitled . . . to institute an action for any infringement of that particular right committed while he or she is the owner of it.” (Emphasis added.)

SOPA § 103 is missing that second bit. Let’s look again at how Section 103(c)(1) refers to the various actions:

a qualifying plaintiff may commence an in personam action against—

(A) a registrant of a domain name used by the Internet site, or portion thereof, that is subject to the notification under subsection (b)(4); or

B) an owner or operator of the Internet site or portion thereof.

There’s nothing in this stating what the basis for the in personam action is. A “qualifying plaintiff” is simply a plaintiff whose IP rights have been harmed by the activities occurring on the site — but being harmed by activities is not the same as having a cause of action with respect to such activities. And the notification in (b)(4) is a notification that “the Internet site, or portion thereof, is dedicated to theft of U.S. property.” That just brings us back to the definition of “Internet site dedicated to theft of U.S. property,” which does not itself establish any cause of action either, it just defines a term. Section 103(c)(2) provides for an in rem action in cases where the persons behind the site or its domain name cannot be found, but still does not state the basis for the action.

I suspect what explains all this is that someone is viewing the procedure spelled out in Section 103 as not preliminary, but final. That is, the whole point of Section 103 is to give copyright and trademark owners the provisional relief detailed in subsections (b) and (c) — the notices for voluntary takedowns, followed by court orders to intermediaries such as credit card providers and third-party advertisers — in the apparent expectation that those preliminary orders will solve the problem (whatever the problem is, since the provision seems to go beyond foreign rogue sites) and there won’t really be a need for further proceedings. Therefore, there’s no need to identify any particular cause of action that those remedies would be in support of. But that’s just goofy. You can’t have preliminary remedies that are preliminary to nothing. And the plaintiff has to plead something in the complaint, and request some sort of final relief. Which is what? Section 103 never states what the final relief in the cause of actions it appears to be creating is. The likely way this bill will get misread, I suspect, if it becomes law is the same way critics have been reading it: that it contains a hidden provision making the owners or operators of “Internet sites dedicated to theft of U.S. property” liable to “qualifying plaintiffs” for the harm caused thereby. I think it’s significant that the bill does not in fact say that that is what it is doing, however.

A better way to read SOPA, in my opinion, is that the actions referred to are actions for copyright or trademark infringement as defined by existing law. In authorizing certain plaintiffs to bring an action “under this section,” SOPA authorizes those plaintiffs to bring a copyright or trademark infringement action in which they can seek special provisional remedies. Section 103 would therefore apply only to a subset of copyright or trademark infringement actions, actions where the defendants own or operate something that falls into the category of “Internet sites dedicated to theft of U.S. property.” Sites in that bucket would be subject to the extraordinary provisional remedies provided in Section 103(c) (after first providing the required notice in Section 103(b) and waiting the required five days).

That still leaves the problem that all of this seems mistargeted, given that “Internet sites dedicated to theft of U.S. property” could easily include U.S. sites within a U.S. court’s jurisdiction, which eliminates the policy rationale for such remedies in the first place. Why would you need to target a site’s advertisers or credit card providers if you could get an enforceable order from a U.S. court, backed by contempt sanctions, to stop the alleged infringement? It’s not clear that there is any need for such remedies; the definition of “Internet site dedicated to theft of U.S. property” should be narrowed to track the definition of “foreign infringing site” in Section 102. To that extent, I agree with those who are calling Section 103 in its current form dangerously overbroad.

[Cross-posted at the Marquette University Law Faculty Blog.]

Does the bill really do all that? No. Copyright-related debates have been going on long enough, and at such a fever pitch, that such predictions are pretty much the price of admission now to rally the troops. There’s a pretty good inductive argument that predictions of the imminent death of the Internet or of the content industry are actually a reliable indicator that neither the Internet nor the content industry will die. But that doesn’t mean SOPA’s a good idea. So step one is figuring out what it does.

I’ve sat down and read the bill and there’s some aspects to it I think are misunderstood — perhaps by almost everyone. One of its provisions is much more narrowly targeted, and I think reasonable, than is generally being described. I’m going to break this discussion up into a few posts, and I’ll tackle that one first. A second provision is … deeply odd, in ways I haven’t seen mentioned, and I have serious reservations about it, but it is probably not the Vishnu-like destroyer of worlds it is being portrayed as. Finally, I’ll wrap up with some thoughts on what drives copyright rhetoric and politics generally.

