Joseph Rosenberg*

It’s 2020 and Boston has become a haven for homicide. In the past five years, the number of homicides in the city has tripled, and the rate of unsolved homicides has climbed from an already-high 53%[1] to nearly 80%. The Boston Police Department (BPD) has seen enough. Believing that an uptick in drug trafficking is responsible for the corresponding uptick in homicides—and left behind by its inability to break into the traffickers’ encrypted devices and communications—the BPD has a potential solution to its unsolved homicide problem: drones.

The BPD wants to use a fleet of up to fifty surveillance drones to fly in public airspace, take twenty-four-hour video footage, and simultaneously follow up to fifty people whom the BPD suspects may be involved in illegal drug sales and responsible for the increased violence. The drones would be equipped with an auto-erase function such that they would retain only the most recent forty-eight hours of surveillance footage. No BPD officer would monitor the auto-piloted drones, which would “lock on” to their targets using facial recognition software and soft biometric factors. The drones would be large enough to be seen from the ground with the naked eye, but small enough that one would need binoculars to be sure the drones were not, say, high-flying birds.

Before assigning a drone to surveil a target, the BPD would have to obtain an initial warrant (an “observation warrant”). A court would issue the observation warrant upon the BPD’s showing of a reasonable and articulable suspicion—a standard lower than probable cause—that the drone could be expected to collect evidence of the target’s criminal wrongdoing. An observation warrant would enable a drone, constrained to a virtual perimeter by geofencing its airspace, to surveil its target only in public places and not in constitutionally-protected areas such as homes, religious centers, and political rallies. The observation warrant would expire after fifteen days.

Although the observation warrant would sanction targeted drone surveillance, the BPD would not have contemporaneous access to the surveillance footage. After becoming aware of a crime, the BPD would have to obtain a second warrant (a “viewing warrant”) to view a particular drone’s surveillance footage. A court would issue this viewing warrant only on a showing of probable cause that a particular drone’s footage contained evidence of a crime. The viewing warrant would entitle the BPD to the drone’s previous forty-eight hours of footage, the rest having been erased. Should an observation warrant expire without the BPD’s seeking a viewing warrant, the BPD would be “locked out” of obtaining a second observation warrant on that particular target for ten days. If the BPD later targeted the same individual and the observation warrant were again fruitless, the BPD would be locked out for twenty days, and so on.

The BPD would be the only party with access to the drone surveillance footage and it would not be shared with anyone except in a case of emergency, such as an imminent threat to national security. The sole exception to this rule is as follows: a target himself, with a court order, could request drone footage in the event of an allegation of police misconduct against him. The forty-eight-hour auto-erase feature might limit this exception’s helpfulness. But an observant target might notice the drone surveilling him, and thus know to seek such an order in case of police misconduct against him. Another conceivable remedial measure would involve the drone surveillance footage being sequestered with a third party after forty-eight hours. This footage would be accessible only by citizens seeking to challenge police misconduct.

The BPD anticipates that its proposed drone surveillance scheme will help reduce Boston’s unsolved homicide rate, but acknowledges its cost to Bostonians’ privacy. Thus, the BPD proposes that it be allowed to use its drone scheme only in years when the previous year’s unsolved homicide rate was above a certain percentage, which the state legislature would prescribe (perhaps the average national unsolved homicide rate of about 33%[2]).

I. Would the BPD’s proposed scheme intrude upon Bostonians’ Fourth Amendment protections?

A Fourth Amendment search is implicated when a person’s subjective expectation of privacy—which society recognizes as reasonable—is violated.[3] Thus, law enforcement may conduct warrantless surveillance on a person only when that person has no reasonable expectation of privacy. There is generally no reasonable expectation of privacy in public places,[4] and surveillance in public areas is typically not a Fourth Amendment search, even when that surveillance occurs from public airspace, at least up to 1,000 feet.[5] Massachusetts law operates under the same standard.[6] As long as the BPD’s drones operate only in public airspace and do not peep inside constitutionally protected areas, their collection would be lawful. Accordingly, where a law enforcement officer could obtain the information by naked eye observation, the BPD’s drones would be allowed to do the same.[7]

Video surveillance by closed circuit television (CCTV) provides a good point of comparison. Today, such surveillance is pervasive and generally does not require a warrant.[8] For example, following in the footsteps of London—and Britain, generally[9]—New York has instituted a sophisticated video surveillance system that constantly monitors lower Manhattan.[10] More cities are sure to follow. The pervasiveness and legality of video surveillance collection suggests, by analogy, that the BPD’s drones would be allowed to take surveillance footage without a warrant, too. Even so, the BPD is committed to obtaining a pre-surveillance observation warrant.

