Balkinization  

Friday, August 10, 2018

Podcast on John Bingham

Gerard N. Magliocca

Yesterday the National Constitution Center released an hour-long podcast that I did with Kurt Lash on John Bingham and the Fourteenth Amendment. You can find the podcast here.

Wednesday, August 01, 2018

What's the deal with 3-D plastic guns -- and what's the Freedom of Speech got to do with it? [UPDATED Aug. 2]

Marty Lederman


There's been an explosion of stories in the past few days about how the Trump Administration is allegedly about to permit people to create "3-D plastic guns" that can't be detected by metal detectors; about a suit by several states to stop the Trump initiative; and about a district court's injunction yesterday, which some critics are (wrongly) describing as a vast "prior restraint" on Internet "speech."

There is, to say the least, a lot of confusion and misinformation swirling around out there about this topic--especially about what the law does and does not prohibit and about what, if anything, the First Amendment has to do with the creation of guns.  

Herewith, then, a modest effort to offer some clarification.  It's entirely possible some of what follows is mistaken; therefore I'd appreciate any corrections or qualifications, and I'll edit the post accordingly when I receive more accurate information about the technology or the state of play.

What federal law prohibits

There is no federal law prohibiting persons from manufacturing otherwise lawful firearms at home "solely for personal use"—by use of a 3-D printer or otherwise.  (Federal law doesn't even require registration of such homemade weapons.)  Nor is there any federal law prohibiting the sale, manufacture or distribution to U.S. persons of computer programs that, when used as designed, result in the creation of such a 3-D firearm. 

On the other hand, it's unlawful to make at home any firearms that can't lawfully be possessed, such as machine guns and some rifles.  Most pertinent for present purposes, the federal Undetectable Firearms Act makes it illegal to manufacture or possess a weapon that’s undetectable by walk-through metal detectors, such as a weapon made entirely of plastic that has no metal parts.

An organization called “Defense Distributed” has developed computer aided design (“CAD”) files consisting of computer code that, when downloaded, can direct a 3-D printer or computer numerical control (CNC) mill to automatically create “3-D printed weapons”—a plastic lower receiver of a rifle or a fully functional single-shot plastic pistol—at the virtual push of a button.  [UPDATE:  A former student helpfully informs me that the CAD file for an M-16/AR-15 lower receiver can't be used to produce a fully functioning, non-detectible plastic rifle because the upper receiver with which the 3D-printed lower receiver must be paired includes a metal barrel, which of course can't be printed.  A fully functional plastic pistol, however, is a possibility.]

So, to the extent someone uses the Defense Distributed files to create a firearm detectable by metal detectors—such as one to which a six-ounce piece of metal is inserted—that person is not violating federal law (and probably isn’t violating any state laws, either, although I haven't examined that question).  According to the federal government, however (see footnote 6 of this Declaration), the files can also be used to make an operable firearm without any metal, i.e., a weapon that metal detectors cannot detect.  And, indeed, there's not much of a reason to go to the time and expense of making a 3-D plastic firearm (which is more fragile and less reliable than ordinary weapons) other than to escape metal-detector detection.  So presumably this will describe most of the guns made with the Defense Distributed files.  If an individual creates that kind of 3-D weapon, with Defense Distributed’s files or otherwise, she violates federal law.

As of now, federal law does not generally prohibit Defense Distributed from selling or distributing its CAD files to other persons, notwithstanding that such files have the potential to automatically create unlawful weapons (depending on how the recipient uses them).  (When I say “automatically” here, I simply mean in an automated manner, without the intervention of any human creativity, industry or material effort beyond inserting the files and pressing a button or two.)

However, the federal government does prohibit the export of 3-D weapons to persons overseas, and until the end of last week it also prohibited the export of the computer files that, when used as designed, create such weapons.  Such CAD files, explained the government in a recent court case, “constitute the functional equivalent of defense articles: capable, in the hands of anyone who possesses commercially available 3D printing equipment, of ‘automatically’ generating a lethal firearm that can be easily modified to be virtually undetectable in metal detectors and other security equipment.”

What’s more, the federal government considers the posting of such (downloadable) computer files on the Internet to be such an “export” because, of course, that’s functionally equivalent to handing foreign persons the files, since all they have to do is download them and use them as instructed in a computer in order to create weapons.

