Author: Margot Kaminski


The Leaked TPP: Some notes, and Criminal Copyright

Today Wikileaks released the latest leaked draft of the IP chapter of the Transpacific Partnership Agreement (TPP). The TPP is a plurilateral trade agreement being negotiated between select countries across the Pacific, and is the latest in a longish line of free trade agreements that address IP law in great detail.

Free trade agreement negotiations are subject to a surprising amount of secrecy, but select industry advisers have access to detailed US negotiating objectives. Today’s leaked text, dated May of 2014, allows public interest groups and the general public a chance to play catch-up to US industry advisers.

I suspect that the drafts are often leaked, or “pleaked“,  by negotiating partners seeking to activate US public interest groups more sympathetic to their negotiating positions than to the positions proffered by the US. The leaked text shows evidence that such pleaking can work: more controversial provisions from older leaked drafts, including one in direct conflict with the Supreme Court’s decision on first sale doctrine in Kirtsaeng, have been removed. And the repeated leaking raises the question of why the US doesn’t voluntarily open the process up earlier- if leaking is now so prevalent that we’re going to see what’s in the agreements anyway, why insist on the dirty haze of secrecy and keep the public one step behind?

A number of people at public interest organizations (KEI, Public Citizen) have criticized the leaked text’s patent provisions for blocking access to medicines. The text also contains extensive provisions on geographical indications, and public performance rights, each of which raise significant issues (especially performance rights in light of the controversial Ninth Circuit decision in Garcia– for more on this issue in the international context, see Jacob Victor’s just-released essay).

The leaked draft also contains language on trade secrets that could be read to require a private cause of action, and may end up being used to launder federal law in the current debates over whether the U.S. should create a federal trade secrets private cause of action.

But the portion of the draft I would like to highlight is its provisions on criminal copyright law. Criminal copyright is a large part of what got ACTA rejected in the EU. As many have noted, the United States does not have the greatest interface between its copyright law and free speech rights, thanks to Supreme Court decisions in Eldred and Golan. As a consequence, Congress has been able to legislate into existence criminal copyright law that impinges on free speech values both by defining a low level of infringement as criminal, and by employing enforcement tools (such as the seizure of websites) that restrict freedom of expression.

The leaked draft shows a fight between the US and Canada over the scope of criminal copyright infringement and enforcement. Canada wishes to clarify that countries may restrict the criminalization of infringement to truly commercial scale infringement (see fn 183), while the US, per our domestic statute (the NET Act), wishes to reach noncommercial acts. The US fought China on this exact issue at the WTO and lost, so has been using free trade agreements to try to raise the TRIPS standard. The TPP parties are evidently debating whether to include the following: “For greater certainty, “financial gain” does not obligate a Party to provide criminal procedures and penalties in cases of de minimis infringements.” My guess, given past FTA language, is that the US comes down strongly against this footnote.

The underlying standard for criminal copyright infringement matters because it brings with it a host of enforcement tools, and active involvement by the state, thus implicating privacy in addition to free speech.

And when the low standard gets exported from the US, it can result in unintended consequences abroad. In Colombia, which enacted its current criminal copyright law pursuant to its free trade agreement with the US, a graduate student was arrested for posting a fellow scholars’ academic paper online without permission. Annemarie Bridy points to this prosecution as the consequence of US trade policy. And while the EFF is optimistic that the student should be acquitted under Colombian law, the case shows that free-speech fears over criminal copyright enforcement are real.

Until enhanced criminal copyright law comes off the US trade agenda, free trade agreements will continue to face opposition from those concerned about free expression and privacy online.


Drones and Newsgathering at the NTSB

Who would have predicted that a First Amendment amicus brief on behalf of national news organizations would be filed at the National Transportation Safety Board (NTSB)?

News Media filed the brief a few days ago in support of drone filmmaker Raphael Pirker. Pirker flew a drone, made a video, distributed it, and was fined by the FAA. In response, Pirker challenged whether the FAA’s notice banning the commercial use of even small drones meets the criteria for valid rulemaking. An ALJ invalidated Pirker’s fine, and the NSTB is currently reviewing the decision. This is an administrative law case about rulemaking and FAA definitions. But the News Media brief shows that as the FAA, Congress, and states decide how to regulate drone use, First Amendment concerns will inevitably be raised.

Drone regulation implicates the First Amendment because drones carry recording devices. A number of courts have recognized a limited First Amendment right to record, although it’s important to also note that some courts have not. As drone regulations are enacted, courts will need to figure out how broadly the right to record extends, and which government regulations do and don’t implicate it.

The News Media amicus brief takes a slightly different approach.  It urges the NTSB to “safeguard the public’s First Amendment interest in the free flow of information.” A number of cases, most notably Branzburg v. Hayes, speak to the importance of protecting newsgathering. Newsgathering protection is really protection of free speech infrastructure. Protecting speech without protecting free speech infrastructure could result in government use of other regulatory tools like taxes or roadblocks to shut down speech as effectively as censorship.

Unfortunately, the News Media brief overstates its case. The NTSB should certainly consider the First Amendment implications of FAA rules. But the FAA’s general ban on commercial drone use is just that: a general ban. The brief recognizes that the FAA’s “de facto policy” is the “almost complete prohibition on the civilian use of UAS for any purpose, including First Amendment purposes.” By this argument, the government isn’t targeting the press; it’s including them with everybody else.

Generally applicable regulations, such as labor laws, can be applied to the press. “The publisher of a newspaper has no special immunity from the application of general laws.” (AP v. NLRB) But if the government uses such laws to specifically target free speech infrastructure, ie with press-specific taxes, that targeting violates the First Amendment (Grosjean v. AP). The News Media brief suggests that the press deserve an exception from a generally applicable rule, not that the FAA has impermissibly targeted the press.