One major difference between the BPD’s proposed drone program and a CCTV program is that a CCTV is stationary and picks up whatever is in its view, while a drone is mobile and follows a particular individual. But tracking an individual with surveillance technology in public is not automatically forbidden, even without a warrant. While United States v. Jones indicates that there may be some limit on how long such tracking may continue,[11] the BPD’s proposed scheme would probably comply with Jones. First, the BPD’s observation warrant is limited to fifteen days, which is just over half of the four-week period that the court in Jones found to be constitutional. Second, the observation warrant’s invasion of privacy is minimal because the BPD would be unable to view the drone’s footage without obtaining a viewing warrant based on probable cause. Third, the BPD would be “locked out” for increasing increments of time from obtaining an observation warrant for any particular target when previous observation warrants on that target have been fruitless. In other words, the BPD would be unable to follow a target endlessly. Rather, the BPD would be required constantly to show that its targeting is justifiable. The onus would be on the BPD to choose fruitful targets to surveil.

In addition, it is worth noting that roving wiretaps—which follow people rather than devices—on domestic targets are legal under Title III; they have been found to satisfy the Fourth Amendment’s particularity requirement and not to resemble an unconstitutional general search.[12] The legality of roving wiretaps, then, also supports the BPD’s drone scheme as outlined above.

The double warrant requirement, operational limitations, and minimization procedures described above would be strong bases on which the BPD can withstand any Fourth Amendment challenges to its drone policy. But the BPD scheme’s probable legality is only one consideration on the scheme’s ledger of pros and cons. In the area of technology and intelligence gathering, “shifting social norms, combined with . . . technological developments, threaten to make laws [governing intelligence gathering] irrelevant” or obsolete.[13] The normative tensions in the BPD’s proposed scheme deserve consideration as well.

II. Is the BPD’s proposed surveillance scheme justified and palatable?

A. How helpful would the BPD’s scheme be, really?

A threshold consideration is whether the BPD’s surveillance scheme would be helpful enough in solving homicides to justify its intrusion into the privacy of Bostonians. Critics of NSA surveillance programs argue that the benefits of the NSA’s powerful tools of collection in reducing terrorism are insufficient to justify their encroachments on the privacy of a large number of people.[14] It is likely that if the BPD’s scheme halved the percentage of unsolved homicides, Bostonians would object less to the scheme’s additional intrusion on their privacy. However, the debate over the scheme’s efficacy should be considered from an ex ante perspective in order to get at the normative tensions without the “taint” of ex post success (or failure).

Critics of the BPD’s surveillance scheme are likely to point out that the BPD already has robust CCTV surveillance.[15] The BPD’s proposed scheme, they might argue, would succeed simply in generating a further deluge of data to review, which can lead to missing important information and, therefore, a security concern of its own.[16] Indeed, in addition to the BPD’s own CCTV surveillance, cooperation with or legal compulsion of private parties with CCTV surveillance at homicide scenes might be just as effective as drone surveillance in identifying a killer.[17] And as a technical matter, CCTV cameras are more stable than drones, which need recharging.

But the BPD’s scheme would be at least as good as an aggregated CCTV scheme—and probably better. Drones can fly for practically unlimited periods of time between charges,[18] and they can take extremely high-resolution footage (which can even be equipped to recognize faces)[19] rather than CCTV’s often grainy footage.[20] Additionally, public knowledge of the BPD’s drone scheme, as well as the drones’ slight visibility from the ground, may actually deter crime.[21] When investigating a crime, the BPD could still pursue CCTV leads, but would also have a potentially fruitful repository of footage that it could examine after showing probable cause. The repository would not suffer from the deluge-of-data problem, because the BPD would have at most fifty drones in circulation. Moreover, once probable cause was obtained, the drones would provide not only linear video footage but also metadata about their targets’ locations, allowing a BPD officer to sift through the drone footage quickly to the area of interest.

A major additional benefit of the BPD’s drone surveillance scheme concerns the BPD’s human resources. The BPD’s drones would be auto-piloted: they would “lock on” to their targets using facial recognition from state-owned photographs such as a driver’s license, and soft biometric factors[22] but would be constrained in space by geofences so that they would not track their targets into a constitutionally-protected area.[23] Thus, BPD officers would be free from surveilling dangerous streets and would stay out of harm’s way.

Another consideration justifying the BPD’s proposal is that drones are now used in increasing numbers for many purposes, including commercial delivery,[24] agriculture,[25] and recreation.[26] And even though most states restrict drone usage, local and state law enforcement agencies are undoubtedly allowed to use drones to some extent.[27] To what extent is a gray, developing area of the law. In Massachusetts, for example, the BPD recently purchased three drones to augment security at public demonstrations, help process and clear crime scenes, and aid otherwise-lawful surveillance.[28] The purchase was made without public discussion and the ACLU argued vigorously against it.[29] In Connecticut, the state legislature recently considered (and rejected) a bill that would have allowed law enforcement to use drones outfitted with deadly weapons.[30]

While true that there are “more restrictions on technologies . . . utilized primarily by law enforcement than those that are in wide use,”[31] the BPD might argue that—considering the wide civilian and law enforcement use of drones—not allowing the BPD to use drones for surveillance would unfairly hamstring the BPD. As drones become ubiquitous, people’s expectations about potential drone surveillance will shift, and their Fourth Amendment protections from drones might, too: the “contours of what is reasonable under the Fourth Amendment may adjust as people’s expectations of privacy evolve.”[32]