In effect, this export regulation prohibited the undifferentiated Internet posting of the Defense Distributed CAD files, because there’s no practical way to limit such postings so that only persons in the United States can download the code files.  (At the same time, as noted above, Defense Distributed has always been free to disseminate the computer files in public or private domestic fora, including via the mail or any other medium that does not provide persons overseas the ability to obtain or download the files.)

The “freedom of speech” red herring

Defense Distributed challenged the State Department's Internet-posting ban as an alleged abridgment of its freedom of speech.  Its theory is based on the idea (as it argues in a brief) that its computer code “is expressive in that it can be read and edited by humans, who can also understand and adjust its output.”

It may be right about that fact:  Some small number of human beings can in some sense “read” and “edit” this code, just as there are some who can “read” most computer programs; the code might even have certain properties associated with a language.  That’s not the primary purpose of posting it to the Internet, however:  Presumably Defense Distributed expects and intends that 99.99 percent of the people who download it will not “read” or “edit” its code but will instead simply do with it what we all do with computer files every day, namely, stick them or download them into a computer so that they can perform the technological functions for which they were designed—in this case, to create operational weapons at the click of a mouse, without conveying any information to anyone.

More importantly for constitutional purposes, even if the sharing of such files can (in rare cases) be "speech" because it is done for the purpose of sharing information with others, the government’s reason for regulating the distribution of the code is not to suppress any potential “informational” value in those rare cases where a recipient might “read” it.  The government's objective is, instead, to prevent persons overseas from obtaining a tool of production the physical properties of which cause the easy creation of non-exportable weapons.  (This is not, in other words, a restriction on speech because it (as Noah Feldman put it) "instructs the public how to commit a crime."  I agree that, in such a case (see, e.g., the case of the Progressive's publication of bomb instructions, or the "Anarchist Cookbook" controversy), there'd be very serious First Amendment constraints, at least absent proof of an intent to help facilitate crimes--see this DOJ Report, the constitutional analysis of which I helped draft.)

As the Department of Justice wrote in one of its briefs in Defense Distributed’s recent constitutional challenge:

Even if plaintiffs are correct that computer code can serve as “an expressive means for the exchange of information and ideas about computer programming,” Junger v. Daley, 209 F.3d 481, 485 (6th Cir. 2000), that potential exchange of ideas is not the basis for the government regulation here. While computer programs can in some cases be read by human beings (such as other computer programmers) and thus convey information, the computer files at issue are subject to regulation because they facilitate automated manufacture of a defense article.  The State Department is concerned here with the use of the data files at issue by machines, not with their ability to express a message to humans.  Although a person must direct a machine to read the files, “this momentary intercession of human action does not diminish the nonspeech component of [computer] code.” [Universal City Studios v.] Corley, 273 F.3d [429,] 450 [2d Cir. 2001].  (Emphasis added.)

At most, therefore, the fact that the State export regulation (including the limit on Internet posting) would affect the comparatively very rare case of a distribution and use of the code for informational purposes, that incidental, rare impact would subject the regulation to, at most, “intermediate” scrutiny under U.S. v. O’Brien (the classic precedent involving the regulation of conduct that some might engage in for expressive purposes), and should easily withstand such scrutiny, in the same way that a law prohibiting the posting of readily downloadable computer “virus" codes would easily pass First Amendment muster.[1]

The State Department’s decision to rescind its restriction on posting of the code, and the current litigation

OK, but even if I’m right that the State Department’s longstanding prohibition on the export—and Internet posting—of gun-creating software does not violate the Free Speech Clause, the Department nevertheless has now chosen to rescind that regulation.  Last Friday, it posted on its website a notice stating that “the Acting Deputy Assistant Secretary for Defense Trade Controls has determined that it is in the interest of the security and foreign policy of the United States to temporarily modify” the U.S. Munitions List (USML) to “exclude” Defense Distributed’s CAD files.  The notice called this a “temporary modification” that was to “remain in effect while [a] final rule” to the same effect “is in development.”

As far as I know, the State Department has the legal authority to remove the CAD files from the List, thereby permitting their export and Internet posting.