The centrality of “the press” in this argument also should give pause. The brief teems with press exceptionalism. It asks the NTSB to consider the First Amendment newsgathering rights of “professional news organizations” (at 9) and “accredited news media” (at 10). We all know this is not the way news is gathered now. It’s not even how drone journalism has worked. Many newsworthy drone videos have been crafted by hobbyists (at 13). But the brief argues that “the use of UAS for newsgathering should receive greater protections than those afforded to hobbyists and commercial users,” against the backdrop that the brief understands “newsgathering” to be newsgathering by professional organizations (at 12). This just won’t (pardon the expression) fly. Sometimes the government creates special exceptions for the institutional press (at 10-11), but institutional press access is often established in situations where the government just can’t let everyone in. FAA regulation of drones isn’t quite the same. In some scenarios, the government can’t let everyone in at once for safety reasons, but in others, that limitation won’t be at play. Encouraging the FAA to carve out exceptions for professional newsgatherers and not for hobbyists potentially creates rather than solves a First Amendment problem. It could discriminate between actors, and likely discriminate between viewpoints.

The problems in the brief don’t mean that the First Amendment has no role to play here. The two most interesting questions raised by the brief are as follows: should the FAA be able to fine newspapers that distribute drone hobbyist footage? (at 21) And might the FAA’s decisionmaking be an example of impermissibly ad hoc and opaque delegation? (at 23)

The first question- whether the FAA can fine newspapers that distribute hobbyist footage- strikes at the challenging core of Bartnicki v. Vopper. A broad reading of Bartnicki is that government regulations of the distribution of information are subject to strict scrutiny and thus likely will be struck down by the First Amendment (see also Stevens). A narrower reading is that the government cannot regulate the distribution of information of public interest (vs private info), as long as the publisher took no part in the illegal obtaining of that information. The fact pattern suggested here- a newspaper publishing information legally obtained by a hobbyist under the FAA’s model aircraft exception- could force a reexamination of what Bartnicki means by “obtained lawfully”.  The hobbyist’s making of the video is not itself unlawful, but it’s unlawful to fly a drone for business purposes, so the newspaper’s participation in distributing the video arguably makes both that distribution and the initial drone flight unlawful. It’s a mess.

The question of whether the FAA’s decisionmaking process is impermissibly ad hoc, opaque, and subjective is similarly fascinating. Opaque, ad hoc, and subjective policymaking has been found to violate the First Amendment because it leaves too much discretion, and thus room for discrimination, in the hands of local authorities (the brief cites the Ninth Circuit case Foti v. City of Menlo Park). The FAA’s ban on commercial drone use doesn’t seem to raise this issue, since the FAA bans everyone. But the FAA’s drone licensing process for universities might raise this concern (huge caveat: I don’t know that much about it). Or if Congress or the FAA were to put in place a broader drone licensing system, it would have to make sure that FAA officials aren’t giving licenses based on the content or viewpoints of particular organizations.




Fixing International IP Capture? Some Problems with the Public Interest Trade Advisory Committee (PITAC)

U.S. Free Trade Agreements contain skewed intellectual property law because the Office of the U.S. Trade Representative (USTR) has been captured through asymmetric information flow.  I’ve written about this information flow problem at the USTR and its substantive consequences. The USTR’s lack of voluntary transparency, combined with its use of FOIA exemptions, exclusion from the APA, and exception from Federal Advisory Committee Act (FACA) requirements, has led to an advisory system that privileges access by and input from a subset of domestic IP policy actors.  This affects the IP law we export.

Since my paper was released, the Washington Post covered the USTR’s capture and published a helpful infographic showing the membership makeup of all trade advisory committees.  The New York Times published an oped criticizing the USTR advisory system.

USTR Michael Froman responded by announcing the formation of a new Public Interest Trade Advisory Committee (PITAC).  The PITAC, Froman announced, will include NGOs, academics, and other public interest groups.  It will “provide a cross-cutting platform for input in the negotiations.”  The call for nominations suggests that the membership will be more limited than Froman’s remarks indicated, requiring that applicants represent a U.S. organization with “a demonstrated interest in international trade.”  This will likely exclude most academics, and may exclude some NGOs.

Nonetheless, the PITAC is a step in the right direction.  The USTR has finally formally recognized that there’s a problem—the trade negotiating process suffers from both information capture and a legitimacy crisis.  The USTR has taken a public step towards addressing these problems.  Unlike previous efforts by the USTR, which amounted not to transparency but to transparency theater, the PITAC will give more public interest representatives access to negotiating texts, and that is unquestionably a good thing.

But you don’t solve information capture just by increasing access to the text.  Information capture occurs because of a combination of what information comes out, what information goes in, the kinds of access granted to negotiators, and the ability to make negotiators listen.  The structure of the USTR advisory committee system means that the PITAC is a weak fix, not a strong one.  As Sean Flynn points out, the USTR made the decision to isolate public interest groups on the PITAC rather than integrate them into chapter-level industry trade advisory committees.  This decision limits the PITAC’s efficacy, and means the PITAC is more an aesthetic than a substantive fix.

First, there’s the issue of overall numbers.  The Washington Post infographic shows that industry voices currently make up 85% of the membership of trade advisory committees.  Adding one public interest committee, which will likely have no more than 35 members, is only a small step towards achieving balanced input overall.