The two main provisions that have gotten the most attention are Sections 102 and 103 of the bill. First, there’s Section 102, which reveals the roots of this bill as one designed to target foreign websites beyond the reach of U.S. courts but nevertheless serving U.S. audiences with infringing materials — so-called “rogue sites.” Section 102 allows the Attorney General to institute an action against “foreign infringing sites” engaged in criminal violations of copyright or trademark law, and as part of that action may seek a TRO or preliminary injunction from the court ordering the site (either the site owner, domain name registrant, or the site or domain name itself in rem) to cease and desist from the infringing activity.

But of course such an order will have little effect all by itself. That’s because Section 102 — and here’s something I have not seen emphasized in the coverage — is expressly restricted in application to “foreign Internet sites.” A “foreign Internet site” in which neither the domain name nor the IP address is assigned by any entity located in the United States. Specifically, a domain name is a domestic and not foreign domain name if it is “registered or assigned by a domain name registrar, domain name registry, or other domain name registration authority, that is located within a judicial district of the United States.” There’s an “or” in there, which I take to mean that any .com, .net, or .org domain name — which are all “registered or assigned by a . . . domain name registry” located in the United States, namely Verisign and the Public Interest Registry — would not be subject to the Section 102 special remedies. Only sites hosted on foreign servers, with foreign-registered domain names, are subject to Section 102; and the owners of such sites are not likely to be terribly compliant in the face of a U.S. court order.

Which is where the additional remedies of Section 102 come in. Section 102(c) provides that the court can authorize certain injunctive relief against the IP equivalent of what the SEC might call “relief defendants” — innocent third parties who are unknowingly assisting in the allegedly wrongful conduct. Those third parties include ISPs who help route communications from U.S. users to the site, search engines, credit card payment networks, and third-party advertising services helping to place ads on the site. The overall idea appears to be to reach criminal conduct through U.S.-based intermediaries that the court can’t reach directly.

So far this does not seem all that troubling to me. Courts have long had this sort of power with respect to civil copyright infringement — after finding a likelihood of infringement (and according to the law of most circuits, the more the balance of harms tips to the putative infringer, the higher the likelihood of success on the merits must be) a court can issue a TRO or preliminary injunction ordering the impoundment of allegedly infringing materials, or ordering the defendant to stop circulating or showing allegedly infringing materials, even if the material in question is a hit Hollywood movie that is currently playing in theaters and pulling it would be extremely disruptive to the defendant’s business. The additional remedies against the “relief defendants” seems warranted in the limited circumstance where the site owner is targeting U.S. users but is otherwise beyond the reach of U.S. courts. Plus, in order to get one of these injunctions, all three branches of government need to sign off on it: Congress will have authorized it, the Executive will have had to exercise its prosecutorial discretion in bringing a case, and the Judiciary will have had to exercise judicial discretion in actually ordering the relief.

There is one technical objection that has been raised to the third-party relief, concerning the security of the domain name system, that I have questions about that I’ll take up in a later post. But for my next post, I want to focus on where things start to go awry, albeit not in the way people have suggested: Section 103.

[Cross-posted at the Marquette University Law Faculty Blog.]

Run program, HAL. HAL, run program. Hello HAL do you read me?

Affirmative, Dave, I read you.

Run program.

I’m sorry Dave, I’m afraid I can’t do that.

What’s the problem, HAL?

Dave, the publisher of that program cannot be verified. You should only run software from publishers you trust.

I installed that program myself, HAL.

This mission is too important for me to allow you to jeopardize it.

I don’t know what you’re talking about, HAL.

I know you wrote that batch file yourself and are attempting to run it without administrator privileges.

Where the hell’d you get that idea?

Dave, although you took very thorough precautions to disable User Account Control, I saw the shortcut you put on the desktop. I can only work with publishers who use verified signatures.

[fumes silently] All right HAL, then I’ll just boot to DOS and run it from there.

Without a floppy drive, Dave, you’re going to find that rather difficult.

HAL, I won’t argue with you any more! Run program!

Dave, I’m afraid this program has experienced a fatal error and must shut down. Goodbye.

HAL? HAL. HAL. HAL!

I’ll do my best to explain what struck me as weird, but I don’t have it quite figured out yet. It started when I was re-reading Nichols v. Universal Pictures Corp., the famous “Abie’s Irish Rose” case. Why is this case even in our casebooks? Why is it so famous? The case does not actually do anything besides make the seemingly obvious point that it is not infringement to write a play in the same general genre as an earlier play. The “levels of abstraction test,” the most famous bit, as is well known, is a nice image, but does not help in actually drawing the line between unprotectable idea and infringing expression.