On Monday, however, eight states and the District of Columbia filed a suit in the Western District of Washington seeking to enjoin State’s “temporary modification” of the USML.  They allege that if and when State allows Defense Distributed to post its files to the Internet, it’s foreseeable that many more persons will use the code to easily create undetectable plastic weapons, which will in turn lead to harm for the residents of the affected states. 

Yesterday, Judge Robert Lasnik granted a temporary restraining order that enjoins the federal government defendants from implementing or enforcing the “temporary modification” of the USML, and requires them to “preserve the status quo ex ante as if the modification had not occurred.”  He also scheduled a hearing for next Friday [UPDATE:  now rescheduled for August 21] on the question whether to transform the TRO into a preliminary injunction.

Contrary to the feverish alarms of Defense Distributed’s counsel and others, Judge Lasnik did not issue a prior restraint on anyone’s speech; indeed, he didn’t enjoin any private parties at all.  He merely enjoined the federal government from removing particular items from the Munitions List, thereby requiring the government to preserve the regulatory status quo that’s been in place for years.

Despite the absence of a real First Amendment problem, however, it’s not clear whether the injunction against the State Department will last very long, for at least three reasons. 

First, there’s some question whether the states (and DC) have standing to sue to challenge the revocation of an export regulation that limits the conduct of private parties.  Judge Lasnik concluded that they did have such staning, at least “[f]or purposes of this temporary order,” because “[t]he States and the District of Columbia have a clear and reasonable fear that the proliferation of untraceable, undetectable weapons will enable convicted felons, domestic abusers, the mentally ill, and others who should not have access to firearms to acquire and use them.”  Whether that reasonable fear is sufficient to establish Article III “injury in fact” for the states and DC will be a question to watch as the case proceeds.

Second, there’s the question whether the merits of the suit are valid.  The states and DC rely primarily upon a procedural claim:  They argue not that the State Department can’t rescind an item on the USML, but instead that the President must give Congress 30 days’ notice before State does so, pursuant to 22 U.S.C. § 2778(h), which provides that “the President may not remove any item from the Munitions List until 30 days after the date on which the President has provided notice of the proposed removal to the Committee on International Relations of the House of Representatives and to the Committee on Foreign Relations of the Senate in accordance with the procedures applicable to reprogramming notifications.”  The government argues that the things it was removing from the list are not “items” covered by this notification requirement.  I don’t know whether DOJ is right about that, but it’ll obviously be a contested issue going forward.[2]

Third, the Executive can in any event ensure that the court’s injunction is limited to a month.  If the President does give Congress notice, there’s little question that State can remove Defense Distributed’s files from the USML 30 days later, thereby allowing Internet posting of the code—or State could do so, anyway, if that reversal is not arbitrary or capricious.[3]  Ultimately, then, it’ll probably be up to Congress (and the President’s veto pen) whether to do anything further to prevent the widespread distribution of the files.  And as to that question, “freedom of speech” shouldn’t be a factor.  

[UPDATE 1:  Reportedly the State Department made the decision to amend the USML and allow the posting of Defense Distributed’s files without consulting with the President, who's "glad" the district judge" enjoined the initiative so that he (the President) can "study" the issue.  

Also, there is legislation pending in Congress (see, e.g., the Senate version) that would (if my understanding is accurate) amend the Undetectable Firearms Act to prohibit the manufacture and possession of virtually all the products produced by 3-D printers, such that any actual use of Defense Distributed's CAD files would be unlawful.  The NRA successfully lobbied to stymie such legislation five years ago, and it is (so far) languishing in the Congress again.  Perhaps the current controversy will alter the political viability of the legislation.  Or perhaps not.  Moreover, as far as I can tell the legislation is not designed to regulate the distribution of the computer files with which someone could manufacture the weapons in question--only to prohibit their use.

On Tuesday, however, Senator Nelson introduced S.3304, legislation that would make it "unlawful for any person to intentionally publish, over the Internet or by means of the World Wide Web, digital instructions in the form of Computer Aided Design files or other code that can automatically program a 3-dimensional printer or similar device to produce a firearm or complete a firearm from an unfinished frame or receiver."  Nelson requested unanimous consent on the bill, but Senator Lee objected.  It'll be worth watching what becomes of S.3304.]