Second, the PITAC won’t meet as frequently with USTR negotiators as industry advisory committees do.  This is because it is a different category of committee—a mid-level, or “tier two,” advisory committee.  Industry Trade Advisory Committees meet with USTR negotiators an estimated six times a year (see How Frequent and Relevant are the meetings?).  By contrast, the mid-level committee on Trade and the Environment meets “as needed at the call of the U.S. Trade Representative.”  According to the AFL-CIO President Richard Trumka, the mid-level Labor Advisory Committee meets one-third as often as the industry advisory committees, just twice a year.

The third problem with isolating public interest groups on the PITAC is that they will not be able to perform the important role of conducting oversight over advice from industry advisers.  Most advice from industry advisory committees is informal oral advice given to negotiators.  PITAC members will not be privy to this oral advice, and so their role will be to react rather than converse.  PITAC members won’t know what industry groups ask for and counter that advice; they’ll only be able to search for changes in the text.

The fourth problem with the PITAC is that it encompasses too many topics.  This will dilute its members’ voices, and prevent them from developing relationships with chapter negotiators the way industry advisory committee members do.  The IP industry advisory committee focuses on the IP chapter, and presumably meets with the IP chapter negotiator.  The PITAC is not focused on a particular chapter or topic.  It will be a “cross-cutting” platform, in Froman’s words.  It thus won’t meet regularly with specific chapter negotiators.  And the message of individual public interest representatives on the PITAC won’t be as loud as the message of an entire (imbalanced) industry advisory committee.

The USTR already has a mid-level advisory committee on Trade and the Environment, where all members weigh in specifically about environmental concerns.  Instead of lumping together all remaining public interest concerns onto one catch-all committee, the USTR might consider creating narrower and consequently more effective public interest committees dedicated to particular issues, like digital civil liberties or public health.

Fifth, and this is a much bigger issue, the PITAC may face serious hurdles in getting the USTR to listen to its advice.  How effective can public interest groups be without transparency, without the ability to mobilize their constituents on the details of the text?  It may not just be a question of who is at the table, but how well the players at the table are able to make their voices heard.  Truly listening to the public interest may take a different system of governance.

I maintain what I concluded in my paper: Congress has a role to play here.  Congress could empower PITAC members by including public interest negotiating objectives and making those objectives enforceable—if the USTR does not listen to public interest groups and comply with negotiating objectives, it could get kicked out of a re-enacted Fast Track.

The bigger problem, as I’ve noted, is that the usual Congressional scheme governing information flow doesn’t currently apply to the USTR.  That is the source of these problems, and that is what needs to be addressed.  Congress can and should weigh in about the lack of transparency at the USTR.  Not all texts need to get released, but the current level of secrecy is untenable.  It leads to a lack of trust in the system and to bad policymaking on the ground.  And Congress could address the information capture problem head-on, by asking for balanced membership of advisory committees overall (okay), or balanced membership at the level of each individual industry advisory committee (preferable).

Many of these fixes could be done by the executive branch: the USTR could start voluntarily releasing texts, adding members to balance the viewpoints on individual industry advisory committees, and creating new mid-level advisory committees dedicated to empowering public interest voices in more specific subject matter areas.  But the USTR hasn’t done these things. The PITAC is palliative, and a positive step, but it’s not a harbinger of systemic change.


The TPP and Copyright

This is my second post about the leaked draft of the Trans-Pacific Partnership Agreement.  Here, I address some of the copyright provisions. This is not an exhaustive analysis, and I’ve tried to make it complimentary to what’s already out there. For analysis of additional provisions, see KEI, Public Citizen, Ars Technica, Kimberlee Weatherall for an Australian perspective, and EFF. [Note: I have updated this post .]

First, there are major splits between countries. The United States consistently takes the position of pushing for stronger IP, while others—especially Canada and New Zealand— advocate a more balanced approach.  The divisions are particularly prominent when it comes to preliminary statements about the public domain and public health, none of which the U.S. supports.  These analyses are on one subset of proposed language, not finalized language, and a lot of the agreement could change. I focus my analysis primarily on the U.S. proposed language.

Second, the U.S. proposals look very similar to our past Free Trade Agreements, including the earlier texts of ACTA.  Whatever message was sent when the EU refused to ratify ACTA, the U.S. Trade Representative did not receive it.  Many of the provisions that appeared in ACTA are also areas of proposed reform in the United States—the U.S. proposed text for the TPP would internationally bind this country and prevent many proposed reforms to our copyright law.

Third, and this is an important point: U.S. proposals are less balanced than U.S. domestic law. As I noted in my previous post, this is because a subset of IP industries provide the USTR advice, leaving out important balancing viewpoints and sometimes misrepresenting U.S. law.

Similarities to ACTA:

Intermediary liability (Article QQ.I.1): As in ACTA, the central copyright argument in TPP is over what kind of intermediary liability regime will be internationally required.  Currently, online intermediary liability is not governed at an international level.  Michael Geist has done a great job of outlining the basic differences in the TPP proposals for intermediary liability. The United States is pushing for notice-and-takedown, with a coalition of countries, including Canada and New Zealand, pushing instead for a notice-only provision.  These regimes offer different levels of due process for Internet users.

Statutory damages (ARTICLE QQ.H.4.Y(15)): As in ACTA, countries are fighting over statutory (or “pre-established”) damages: damages awarded to copyright owners without a showing of actual harm.  Australia opposes this provision because it does not have statutory damages. Statutory damages are an often-criticized part of the U.S. regime, enabling disproportionately large awards in copyright cases.

Criminal liability (Article QQ.H.7(2)): As in ACTA, countries have split over the definition of criminal copyright infringement.  The U.S. proposal, like U.S. law, pushes to criminalize very low level infringement, to go after even low level filesharing.  Other countries want to keep it at the international requirement, which requires a motive of financial gain or commercial advantage.  The United States lost a recent WTO case on criminal copyright, and has since been trying to ratchet up international criminal copyright law through free trade agreements.

