Professor Ken Grady of Michigan State University College of Law published a compelling article this morning about the decline of excellence in the craft of lawyering. In it, he asks a series of questions that lawyers and legal educators alike should consider:

What does it take to become a lawyer who excels at what he or she does? What type of commitment is involved? What sacrifices must that lawyer make? What are the consequences of not excelling at the craft? Why aren’t the lawyers at the other firms, all of whom charge premium fees for their services, excelling at their craft?

Underlying all of these questions is another, even bigger one: what is excellence? 

Attempting to define what it means to be an excellent lawyer reveals many of the anxieties of modern legal practice. An attorney may work hard, bill the requisite number of hours, win a few cases, and still lie awake at night wondering, “am I excellent?” Or in other words, am I thriving in my field, or just surviving? 

Excellence, then, means more than just working hard and billing hours. Prof. Grady’s article explicitly recognizes this. He identifies effective legal writing, knowledge of clients’ businesses, and the development of deep expertise in novel areas such as quantitative analysis as paramount.

How can law schools inculcate excellence? 

The next obvious question is what law schools can do to prepare students to become excellent attorneys. Prof. Grady does this, actively, by teaching courses geared toward legal tech and supporting the LegalRND program at Michigan State, which I hear is truly excellent.

The first step for law schools is to be frank about the realities facing the profession.

Graduating law students must be savvy about the limits and pitfalls of traditional legal practice—including the billable hour model. Recently, Thomson Reuters’ Legal Executive Institute declared the billable hour model “effectively dead.” By 2020, alternative fee arrangements will play a key role in the attorney-client relationship, with serious implications for the profession as a whole. Even in commercial litigation, firms are increasingly being asked to share in the risk of failure. Internally, firms have adopted systems to calculate the value of legal services being delivered. Over 90% of firms with more than 250 attorneys have already adopted such practices, and a trickle-down effect is likely as the profession moves into the third decade of the twenty-first century. Meanwhile, law schools operate under a Langdellian model of legal education developed by Harvard in the 1800s. To the degree that law schools have attempted to emphasize practical skills, many do so using a model that caters to what BigLaw looked like twenty years ago. We can, and should, do better.

Pursuing excellence requires a mindset shift, one that isn’t bolstered by attending a traditional law school. You see, the real enemy of excellence isn’t mediocrity—nobody wants to be mediocre—but fear. Fear tells us to follow the safest path, that already trod by others. Fear tells us we aren’t good enough to be different, or that if we try, we might fail. Law school does nothing to abate, and much to validate, this sort of fearful thinking that precludes excellence. Keeping up with one’s peers becomes a fear of falling behind. Excessive debt acts as a disincentive to taking professional risks. It’s no wonder that legal education has been criticized for stymieing creativityIf you want to pursue excellence, give up on perfectionism.

Similarly, lawyers at firms large and small aren’t shirking from excellence because they’re lazy. As Prof. Grady notes in his article, it’s more frequently because they’re simply too busy, or think they’re too busy, to learn about data or their clients’ businesses. These activities aren’t billable, so they don’t matter. I’ve been lucky enough to work in environments where this attitude is changing, but it’s certainly fairly pervasive among practicing attorneys. Law schools can play a role in reversing the trend.

The legal field is weeding out mediocre lawyers; at the same time, the paths to achieving excellence as an attorney are multiplying. The profession is becoming increasingly multipolar as the hegemonic influence of BigLaw declines, with multiple actors in the legal system positioning themselves such that they are better-suited to take on tasks traditionally done by associates at large firms. This may result in a lot of fear among the powers-that-be. Ultimately, we need to seriously consider what excellence looks like — as attorneys, and to our clients — while at the same time we seek to cultivate it. It’s going to take a major wake-up call for many lawyers.

But for those of us just entering the field, the times couldn’t be more exciting! What it means to be a lawyer, an excellent one, will change more in the next twenty years than the past two hundred. Personally, I can’t wait to see what the future holds.

Given the powerful influence of Silicon Valley and the dysfunction of the current political moment, it’s likely that we’ll only get one chance at seriously regulating Facebook. The policy solutions that emerge from the Facebook Freakout of 2018 must include changes to the ways in which the government can use personal data. 

Missing from the discourse surrounding the Cambridge Analytica controversy—much of which has centered on whether Facebook breached its own terms of service (it probably did) and the degree to which the 2016 election turned on the disclosure (it probably didn’t)—is the recognition that ultimately, this is a story about the abuse of power. In fact, while Mark Zuckerberg prepares to potentially testify before governments on both sides of the Atlantic, the parent company of Cambridge Analytica is developing contacts with officials in the American government and military. From the Washington Post:

“The company, SCL Group, has hired additional staffers who are working out of a new office down the street from the White House. It has in recent weeks pitched officials in key national security agencies on how its technology could be used to deter terrorism, bolster the military’s capacities as it prepares for a possible buildup and help assess attitudes about immigrants.”

In other words, the same data used to influence the 2016 election, and algorithms derived therefrom, are potentially being used as political tools to drum up support for new policing and immigration strategies. All policy disagreements aside, this shift seems to cross the line from effective campaigning to Orwellian propaganda.

Worse still, the very use of microtargeting in political messaging is inherently anti-democratic, a point which can easily be missed in discussions of technical minutiae and Facebook’s terms of service. Cathy O’Neil, whose book Weapons of Math Destruction should be required reading for all those interested in data science, recognized the problem even before the 2016 election:

“The result of these subterranean campaigns is a dangerous imbalance. The political marketers maintain deep dossiers on us, feed us a trickle of information, and measure how we respond to it.  But we’re kept in the dark about what our neighbors are being fed. This resembles a common tactic used by business negotiators. They deal with different parties separately so that none of them knows what the other is hearing. This asymmetry of information prevents the various parties from joining forces—which is precisely the point of a democratic government.”

Politicians—and now, government institutions using the algorithms developed using Facebook data—can leverage informational asymmetry in order to “become all things to all people.” And here we divert from the model of the Orwellian government as “Big Brother”: a monolithic entity with a universal dogma ascribed to by all citizen-adherents. The new political nightmare is the government as chimera, subtly shaping dominant narratives in order to appeal to the preferences and anxieties of each individual. It’s ideology à la carte. 

Deleting Facebook won’t solve the problem of political microtargeting as there’s already a large enough data set that any one individual’s participation is largely irrelevant. The coming regulatory response to the Cambridge Analytica scandal will likely focus on the circumstances under which user data can be shared with third parties. This doesn’t go far enough.

Instead, reformers should advocate an approach that addresses both the collection and sharing of user data as well as the use of said data in the algorithms that increasingly govern private relations in society, including exposure to political messaging. O’Neil calls these algorithms “Weapons of Math Destruction.” 

A complete regulatory response would start by requiring that those who wield these digital WMDs open their logic to public scrutiny by making the code open source. When the government uses algorithms in decision-making or political messaging, a Freedom of Information Act request for the underlying code and data should be granted as the public has a legitimate, cognizable interest in understanding how these functions operate. While regulations are largely doomed to play catch-up in the face of the rapid growth of Big Data, public access can at least allow independent auditing and feedback to ensure that governmental and private entities do not exploit its power.