[UPDATE 2:  Scott Greenfield, mistakenly assumes, as do others, that my argument depends upon denying that "code is speech."  Any careful reader of this post, however, will understand that that's not the case.  I concede that code can be expressive to some "readers," and even that some persons might (conceivably) post code to the Internet for expressive purposes (although that's obviously not Defense Distributed's primary objective), such that the posting might be "speech," just as flag-burning might be speech.  But code-posting is not only (or even primarily) speech [does Adobe post Acrobat online in order to "speak" to me?] and, more to the point, the government's effort to regulate its posting is not based upon its potential to "speak" to any individuals.  It is, instead, based upon its functional capabilities. 

This part of Greenfield's post demonstrates the point nicely:

Some code causes the Constitution to appear on our computer screen. Some code causes Lederman’s law review articles to appear at SSRN. Some code makes a cute kitten pic show up. And some code provides the means to cause a 3D printer to create a gun.  
Let's play "Which of These Things is Not Like the Others?"

Yes, of course code not only can itself be expressive, but can also facilitate other speech (just as ink and paper can do).  And if the government were trying to restrict the posting of the CAD files because it resulted in certain speech popping up on my computer screen (even my own law review articles!), then that would almost certainly be unconstitutional.  But the ordinary and intended use of this code doesn't simply result in "cute gun pics" appearing on users' computer screens:  It "cause[s] a 3D printer to create a gun"!  And that's why the government restricts its indiscriminate distribution.  Hence, O'Brien scrutiny applies (at most), and the regulation does not violate the First Amendment, whether or not the code is, or can be, "speech."]



[1] Although the State Department has (unwisely, in my view) denominated regulated software as a category of “technical data,” see 22 CFR § 120.10(a)(4), it restricts the export of such software not because of its informational value, i.e., its value as data, but because of its physical functionality, i.e., its capacity to cause computers to create firearms.  By contrast, State does restrict the export of some “information . . . required for the design, development, production, manufacture, assembly, operation, repair, testing, maintenance or modification of defense articles,” such as “blueprints, drawings, photographs, plans, instructions or documentation,” id. § 120.10(a)(1).  (Think, for example, of someone who does not provide already manufactured defense articles to Chinese officials, but shares with them closely held instructions on how to manufacture them:  The federal government generally prohibits such information-sharing with foreign officials because of its informational impact.)  Unless practice has changed since I left government, State does not generally prohibit public Internet posting of such “technical data” that are regulated because of their informational value—and it would raise serious First Amendment questions if it did so, as OLC concluded in 1981.  But State regulates the posting of software that can create defense articles through ordinary technological means without regard to any informational function it might have—and that regulation is no more constitutionally problematic than the regulation of the export of the defense articles themselves (e.g., certain firearms), even though some people might be able to extract information, and learn, from the study of such articles and software.  

[2] The states also argue that Section 1(n) of Executive Order 13637 requires the concurrence of the Secretary of Defense before any change in USML designations.  Even if that Executive Order precondition applies here, however, I’d be surprised if the failure to obtain the SecDef’s concurrence is a justiciable matter.  See, e.g., Section 6(c) of that Order (“This order is not intended to, and does not, create any right or benefit, substantive or procedural, enforceable at law or in equity by any party against the United States, its departments, agencies, or entities, its officers, employees, or agents, or any other person.”).

[3] The plaintiffs argue that the decision to do so is, in fact, arbitrary and capricious:  “Defendants have provided no explanation for the Government’s complete reversal of its position on the files at issue, including its action to grant a ‘temporary modification’ to exclude the files from ITAR jurisdiction and to issue a letter stating that the files are exempt from ITAR’s export licensing requirements. The Government has released no reports, studies, or analyses to explain why downloadable guns should be removed from ITAR regulation. It appears that Defendants have also failed to consider or acknowledge the serious national security concerns or the threat to public safety posed to the States, created by the dissemination of these files.”




Tuesday, July 31, 2018

Taking Section Two of the Fourteenth Amendment Seriously

Gerard N. Magliocca

Building on Professor Fishkin's post, my new article on Section Two of the Fourteenth Amendment was published today. Section Two gets far less attention than Section One, but Section Two is vital for two reasons. The first is that this is the source of the rule that all persons shall be counted for purposes of national representation. (Litigation is pending in the Southern District of New York on whether the Commerce Department's proposal to add a citizenship question to the next census is unlawful in part because that decision was made to discourage an accurate count of non-citizens).