Criminal Intermediary liability (Article QQ.H.7(6)): As in ACTA, the TPP criminalizes aiding and abetting infringement. This allows authorities to go after websites under criminal law, rerouting around protections from civil liability. Coupled with a low underlying threshold for criminal infringement, this could sweep in a lot of activity. Thanks to Kimberlee Weatherall for catching this.

Asset Forfeiture (Article QQ.H.7(c) p. 79): As in ACTA, the TPP allows the seizure of “any related materials and implements used in the commission of the alleged offense”. In the United States, asset forfeiture has been used to seize websites before trial.

DRM/Technological Protection Measures (Article QQ.G.10): As in ACTA, the United States is trying to export the DRM provisions of the DMCA, as its version of implementation of the WIPO Copyright Treaty. As I note below, however, what’s proposed is a worse version of what we have at home. The recent conflict over cell-phone unlocking shows that this is likely to be the focus of domestic policy reform. Some suggest that the DRM provisions here criminalize unlocking even when the underlying purpose is fair use.

De minimis border measures (the “Ipod search”)( Article QQ.H.6(8)): As in ACTA, the United States is pushing to shrink the “personal use” border exception that currently exists in international law, in TRIPS Article 60.  For now, however, it looks as though the de minimis provision is functionally the same as TRIPS, due to a strategic footnote (235).  This is something to watch.

Account termination (Article QQ.I.1(p. 88)): In U.S. copyright law, there’s a provision requiring online intermediaries to have a termination policy in place for repeat copyright infringers. Similar language has been proposed in the draft TPP. As Annemarie Bridy has pointed out, this language may have different implications in different countries.  It probably does not require graduated response.  However, its inclusion can also be read as endorsing recently developed private-ordering graduated response in the United States: agreements between private companies to kick infringing users offline. And transplanting this language into other countries might lead them to push for similar or more draconian policies.

Things that differ from US law:

Fair Use/Exceptions and Limitations language (Article QQ.G.Y): There‘s no fair use. The broadened copyright exceptions language that the USTR bragged about earlier this year is not very broad. And it’s not fair use.  This may harm our exporting businesses.  Google just won the Google Books case, where Google books was found to be fair use; TPP shows no evidence of extending that kind of exception abroad.

International First Sale Doctrine (Article QQ.G.3,  Article QQ.G.17):  Despite the fact that the Supreme Court held in Kirtsaeng in March that first sale doctrine applies abroad—and trumps the importation right—the USTR is exporting the opposite. The USTR also opposes saying that countries are encouraged to establish international exhaustion of rights. All other countries (except Canada, which is neutral) want international first sale doctrine. It’s not clear to me why the USTR thinks it can bind the US to law contrary to Kirtsaeng—and this draft is from months after the Kirtsaeng opinion came down. I’m happy to be convinced otherwise, but I can’t see how the two are reconcilable.

Temporary reproductions (Article QQ.G.1): Despite the fact that there is a circuit split over whether temporary reproductions are considered “fixed” enough to be copyrightable, the USTR proposes exporting a requirement that temporary reproductions are covered by copyright.

Standard technical measures (Article QQ.I.1(p. 88)):  In the DMCA, there is a requirement that intermediaries accommodate “standard technical measures.”  However, those measures must be arrived at through a multi-industry process that is open and fair. In practice, this means that there are no standard technical measures in the United States. The words “multi-industry” and “fair” get left out of the TPP, which means that standard technical measures may get developed  abroad, and online intermediaries will have to accommodate them there, where they don’t have to at home.

Criminal liability ((Article QQ.H.7(2)):  U.S. criminal copyright law contains a numerical threshold for criminal copyright infringement that is done without commercial motivation- $1000 of infringement in 180 days. The TPP proposal substitutes the word “significant” for an actual number. This difference could go either way.

Notice-and-takedown misuse & attorneys’ fees (Article QQ.I.1 (p.89)): In the DMCA, people who misuse notice-and-takedown can be sued, and are subject to attorneys’ fees.  Even though the US proposal clearly asks for attorneys’ fees in other places, it does not include attorneys’ fees in the material misrepresentation provision.

Reverse engineering exemption to TPM/DRM (Article QQ.G.10): Thanks to Jonathan Band for pointing this one out to me originally. The reverse engineering exception to DRM exported by the US is not as broad as the one available domestically.

Privacy (Article QQ.I.1): TPP barely mentions privacy.  There are some discussions of a no-duty-to-monitor provision, but no broader assertion of privacy principles that I can find.  We don’t normally think of U.S. law as containing a nod to balancing copyright enforcement with privacy, but it turns out that 512(m)(2), which says providers shall not access content contrary to law, was intended to bolster the “no monitoring necessary” requirement to mean “no monitoring necessary, AND don’t violate wiretap law.” This didn’t make it in.


Capture, sunlight, and the TPP leak

Yesterday, Wikileaks leaked the draft IP chapter of the Trans-Pacific Partnership Agreement (TPP).  This is the first of two posts I’ll make on the significance of the leak.  In this post, I discuss why the leak matters from the process perspective.  In the next, I’ll point out detailed substantive issues in the draft’s copyright provisions.  Unsurprisingly, the process and the substance are closely intertwined.

Up until this leak, only a subset of domestic IP stakeholders has had access to the TPP text.  The U.S. Trade Representative has shown the draft text to its closed advisory committees, but not to anybody else.  Content industries and pharmaceutical industries sit on the IP advisory committee.  Internet industries, smaller innovators, generics companies, and public interest groups do not.