Microtargeting is both nefarious and insidious because it plays to our deepest unrecognized fears and biases. For individuals, awareness of microtargeting should encourage us to scrutinize what comes across our screens and reflect on whether our own predilections are making us vulnerable to political manipulation. Hopefully, the present controversy will serve as a wake-up call to those who would prefer to ignore the dark side of Big Data and the information we share online. What the Cambridge Analytica controversy and Weapons of Math Destruction thesis reveal is that in a world where out information is used against us and others, “I have nothing to hide” is no longer an adequate justification for ignoring issues of privacy and social media.

On February 5th, 2018, the United States Court of Appeals for the Ninth Circuit ruled that California wage and hour laws may protect workers on fixed drilling platforms in the outer continental shelf.[1] The case, Newton v. Parker Drilling Mgmt. Serv., Ltd., involves a putative class action brought by an oil rig worker employed off the coast of Santa Barbara. Plaintiff Brian Newton worked fourteen-day shifts comprised of twelve-hour “on duty” periods followed by twelve-hour “controlled standby” periods. His principal claim against the defendant, Parker Drilling, is that he was not properly compensated under state law for the standby hours spent on the rig. The district court previously dismissed Newton’s state law claims, holding that the Fair Labor Standards Act (FLSA) provides the exclusive remedy for employees working on the outer continental shelf (OCS).

The dispute centers on the proper interpretation of the Outer Continental Shelf Lands Act (OCSLA), 43 U.S.C. § 1333(a)(2)(A), which states that “[t]o the extent that they are applicable and not inconsistent with this subchapter or with other Federal laws . . . the civil and criminal laws of each adjacent State, now in effect or hereafter adopted, amended, or repealed are declared to be the law of the United States for that portion of the subsoil and seabed of the outer Continental Shelf, and artificial islands and fixed structures erected thereon, which would be within the area of the State if its boundaries were extended seaward to the outer margin of the outer Continental Shelf.” Thus, the precise question is whether the additional protections offered by the California Labor Code are “inconsistent with” the FLSA, which does not recognize the sort of claim brought by Newton. In general, California labor law offers powerful tools to employee-litigants; these include the Private Attorneys General Act, which authorizes aggrieved employees to file lawsuits to recover civil penalties on behalf of themselves, other employees, and the State of California for Labor Code violations.[2]

The OCSLA determines what laws are applicable to installations which are permanently or temporarily attached to the seabed of the outer continental shelf.[3] The shelf is fully within federal jurisdiction and governed by federal law, except that state laws may be raised to the status of “surrogate” federal law under certain conditions provided for by the OCSLA. Before the Ninth Circuit, the Newton defendant argued that only those state law provisions which are necessary to fill in a gap or void in federal law are incorporated by the Act. Since the FLSA is a complete body of federal law unto itself, such a reading would foreclose state law employment claims.

Yesterday, the Ninth Circuit rejected this restrictive interpretation of the phrase “applicable and not inconsistent.” First, the court held that the term “applicable” does not imply the need for a gap in federal law. Rather, the word simply requires that the state statute is capable of application to the subject matter of the dispute. If Congress explicitly desired that state law would only serve to fill gaps in federal maritime law, the drafters of the OCSLA could simply have replaced the word “applicable” with “necessary.” Congress’ failure to do so, according to the Ninth Circuit, supports a more expansive interpretation of § 1333(a)(2)(A).

More interesting than the Ninth Circuit’s textual analysis of the OCSLA, however, is its discussion of legislative history. Departing from the Fifth Circuit, which held that the primary aim of the statute was to assert the supremacy of federal law over the OCS, the court highlighted passages from the OCSLA’s legislative history which note that merely extending admiralty law to installations on the OCS would be inadequate to protect workers.[4] In any event, the Ninth Circuit found that the legislative history did not overcome the ordinary meaning of “not inconsistent” or furnish a basis for reading the word “necessary” into § 1333(a)(2)(A).

This left the court with a very basic question: is the California labor code “inconsistent with” the Fair Labor Standards Act? The defendant Parker argued that the laws necessarily conflict because the Fair Labor Standards Act does not count “controlled standby” as hours worked.[5] Instead of seeing this as an inconsistency, however, the Ninth Circuit conceived of the FLSA as a floor above which state labor code provisions may establish a protective ceiling. In other words, a more protective state standard is still “not inconsistent with” the Act. This conception finds support in the FLSA’s savings clause.[6] In light of its holding that the FLSA and OCSLA did not combine to preempt Newton’s state law claims, the court vacated the district court’s order dismissing the action and remanded the case for further proceedings.

Because California has a long coastline (and generous employment laws), the Newton decision has the potential to create a new wave of collective employment litigation in the Pacific outer continental shelf. As the case progresses to the class action stage, observers will be watching to see how the court disposes of novel issues of commonality, typicality, and representativeness, all of which are complicated by the offshore nature of the work. Newton will also raise the specter of similar actions being brought under the state labor law provisions of Washington, Oregon, and Alaska. For offshore employers subject to the OCSLA—especially those operating adjacent to California waters—the shifting judicial winds may indicate a gathering storm.

[1] Newton v. Parker Drilling Mgmt. Serv., Ltd., –F.3d–, No. 15-56352, 2018 WL 706490 (9th Cir. Feb. 5, 2018).

[2] Cal. Lab. Code § 2698 et seq.

[3] Rodrigue v. Aetna Cas. & Sur. Co., 395 U.S. 352, 355 (1969) (“The purpose of the [OCSLA] was to define a body of law applicable to the seabed, the subsoil, and the fixed structures such as [drilling platforms] on the outer Continental Shelf.”).

[4] See 99 Cong. Rec. 6963 (1953) (arguing that under pure admiralty law, “[t]he so-called social laws necessary for protection of the workers and their families would not apply. I refer to such things as unemployment laws, industrial-accident laws, fair-labor-standard laws, and so forth. It was necessary that the protection afforded by such laws be extended to the outer Shelf area because of the fact that ultimately some 10,000 or more men might be employed in mineral-resource development there”).

[5] See Mendiola v. CPS Security Solutions, Inc., 60 Cal. 4th 833, 837 (2015) (comparing the FLSA and California law on this question).

[6] See 29 U.S.C. § 218(a) (“No provision of this chapter or of any order thereunder shall excuse noncompliance with any Federal or State law or municipal ordinance establishing a minimum wage higher than the minimum wage established under this chapter or a maximum work week lower than the maximum workweek established under this chapter . . .”).

Under federal copyright law, a copyright holder has the right to control the public performance of their works, including audio reproductions.[1] A “public performance” includes playing a music artist’s song in a club.[2] A quartet of licensing entities—BMI, ASCAP, SESAC, and GMR—holds the licenses to nearly all commercial music, having purchased performance rights from songwriters. These are commonly referred to as “performing rights organizations.” Businesses seeking a license to play a piece of music must first negotiate with one or more of the organizations to obtain performing rights. The question that frequently arises is whether purchasing the rights to a song from one of the licensing entities allows a business to play/perform that music when there were multiple songwriters and not all of them are represented by the organization. The traditional answer to this question is no—a business must have a license from the performing rights organization that owns the performance rights of each songwriter. This is described as fractional licensing.