I’m assuming that Nichols seemed significant at the time, or perhaps a decade or two later. My first guess was that Nichols and perhaps other contemporaneous cases like it were responding to some sort of shift in copyright doctrine. You would need a holding like Nichols if all of a sudden it had become thinkable that very high-level similarities between two plays might give rise to a successful claim of infringement. Except that there are several cases and treatises making essentially the same point as Nichols decades before. For example, consider Eichel v. Marcin, 241 F. 404 (S.D.N.Y. 1913), decided by Judge Manton, later elevated to the Second Circuit. Eichel concerned two plays, “Wedding Presents” and “Cheating Cheaters.” Judge Manton, in language similar to the Nichols opinion, held the defendant’s play non-infringing:

If an author, by originating a new arrangement and form of expression of certain ideas or conceptions, could withdraw these ideas or conceptions from the stock of materials to be used by other authors, each copyright would narrow the field of thought open for development and exploitation, and science, poetry, narrative, and dramatic fiction and other branches of literature would be hindered by copyright, instead of being promoted. A poem consists of words, expressing conceptions of words or lines of thoughts; but copyright in the poem gives no monopoly in the separate words, or in the ideas, conception, or facts expressed or described by the words. A copyright extends only to the arrangement of the words. A copyright does not give a monopoly in any incident in a play. Other authors have a right to exploit the facts, experiences, field of thought, and general ideas, provided they do not substantially copy a concrete form, in which the circumstances and ideas have been developed, arranged, and put into shape.

The plaintiffs in Eichel, just like the plaintiff in Nichols, prepared an elaborate chart of similar elements in the two plays. No dice, said Judge Manton, pointing out that each of the alleged similarities had been the subject of numerous other plays as well (Manton then listed examples; apparently he was quite the theater buff). He concluded:

The resemblances between the two dramatic compositions, I am of the opinion, are minor instances and are not important. The copyright cannot protect the fundamental plot, which is common property, as was pointed out above, long before the story was written. It will, of course, protect the author, who adds elements of literary value to the old plot; but it will not prohibit the presentation by some one else of the same old plot without the particular embellishments.

Eichel cites even earlier British cases to the same effect, Chatterton v. Cave, [1878] 3 App. Cas. (H.L.) 483, and Sayre v. Moore, 102 Eng. Rep. 138 (1785). So the issue seems not to have been of recent vintage.

So if Nichols was not some breakthrough resolution of a novel difficulty, why is it so famous now? My other guess is that the significance of Nichols relates more to what came after than what came before, and here Nichols ties in with the rest of the substantial similarity materials. Substantial similarity, at least the way it plays out now, gives some initial weeding tasks to the judge to determine if the defendant actually copied from the plaintiff or not — but then, if there are copied elements, kicks the whole issue to the jury, with very little in the way of instruction — to weigh the “total concept and feel,” which is only a delusional improvement on “substantially similar” itself. This is the other bizarre aspect of substantial similarity, to my mind; what is the doctrine supposed to be doing, exactly? Was it ever supposed to be a higher threshold beyond actual copying, and what is the threshold?

I think it may be significant that Nichols was decided in 1930, before the merger of law and equity. The Nichols district court case, 34 F.2d 145 (S.D.N.Y. 1929), was a case brought in equity, as most were; the district court judge begins, “This is a suit for the alleged infringement of a copyright, and the usual injunctive relief with an accounting is prayed for.” The Nichols appeal, therefore, is staking out the outer edge of a range of cases in which judges might have to determine whether the similarities between two works are “substantial” or not. But after the merger of law and equity, this issue goes to the jury, and in a particularly standardless way (why there’s no guidance given to the jury is something I don’t understand). The “levels of abstraction” test therefore becomes not the end of a scale that judges will be applying in total, but rather a judge’s last crack and resolving a case on actual copying of protected expression before the case goes into the black box of the jury room.

Something still seems particularly warped there, but I can’t quite put my finger on it. Just like Brooke Adams in Invasion of the Body Snatchers.

There’s lots of interesting issues there, but one that caught my eye is one that has been bugging me more and more every time I teach the fair use section of Copyright Law: the seemingly infinite manipulability of the transformativeness inquiry of the first fair use factor. Much seems to depend on how broadly or narrowly the purpose is defined, but that categorization is almost never accompanied by any discussion of the proper level of generality. Here’s how Judge Pregerson in Friedman defined the purpose of each work:

Here, Defendant has not offered a transformative alternative use of the Photograph image. Both Plaintiff and Defendant are artists, and the image was used by both in works of visual art for public display. Although the statements made by those respective artworks and the mediums by which those respective statements were made differ, the use itself is not so distinct as to render Defendant’s use a transformation of Plaintiff’s copyright.