This is no accident.  When Congress established the trade negotiating system, it exempted the Trade Representative from requirements of an open government law that was enacted to prevent agency capture, the Federal Advisory Committee Act (FACA).  FACA requires transparency, a limited term length for industry advisory committees, and balanced membership on those committees.  The trade advisory committee on IP does not have balanced membership.  It is not subject to public oversight.   And it has an extended term limit, at the discretion of the USTR.  Other open government laws also don’t apply—the USTR exempts itself from the Freedom of Information Act (FOIA) by claiming a national security exemption, and the Administrative Procedure Act (APA) doesn’t apply to international lawmaking. As a consequence, the U.S. role in international IP lawmaking is captured through one-sided industry advice.

The USTR is supposed to be exporting US IP law.  But what the USTR exports is not US law.  The USTR paraphrases US law; it doesn’t export our statutes.  This process allows information capture to have substantive consequences.

The paraphrasing USTR sends out is worse than our law. It’s less balanced, and it’s missing things. In places, the USTR misrepresents that one side of a current circuit split is the authoritative word on issues that deeply divide domestic constituents.

These skewings and omissions are not an accident. Some are directly requested by the IP advisory committee—you can see the requests in past advisory committee reports on other free trade agreements.  Others are the result of the USTR’s failing to consult public interest, academics, or opposing industries, who would point out what must be included, what’s wrong, and which omissions matter.

Balanced advice is necessary at the level of the text.  The USTR recently conducted a series of “stakeholder phone calls,” where it takes questions on broad policy issues.  These calls are transparency theater, not transparency.   They lump together stakeholders on an impossibly wide range of issues—from dairy to textiles to IP.  Discussions are high-level, where many of the problems with USTR’s policies are textual problems.  The devil, in law, is very much in the details.

The USTR is captured in other ways, too.  There is a significant revolving door problem. USTR negotiators come from and leave for employment in the same IP industries that sit on its advisory committee.  But the crux of the problem is that nobody outside of the advisory committees sees the text.

This is why the TPP leak is so important: it levels the playing field.  It lets those who aren’t on the advisory committees, like me, find and point to misrepresentations the USTR is making.  If there is one thing I’d say to our negotiating partners, it’s this: find yourself a good U.S. lawyer who can tell you what is actually in U.S. IP law.  And let them read what you’re negotiating.

Our domestic IP law is the result of democratic political process.  Even if that process has been subject to significant collective action problems, it still shows compromise.  Constituents on opposing sides of an IP issue arrive at legal meaning through statutory and regulatory compromise, and litigation.  Courts play an important role in reinserting balance into the system.  The international agreements we negotiate and sign are produced by a closed and captured process.  What little balance we achieve domestically is not what we are sending abroad.

This approach will have consequences.  As many rightfully point out, it will harm developing countries.  But it will also harm domestic constituents.  Google now makes over 50% of their profits overseas, and we export copyright law that will make it more difficult for them to operate abroad.  What’s more, the Internet is global—digital copyright law made abroad affects online content available to people here.  And the law the USTR exports affects our policymaking process at home.  It binds us to detailed IP law created by advising incumbents.  Deviations from that law may result in threats of trade sanctions.

Killing fast track is one way to change the system.  If fast track is stopped, Congress will have much more input into the trade process.  It will be able to amend trade agreements.  But this is a blunt instrument, and it might kill international trade.  A more tailored combination of changing the advisory system, giving the USTR precise and enforceable negotiating objectives, and increasing transparency could produce the changes we need.  This is why I signed the law professors’ letter calling for transparency in trade, issued today. One leak won’t fix things, but it’s certainly a start.

Sunlight is the best disinfectant. Many thanks to Wikileaks for doing what we rightfully expect our government to do.


First Sale and the Specter of “International Exhaustion”: Kirtsaeng comes down

The Supreme Court issued its decision today in Kirtsaeng v. Wiley, addressing whether the first sale doctrine, which prevents copyright owners from further controlling works after sale, applies to works legally manufactured abroad. First sale doctrine allows libraries and second hand bookstores to operate without fear of copyright consequences. Kirtsaeng was a student in the U.S. who made money off of reselling Wiley textbooks that had been published in Thailand. Wiley argued that the first sale doctrine as codified by Congress in 17 USC 109(a) does not apply to works manufactured and sold abroad. The Court disagreed.

This post summarizes the opinion, talks briefly about the conflict that exists between US law and language in our Free Trade Agreements, and then makes a few guesses of what will happen next.

The opinion:

Justice Breyer wrote the majority opinion (6-3), which to those following SCOTUS copyright cases says it all. The majority held that authorized works manufactured abroad are subject to the first sale doctrine, so Kirtsaeng did not infringe copyright by reselling the legally purchased works in the United States. In other words, when coupled with Quality King, Kirtsaeng determines that 17 USC s. 602(a)(1) creates an importation right as part of the distribution right of s. 106, which in turn is limited by first sale in s. 109(a); thus 602(a)(1) is limited by 109(a). (As I’ll discuss below, it is important that this is all statutory interpretation.)

Breyer trumpeted the importance of first sale, emphasizing the “practical copyright-related harms with which a geographical interpretation would threaten ordinary scholarly, artistic, commercial, and consumer activities.” In the absence of first sale doctrine, libraries, used book stores, museums, and retailers would all struggle to trace title for used books and other works originating abroad. As my colleague Christina Mulligan has noted, investigating the licensing history of a work can be costly.

Breyer also got a chance to reorient copyright doctrine from the rights holder to the consumer, remarking that “American law too has generally thought that competition, including freedom to resell, can work to the advantage of the consumer” (at 18). The opinion evinced little sympathy for publishers, explaining that there is  “no basic principle of copyright law that suggests that publishers are especially entitled to… charge different prices for the same book in different geographic markets” (at 31).