The other reason Section Two is worth studying is that the system used in 2011 to reapportion representatives among the states was unconstitutional. Reapportionment is not as glamorous as gerrymandering. Nonetheless, the question of how representatives are allocated among the states matters and the constitutional violation is hiding in plain sight. As the Article explains, Section Two of the Fourteenth Amendment says that the states shall be penalized in their representation if certain conditions are met. The reapportionment statutes, however, bar the imposition of any such penalty. These statutes are therefore invalid. (There is more to the article than this syllogism, of course.)

Unless Congress corrects this defect by 2021, it is my hope that some states will challenge the next reapportionment on constitutional grounds using the theory laid out in the paper.


Monday, July 30, 2018

Count All the People

Joseph Fishkin

The Fourteenth Amendment sets out a simple rule for the interstate allocation of political power in the United States: “Representatives shall be apportioned among the several States according to their respective numbers, counting the whole number of persons in each State, excluding Indians not taxed.” With this sentence, 150 years ago this month, the Constitution obliterated the three-fifths clause. Later, just under a century ago, we incorporated all Native Americans into the body politic by statute, so there are no more “Indians not taxed.” These changes have wrought a beautifully straightforward rule: our federal government allocates political power according to “the whole number of persons.” Not just the citizens. Not just the adults. Not just the people with valid immigration status, or the people a state has decided in its wisdom to make eligible to vote. All persons.

This hard-won bedrock constitutional principle structures the entire American political order under our feet. But that doesn’t mean Americans are all aware of it or understand it. To some Americans today (especially a lot of Republicans) it seems kind of crazy that we count non-voters such as children and even non-citizens, regardless of immigration status, in allocating political power. If they can’t vote, these critics ask, why do they count?

That way of thinking is now underwriting a flurry of profoundly ill-conceived activism by Republicans across the country who are aiming in a variety of ways to circumscribe the set of persons who count in our processes for allocating political power. One line of attack is a political lawsuit aiming to persuade the Census Bureau to violate the Constitution by refusing to count undocumented immigrants at all. (The lawsuit of course claims that the Constitution requires, rather than prohibits, this never-before-tried exclusionary practice.) Another line of attack aims to redraw state and local district lines so that instead of counting all the people, they count only citizens or eligible voters or some other measure. The highest-profile line of attack to date emanates from within the Trump administration, which is attempting to add a citizenship question to the Census in an effort to make such exclusionary redistricting easier for states and localities—while at the same time, in a fortuitous nativist twofer, suppressing the Census count itself in immigrant communities. The partisan stakes in all this are mostly pretty obvious (although not entirely). But sometimes the commentary about partisan effects obscures the underlying constitutional principle at stake.

Today in the Washington Post, I have an op ed offering a defense of the principle that we should count all the people. It’s pretty short; the defense is condensed. If you’re interested in a more in-depth discussion, you can take a look at this essay I recently published in a symposium in the William & Mary Law Review. It seems to me that we are now in the early stages of a twenty-first century reckoning with the question of who counts—on the Census, in apportionment, and across federal and state redistricting. The core underlying questions across these domains are fairly consistent. I think there are strong justifications for leaving in place the settlement reached in 1789 and 1868, later magnified outward by the core reapportionment revolution decisions of 1964: we count all the people, and we all count equally. I also think there is far more work to be done in this area. This debate is just beginning. I write this blog post in the hope that some readers may decide to write in this area.

Sunday, July 29, 2018

Understanding the Cycles of Constitutional Time, or, Why Things Are Going to Get Better

JB

I have posted my latest article, The Recent Unpleasantness: Understanding the Cycles of Constitutional Time, on SSRN. This is the written version of the 2017 Addison C. Harris Lecture  at the University of Indiana.  Here is the abstract:

This article, originally given as the 2017 Addison C. Harris Lecture at Indiana University, analyzes recent events in terms of three great cycles of change in American constitutional history. The first is the cycle of the rise and fall of political regimes. The second is the cycle of polarization and depolarization. The third is the cycle of the decay and renewal of republican government--the cycle of constitutional rot. Each of these cycles operates on a different time scale. Their interaction generates "constitutional time."

Many commentators worry that the United States is in a period of constitutional crisis, or that American democracy is doomed. These fears, although understandable, are overstated. America is not in a constitutional crisis, although it is suffering from a fairly severe case of constitutional rot, connected to rising polarization and economic inequality.