B. Does restricted access to device content justify an increase in physical surveillance?

Many prominent law enforcement officers believe that device encryption poses a serious problem for law enforcement’s ability to prevent and solve crime because such encryption is responsible for criminals’ “Going Dark.”[33] Former FBI director James Comey describes the Going Dark problem this way: “We have the legal authority to intercept and access communications and information pursuant to court order, but we often lack the technical ability to do so.”[34] And in the wake of last March’s Westminster attacks, U.K. Home Secretary Amber Rudd decried: “We need to make sure that organizations like WhatsApp . . . don’t provide a secret place for terrorists to communicate with each other.”[35] The Paris attackers in 2014 also used WhatsApp to conceal their plans.[36] Local law enforcement is affected, too; the New York County District Attorney recently complained that he had 175 phones—with potentially pertinent information to homicide and other investigations—that he couldn’t unlock.[37] Similarly, in this hypothetical criminals in Boston have been Going Dark, and the BPD believes that encrypted phones and messaging have hindered its ability to solve homicides. Regardless of the concern’s validity, the Going Dark corollary—that encrypted communications apps are to blame for terrorist attacks—heaps difficult political pressure on device manufacturers and communications software companies.

A robust opposition believes that Going Dark is not accurately termed, and does not present a new and unique problem to law enforcement.[38] In a recent article, Orin Kerr and Bruce Schneier suggest encryption workarounds that law enforcement can employ to defeat device encryption.[39] Indeed, even if Going Dark were as big of a problem as its proponents claim, law enforcement already has more surveillance power than ever before.[40] To name two examples, today over seventy state and local police departments use Stingrays—possibly unlawfully without a warrant[41]—to gather information on everyone carrying a cellphone within a certain area,[42] and CCTV video surveillance is ubiquitous.[43] Finally, history urges caution in surrendering surveillance power to law enforcement agencies; such power has often been abused.[44] Thus, the rise of encryption can be seen in some ways as a return to the status quo.

Crime is up in Boston, however, with more than half of homicides going unsolved. Encryption has frustrated the BPD’s attempts to discharge some of its basic duties—preventing homicide and, when it happens, bringing the perpetrators to justice. The BPD will thus argue that Massachusetts should allow a corresponding increase in physical surveillance to help pull back the curtain of encryption. On the other hand, the BPD is also interested in protecting and preserving Bostonians’ civil liberties and does not wish to abuse its proposed drone surveillance program (or even for the program to have the appearance of being abuse-able). It has offered numerous reasonable compromises and minimizations to appease civil liberties groups and to ensure proper use of its drone scheme.

C. Minimization Procedures

The BPD believes that its drone surveillance scheme will help reduce the unsolved-homicide rate in Boston, and does not wish to use the drone footage for any other purpose. The BPD understands, however, the inflammatory nature of drones buzzing about Boston and the appearance of the potential for corruption. In response to these issues, the BPD has proposed minimization procedures to assuage citizens concerned about privacy. As described supra, the BPD would commit to: (1) capping the number of drones at fifty; (2) using the drone scheme only when the unsolved homicide rate rises above a certain percentage; (3) requiring observation warrants—on a standard of reasonable and articulable suspicion—which would be limited to fifteen days with step-wise “locking out”; and (5) equipping the drones to follow only their approved target and with geofencing technology.

These ex ante restrictions would force the BPD to prioritize the most likely criminals and not to tail fruitless targets forever—conduct that would essentially amount to general, blanket surveillance on thousands of targets.[45] Recognizing that actually looking through the surveillance footage is more intrusive on privacy than the initial collection, the BPD also commits to ex post minimizations: (1) outfitting drones with an auto-erase function to retain only the most recent forty-eight hours of surveillance;[46] (2) requiring viewing warrants based on probable cause; (3) never selling or sharing surveillance video unless in case of emergency, such as a threat to national security; and (4) making available to a plaintiff alleging police misconduct, upon successful application to a judge, video from a surveillance drone pertaining to that alleged misconduct. These restrictions and minimization procedures stand in stark contrast, for example, to the ability of multiple federal agencies to query NSA’s 702 database.[47]

Combined, the BPD’s ex ante restrictions and ex post minimizations are comparable to those approved in a recent Northern Illinois case.[48] There, the court imposed three requirements on the requesting law enforcement agency’s use of a Stingray: first, law enforcement would have to ensnare as few incidental people as possible; second, all extraneous data had to be destroyed within forty-eight hours; and third, no information about third parties was to be used in any way.[49]

The BPD’s mitigations and minimizations would ensure accountability and proper oversight. A common critique of the U.S. intelligence community is that, even though all three branches are nominally involved, those charged with overseeing the intelligence community’s activities are not informed enough to do so effectively in practice.[50] The BPD’s scheme would not suffer the same problem; rather than relying on murky statistics[51] and courts that proceed in secret, the BPD would obtain warrants in public courts for drones flying only in public airspace (although targets of warrants would remain sealed until the warrants expired).[52] The two-step warrant process would properly involve Massachusetts courts, and, because the public would be aware of the BPD’s scheme (and because state legislatures actively regulate law enforcement’s drone use),[53] the Massachusetts state legislature, too, would serve as an efficient check, resulting in a scheme that both privacy and security advocates would respect.