However, in recent years, the government has sought to impose full-work licensing which would allow any copyright holder in a song that has multiple owners to license the song on behalf of all copyright holders—this would mean that a deal with any of the licensing groups (ASCAP, BMI, etc.) would suffice to protect an individual or entity seeking to license a particular song.[3] For the performance rights organizations, full-work licensing could create problems. On September 16, 2017, the District Court for the Southern District of New York ruled that consent decrees entered into by ASCAP and BMI “neither barred fractional licensing nor required full-work licensing.”[4] The Second Circuit affirmed this ruling on December 19, 2017.[5] Unless this ruling is successfully appealed, businesses must negotiate licensing agreements with each of the performing rights organizations.

The Case: United States v. Broadcast Music, Inc. 

The dispute between the United States and BMI centers on the interpretation of consent decrees entered into as a result of previous antitrust litigation. In the antitrust cases, plaintiffs challenged the blanket licenses sold by the performing rights organizations as illegal price fixing and a concerted refusal to deal.[6] The consent decrees hold that BMI may not discriminate among licensees—it must offer to anyone, at a reasonable fee, a “license for the right of public performance of any, some or all of the compositions in [its] repertory.”[7] In its appeal brief, the United States seized on the word, “the,” preceding “right of public performance” to argue that BMI must offer licensees the right to play a song, even though one of the songwriters is not signed up with BMI.[8] In other words, the right must be complete at the time when it is offered, rather than fractional.

The Second Circuit rejected this textual argument, holding that “the blanket license itself does not necessarily confer a right of immediate public performance: the license covers all the rights held by the [performance rights organization] regardless of whether those rights are valid or invalid, exclusive or shared, complete or incomplete.”[9] For guidance, the court looked to the use of “the right of public performance” as a term of art in the Copyright Act through the lens of industry practice at the time the consent decrees were entered into.[10] The Copyright Act allows partial licensing by copyright holders, as well as separate ownership by individual co-owners, and this was standard industry practice in the music business at the time of the most recent amendment to the consent decree between the United States and BMI.[11] Finally, the court held that the DOJ could not, by invoking the goals of the Sherman Act, “read an additional requirement into the decree to advance . . . procompetitive objectives.”[12] Rather, the terms of the decree itself govern the relief that may be granted, even if there are substantive policy reasons supporting its reinterpretation.[13] Full-work licensing, though it might benefit consumers, will have to wait.[14]

The Future of Commercial Performance Rights

But what if the government had won? At present, commercial users of music generally have licenses with all of the major performance rights organizations. This circumvents the issue of fractional licensing by allowing the commercial user to piece together a full license even when there are multiple songwriters so long as each is signed up with one of the licensing entities. But this state of affairs has obvious anti-competitive characteristics. Full-work licensing, on the other hand, would change things significantly—a business could sign up with one of the major organizations and receive the rights to each song in that organization’s catalogue, whether or not the organization has rights from each songwriter. It would then be up to the major performance rights organizations to create an arrangement wherein, for example, BMI could pay ASCAP the partial royalties from the use of a given song with rights-holders represented by both organizations.[15] According to the professional rights organizations, this would result in “disarray.”[16] In any event, it would be difficult to predict how the major licensing entities would react and whether their collective reaction would actually benefit commercial music users as the government seeks.

In any event, shakeups in the music rights industry are ongoing and impact the analysis of full-work licensing versus partial licensing. For example, the traditional balance of power among the quartet of rights organizations has been upended by two developments: first, the emergence of GMR, a smaller entity that has more preferential licensing terms for artists, and second, the increasing resistance of artists to signing up with a licensing entity at all. While both trends put pressure on the larger performance rights organizations, the second is importantly notable as it shows how “the licensing of performing rights is starting to become much more fractured.”[17] From an economic standpoint, this fracturing will only increase transaction costs across the commercial licensing spectrum, even as it creates individual gains for new actors.

Yet it doesn’t necessarily follow that the imposition of full-work licensing will result in a “better” market for commercial music licenses. For example, SESAC and GMR argued as amici curiae before the Second Circuit that full-work licensing resulting in royalty transfers between the rights organization would defeat the purpose of an artist’s choice of a particular organization as they would still be obligated to receive a portion of the royalties collected by another licensing entity.[18] This would be true despite a lack of contractual privity between the artist and the other entity.

In a sense, then, the government’s interpretation of the BMI consent decree would simply flip the landscape of commercial music rights; the artists, rather than the commercial users, would be beholden to all of the performance rights organizations in the event that they share the rights to a particular song. Taking a charitable view of the government’s position, however, there may be reason to believe that songwriters could more easily align their performance rights as to a piece of music than commercial users. Under full-work licensing, the incentive to do so would be even greater. But the fact that most songwriters are already signed up with a performance rights organization presents a significant barrier to such an alignment, as there are practical difficulties and contractual arrangements that would be disturbed in the process.[19]

Ultimately, the state of affairs surrounding commercial music licensing results from an information gap—artists cannot themselves monitor whether their rights are being violated by commercial users and cannot contract individually with each potential user, so they must rely on a third party performance rights organization to ensure that they are receiving deserved royalties. If this information gap were bridged, it would no longer be necessary for artists to sell their rights to a member of the licensing quartet to act as a middleman. Indeed, blockchain companies are already seeking to use technology to achieve this goal.[20] But given the stranglehold that the performance rights entities currently maintain over the music industry, there will be many challenges to such an effort going forward. Commercial users are presently left with few options as a result of the Second Circuit opinion in United States v. Broadcast Music, Inc. At least for now, the song remains the same.

Thanks for reading!

Citations: 

[1] 17 U.S.C. § 106(4); 17 U.S.C. § 106(6).

[2] See H.R. Rep. No. 94-1476, at 64-65 (1976) (discussing Congress’s intention to capture “semipublic” places such as “clubs, lodges, factories, summer camps, and schools” within the public performance right); ABC, Inc. v. Aereo, Inc., 134 S. Ct. 2498, 2506-07 (2014) (interpreting the public performance right in the content of a cable system’s performance of audiovisual works to the public).

[3] Paula Parisi, ASCAP Joins BMI in Dept. of Justice Licensing Smackdown, Variety, Aug. 24, 2017, http://variety.com/2017/biz/news/ascap-joins-bmi-in-dept-of-justice-licensing-smackdown-1202538745/.

[4] United States v. Broadcast Music, Inc., 207 F.Supp.3d 374, 377 (S.D.N.Y. 2016).

[5] United States v. Broadcast Music, Inc., No. 16-3830-cv, 2017 WL 6463063 at *1-3 (2d Cir. 2017).

[6] See Broadcast Music Inc. v. Columbia Broadcasting System, Inc., 441 U.S. 1, 6 (1979).

[7] Brief of Appellant, United States v. Broadcast Music, Inc., No. 16-3830-cv, 2017 WL 6463063 (2d Cir. 2017), at *27.