In the broadest language of the opinion, Breyer explained that a “copyright law that can work in practice only if unenforced is not a sound copyright law. It is a law that would create uncertainty, would bring about selective enforcement, and if widely unenforced, would bred disrespect for copyright law itself” (at 24). This language can be applied to a whole host of current copyright policy, including the little-enforced criminal copyright standard created by the NET Act.

Conflict with Free Trade Agreements (FTAs):

The US Trade Representative (USTR) has bound us to a standard that is now in conflict with Supreme Court caselaw. My co-author Sean Flynn correctly pointed out this morning that Kirtsaeng conflicts with proposed language in the Trans-Pacific Partnership Agreement and language adopted in earlier Free Trade Agreements. In the U.S.-Jordan FTA, for example, Article 4(11) prohibits the “importation into each Party’s territory of copies of works and phonograms, even where such copies were made with the authorization of the author.” This directly conflicts with the Kirtsaeng-Quality King holding that importation is allowable when works were legally purchased abroad.

Similarly, both the Anti-Counterfeiting Trade Agreement (ACTA) and the TRIPS Agreement contain broad border measure provisions that are implicated by Kirtsaeng. Both agreements assume that countries will employ broad border measures in the name of copyright enforcement. Now those border measures must be limited to pirated goods only, not including legally manufactured goods that weren’t authorized for importation.

Each agreement contains “de minimis” provisions that limit border measures from applying to smaller importations. Under TRIPS, copyright-related border measures need not be applied to “small quantities of goods of a non-commercial nature contained in travellers’ personal luggage or sent in small consignments.” ACTA, by contrast, requires that border measures be applied to “goods of a commercial nature sent in small consignments.” This means that under TRIPS, border guards can search everything except for small consignments and personal luggage to determine if goods are infringing. Under ACTA, border guards were required to search through “small consignments” like the ones Kirtsaeng received, to determine if these consignments are “commercial” or “noncommercial” in nature.

Kirtsaeng suggests that these de minimis provisions are now far less significant because border measures should apply to less activity. ACTA’s requirement that border measures must apply even to “goods of a commercial nature sent in small consignments” must now be limited to goods that are copyright infringing, not goods that are unauthorized for importation.

What happens next:

What will happen after Kirtsaeng?

Kagan’s concurrence suggests one outcome: Congress could legislate around the Court’s opinion and re-establish a broader importation right not subject to first sale. But if done carelessly, new legislation will present exactly the same problems for secondary markets such as libraries and museums: the cost of determining whether a good has been legally imported would rise again, discouraging resale markets. As James Grimmelmann tweeted this morning: “If you want national market segmentation, prohibit unauthorized imports, but keep your fingers off resale of the items once they’re here.” I would add that if Congress creates a new prohibition against unauthorized imports, it should clarify that (1) downstream purchasers of legally manufactured but unauthorized imported goods will not be liable for copyright infringement and (2) the de minimis importation of goods manufactured abroad will still be allowed, under TRIPS Article 60.

Another likely outcome post-Kirtsaeng is that rights holders will continue to expand the trend of licensing copyrighted goods rather than selling them. Breyer makes it very clear that first sale applies to owners only. A “lessee of a copy will not receive “first sale” protection but one who owns a copy will receive “first sale” protection, provided, of course, that the copy was “lawfully made” and not pirated” (at 14). Many copyrighted goods are now licensed, not sold, thus working around Kirtsaeng. Amazon, for example, notoriously does not allow you to “purchase” territorially licensed books on the Kindle from other geographical territories.

Finally, there’s the question of what the practical outcome will be if Congress does not legislate. Joe Karaganis has done great work showing that piracy occurs in emerging economies when there is a “failure of affordable access to media in legal markets.” If publishers and other rights holders try to avoid the inevitable arbitrage by raising prices in emerging countries to match prices in the U.S., they may stimulate more piracy in those countries.


Lie to me: the First Amendment in US v. Alvarez

The Supreme Court had a busy day yesterday, and in the wake of healthcare, there’s a risk of overlooking an important addition to this Court’s First Amendment jurisprudence: U.S. v. Alvarez.

In short, the Court found that Congress can’t send you to jail just for lying. Alvarez confirms that this Court is extremely reluctant to create new FirstAmendment exceptions, and has a speech-protective understanding of the marketplace of ideas. Alvarez also leaves open some interesting questions, both doctrinal and practical.

Alvarez was prosecuted under the Stolen Valor Act (18 USC  s. 704) for lying about having received the Congressional Medal of Honor. What made this case particularly interesting, and probably what split the Court, is that Alvarez did not lie to gain money, or to get a job. He didn’t lie for any apparent reason. He just lied.

The Court split 4-2-3, with six affirming the Ninth Circuit and finding the Act unconstitutional. Justice Kennedy wrote the plurality, Justice Breyer wrote the concurrence (joined by Justice Kagan), and Justice Alito rather unsurprisingly wrote the dissent.

The plurality forcefully reiterated what the Court articulated two years ago in U.S. v. Stevens (2010): content-based restrictions on speech are subject to strict scrutiny, with limited exceptions that have been clearly established in prior caselaw. What was (again!) at stake in this decision was whether the First Amendment protects all speech except for the familiar carveouts, or presents an “ad hoc balancing of relative social costs and benefits” with each new proposed exception (at 4, quoting U.S. v. Stevens (2010)).

The plurality went the First-Amendment-protective route. Its “historic and traditional categories” of First Amendment exceptions present a familiar roster:  obscenity, fighting words, incitement, and the rest. False speech as false speech is not one of the historical exceptions, and the plurality made it perfectly clear that it does not plan to add to the list. In Stevens, then, the Court said what it meant about not intending to add to historical First Amendment exceptions. Future brief-writers would do well to keep this in mind.