The BPD’s misconduct concession, in particular, is a nod to the civil liberties community and a strong, built-in check on the BPD’s scheme. A particularly close analog is law enforcement body cameras, which are gaining popularity among law enforcement agencies.[54] Civil liberties groups generally support body cameras for their function of holding police officers accountable.[55] But body cameras can pose problems to privacy, too, such as incidentally taping unwitting non-targets, recording particularly sensitive situations, and retaining data for long periods of time.[56] The BPD’s scheme provides the benefits of body cameras and addresses many of the downsides. For example, the BPD already wants to adopt minimizations limiting the retention of video surveillance to forty-eight hours and restricting its drones to public spaces, erasing the threat of catching particularly intimate and sensitive moments on tape.

D. Finally, is there a First Amendment concern with the BPD’s program?

Government surveillance can chill First Amendment rights, both explicitly—by actually attempting to influence politics[57]— and less explicitly, such as by monitoring political rallies or tracking people’s movements.[58] This chilling effect has long been a worry: the Church Committee noted it in 1976.[59] More recently, in Jones, Justice Sotomayor described the effect:

GPS monitoring generates a precise, comprehensive record of a person’s public movements that reflects a wealth of detail about her familial, political, professional, religious, and sexual associations . . . . Awareness that the government may be watching chills associational and expressive freedoms.[60]

Critics of the BPD’s scheme might levy a similar worry: drones zipping around the sky, even if monitoring only public places and not especially visible to the naked eye from ground level, are creepy, and merely knowing they are there, somewhere, will affect people’s behavior. These effects can be measurable, such as by tracking the Google search trends of questionable words that ensued after the Snowden revelations[61] and surveying writers—who must feel free to share unpopular ideas for the proper functioning of our democracy—about their self-censorship.[62]

While it is possible that the BPD’s scheme would contribute to a chilling effect, so, too, would a uniformed police officer standing on a corner. Indeed, the First Amendment does not impose substantive or procedural limitation on good faith criminal investigations beyond those guaranteed by the Fourth and Fifth Amendments.[63] Often, too, claims of a chilling effect that result from a fear of surveillance and that require a court to “review actions of the political branches in the field[] of intelligence gathering” are not actionable.[64] Thus, as a purely legal matter, the BPD’s drone scheme would likely not be subject to a First Amendment challenge.

Normative considerations, though, require more attention: the important question is whether the marginal benefit of the BPD’s scheme would outweigh its marginal cost in potential chilling effect. In truth, the BPD’s scheme is comparatively modest: some other police departments already use drones for more invasive and less targeted uses than the BPD proposes.[65] Furthermore, in incorporating geofence technology,[66] the BPD is amenable to preventing its drones from traveling near sites particularly sensitive to a chilling effect, such as places of worship and political rallies. The state legislature or courts would articulate what those places might be.

In addition, people might worry that drones zipping overhead would catch them doing legal things they simply do not want known, such as walking into a sex shop or visiting a particular type of doctor. Even people who say they have nothing to hide and so support robust surveillance might balk at these activities being caught on camera.[67] The BPD’s scheme, though, would not present a major problem of catching non-targets doing anything, because the BPD’s minimization of 48-hour mandatory erasure and its highly-tailored targeting ensure that incidental collection would almost never be viewed. In fact, the BPD’s scheme is less invasive than a stationary CCTV’s video surveillance, which can be retained for long periods of time.

III. Conclusion 

The BPD’s scheme is well-intentioned, contains numerous ex ante restrictions and ex post minimizations, and will likely help to solve serious crime. It thus balances privacy and security. Although in some ways intrusive on privacy and possibly chilling on Bostonians’ behavior, a balance must be struck, and the BPD’s scheme strikes it reasonably. The adoption of technologies like drones and plans like the BPD’s should be the subject of an open and robust conversation among law enforcement, civil liberties groups, and the legislative branch. That conversation—along with ongoing, neutral, and open judicial oversight—should ensure that law enforcement, privacy advocates, and ordinary citizens can be both safe and free.


*Joseph Rosenberg is a second-year student at Harvard Law School.

[1] In 2015 and 2016 combined, the BPD did not solve 46 of 87 homicides. See Unsolved Homicides 2016, BPD News, http://bpdnews.com/2016-unsolved-homicides (last visited Dec. 22, 2017); Unsolved Homicides 2015, BPD News, http://bpdnews.com/solved-homicides http://bpdnews.com/2015-unsolved-homicides (last visited, Dec. 17, 2017); Solved Homicides 2016, BPD News, http://bpdnews.com/2016-solved-homicides (last visited Dec. 28, 2017); Solved Homicides 2015, BPD News, http://bpdnews.com/2015-solved-homicides (last visited Dec. 28, 2017).