[8] Id. at *31.

[9] Broadcast Music, Inc., 2017 WL 6463063 at *2.

[10] Id. at *2-3; see 17 U.S.C. §§ 106(4), 201(a).

[11] Id. (“Each individual co-owner has a right to public performance, and such a right is not associated specifically with “full-work” licensing or with an indivisibility principle.”).

[12] Id. at *3.

[13] Id. (quoting United States v. Armour & Co., 402 U.S. 673, 683 (1971)).

[14] Id. (“If the DOJ decides that the consent decree, as interpreted by the district court, raises unresolved competitive concerns, it is free to move to amend the decree or sue under the Sherman Act in a separate proceeding.”).

[15] See Brief of Amici Curiae Sesac, Inc. and Global Music Rights, LLC in Support of Affirmance, United States v. Broadcast Music, Inc., No. 16-3830-cv, 2017 WL 6463063 (2d Cir. 2017), at *21-22 (“Under the government’s new decree interpretation, a mechanism would have to be forged—presumably through some direct discussions among competing PROs—on how to account to each other and their songwriter affiliates and members for the fractional interests each represents.”).

[16] Id.

[17] Coe W. Ramsey, Music Performing Rights Organizations and the “Full-Work” vs. “Fractional” Licensing Dispute: Government Seeks to Overturn Fractional Licensing Decision, Lexology, May 23, 2017, https://www.lexology.com/library/detail.aspx?g=5169651d-513b-4047-8263-9a994f0b8193.

[18] See Brief of Amici Curiae Sesac, Inc. and Global Music Rights, LLC in Support of Affirmance, United States v. Broadcast Music, Inc., No. 16-3830-cv, 2017 WL 6463063 (2d Cir. 2017), at *22-23 (“For example, a songwriter who has left BMI for GMR but co-writes with a BMI writer would be required to accept royalty payments and accounting from BMI, despite the fact that the former BMI writer has no contractual privity with BMI and no ability to audit or otherwise confirm that the royalty payments are accurate. Smaller PROs like GMR are attractive to songwriters for the very reason that they offer their writers audit rights and other unique contractual benefits. Not only would the former BMI writer in the example above be deprived of, among other things, the benefit of the audit right provided in the writer’s GMR agreement, but the former BMI writer would also have no mechanism by which to challenge payments received by BMI.”).

[19] See id. (describing these reliance interests with reference to both the rights organizations and songwriters).

[20] See Imogen Heap, Blockchain Could Help Musicians Make Money Again, Harvard Business Review, Jun. 5, 2017, https://hbr.org/2017/06/blockchain-could-help-musicians-make-money-again.

 

The New York Patrolmen’s Benevolent Association (PBA) is suing the NYPD and Mayor de Blasio in order to block the release of body camera footage without a court order. The lawsuit is premised on New York’s state FOIA, which exempts “police personnel files” from release to the public. I will be addressing this exemption in depth in a forthcoming law review article, but for now the New York lawsuit brings up important concerns worth addressing.

The New York law states in pertinent part:

“All personnel records used to evaluate performance toward continued employment or promotion . . . shall be considered confidential and not subject to inspection or review without the express written consent of such police officer . . . except as may be mandated by lawful court order.” N.Y. Civ. Rights Law § 50-a.

According to the PBA, this provision prevents the unilateral release of body camera footage by the NYPD or the Mayor’s office, which has occurred in previous instances where body camera footage shed light on police shootings. In principle, this makes sense: unilateral, ad-hoc release of redacted footage without guidelines for what sort of footage should see the light of day is not a firm basis for ensuring freedom of information. However, the remedy that the PBA seeks—a blanket ban on disclosure without a court order—goes too far in the other direction. Rather, it is up to the New York legislature, and other state legislatures around the country, to ensure that the framework for the release of body camera footage is coherent and serves the public interest in ensuring that community policing strategies and internal affairs investigations are above reproach (once my article is out they can simply borrow from my model statute!). Using the “police personnel files” exemption to swallow up all on-duty footage and prevent it from seeing the light of day runs contrary to this robust public interest.

The PBA lawsuit highlights two important questions: first, what are “police personnel files?”; and second, why prevent the public from looking at them? In regard to the first question, different courts around the country have come to different conclusions based on their view of the purpose of open records acts and the facts presented in a particular case. Often, judgments regarding a state FOIA’s text encompass substantive policy determinations because “personnel file” is a label, not a description. In other words, a local police department can slap a “personnel files” label on a cabinet and viola!—its contents are not subject to public scrutiny. Further, who’s to say what records will be “used to evaluate performance toward continued employment or promotion?” Body camera footage of a cop using excessive force could, and probably should, impact continued employment or promotion. At first glance, the PBA’s argument has significant textual support.

But again, this raises the specter of the exception swallowing the rule. There must be an outer limit or bright line identifying what is a personnel record and what is simply a record. Thus the following principled distinction: files containing predecisional opinions regarding an officer’s performance are “personnel files” not subject to public disclosure, but objective reports of an officer’s on-duty actions are subject to release under the FOIA. Under this model, body camera footage would be releasable, but not a supervisor or fellow officer’s opinion as to whether the officer wearing the camera engaged in misconduct.

This model protects the interests that the New York legislature sought to accommodate when it passed § 50-a, and brings us to the second question of why we would have a police personnel files exemption in the first place. In reality, there is a risk that opening up personnel files to the public eye will make police officers less likely to accurately opine whether their fellow officers engaged in misconduct. Apparently even cops don’t want to be known as “narcs.” Objective reports of on-duty conduct, such as body camera videos, do not fall under this justification for non-release. Nobody is giving an opinion as to whether an officer should continue to be employed or be promoted. As such, the opinionated/objective distinction gives weight to the phrase, “used to evaluate performance toward continued employment or promotion” in addition to limiting the personnel files exemption to a point where it reasonably serves the public interest undergirding its enactment. The NYPD’s argument that unredacted video is a personnel file, but redacted video is not, does not find a basis in governing law (not to say that it won’t be successful in the judicially-crafted “upside down” of state FOIA exceptions).

But what about the privacy interests of the individual police officers? In response to this concern, one need only note that government officials generally have no expectation of privacy in the public performance of their official duties. And while the PBA suit claims that individual citizens’ privacy concerns are implicated by the release of footage, redaction of body camera video can effectively eliminate this concern unless the identity of the person being videotaped is already generally known to the public at the time of release, in which case they too lack a cognizable privacy interest. As such, this justification will generally not prevent the release of body camera footage to the public.

The objective/opinionated distinction places a meaningful limit on the “police personnel files” exemption, but does not address the precise issue that led to PBA’s suit—when can a police department (or the Mayor’s office) unilaterally release body camera footage? The ideal solution, releasing each and every video after redacting the personal details of individuals contained therein, is currently too resource-intensive to be feasible. For the foreseeable future, this means that ad-hoc release by police departments is likely inevitable, which is unfortunate because such releases are unlikely to occur unless release is in the department’s interest. Again, this is a problem, but the “solution” of blocking all release absent a court order throws the baby out with the bathwater.