Our current difficulties are a temporary condition. They stem from the fact that the Reagan regime that has structured American politics since the 1980s is dying, but a new regime has yet to be born. This is a difficult, agonizing, and humbling transition; and its difficulty is enhanced by the fact that, unlike the last transition, it occurs at the peak of a cycle of polarization and at the low point of a cycle of constitutional rot. For that reason, the transition to a new political regime is likely to be especially difficult. But we will get through it. And when we get through it, about five to ten years from now, American politics will look quite different.

Political renewal is hardly foreordained: it will require persistence and political effort. The point of this lecture is to offer a bit of hope in difficult times. If people misunderstand our situation, and conclude that American decline is inevitable, they may unwittingly help to make that fate a reality; but if they understand the cycles of constitutional time, they may come to believe that their democracy can be redeemed, and do their part to realize that worthy goal.

Wednesday, July 25, 2018

Accountability for the Internet of Torts

Guest Blogger

Rebecca Crootof

Tort law has always shaped political economy in the wake of technological developments. Sometimes it operates to protect the powerful; sometimes it intervenes in power relations to correct new imbalances. The history of tort law can be understood as a series of case studies in how new technologies enable new conduct and harms, and in how judges and legislatures changed the law to address the resulting power dynamics between industry and individuals. The concept of ultrahazardous activities, the creation of no-fault workers’ compensation and motor vehicle insurance, and the rise of mass tort litigation can all be partially traced to underlying technological changes and accompanying social shifts.

Today, we are at the inflection point of another such transformation. In an earlier post, Introducing the Internet of Torts, I discussed how Internet of Things (IoT) companies are able to create and impose their own contractual governance regimes. They use terms of service to displace the law of the state, and they employ technological self-help to enforce their rules. Furthermore, the physicality of IoT devices increases the likelihood of consumer property damage and physical harm when companies discontinue service or otherwise engage in digital repossession. In this post, I will use prior tort law revolutions as a springboard to discuss how new products liability law and fiduciary duties could be used to rectify this new power imbalance and ensure that IoT companies are held accountable for the harms they foreseeably cause.

Read more »

Tuesday, July 24, 2018

Introducing the Internet of Torts

Guest Blogger

Rebecca Crootof

Once upon a time, missing a payment on your leased car would be the first of a multi-step negotiation between you and a car dealership, bounded by contract law and consumer protection rules, mediated and ultimately enforced by the government. You might have to pay a late fee, or negotiate a loan deferment, but usually a company would not repossess your car until after two or even three consecutive skipped payments. Today, however, car companies are using starter interrupt devices to remotely “boot” cars just days after a payment is missed. This digital repossession creates an obvious risk of injury when an otherwise operational car doesn’t start: as noted in a New York Times article, there have been reports of parents unable to take children to the emergency room, individuals marooned in dangerous neighborhoods, and cars that were disabled while idling in intersections.

This is but one of many examples of how the proliferating Internet of Things (IoT) enables companies to engage in practices that foreseeably cause consumer property damage and physical injury. But how is tort law relevant, given that these actions are authorized by terms of service and other contracts? In this post I’ll elaborate on how IoT devices empower companies at the expense of consumers and how extant law shields industry from liability. In a future post, Accountability for the Internet of Torts, I’ll discuss what we can learn from prior tort law revolutions about how the law might evolve to hold these companies accountable. Overarchingly, a political economy perspective highlights how technological developments are not neutral—they enable new conduct, new relationships, and new kinds of harm that disproportionately affect the poor—and how law can be used either to preserve or correct resulting power imbalances.

Three characteristics of IoT devices—their ability to collect personalized data, their capacity for communication with a cloud-based service provider, and their physicality—combine to form a product that is simultaneously an object and an ongoing service. And, in most cases, an IoT device’s utility depends on the perpetual provision of that service: without Alexa Voice Service, an Amazon Echo is merely an expensive doorstop. As a result, instead of an association that ends with the purchase of an item, consumers now enter into in an ongoing relationship with IoT companies.