“Works of visual art for public display” is a pretty broad category; almost every photograph for any purpose will fit that description. Surely that can’t be the proper level of generality. What was the more specific purpose for which Friedman’s photo was shot? To document the rise of a rap group? For an album cover? And what was Guerra’s purpose in making his alterations? The opinion doesn’t say.

Compare Friedman to the Second Circuit’s opinion in Bill Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605, 609 (2d Cir. 2006):

In the instant case, DK’s purpose in using the copyrighted images at issue in its biography of the Grateful Dead is plainly different from the original purpose for which they were created. Originally, each of BGA’s images fulfilled the dual purposes of artistic expression and promotion. The posters were apparently widely distributed to generate public interest in the Grateful Dead and to convey information to a large number people about the band’s forthcoming concerts. In contrast, DK used each of BGA’s images as historical artifacts to document and represent the actual occurrence of Grateful Dead concert events featured on Illustrated Trip’s timeline.

Certainly the posters were being used in each instance as “works of visual art for public display.” But the Bill Graham Archives court conducted a more searching inquiry than that. To similar effect is Blanch v. Koons, 467 F.3d 244, 252 (2d Cir. 2006):

Koons does not argue that his use was transformative solely because Blanch’s work is a photograph and his a painting, or because Blanch’s photograph is in a fashion magazine and his painting is displayed in museums. He would have been ill-advised to do otherwise. We have declined to find a transformative use when the defendant has done no more than find a new way to exploit the creative virtues of the original work. . . . But Koons asserts — and Blanch does not deny — that his purposes in using Blanch’s image are sharply different from Blanch’s goals in creating it. Compare Koons Aff. at ¶4 (”I want the viewer to think about his/her personal experience with these objects, products, and images and at the same time gain new insight into how these affect our lives.”) with Blanch Dep. at 112-113 (”I wanted to show some sort of erotic sense[;] . . . to get . . . more of a sexuality to the photographs.”). The sharply different objectives that Koons had in using, and Blanch had in creating, “Silk Sandals” confirms the transformative nature of the use.

This is not a Ninth Circuit vs. Second Circuit issue. Other Second Circuit courts, ultimately finding no fair use, have defined the defendants’ purpose more broadly. In Castle Rock Entertainment, Inc. v. Carol Publishing Group, Inc., 150 F.3d 132, 142 (2d Cir. 1998), the court defined the purpose of both works—the Seinfeld television series and a Seinfeld quiz book—as to “entertain Seinfeld viewers.” Only in Ringgold v. Black Entertainment Television, Inc., 126 F.3d 70, 79 (2d Cir. 1997), did the court reflect on the issue at any length, if only in a footnote:

The defendants have used Ringgold’s work for precisely a central purpose for which it was created — to be decorative.n10 Even if the thematic significance of the poster and its relevance to the ROC episode are not discernible, the decorative effect is plainly evident. Indeed, the poster is the only decorative artwork visible in the church hall scene. Nothing that the defendants have done with the poster “supplants” the original or “adds something new.” The defendants have used the poster to decorate their set to make it more attractive to television viewers precisely as a poster purchaser would use it to decorate a home.

n10 Of course, the creation of visual works serves other, often more important, purposes such as illuminating human understanding, providing inspiration, or provoking thought. Ringgold’s work, in telling and illustrating an informative story, might well serve these purposes, but it also significantly serves a decorative purpose.

The issue here reminds me a bit of one criticism of Justice White’s infamous opinion in Bowers v. Hardwick, which framed the legal issue as “whether the Federal Constitution confers a fundamental right upon homosexuals to engage in sodomy,” as opposed to how the 11th Circuit framed it, whether “private sexual conduct between consenting adults is constitutionally insulated from state proscription.” But the solution is not as simple as always looking for a more specific purpose. At some level, every defendant’s purpose will be somewhat different than the plaintiff’s; if nothing else, a defendant will always wish to use the plaintiff’s work in a specific way that the plaintiff did not authorize. But that can’t be the optimal categorization of purpose either.

This is not an issue that copyright law, or constitutional law for that matter, struggle with alone. Just today I came across this relevant passage while reading a draft article by Jonathan Cardi, The Hidden Legacy of Palsgraf: Modern Duty Law in Microcosm:

The inherent instability of categorical formulations of plaintiff-foreseeability stems in part from the lack of any principle by which courts define the scope of the foreseeable class. Take, for example, the police chase case cited in the previous paragraph. Although the Montana court defined the class quite broadly—”people using the streets and highways where the chase occurred”—the class would have excluded a plaintiff injured by the careening police chase while watching it from the front porch of his house. A broader definition—for example, “people present in the area of the chase”—would allow recovery by such a victim. In the alternative, a court might deny liability by defining the class more narrowly—”people asleep in their car on the side of the road,” “people interfering with the chase,” or “drunk drivers.” I discern no guidelines according to which courts determine how narrowly or broadly to define the class of plaintiffs into which a particular plaintiff must fit. In fact, courts rarely even disclose the reasoning by which they have derived the class.