But access to body camera footage is not enough to address the prevalent issue of police misconduct. What advocates of police reform truly need is uniform data collection allowing for apples-to-apples comparisons between departments in different communities. This would allow us to truly understand the scope of the problem and could theoretically identify “bad apples” in addition to providing useful information about where and when a department chooses to allocate its resources. Rejecting the expansion of the “police personnel files” exemption represents a necessary first step in the right direction.

See also:

My forthcoming article, “To Protect, Serve, and Inform: Freedom of Information Act Requests and Police Accountability,” to be published in an upcoming issue of the Texas Tech Administrative Law Journal.

http://www.nydailynews.com/new-york/pba-suing-de-blasio-block-nypd-release-bodycam-footage-article-1.3746361

https://www.politico.com/states/new-york/city-hall/story/2015/08/lawmakers-push-public-access-to-police-camera-footage-024366

Thanks for reading!

Like a wedding guest who shouts, “I object!” at the ceremony, those who appeal the certification of a class action settlement are frequently viewed as officious intermeddlers who would best go away.[1] But federal law protects the rights of those who disagree with the terms of a class action settlement, recognizing the potentially divergent interests of class members—whose individual claims are often dwarfed by the attorney’s fees awarded in a given case—and their counsel.[2] Class members must be informed of their right to opt out of the action and proceed individually,[3] and a district court is obligated to conduct a fairness hearing in which the court reviews the settlement to determine whether it is fair, adequate, and reasonable under the circumstances.[4] This determination is entitled to significant deference under an abuse of discretion standard.[5] Still, there is a broad right of appeal for all members of the class, rather than just named parties.[6]

The broad right of appeal of class action settlements creates disproportionate incentives for individual, unnamed members of the class to file frivolous appeals of a district court’s ruling that a settlement is fair, adequate, and reasonable under the circumstances.[7] A single plaintiff can potentially hold up millions in settlement funds while an appeal is litigated, creating an economically-measurable detriment to the other class members which can be expressed in terms of the lost time value of the settlement funds; this encourages the parties to simply settle with the individual plaintiff in order to avoid a lengthy appeal and protect the interests of the class as a whole.[8] The result of these competing incentives has been referred to as a “scandalous state of affairs,” wherein “collusion and inadequate representation are everyday features of the class action world.”[9]

Thankfully, courts have found a solution to this serious problem—namely, the appellate and supersedeas bonds authorized by Rule 7 and Rule 8 of the Federal Rules of Appellate Procedure.[10] These bonds are intended to guarantee the availability, in advance, of funds for cost-shifting on appeal.[11] Rule 7 authorizes a district court to “require an appellant to file a bond or provide other security in any form and amount necessary to ensure payment of costs on appeal.” Generally, this amount includes the costs incurred to the class as a result of litigating the appeal, including court costs, the cost of notifying the class of the appeal, and anticipated attorney fees if those are awardable under the statute creating the plaintiffs’ cause of action.[12] Under Rule 8, a district court may require an appellant to post a supersedeas bond in order to remunerate the lost value of the settlement to the class in the event that the settlement is upheld.[13]

Once an appellee submits a motion requesting an appeal bond, the district court will determine whether a bond is appropriate and its amount.[14] The first question turns on a four factor test that considers “(1) the appellant’s financial ability to post a bond; (2) the risk that the appellant would not pay appellee’s costs if the appeal is unsuccessful, (3) the merits of the appeal, and (4) whether the appellant has shown any bad faith or vexatious conduct.”[15] Factor one is presumed in the absence of specific evidence of significant financial hardship,[16] and factor two favors imposing a bond whenever an appellant-objector resides outside the jurisdiction of the court.[17] Bad faith or vexatious conduct can be shown by an appellant’s choice of counsel, specifically whether appellant’s attorney is a “professional, or serial [class action] objector.”[18] The third factor—the merits of the appeal—is generally weighted against appellants because the district court’s determination that a settlement is fair, adequate, and reasonable under the circumstances is entitled to significant deference.[19] Finally, courts have cited attorneys’ professional histories—for example, whether the attorney has filed settlement appeals later deemed frivolous or withdrawn an appeal once a bond was imposed—in decisions finding “bad faith or vexatious conduct” and imposing a bond.[20]

Bond amounts in class action appeals have been set at hundreds of thousands,[21] and even millions,[22] of dollars. Generally, the amount will depend on the size of the class, the court’s estimate of how long the appeal will take, whether the jurisdiction allows the Rule 7 component of the bond to include attorney fees, and whether the objector can make a showing of financial hardship such that the appellees’ requested bond amount would constitute an impermissible barrier to appeal.[23] An appellant who refuses to pay the bond will forfeit his or her right to appeal the court’s approval of the class action settlement.[24]

Academic literature on the theoretical validity of class action lawsuits has focused on the relationship and divergent interests between the members of a class and their attorneys.[25] The unnamed plaintiff’s right to appeal is one way to monitor this relationship and prevent abuse; unfortunately, however, this mechanism is itself ripe for abuse. A district court’s decision to post a substantial appeal bond under Rules 7 and 8 of the Federal Rules of Appellate Procedure should be viewed as an opportunity to set the up-front costs of the appeal at a level that will deter frivolous litigants, but not those who present serious challenges to the validity and/or value of a settlement. However, the inherent difficulty of calibrating this exact amount will require district courts to peer through the pleadings and evaluate the merits of the appeal using any tools which are available—including the past conduct of the objector’s counsel and the care with which the court has reviewed the proposed settlement.[26] This is the precise reason why Rule 7 gives the court discretion to set the bond amount.

Class action settlements are an area of the law where traditional relationships (attorney-client, plaintiff-defendant) begin to break down. For example, a motion for an appeal bond may be submitted to a district court under the names of both the named plaintiff and the defendant, and a meddling serial objector may effectively steal from the class in the name of protecting it.[27] Appeal bonds present a solution for reining in frivolous appellate litigation that is contrary to the interests of the class, the defendant, and the court. To extend the metaphor of appellant-as-wedding-objector, the imposition of a substantial appeal bond tells guests that they should hold their peace during the ceremony unless they are willing to pay for the reception.

Citations:

[1] See Lawrence W. Schonbrun, The Class Action Con Game, 20 Regulation 53 (1997) (“Objectors are as welcome in the courtroom as is the guest at a wedding ceremony who responds affirmatively to the minister’s question, ‘Is there anyone here who opposes this marriage?”’); see also Edward Brunet, Class Action Objectors: Extortionist Free Riders or Fairness Guarantors, 2003 U. Chi. Legal F. 403, 472 (2003) (referring to class action objectors as “perhaps the least popular parties in the history of civil procedure”).

[2] John E. Lopatka & D. Brooks Smith, Class Action Professional Objectors: What to Do About Them?, 39 Fla. St. U. L. Rev. 865, 867 (2012) (stating that “[t]his well-known dynamic has prompted Congress and the Supreme Court to design procedural measures intended to protect the class from overreaching by its lawyers”).

[3] See Fed. R. Civ. P. 23(c)(2)(B)(v).