[2] Getting Away with Murder, Economist (July 4, 2015), http://www.economist.com/news/united-states/21656725-police-fail-make-arrest-more-third-nations-killings-getting-away.

[3] Katz v. United States, 389 U.S. 347, 361 (1967); see also Mancusi v. DeForte, 392 U.S. 364, 368 (1968) (applying the “reasonable expectation” test).

[4] See United States v. Knotts, 460 U.S. 276, 281 (1983) (“A person travelling in an automobile on public thoroughfares has no reasonable expectation of privacy in his movements from one place to another.”); Rachel Levinson-Waldman, Hiding in Plain Sight: A Fourth Amendment Framework for Analyzing Government Surveillance in Public, 66 Emory L.J. 527, 529 (2017).

[5] See California v. Ciraolo, 476 U.S. 207, 213 (1986); see also Florida v. Riley, 488 U.S. 445, 450 (1989); Richard M. Thompson II, Cong. Res. Serv., R42701, Drones in Domestic Surveillance Operations: Fourth Amendment Implications and Legislative Responses 7 (2013); Levinson-Waldman, supra note 4, at 541.

[6] See Commonwealth v. One 1985 Ford Thunderbird Auto., 416 Mass. 603, 604 (1993) (refusing to exclude evidence that police obtained—through naked-eye observation and the use of binoculars and cameras—during three helicopter fly-overs at altitudes of 1,500, 800, and 700 feet).

[7] Even though sense-enhancing technology that is not in general public use cannot be used without a warrant to search a constitutionally-protected area, see Kyllo v. United States, 533 U.S. 27, 34 (2001), the BPD’s drones would be monitoring only public places with an observation warrant and so would not be searching constitutionally-protected areas. It is possible, then, that the BPD’s drone footage could be zoomed and enhanced beyond the power of a human eye and still be legal. Further, because drones are arguably in general public use, see infra notes 24–26, the BPD might even have an argument under Kyllo for using the drones to search constitutionally protected areas (even though the BPD would not seek to do so).

[8] See, e.g., Katz, 389 U.S. at 351 (“What a person knowingly exposes to the public . . . is not a subject of Fourth Amendment protection.”); see also United States v. Jones, 565 U.S. 400, 428 (2012) (Alito, J., concurring) (noting that the ubiquity of CCTV monitoring itself influences peoples’ reasonable expectation of privacy). While there is no dispute regarding warrantless video surveillance in particularly public places like squares and thoroughfares, some states do limit long-term video surveillance in more sensitive—but still public—areas. See, e.g., State v. Jones, 903 N.W.2d 101, 113 (S.D. 2017) (holding that police were required to obtain a warrant to install a pole camera that monitored the entrance to the defendant’s home for two months). In any event, the BPD is in the clear because it proposes obtaining a warrant for its drone scheme.

[9] See David Barrett, One Surveillance Camera for Every 11 People in Britain, says CCTV survey, Telegraph (July 10, 2013), http://www.telegraph.co.uk/technology/10172298/One-surveillance-camera-for-every-11-people-in-Britain-says-CCTV-survey.html.

[10] Chris Francescani, NYPD Expands Surveillance Net to Fight Crime as well as Terrorism, Reuters (June 21, 2013), http://www.reuters.com/article/usa-ny-surveillance-idUSL2N0EV0D220130621.

[11] See Jones, 565 U.S. at 412 (acknowledging that temporal limits on warrant-less monitoring may exist, but declining to determine what those limits would be).

[12] United States v. Petti, 973 F.2d 1441 (9th Cir. 1992), cert. denied, 507 U.S. 1035 (1993).

[13] Omer Tene & Jules Polonetsky, A Theory of Creepy: Technology, Privacy and Shifting Social Norms, 16 Yale J. L. & Tech. 59, 74 (2014).

[14] See, e.g., Michael Isikoff, NSA Program Stopped No Terror Attacks, says White House Panel Member, NBC News (Dec. 20, 2013), http://www.nbcnews.com/news/other/nsa-program-stopped-no-terror-attacks-says-white-house-panel-f2D11783588 (explaining the non-existent benefits of “Section 215” collection); see also Oversight and Reauthorization of the FISA Amendments Act: The Balance Between National Security, Privacy and Civil Liberties: Hearing Before the Subcomm. On the Judiciary, 114th Cong. 17 (2016) (statement of Ms. Elizabeth Goitein, Co-Director, Brennan Center for Justice at New York University School of Law) [hereinafter “Goitein Testimony”] (pointing out that efficacy of “Section 702” collection might be overblown).

[15] Heather Kelly, After Boston: The Pros and Cons of Surveillance Cameras, CNN (Apr. 26, 2013), http://www.cnn.com/2013/04/26/tech/innovation/security-cameras-boston-bombings/.

[16] See Goitein Testimony, supra note 14, at 10 (“[T]here is a risk to national security in acquiring too much data . . . [because] [w]hen [humans] . . . are presented with an excess of data, real threats can get lost in the noise,” such as the Underwear Bomber in 2009); see also Bruce Schneier, Why Data Mining Won’t Stop Terror, Schneier On Security (Mar. 9, 2005), https://www.schneier.com/essays/archives/2005/03/why_data_mining_wont.html (noting that even an unrealistically optimistic data mining system would “generate 1 billion false alarms for every real terrorist plot it uncovers”).