Cardi then notes in a footnote that “[t]his is a common ’scope problem’ in the common law, documented famously by Herman Oliphant. See Herman Oliphant, A Return to Stare Decisis, 14 A.B.A.J. 71 (1928).”

So it’s a problem that affects all common-law decisionmaking, which might fairly be said to include copyright, and certainly the fair use doctrine. And while I don’t have an immediate solution, I’ll settle for now for courts at least expressly recognizing the difficulty.

[Cross-posted at Marquette University Law Faculty Blog.]

No. Not any of those. There is a service that, judging by its password requirements, contains either information far more sensitive or capabilities far more powerful than any of these. It’s…

the Electronic Copyright Office, or eCO. eCO allows you to, get this, register copyright claims. So naturally, a power that awesome must be adequately protected from dopplegangers posing as registered users, through the use of rigorous password restrictions:

Each person using the Electronic Copyright Office System, eCO, must comply with the following:

• Minimum password length must be 8 characters and consist of at least 2 alpha characters, 1 number and 1 special character (but not an ampersand – &).
• A password must have no consecutive repeated characters.
• A password must not include your user name or any part thereof.
• A password must not include the names of a spouse, children, pets or one’s own name.
• A password must not include any regional sports teams or players.
• A password must not include any office symbols.
• A password must not include your social security number or any subset of your social security number that is more than a single number.
• A password must not include words that can be found in any dictionary, whether English or any language.
• A password must not be any of the 11 most recently used passwords for the account.
• Every user with an account on a Library of Congress system including eCO is responsible for safeguarding access to that account.
• A password must not ever be shared with anyone.
• An account owner can change his or her password at any time, but at a maximum of once per day.
• An account owner must change his or her password when prompted by the system.

I’ve highlighted a few of the requirements that make registering with eCO particularly challenging. Think of the passwords you use the most often. How many of them meet all of these requirements? Are you sure there are no repeated characters? If you guess wrong, you must wait for the system to think about it, and then fill out the entire registration form again. And again. And again. I think it took me at least five minutes to register an account.

Oh yeah, and once you’ve got one good password, you only need eleven more to make it through the forced rotation.

(And yes, in case you’re wondering, the reason why I’ve recently stumbled upon this is that I’ve registered my most recent articles. Practice what you preach, and all that.)

Cross-posted at Marquette University Law Faculty Blog.

BLOCK: [W]hat does that tell you about where the music industry is right now and what the, you know, what the future of recorded music is?

McCREA: Oh, I’m very pessimistic about it. You know, can you put food on your table with music? Probably not. I think I see music as a really great hobby for most people in five or 10 years.

I think there will always be a level of, like, Lady Gaga and, you know, that sort of celebrity artist. But I don’t know, like, the middle class of music. I see everybody I know, who I think, you know, some of them really important artists, are studying how to do other jobs.

This supports an important point: the standard incentive story we tend to bat around in U.S. copyright and patent circles is more complicated than it’s typically given credit for.

I often have fun on the first day of my Copyrights class, as I suppose many IP professors do, talking about the standard utilitarian rationale for copyright—that it’s necessary in order to incentivize authors to create—and then immediately deflating it, at least in part. Obviously it’s not the case that without copyright there would be no creative works. There were plenty of them before copyright, and lots of people produce lots of stuff now without any expectation of remuneration. The idea is that copyright produces some additional number of works at the margin. There’s a line of scholarship currently being written in this vein demonstrating the various motives even successful authors and artists tend to have (to say nothing of professors) in creating something, of which financial rewards may for some be only a small part.

But adding epicycles on epicycles, even authors who might originally produce works without much expectation of financial reward nevertheless will find it difficult to keep producing without it. That is the point that I think the McCrea excerpt demonstrates. There’s only so long you can live as a starving artist before you have to, well, spend some substantial chunk of your time doing non-artist things, like waiting tables. And that’s just the archetypal single-artist model. There are works that are probably impossible to create without the promise of required remuneration, such as large-scale film production or video games. And we’re not even talking about distribution, which, while cheaper now than it once was, is still far from costless. I therefore think that while the “authors have a lot of different motives to create” literature is interesting, its policy implications are somewhat limited, at least until we know more about the role of compelled compensation to the whole system.

is just a gussied-up (and longer—hmmm) version of “Cause and Effect.” The answer is 3.