[4] See Fed. R. Civ. P. 23(e)(2).

[5] See Fidel v. Farley, 534 F.3d 508, 513 (6th Cir. 2008) (“We review a district court’s approval of a settlement as fair, adequate, and reasonable for abuse of discretion.”).

[6] See Devlin v. Scardelletti, 536 U.S. 1, 14 (2002) (holding that an objecting nonnamed class member is a “party” to the action and therefore has a right to appeal a final judgment approving a settlement).

[7] See Lopatka & Smith, supra note 2, at 868 (“It is this broad right of appeal that enables a professional objector to find nonnamed class members willing to lend their names to a dubious objection and a meritless appeal in the expectation that class counsel will pay a handsome sum to make the complainants go away.”).

[8] See Susan P. Koniak & George M. Cohen, In Hell There Will Be Lawyers Without Clients or Law, 30 Hofstra L. Rev. 129, 155 (stating that the situation is “ripe for abuse” because “[i]t is in the interest of all the participants in the class action—save the absent members of the class—to settle class actions by collusively transferring money from the class to class counsel”).

[9] Id.

[10] See Fed. R. App. P. 7; Fed. R. App. P. 8.

[11] Fed. R. App. P. 7.

[12] Compare In re Cardizem CD Antitrust Litigation, 391 F.3d 812, 817 (6th Cir. 2004) (ruling that attorney fees may be properly included in the amount of a Rule 7 bond where a statute includes attorney fees as part of the costs that may be taxed on appeal) with Tennille v. Western Union Co., 774 F.3d 1249, 1255 (10th Cir. 2014) (citing In re Am. Presidential Lines, Inc., 779 F.2d 714, 716 (D.C. Cir. 1985)) (limiting the amount of the bond “to only costs listed in [Federal Appellate Rule] 39”).

[13] Fed. R. App. P. 8.

[14] See Gemelas v. Dannon Co., No. 1:08 CV 236, 2010 WL 3703811 at *1 (N.D. Ohio Aug. 31, 2010).

[15] Id. (citing Tri-Star Pictures, Inc. v. Unger, 32 F.Supp.2d 144, 147-50 (S.D.N.Y. 1999)).

[16] In re Cardizem, 391 F.3d at 818 (“It is [the appellant’s] burden to demonstrate that the bond would constitute a barrier to her appeal.”).

[17] In re Polyurethane Foam Antitrust Litigation, 178 F.Supp.3d 635, 641 (N.D. Ohio 2016).

[18] See id. (citing Roberts v. Electrolux Home Prods., Inc., 2014 WL 4568632 at *10 (C.D. Cal. 2014)).

[19] Fidel, 534 F.3d at 513 (stating that the standard is abuse of discretion).

[20] Id. (collecting cases).

[21] See, e.g., Gemelas, 2010 WL 3703811 at *2 (requiring a single objector to post a $275,000 appeal bond which included $250,000 for attorney fees).

[22] See Allapattah Serv., Inc. v. Exxon Corp., 2006 WL 1132371 at *18 (S.D. Fl. April 7, 2006) (setting appeal bond at $13.5 million for any individual plaintiff who appealed its judgment that the proposed settlement was fair, adequate, and reasonable under the circumstances).

[23] Seee.g., In re Polyurethane Foam Antitrust Litigation, 178 F.Supp.3d at 645 (reducing the ultimate amount of the appeal bond from $250,000 to $145,463 to avoid “unduly burdening Objectors’ right to appeal”).

[24] See In Re Cardizem CD Antitrust Litigation, 391 F.3d at 818 (“A litigant cannot ignore an order setting an appeal bond without consequences . . . failure to secure an appeal bond can result in dismissal of the appeal.”).

[25] Seee.g., Brunet, supra note 1, at 405 (“The theoretical attack on class actions rests heavily upon the agency cost problem: class members, including their leaders—the representative parties—simply cannot efficiently monitor their attorneys—class counsel.”); See also John C. Coffee, Jr., Rethinking the Class Action: A Policy Primer on Reform, 62 Ind. L. J. 625, 629 (1987) (“[T]he members of the plaintiff class usually have very little capacity to monitor their agents.”).

[26] Seee.g., In re Polyurethane Foam Antitrust Litigation, 178 F.Supp.3d at 643-45 (conducting the balancing test outlined above).

[27] See Koniak & Cohen, supra note 8, at 155 (“Even when objectors and their lawyers have sufficient incentive and funding to challenge the class settlement, however, they are often motivated not by the chance to protect the class from a sellout settlement but by the prospect of being paid off by class counsel and/or the defendant to drop their objections and walk away.”).

Note: this article first appeared on the forum for the Michigan State Law Review, available here: http://www.michiganstatelawreview.org/forum/. Citations may link to the original article.

Cloud Computing and Third-Party Access to Personal Information

Cloud computing offers users the ability to store information and run programs over the internet rather than relying on their personal computer’s local storage. The cloud service provider operates a set of connected servers which distribute tasks among the servers to increase computing efficiency. Google Drive provides one example of cloud computing; the files and documents stored on the drive “exist” on Google’s servers, not the users’. Moving files from physical storage on a personal computer to distributed storage on a connected “cloud” avoids the inefficiency of leaving unused memory on the individual’s computer and creates networking redundancies that can prevent data loss in the event of technological failure. If your laptop crashes, files stored on Google Drive will not be lost. These features will inevitably encourage a shift toward cloud computing. Even now, many of the services we use on a day-to-day basis have cloud-like features that result in third-party access to data. Understanding how these data are stored and used is crucial to effectively regulating cloud services.

Indeed, cloud computing creates serious privacy concerns. Beyond the threat of hacking, there is the possibility that the use of a cloud will subject the user’s private information to third-party subpoenas, even without notice to the user. This is due to the antiquated framework of the Stored Communications Act, which addresses the circumstances under which a court may compel the disclosure of stored wire and electronic communications and transactional records. The Act is notoriously confusing and subject to conflicting interpretations by reviewing courts, and has even prompted calls for Congressional action.[1] It limits the government’s ability to compel disclosure of communications stored by a third  party and regulated the voluntary disclosure of such information. In doing so, the Act distinguishes between electronic communications services (ECS) and remote computing services (RCS), with the former receiving additional protection.

Categorizing Cloud Services Under the Stored Communications Act

In order to qualify as an “electronic communications service,” a platform must provide the ability to send or receive electronic communications and keep those communications in “electronic storage.” But the Act defines electronic storage narrowly as only encompassing “any temporary, intermediate storage of a wire or electronic communication incidental to the electronic transmission” and the storage of communications for backup purposes. The definition contemplates unopened emails and potentially not much else. In reality, Google Drive and other cloud computing services are not merely passive receptacles for data—they use user data to sell targeted advertisements, generate profiles of users, and even simply to look for ways to improve their services. Their very purpose is to provide permanent, rather than temporary, storage. This takes them out of the category of electronic communication services. By contrast, a “remote computing service provider” is an entity that provides to the public “computer storage or processing services by means of an electronic communications system.” This definition applies to the majority of cloud applications.