We’ve seen how connected products enable industry overreach before. Digital tech companies have long employed terms of service and digital rights management technologies to limit how consumers use purchased products—say, to keep consumers from sharing music or accessing an e-book. Borrowing tactics from earlier digital tech companies’ playbooks, IoT companies are using their terms of service to lock consumers into contractual governance regimes, thereby supplanting the “law of the state” with the “law of the firm.” They can then engage in digital repossession and other forms of technological self-help to enforce these contracts, sidestepping the state’s checks on unfair contractual provisions.

Such practices are concerning enough in the digital world, but they have even more sinister implications in the IoT context. IoT companies can harness devices’ extensive surveillance capabilities to impose and monitor compliance with increasingly invasive terms; they can condition necessary security and software updates on consumers’ assent to contract modifications; and they can digitally repossess items by remotely impairing or completely disabling devices.

But there’s also an entirely new problem: Because an IoT device interacts with the physical environment, there is an increased risk that consumers will suffer property damage and physical harm should a company digitally repossess it. Smart fridges are marketed as being able to warn you of food spoliation, but a disconnected one might increase your chances of food poisoning. You might sleep soundly, trusting an IoT baby monitor, senior lifeline, home security system, or fire alarm to notify you of a problem—but should a company remotely deactivate the alert system, your reliance could lead to tragedy. Your front or garage door could be left open because you left a bad review on Amazon. And IoT medical devices make the risks of digital repossession all the more obvious. If, as Ryan Calo has quipped, robots are “software that can touch you,” IoT devices are contracts that can hurt you. And these harms are most likely to fall on the poor, as they will disproportionately suffer the consequences of digital repossessions.

Classically, an injured individual could bring a tort suit to seek compensation for harm. But in addition to social and practical deterrents, a would-be plaintiff suffering from an IoT-enabled injury faces three significant legal hurdles. First, there is likely a contract between the consumer and the company that authorizes the digital repossession. Contract law does not price harms: it leaves that determination to the bargaining parties, sometimes to the extent that it “launder[s] injustice.” Of course, companies cannot overreach in setting their terms, especially when consumers have limited ability to negotiate those terms. When they do, the contract can be declared unconscionable, and tort law will operate as a backstop to price the harms. Absent a better understanding of how IoT-enabled harms scale, however, judges are unlikely to declare clauses limiting liability unconscionable when evaluating individual cases.

Even if the contract is struck, a plaintiff will still need to prove breach of a duty and causation. But there is little clarity about what duties an IoT company owes users. And proving causation will be fraught, as the doctrine of intervening causes will likely relieve the company of liability. A digitally booted car doesn’t hurt you—it is the other car that hits you when you’re trying to escape an intersection that is the direct cause of your injury.

In short, IoT devices enable both familiar and new opportunities for harmful industry overreach, with the added twist that these practices can now cause physical harm and property damage. Simultaneously, there is little government oversight or routes of recourse for injured consumers under extant law.

But law can evolve. As with prior, technological change, the proliferation of IoT devices will necessitate a reconsideration of familiar liability analyses. In my next post, I will discuss how expanded understandings of duty and causation could correct the IoT-enabled power imbalance between companies and consumers.

Crossposted at Law and Political Economy

This post is adapted from a forthcoming paper, “An Internet of Torts.” Rebecca Crootof is a Research Scholar and Lecturer in Law at Yale Law School and the Executive Director of the Information Society Project.


Monday, July 23, 2018

The Rulification of Penalty Kicks—and a Reform Proposal

David Pozen


The 2018 soccer World Cup was the first to use Video Assistant Referees (VAR). VAR allows decisions by the head referee involving goals, penalties, direct red cards, and “mistaken identity” to be reviewed, immediately afterward, with the aid of video footage. Not coincidentally, the 2018 World Cup was also the first to feature upward of twenty penalty kicks. At the 2014 tournament in Brazil, a total of thirteen penalty kicks were called, not including shootouts. In Russia, the number was twenty-nine.

The criteria for awarding penalty kicks have not changed. According to the official laws of the game, if a player commits a foul punishable by a direct free kick inside her own penalty area, “[a] penalty kick is awarded.” Each and every time a player inside this zone pushes an opponent, trips an opponent, handles the ball deliberately (unless she is a goalkeeper), and so on, the opposing team gets a penalty kick.