[17] For example, in the wake of the Boston Marathon Bombing in 2013, one of law enforcement’s first moves was to acquire surveillance footage from the approximately 200 businesses near the bombs’ detonations. See Patrick J. Kiger, How They Identified the Bombers: A Timeline from Event to Capture, Nat’l Geographic (Apr. 1, 2014), http://channel.nationalgeographic.com/inside-the-hunt-for-the-boston-bombers/articles/how-they-identified-the-bombers/.

[18] See, e.g., Will Knight, This Surveillance Drone Never Needs to Land, MIT Tech. Rev. (Nov. 5, 2015), https://www.technologyreview.com/s/543196/this-surveillance-drone-never-needs-to-land/; Evan Rodgers, Laser Charging Gives Lockheed Martin’s Stalker Drone ‘Practically Unlimited’ Flight Time, Verge (July 12, 2012), http://www.theverge.com/2012/7/12/3154543/lockheed-martin-stalker-drone-48-hour-laser-power.

[19] Levinson-Waldman, supra note 4, at 540.

[20] Id. at 542–43.

[21] Cf. Edward K. Cheng, Structural Laws and the Puzzle of Regulating Behavior, 100 Nw. U. L. Rev. 655, 681 (2006) (discussing deterrent effect of “speed traps”). This observation gives rise to a potential drone usage which is largely beyond the scope of this article: the introduction of dummy drones that would not be equipped with video surveillance technology but which would be visible to the target. Perhaps such a scheme would deter or prevent a target from engaging in crime.

[22] See Levinson-Waldman, supra note 4, at 542–43; see also Sandee LaMotte, The Other ‘Fingerprints’ You Don’t Know About, CNN (Dec. 4, 2015), http://www.cnn.com/2015/12/04/health/unique-body-parts/; Dan Moren, 7 Surprising Biometric Identification Methods, Popular Sci. (Dec. 30, 2014), http://www.popsci.com/seven-surprising-biometric-identification-methods.

[23] See Levinson-Waldman, supra note 4, at 607–08; see also Eddie Schmid, Geofences and Responsible Drone Flight, Autel Robotics (Dec. 17, 2016), https://www.autelrobotics.com/blog/geofences-and-responsible-drone-flight/.

[24] Amazon’s Prime Air drone made its first demo delivery on March 24, 2017. See James Vincent, Watch Amazon’s Prime Air Drone Make its First Demo Delivery in the US, Verge (Mar. 24, 2017), http://www.theverge.com/2017/3/24/15047424/amazon-prime-air-drone-delivery-public-us-test-mars.

[25] See, e.g., Michal Mazur, Six Ways Drones are Revolutionizing Agriculture, MIT Tech. Rev. (July 20, 2016), https://www.technologyreview.com/s/601935/six-ways-drones-are-revolutionizing-agriculture/.

[26] See, e.g., Troy A. Rule, Drone Zoning, 95 N.C. L. Rev. 133, 137–38 (2016) (“Global sales of small drones increased by 167% from 2013 to 2015, and sales are only expected to climb over the next several years.”).

[27] See Local and State Drone Laws, Ctr. for the Study of the Drone at Bard Coll., http://dronecenter.bard.edu/state-and-local-drone-laws/ (last visited December 19, 2017) (counting 31 states by March 2017 that had enacted drone rules, some of which restricted permissible law enforcement usage).

[28] Molly Boigon, Bill Evans Defends the Police Department’s Purchase of Drones, WGBH News (October 17, 2017), https://news.wgbh.org/2017/10/17/local-news/bill-evans-defends-police-departments-purchase-drones.

[29] Id.

[30] Barbara Goldberg, Connecticut May Become First U.S. State to Allow Deadly Police Drones, Reuters (Mar. 31, 2017), http://www.reuters.com/article/us-connecticut-drones-idUSKBN1722SO.

[31] See Levinson-Waldman, supra note 4, at 551.

[32] Thompson II, supra note 5, at “Summary.”

[33] James B. Comey, Dir., Fed. Bureau of Investigation, Address at the Brookings Institution: Going Dark: Are Technology, Privacy, and Public Safety on a Collision Course? (Oct. 16, 2014), https://www.fbi.gov/news/speeches/going-dark-are-technology-privacy-and-public-safety-on-a-collision-course.

[34] Id.

[35] Gregory Katz, UK: Attacker used WhatsApp, firm must help police get access, Associated Press (Mar. 26, 2017), https://apnews.com/52aa182132724f5b8656af4c052673d3/uk-police-still-think-westminster-attacker-acted-alone.

[36] Sebastian Rotella, ISIS via WhatsApp: ‘Blow Yourself Up, O Lion’, ProPublica (July 11, 2016), https://www.propublica.org/article/isis-via-whatsapp-blow-yourself-up-o-lion.