If the distinction doesn’t immediately make sense, that’s because it stems from a 1980s conception of internet technology and architecture. At that time, email was understandably seen as analogous to a wire or telegraph function. Users would sign into their email client and download emails to their computers, where they would be locally stored. Then, the email service provider would delete its backup copy after a certain period (thus the statute’s reference to “temporary, intermediate storage”). Back then, prior to the rise of the personal computer, information was mostly stored on remote third-party mainframes. As such, the functions of information transmission and storage were separate. Today, this is largely no longer the case. Gmail, for example, both transmits and stores users’ emails. In other words, the basic distinction upon which the SCA rests has been rendered obsolete by technological innovation.[2] This has created a widening disconnect between the factors courts must consider in determining the privacy rights of users and those that actually inform users’ subjective expectations of privacy.

Updating the Stored Communications Act to Accord with Modern Usage of the Internet

Short of stretching the definition of electronic communications services under the SCA, courts are left with few tools to protect information maintained on the cloud. In Theofel v. Farley-Jones,[3] the Ninth Circuit held that an opened e-mail remains in “electronic storage” and is protected by the Act’s ECS provisions. The decision expanded the scope of protections afforded by the SCA, but has received significant academic and judicial criticism for running contrary to the text of the Act and falsely assuming that the user downloads a copy of their emails. Still, because the Ninth Circuit hosts a number of technology companies offering services under the SCA, the Theofel decision has some functional weight.

But judicial interpretation alone will not fix the problems with the Act.

The SCA itself provides for circumstances where an individual’s information may be given up without his or her knowledge and lacks a suppression remedy in the evidentiary context. This means an aggrieved party must file a separate action to enforce their rights under the Act, which can be difficult or not worth litigating. Scholars have proposed eliminating the distinctions between ECS and RCS and instead focusing on the type of file at issue.[4]

Locally-hosted mainframe servers are presently a far better solution than the cloud in terms of offering Fourth Amendment protection since they do not rely on the transmission of data to a third party. But these are less efficient and generally too costly for individual users. Because the current scheme does not protect the public, Congress should reform the SCA and update its basic framework to accord with the modern uses of technology. Until that happens, cloud services will subject users to the Faustian bargain of giving up privacy in favor of efficiency and additional capacity.

References:

[1] Hien Timothy M. Nguyen, Cloud Cover: Privacy Protections and the Stored Communications Act in the Age of Cloud Computing, 86 Notre Dame L. Rev. 2189, 2191 (referring to the statute as “confusing”); see also Konop v. Hawaiian Airlines, Inc., 302 F.3d 868, 874 (9th Cir. 2002) (“We observe that until Congress brings the laws in line with modern technology, protection of the Internet . . . will remain a confusing and uncertain area of the law.”).

[2] Or considered another way, the analysis reveals the dirty secret of cloud computing, which is that it basically represents a scaled-up version of the central mainframe. Indeed, the largest difference between a mainframe and a “cloud” is where the physical hardware is located. See Caitlin Hughes, Mainframes: Are They More Secure Than the Cloud?, Advanced Software Products Group, Feb. 11, 2015, http://aspg.com/mainframes-secure-cloud/#.Wc2OgWhSyMo (“The biggest difference between cloud and on-premise computing is that on-premise computing allows you to maintain control over the hardware and everything that runs on it.”).

[3] 341 F.3d 978 (9th Cir. 2003), amended by 359 F.3d 1066, 1075-76 (9th Cir. 2004).

[4] See Orin S. Kerr, A User’s Guide to the Stored Communications Act, and A Legislator’s Guide to Amending It, 72 Geo. Wash. L. Rev. 1208, 1235 (2004).

robot adr

The Rise of Artificially Intelligent Arbitration

Can a computer resolve a legal dispute? At first blush, the idea sounds ludicrous—even dystopian. But AI technology currently being developed seeks to predict how a judge will rule given a set of facts. Why not cut out the middleman, and simply have computer programs that adjudicate?

In the context of commercial dispute resolution outside traditional legal fora, the prospects for adjudication by algorithm, or even AI, are bright. Considering the main goals of alternative dispute resolution (ADR)—reducing costs, promoting adjudicatory expertise, and preserving business relationships between the parties—artificially intelligent arbitration will increasingly come into focus as the technology allowing digital decisionmaking develops. The rise of “smart contracts,” which are written in terms of machine-readable “if-then” statements, will also enable artificially intelligent arbitration.

Due Process and Contracts that Speak for Themselves

In my most recent post on legal automation, I wrote that the true value of hiring an attorney lies in making a real person responsible for your problems. In the context of commercial contracts, however, there is a particular drive to take the human element of interpretation out of the dispute. The goal, in other words, is to let the contract speak for itself. Technology makes this possible in a much more literal way than could ever have been envisioned by the framers of traditional, “four corners” contract law.

Arbitration by AI also sidesteps some of the due process concerns that would prevent, for example, replacing federal district court judges with supercomputers. Parties in digital arbitration would consent to the dispute resolution process in advance and a court could review the decisions of the “arbitrator” under the same principles as current arbitration jurisprudence. Arbitration rules, such as the American Arbitration Association’s Large, Complex Commercial Disputes procedures, can be easily adapted to fit AI adjudication. Parties could limit the issues that go before the digital arbitrator. For example, if the parties wanted to argue about discovery in front of a person, rather than a computer, they could so contract.

The Value of a Human Judgement

What is the value of human judgement in adjudication? It’s easy to imagine questions of law or fact that can be best resolved by a computer, such as the calculation of damages or applicability of certain contractual language to a particular interpretive issue. More difficult judgements involve considerations of justice, such as the weight that should be given to parties’ bargaining power in determining consent or the materiality of information to a particular dispute. Short of hooking witnesses up to a heart monitor during deposition, it’s nearly impossible to rely solely on objective information in determining the credibility of testimony. Over time, however, an artificially-intelligent arbitration program could conduct a number of these adjudications and “learn” how to balance considerations of justice and fairness. AI adjudicators are only as cold as their programming.

Over repeated iterations, AI will gain sufficient experience to enable it to conduct all but the most difficult, fact-intensive adjudications. Consequently, the value of a human adjudicator will diminish. Human arbitrators will assist, rather than direct, the process of submitting evidence and conducting interrogatories. Again, all this is consistent with the movement aimed at reducing the human element in contract law. Moreover, the decision to use automation will be driven by the same consideration that primarily drives arbitration in the first place: cost.

Recommendations for Automated Arbitration

Arbitration rules should provide for judicial review in exceptional circumstances or under a deferential standard such as “clear error.” In order to facilitate judicial review and preserve notions of due process, the program should issue a reasoned judgement in natural text; this would form the basis for a record on appeal. An arbitration program should also be able (similar to a jury) to certify questions to the parties or an overseeing judge. There should be a presumption that certain issues, such as contract formation, are to be handled by a human being. Having a computer determine whether you have consented to its authority crosses the line from efficient to dystopian. In the end, automated arbitration will succeed in contexts where the parties seek to eliminate the ever-fallible human element. In some sense, then, this is but the logical endpoint of the alternative dispute resolution movement.

light up brain

The Specter of Artificial Intelligence

In 2014, General Counsel for IBM predicted that the company’s artificial intelligence software, Watson, would be able to pass a multistate bar exam “without a second thought.” While Watson hasn’t sat for the bar yet, the possibility that lawyers might be replaced by computers has long been a nightmare for many attorneys. The release of ROSS, Watson for legal services, makes what was once the subject of fiction seem not so far-fetched.