In legal-theory parlance, the criteria for awarding penalty kicks are “rules” rather than “standards.” They are clear and precise—not completely clear or precise, as terms like “trip” and “push” go undefined, but relatively so—and they give little discretion to the referees who enforce them. Other laws of soccer were designed from the outset to be flexible and context-sensitive: for instance, the laws empowering referees to give yellow cards for “reckless challenges” and red cards for “excessive force.” The laws governing penalty kicks are not like that. If one of an enumerated list of behaviors is found to have occurred, a penalty kick follows.

This, at least, is the law on the books. The law in action has long been different. As all soccer devotees know, referees sometimes decline to award penalty kicks in situations where the formal laws suggest they are mandatory. If, say, the collision in the box looks innocuous or inadvertent, or if the fouled player was unlikely to score anyway, or if the incident takes place near the end of a close contest, many referees seem more inclined to let play continue. Informing these judgments are intuitions about soccer justice and an appreciation that in such a low-scoring game, the penalty kick is a draconian sanction—leading to a goal more than two-thirds of the time. It’s roughly comparable to a basketball referee awarding one team fifty foul shots, all in a row.

With their decisions subject to review by a phalanx of off-field “assistants,” however, the referees in Russia no longer felt free to apply their situation sense and to refrain from awarding penalties that may have been technically warranted but seemed unduly harsh, given the circumstances. France’s second goal in the final was arguably a case in point. On a few occasions, video review revealed that an apparent foul had not in fact occurred and led to the reversal of a penalty kick that had been whistled on the field, as with Neymar’s dive against Costa Rica. But overall, by subjecting referees to real-time, panoptical scrutiny, VAR made the policing of penalties more severe as well as more mechanical.

The introduction of VAR thus exposed a gap between the law on the books and the law in action. And the effect was to rulify the adjudication of penalty kicks. Under the gaze of FIFA’s all-seeing thirty-three broadcast cameras, a nuanced standard that had developed over many years without ever being written down—a standard that prioritized the punishment of blatant fouls and denials of goal-scoring opportunities—gave way to a comparatively rigid rulebook that recognizes no distinctions among more and less “penalty-worthy” trips, pushes, or the like. Transparency left less room for subtlety. Codified law swallowed custom.

Is this a good or a bad thing? I am inclined to be negative about this aspect of VAR. (The earlier introduction of goal-line technology, in contrast, did not undermine any customs of refereeing or introduce any delays in play, and strikes me as a boon for the game.) Although penalty kicks may now be called more consistently, they will also be called more frequently and mindlessly. FIFA’s president insists that “VAR is not changing football, it is cleaning football.” Yet we know from other contexts that enforcing longstanding laws more aggressively or literalistically can be a deeply disruptive, if not subversive, act. To “cleanse” soccer of enforcement discretion is to change the sport.

My own instinctual negativity likely reflects some combination of nostalgia, technoskepticism, and aesthetic taste. But the degree to which VAR has divided opinion also reflects, I suspect (loosely in line with Duncan Kennedy’s famous analysis in “Form and Substance in Private Law Adjudication”), different orientations toward rules, standards, expertise, and the rule of law. For those soccer fans who are “rules people” and take clarity, predictability, and impersonality to be the essence of a well-functioning legal system, VAR may seem obviously superior to the opacity and ad hockery of the old regime. For those who place greater trust in the professional judgment of on-field referees and greater emphasis on the avoidance of substantively unfair outcomes, on the other hand, VAR may come across as alienating, crude, even callous.

Perhaps we can bridge some of the space separating these two camps. VAR’s critics must concede that the system has real benefits, particularly in cases where the referee simply could not see what happened on the field while it was happening. To capture these benefits without straightjacketing referees, I wonder whether the laws of the game might themselves be made more standard-like through the use of a new intermediate sanction. For example, they could instruct referees to award ordinary penalty kicks for egregious fouls and fouls that deny a clear goal-scoring opportunity, but indirect kicks or unobstructed direct kicks from eighteen yards out (rather than twelve) for all other fouls in the penalty area.

Any such proposal is bound to be enormously controversial. Whatever their views on its merits, legally and philosophically inclined fans might at least agree that the way VAR has transformed the practice of penalty kicks supplies an interesting case study in the jurisprudence of sport, the instability of rules and standards, and the potential for technological change to disrupt sociolegal norms.

Older Posts

Home