[37] See Alyssa Newcomb, New York DA Says He Can’t Access 175 iPhones From Criminal Cases Due to Encryption, ABC News (Feb. 18, 2016), http://abcnews.go.com/Technology/york-da-access-175-iphones-criminal-cases-due/story?id=37029693; see also The Encryption Tightrope: Balancing Americans’ Security and Privacy: Hearing Before the H. Comm. on the Judiciary, 114th Cong. 133–45 (2016) (statement of Cyrus R. Vance, Jr., Dist. Att’y, N.Y. Cty.).

[38] Matt Olsen et al., Berkman Klein Ctr. for Internet & Soc’y Don’t Panic. Making Progress on the “Going Dark” Debate, 2 (2016), https://cyber.harvard.edu/pubrelease/dont-panic/Dont_Panic_Making_Progress_on_Going_Dark_Debate.pdf (“[C]ommunication channels resistant to surveillance will always exist.”).

[39] See generally Orin S. Kerr & Bruce Schneier, Encryption Workarounds, Geo. L. J. (forthcoming), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2938033.

[40] Peter Swire, The Golden Age of Surveillance, Slate (July 15, 2015), http://www.slate.com/articles/technology/future_tense/2015/07/encryption_back_doors_aren_t_necessary_we_re_already_in_a_golden_age_of.html.

[41] United States v. Lambis, 197 F. Supp. 3d 606, 611 (S.D.N.Y. 2016) (“[E]ven though the DEA believed that the use of the cell-site simulator would reveal the location of a phone associated with criminal activity, the Fourth Amendment requires the Government to obtain a warrant from a neutral magistrate to conduct that search.”).

[42] Stingray Tracking Devices: Who’s Got Them?, ACLU, https://www.aclu.org/map/stingray-tracking-devices-whos-got-them (last visited Nov. 8, 2017).

[43] See supra notes 9 and 15.

[44] See Beverly Gage, What an Uncensored Letter to M.L.K. Reveals, N.Y. Times Mag. (Nov. 11, 2014), https://www.nytimes.com/2014/11/16/magazine/what-an-uncensored-letter-to-mlk-reveals.html (uncovering fraudulent letter from FBI to Dr. Martin Luther King, Jr. in 1964 urging him to commit suicide and threatening him if he did not); see generally American Big Brother: A Century of Political Surveillance and Repression, Cato Inst., https://www.cato.org/american-big-brother (last visited Nov. 8, 2017) (documenting a long history of surveilling domestic political activities).

[45] This aspect of the BPD’s scheme responds to Justice Alito’s worry in Jones. See 565 U.S. at 430 (“We need not identify with precision the point at which the tracking of this vehicle became a search, for the line was surely crossed before the 4-week mark.”).

[46] This forty-eight-hour minimization is eminently reasonable and stands in stark contrast to periods of five years or longer for some types of NSA collection. See Rachel Levinson-Waldman, What the Government Does with Americans’ Data, Brennan Ctr. for Just. 4 (2013).

[47] See, e.g., Goitein Testimony, supra note 14, at 12.

[48] See In re Application of the United States of America for an Order Relating to Telephones Used by Suppressed, No. 15 M 0021, 2015 WL 6871289, at *3 (N.D. Ill. Nov. 9, 2015).

[49] See id.

[50] L. Britt Snider’s account of the Church Committee’s investigation into Operation Shamrock provides an example of the problems for legislative oversight. See L. Britt Snider, Recollections from the Church Committee’s Investigation of NSA, CIA, https://www.cia.gov/library/center-for-the-study-of-intelligence/csi-publications/csi-studies/studies/winter99-00/art4.html (last visited Nov. 8, 2017). There have also been more recent calls for Section 702 oversight. See Arthur Rizer & Daniel Semelsberger, Is Reform on the Horizon for Section 702 Surveillance?, The Hill (Apr. 12, 2017), http://thehill.com/blogs/pundits-blog/technology/328351-is-reform-on-the-horizon-for-section-702-surveillance. The secrecy of the FISC adds another layer of opacity that is particularly eyebrow-raising when one considers that the FISC might act mostly as a rubber stamp. See Dustin Volz, U.S. Spy Court Rejected Zero Surveillance Orders in 2015: Memo, Reuters (Apr. 29, 2016), http://www.reuters.com/article/us-usa-cybersecurity-surveillance-idUSKCN0XR009 (noting that the FISC rejected zero surveillance orders in 2015).

[51] See, e.g., Goitein Testimony, supra note 14, at 17–18 (“Perhaps most strikingly, despite multiple requests . . . the NSA has yet to disclose an estimate of how many Americans’ communications are collected under Section 702.”).

[52] See Nola K. Breglio, Leaving FISA Behind: The Need to Return to Warrantless Foreign Intelligence Surveillance, 113 Yale L.J. 179, 181 (2003) (arguing that the “supersecret domain of the FISC” shields the Attorney General from liability, denies targets the opportunity to challenge investigations, and limits constitutional search considerations).