Eugene Volokh in his Amicus brief feared that if the Court went the route of protecting false speech, the First Amendment would become a patchwork of under-theorized exceptions to that rule. The plurality proved him wrong. It both articulated theoretical underpinnings for existing exceptions that do involve false speech, and took the Government to task for advocating an overly restrictive understanding of the marketplace of ideas.

The plurality walked through two general categories of exceptions to First Amendment protection for false speech. These categories are effectively distinguished from most false speech as “false speech-plus.” Each is not just false speech, but has an additional element.

The first kind of false speech not subject to First Amendment protection is false speech where there is a legally cognizable harm to an individual, such as an invasion of privacy or legal costs. This category includes defamation and fraud (at 7). Robert Post might further add that these kinds of crimes and torts generally take place outside of the public sphere, and so are subject to less First Amendment protection because they involve individual relationships rather than public-facing speech.

The second kind of false speech not subject to First Amendment protection is false speech that impedes a government function (eg perjury or lying to a federal officer), or abuses government power without authorization (eg impersonating a Government officer). Here, no direct injury to an individual is required. The plurality found that these two types of laws are similar because both “protect the integrity of Government processes” (at 9).

The more serious and broad-sweeping theoretical debate resolved by the Alvarez plurality concerns a fundamental understanding of the marketplace of ideas.

In the historical understanding of the marketplace of ideas, speech competes with speech towards the pursuit of “truth” (although truth is more accurately understood as political truth, not just truth in the sense of non-falsity). Thus Volokh is probably correct when he writes that historically, false speech was considered of lower value in the marketplace of ideas than true speech.

However, the present-day understanding of the marketplace of ideas is that it’s impossible to determine which speech has high value, and which speech has low value. Speech competes, and listeners choose what to believe, but there’s no competition towards an absolute truth-in-the-sense-of-non-falsity, or towards higher values that have been officially designated as such. The Court acknowledged as much in Cohen v. California, which often gets misread as being a case about political speech, where it’s in fact about protecting traditionally low-value expression.

The Alvarez plurality explicitly rejects the proposal that false speech is low value speech and thus not subject to full First Amendment protections. “The remedy for speech that is false is speech that is true. This is the ordinary course in a free society.” (at 15)

The plurality thus articulates a speech-protective and autonomy-driven understanding of the marketplace of ideas, where the marketplace is self-correcting, and Congress has no place determining what is true, or good or bad, apart from protecting individuals from legally cognizable harms and from abuse of government structures and government power.

Both doctrinal and practical questions remain after Alvarez, unsurprisingly.

Doctrinally, the question is what type of scrutiny applies to false speech. The plurality employed strict scrutiny, while the concurrence used intermediate scrutiny. It is not clear what the Court will employ in the future.

Using intermediate scrutiny to strike down the Act, it should be noted, creates a strange tension between this case and commercial speech doctrine, which allocates First Amendment protection only to commercial speech that is not misleading. Intermediate scrutiny may also raise questions about trademark dilution, where no competition, commercial harm, or likelihood of confusion need be shown. The concurrence thus struggles with trademark dilution on pp 6-7, where the majority could probably get rid of —or at least restrict the scope of— the trademark problem by applying intermediate strutiny.

Practically speaking, the Act might survive on rewriting. The Act might be rewritten to require that the liar lie for the purpose of receiving a benefit. Alternatively, the Act could be rewritten to penalize lying where the liar benefited from the lie (ie, harm was accomplished as a result of the lie). If the Act were thus rewritten, it’s not clear how the plurality would treat it with respect to historic exceptions and their justifications. It also seems likely that the concurrence would switch sides.

It’s worth noting the implications of Alvarez for the ongoing discussion of anonymous speech, and the use of online personae. If Alvarez had gone the other way, the Court might have made it possible for Congress to prohibit the use of pseudonyms, or “fake names,” online. Lying about your identity is another way of describing choosing to hide your real identity, which would have brought the case into conflict with McIntyre v. Ohio and other doctrine on anonymous speech.  I’m not sure that a good doctrinal distinction could be developed between positively asserting that you are another person , and choosing a pseudonym for the purpose of hiding your identity. For now, at least, thanks to Alvarez, the distinction between legal and illegal pseudonymous behavior appears to rest clearly in the additional element of harm the Court noted must be shown for fraud, or the performance of some other tort or crime.

There is another fast-developing area potentially impacted by Alvarez that the Program for the Study of Reproductive Justice at Yale has been working on all year: the regulation of Crisis Pregnancy Centers, where states require the centers to explain that they are not actually doctors and do not actually provide medical services such as abortion. On this issue, though, I’ll defer to my colleague Jennifer Keighley, who has a piece forthcoming on the matter.

But leaving all this aside, there’s a very simple reason Alvarez was correctly decided.

As Kozinski noted below, people lie an awful lot.




Three thoughts on U.S. v. Jones

Many thanks to Danielle Citron for offering to host this writeup, an earlier draft of which is available here.

The Supreme Court came down yesterday in the GPS tracking case, U.S. v. Jones, in which prominent privacy scholars and members of Yale Law School’s Information Society Project filed an amicus brief this fall.

Scalia’s majority held that the physical attachment of a GPS tracker to a car was a warrantless search in violation of the Fourth Amendment, and thus the Court did not need to reach the question of whether GPS tracking invaded a reasonable expectation of privacy. Sotomayor joined, but described in her concurrence how if they had reached the Katz question, she would have found that a reasonable expectation of privacy had been invaded. Alito’s concurrence, joined by Ginsburg, Breyer, and Kagan, criticized the majority’s physical trespass focus and to conclude that under Katz a reasonable expectation of privacy had been invaded.