Indeed, the specter of artificial intelligence asks fundamental questions about the nature and value of the legal profession as a whole. What does a lawyer do, other than collect information about a client’s problem, search for and find the governing law, and apply it to the facts at hand? Is there a human element to persuasion that simply cannot be automated? And perhaps more to the point for firms: can a robot bill hours?

How Automation Replaces (and Doesn’t Replace) Attorneys

AI will start by automating basic legal tasks—serving in an assistant, not a replacement role. Algorithms can reduce a legal task into a simple set of instructions in order to automate it, “replacing” the lawyer for that individual task. It’s easy to think of tasks that involve collecting information as being susceptible to automation: searching legal databases, reducing decisions down to a holding and precedent, finding splits of authority, etc. But the harder tasks that arguably constitute the real substance of lawyering—advocating for clients, persuasively analyzing conflicting authority, and making judgment calls on litigation strategy—involve an element of information processing, not just collection. This is where things get interesting in terms of automation, raising the specter of data-driven AI that could predict how a particular judge would rule in a given case under defined parameters. Why not go further, and have AI systems conduct arbitration? The technology might not be there yet, but it’s not very difficult to imagine.

There is also an element of equity and access at play in the background of debates over legal AI. Again, consider that automation is most easily able to conduct routine tasks that involve collecting, condensing, and reproducing information. This could have very counterintuitive implications for the balance between large and small firms. On the one hand, automating routine tasks that drown smaller firms in paperwork could level the playing field. But it’s far more likely that automation, like so many other technologies, will lead to an arms race. If the value in terms of cost-savings is as robust as some proponents (and detractors) of automation have predicted, larger firms with upfront capital will be the primary first adopters. They can better shoulder the initial costs in order to realize a long-term benefit from the technology. Given the follow-the-leader nature of Biglaw, once one or two firms demonstrate the value of automation to clients by showing that they can do more with fewer billed hours, widespread adoption by larger firms will be inevitable. If smaller firms and solo practitioners are left behind in the automation race, massive resource disparities will result as their cost advantages erode.

Moreover, even within firms AI may affect the balance of power by reducing the need for associate lawyers whose tasks—often the “grunt work” of the law—are increasingly automated. This is less of an issue for young attorneys, though, because the general trend has already turned toward outsourcing these tasks to non-lawyers. There will always be grunt work at law firms, and since many firms see associates as talent to invest in, the distributive impact of automation will not be qualitatively different from, for example, the shift from paper to digital drafting. The upside for young attorneys is that learning about automation and integrating it into their practice can place them on the cutting edge of legal technology.

Automation and the Law’s Human Element

The power of legal automation is clear. What direction it will take, however, is still undetermined. Visions of a technology-wielding lawyer/cyborg are not yet ripe for serious commentary, though they may haunt the dreams of overworked and indebted law students. In truth, automation is but the logical endpoint of the movement towards unbundling and outsourcing legal services.

The true value of a lawyer lies beyond the mechanistic application of law to fact, something easily forgotten in a world where demonstrating value for services delivered—in other words, the bottom line—is increasingly paramount. Moreover, many of the legal developments of the last century, including the vanishing jury and mandatory minimum sentencing, have served to take the “human element” out of the law. But the law is intrinsically human; it involves real people settling disputes with real implications for their lives and livelihoods. At bottom, retaining an attorney means making someone else responsible for your problems. This is something that AI—no matter how advanced—simply cannot do.

Thus, for lawyers, answering the questions posed by legal AI means standing up for and thinking creatively about our role in a technology-enabled society. It means developing one’s practice in a manner that centers around people, rather than billable fee structures. And finally, it means reckoning with the fact that one day computers might become smarter than us even without replacing us. If and when Watson does pass the bar, it’s unlikely that our jobs will be in immediate jeopardy.

Killings by Police and Structural Reform Litigation

1,154 people were killed by police in 2016—an average of more than three killings per day. #BlackLivesMatter has reinvigorated academic and policy discussions of the problem of police misconduct. In response to the movement, President Obama commissioned the Task Force on 21st Century Policing to recommend police practices that build public trust and strengthen community relationships. The U.S. Department of Justice also began investigating police shootings more frequently, through a process of structural reform litigation.

In structural reform litigation, the United States Attorney General initiates action against local police departments engaged in a pattern or practice of unconstitutional misconduct, seeking injunctive or equitable relief. The Department of Justice investigates the police department and prepares a report, which then forms the basis for a consent decree between the DOJ and the police department. Academic research into the efficacy of these agreements has found them relatively successful. However, the process has been criticized for putting enforcement in the hands of the DOJ, as consent decrees do not confer standing to third parties. Despite these limits, structural reform litigation offers an avenue for forcing notorious police departments to engage in reform. One out of five Americans lives in an area governed by a police consent decree, despite the fact that the DOJ has only investigated roughly 3 departments per year. Thus, structural reform litigation is one way for the federal government curtail the worst police abuses in urban communities.

Enter Jeff Sessions

The appointment of Jeff Sessions to Attorney General is a serious obstacle for advocates of structural reform litigation to reduce police misconduct. As a Senator, Sessions referred to consent decrees as an “end run around the democratic process.” Given that the president has indicated his aversion to federal “interference” with law enforcement, the future of structural reform litigation—and police reform in general—seems bleak.

This is a shame, especially given the magnitude of the problem of police misconduct. Even the mere specter of structural reform litigation encourages departments to proactively update their policies regarding the use of force and the conduct of internal affairs investigations. And because consent decrees result from negotiation between the Department of Justice and the local police department, the process balances the legitimate concerns of police with the need for reform. Scholars have argued that structural reform litigation even pays for itself in the long run, as it reduces the number of claims against the department stemming from officer misconduct.

Unfortunately, the current president seems to view any criticism of police as “vilification.” As such, robust federal enforcement and initiation of structural reform litigation will likely be one of the first political casualties of the new administration. Attorney General Sessions has made his position on the matter clear.  Given that the DOJ alone has standing to challenge a department’s implementation of its consent decree, whatever progress has been made by structural reform litigation is now under threat.

Additional Commentary: 

John Felipe Acevedo, Restoring Community Dignity Following Police Misconduct, 59 How. L.J. 621, 633 (2016).

Sunita Patel, Toward Democratic Police Reform: A Vision for “Community Engagement” Provisions in DOJ Consent Decrees, 51 Wake Forest L. Rev. 793 (2016).

Stephen Rushin, Structural Reform Litigation in American Police Departments, 99 Minn L. Rev. 1343 (2015).

Seth W. Stoughton, Principled Policing: Warrior Cops and Guardian Officers, 51 Wake Forest L. Rev. 611  (2016).