[53] The interplay of Federal Aviation Administration and state drone regulations is mixed and still in flux. See, e.g., Rule, supra note 26, at 135–37. But as of July 25, 2017, forty states have enacted laws regulating drone usage and three more states have enacted resolutions regulating the same. See Current Unmanned Aircraft State Law Landscape, Nat’l Conf. of St. Legislatures (July 25, 2017), http://www.ncsl.org/research/transportation/current-unmanned-aircraft-state-law-landscape.aspx. Although states regulate many aspects of drones, law enforcement is often required to obtain a probable cause warrant, subject to various exceptions, to use a drone. See Michael J. Smith, Regulating Law Enforcement’s Use of Drones: The Need for State Legislation, 52 Harv. J. on Legis. 423, 427–32 (2015).

[54] Law enforcement officers in some forty-one of sixty-nine major United States cities use body cameras in some form. A major manufacturer recently pledged to give all United States police departments body cameras and software for free for a year. See Rachel Lerman, Axon Offers 1-year Free Trial of Body Cameras to All U.S. Police Departments; Rival Vievu Rips Plan, Seattle Times (Apr. 5, 2017), http://www.seattletimes.com/business/technology/axon-offers-to-give-police-free-body-cameras-to-speed-companys-growth/.

[55] See Jay Stanley, Police Body-Mounted Cameras: With Right Policies in Place, a Win for All, ACLU (Mar. 2015), https://www.aclu.org/other/police-body-mounted-cameras-right-policies-place-win-all (“Although we at the ACLU take a dim view of the proliferation of surveillance cameras in American life, police on-body cameras are different because of their potential to serve as a check against the abuse of power by police officers.”).

[56] See Richard Lin, Police Body Worn Cameras and Privacy: Retaining Benefits While Reducing Public Concerns, 14 Duke L. & Tech. Rev. 346, 353–57 (Sept. 12, 2016); see also Police Body Camera Policies: Retention and Release, Brennan ctr. for Just. (Aug. 3, 2016), https://www.brennancenter.org/analysis/police-body-camera-policies-retention-and-release.

[57] See, e.g., Gage, supra note 44 (note from FBI to MLK).

[58] For a recent example relating to NYPD’s surveilling Muslim communities, see Hassan v. City of New York, 804 F.3d 277 (3d Cir. 2015); see also Levinson-Waldman, supra note 4, at 553–54 (documenting law enforcement encroachments on First Amendment rights in California, Colorado, Illinois, Michigan, New York, and Virginia).

[59] Select Comm. to Study Gov’t Operations with Respect to Intelligence Activities, Final Report, S. Rep. No. 94-755, bk. III, at 778 (1976) (noting that although certain information gathered by the NSA was “useful in many ways to conducting successfully NSA’s legitimate communications intelligence functions . . . [t]he massive centralization of this information creates a temptation to use it for improper purposes, threatens to ‘chill’ the exercise of First Amendment rights, and is inimical to the privacy of citizens”).

[60] Jones, 565 U.S. at 415–16.

[61] See Goitein Testimony, supra note 14, at 8 (“After Edward Snowden’s revelations in June 2013, an analysis of Google Trends data showed a significant five percent drop in U.S.-based searches for government-sensitive terms . . . .”).

[62] Id. (“In 2013, PEN America surveyed 528 American writers to learn how the [Snowden] disclosures affected their behavior. Twenty-eight percent reported curtailing social media activities; 24 percent avoided certain topics by phone or email; 16 percent chose not to write or speak on a certain topic; and 16 percent avoided Internet searches or website visits on controversial or suspicious topics.”).

[63] Reporters Comm. for Freedom of the Press v. Am. Tel. & Tel. Co., 593 F.2d 1030, 1054 (D.C. Cir. 1978).

[64] Clapper v. Amnesty Int’l USA, 568 U.S. 398, 409 (2013) (citing Laird v. Tatum, 408 U.S. 1, 11–16 (1972)).

[65] See Abby Sewell, Richard Winton & Melissa Leu, Lancaster Takes to the Skies to Get a View on Crime, L.A. Times (Aug. 25, 2012) (describing L.A. County Sheriff’s Dept. using drone to monitor indiscriminately Lancaster residents), http://articles.latimes.com/2012/aug/25/local/la-me-0825-lancaster-aircraft-20120825.

[66] Lily Hay Newman, Here’s How To Set Up a No-Fly Zone Over Your House, Slate (Feb. 10, 2015), http://www.slate.com/blogs/future_tense/2015/02/10/noflyzone_org_lets_you_geofence_the_area_over_your_house_for_drones_to_avoid.html.

[67] See, e.g., Daniel J. Solove, “I’ve Got Nothing to Hide” and Other Misunderstandings of Privacy, 44 San Diego L. Rev. 745, 751 (2007) (asserting that even those who claim no interest in their own privacy “cannot withstand even a few minutes’ questioning about intimate aspects of their lives without capitulating to the intrusiveness of certain subject matters”) (internal quotation marks omitted).

Image by Andrew Turner.

Joseph Rosenberg

Joseph Rosenberg is a second-year student at Harvard Law School.