There has already been a lot written about this opinion, in the short time it has been out. This post is not intended to be comprehensive, by any means, or to evaluate the potential impact of the majority’s trespass/search analysis, which will take time to sort out. I wanted just to add three thoughts to the fray.

(1) Sotomayor’s opinion is the one that will have the biggest practical impact, and thus to me is the most important;

(2) The most troubling indication for future cases is Alito’s observation that the Katz test is grounded in social norms that increasingly deprioritize privacy, which may lead Alito and possibly others away from the Fourth Amendment in the future, towards deferring to legislation; and

(3) The majority’s treatment of the beeper case, Knotts, suggests a possible theme to explore: maybe down the line the question can become whether third party behavior looks more like that of a traditional police informant, rather than following the blanket Miller rule that third parties can do what they like with information. More on this below.

First: Sotomayor’s concurrence is the one that will have the greatest practical impact. The majority holding adds a line of defense against slipping trackers into overcoat pockets– a scenario Kennedy and Sotomayor both referenced at oral argument– but it has a narrow practical impact, because police don’t need to physically attach a tracker to your car to use GPS tracking. They can just turn on GPS trackers already in your car (OnStar) or on your person (cell phones).

Sotomayor joins the majority on the trespass issue and refuses to reach the Katz question of whether GPS tracking violates a reasonable expectation of privacy, but then she spends the concurrence explaining how she would treat Katz. In a GPS tracking case where no physical attachment occurred, Sotomayor would “ask whether people reasonably expect that their movements will be recorded and aggregated in a manner that enables the Government to ascertain, more or less at will, their political and religious beliefs, sexual habits, and so on.” This concurrence signals precisely what Sotomayor’s outcome will be in the next GPS case, about OnStar or its equivalent: GPS tracking (of a certain magnitude) is a violation of the Fourth Amendment.

The real result of Jones is thus not a narrowly held majority about physical trespass. It is, for precedent-citing purposes, but not for its impact on practice or lower courts. Sotomayor uses her concurrence to clearly signal to cops and judges. The next time one of these cases comes up, Sotomayor will be joining Alito on the reasonable expectation of privacy question. Or she may write an even stronger concurrence, since she suggests that Alito hasn’t adequately taken into account the ease of GPS tracking, or its chills on associational and expressive freedoms. This matters– and matters a lot– because in the interim cases, before another GPS case comes to the Supreme Court, judges are going to pay attention to Jones for any hints of a future outcome. And Sotomayor deliberately showed her hand.

Second, the most dangerous part of these opinions is not the majority’s decision about trespass, but Alito’s conclusion that Katz is tethered to evolving social norms, and his tendency to consequently defer to the legislature on these issues. If this is how Alito views Katz, then as society gets more privacy invasive, he will be more likely to defer to Congress than to invoke the Fourth Amendment. It is not clear how many will follow him on this path.

According to Alito at oral argument, “[t]echnology is changing people’s expectations of privacy. Suppose we look forward 10 years, and maybe 10 years from now 90 percent of the population will be using social networking sites and they will have on average 500 friends and they will have allowed their friends to monitor their location 24 hours a day, 365 days a year, through the use of their cell phones. Then — what would the expectation of privacy be then?”

Alito returned to this idea in the concurrence: “even if the public does not welcome the diminution of privacy that new technology entails, they may eventually reconcile themselves to this development as inevitable.”

This angle on Katz is disturbing, because it conflates knowledge that one is being unreasonably surveilled with acceptance of that surveillance. This perversely incentivizes a dearth of public conversation about privacy violations, in the interest of preserving Fourth Amendment-based privacy. It also fails to recognize any difference between surveillance by private companies and surveillance by the government, indicating that Alito may be slow to follow Sotomayor in her hope to overturn Miller. Combine this view with Alito’s indicated desire to defer to the legislature on these issues, and it may be that by the time the next GPS case comes up, Alito will change sides.

My third point will take a lot more development, and admittedly is a long shot or a merely theoretical one. But the majority’s treatment of the earlier beeper case, Knotts, at 8, looks strangely like a misplaced trust case. The individual who received the chloroform containers with the beeper already placed in them by the previous owner (the chloroform manufacturer) in Knotts couldn’t say that his property had been searched, because the police didn’t search it: the previous owner made the choice to cooperate with the police. Similarly, in U.S. v. White, 401 U.S. 745 (1971), the Court found that when a defendant trusts an apparent colleague and that trust turns out to have been misplaced, no warrant is required for wiretapping.

Maybe, then, the misplaced trust idea could be used in the future as a principle to both limit Miller and address the OnStar problem. Instead of looking only to whether the tracker was placed by the police or placed prior to purchase, courts could look to whether the third party intended to be a police informant in supplying the person with the technology. If you sell a person a car in which you’ve hidden a tracking device because you intend to aid the police in tracking that person, no warrant would be needed. But if as a car manufacturer you’ve provided the person with the same car you sell everybody, and one feature happens to be that you can turn on tracking, police would need a warrant to ask you to turn the tracking on.

This idea is in very nascent stages, and I realize Miller is likely insurmountable until overturned or at least reevaluated. I also realize that misplaced trust and the trespass question are different issues, and the majority gave every indication of understanding the moment of placement of the GPS literally and physically. Finally, I understand the very large potential loophole that all third parties could just claim that they always intend to be police informants on everybody, before any sale occurs. But it’s a thought that might serve to unify trespass with misplaced trust and third party doctrine, and create a limiting requirement of specific prior intent to inform for all three.