Harvard Law Review Harvard Law Review Harvard Law Review

Criminal Procedure

United States v. Touset

Eleventh Circuit Holds that Border Searches of Property Require No Suspicion.

Supreme Court precedent affords more Fourth Amendment protection to digital information than to many other kinds of property.1×1. See, e.g., Carpenter v. United States, 138 S. Ct. 2206, 2217 (2018); Riley v. California, 134 S. Ct. 2473, 2494–95 (2014). Searches that occur at the border, on the other hand, are subject to fewer constraints. Border searches never require a warrant or probable cause, and only sometimes require “reasonable” — really, individualized — suspicion.2×2. See United States v. Montoya de Hernandez, 473 U.S. 531, 537, 541–42 (1985). Within the last six years, two federal courts of appeals have held that border searches of digital information require individualized suspicion.3×3. See United States v. Kolsuz, 890 F.3d 133, 137 (4th Cir. 2018); United States v. Cotterman, 709 F.3d 952, 957 (9th Cir. 2013) (en banc). Recently, in United States v. Touset,4×4. 890 F.3d 1227 (11th Cir. 2018). the Eleventh Circuit squarely rejected these decisions by its sister circuits, holding a forensic search of electronic devices at the border constitutional in the absence of a warrant, probable cause, or individualized suspicion.5×5. Id. at 1229. In doing so, the court remained faithful to the original understanding of the Fourth Amendment, properly emphasized the distinctiveness of the border search context in conducting its balancing analysis, and made a persuasive case for judicial restraint in light of congressional silence.

Several years ago, Xoom, a company that facilitates electronic transfers of money, observed on user accounts “frequent low money transfers to . . . source countries for sex tourism and child pornography.”6×6. Id. at 1230. Xoom alerted the National Center for Missing and Exploited Children and Yahoo, the latter because some of the suspected individuals used Yahoo messaging accounts.7×7. Id. Yahoo investigated, discovered a file containing child pornography in a user email account, and found that the account listed a Philippine phone number.8×8. Id. Yahoo contacted the National Center.9×9. Id. The National Center referred the matter to the Cyber Crime Center of the Department of Homeland Security, which determined after investigation that Karl Touset had on three occasions between March and July of 2013 used a Western Union account to send money to an account associated with the Philippine phone number.10×10. Id. The Department of Homeland Security placed a “look-out” on Touset so that his belongings would be searched when he arrived in the United States.11×11. Id. On December 21, 2014, when Touset disembarked from an international flight, the Customs and Border Protection Agency (CBP) searched his luggage, which contained “two iPhones, a camera, two laptops, two external hard drives, and two tablets.”12×12. Id. CBP found nothing incriminating during a manual search of Touset’s phones and camera but seized the other devices for further inspection.13×13. Id. Subsequent forensic searches of the two laptops and external hard drives revealed that they contained child pornography.14×14. Id. In January of 2018, CBP clarified its digital search policies. “Basic” searches may be suspicionless, but “advanced” searches may be conducted when there is a reasonable suspicion. U.S. Customs & Border Prot., Directive No. 3340-049A, Border Search of Electronic Devices 4–5 (2018), https://www.cbp.gov/sites/default/files/assets/documents/2018-Jan/CBP-Directive-3340-049A-Border-Search-of-Electronic-Media-Compliant.pdf [https://perma.cc/YH2D-DTZW]. The Department of Homeland Security obtained a warrant to search Touset’s home.15×15. Touset, 890 F.3d at 1230. On January 28, 2015, federal agents executed the warrant and arrested Touset.16×16. Id. Evidence obtained from Touset’s home demonstrated that he had sent more than $55,000 to the Philippines to purchase large amounts of child pornography.17×17. Id.

A grand jury indicted Touset on three counts related to child pornography.18×18. Id. at 1231. Initially, Touset pleaded not guilty to all charges.19×19. Id. He then filed a motion to suppress evidence obtained from the forensic border searches and “the fruit of those searches.”20×20. Id. A magistrate judge recommended denying the motion, noting that the parties agreed that forensic searches required reasonable suspicion and finding that the government was reasonably suspicious of Touset.21×21. United States v. Touset, No. 15-CR-45, 2016 WL 1048047, at *3 (N.D. Ga. Mar. 11, 2016). The district court agreed, denying the motion.22×22. Id. at *4, *6. Touset then pleaded guilty to one of the charges, reserving his right to appeal the district court’s denial of his motion to suppress.23×23. Touset, 890 F.3d at 1231. The government dropped the other two charges. Id. The district court sentenced him to 120 months in prison and supervision for life.24×24. Id. Touset appealed.

The Eleventh Circuit affirmed. Writing for the panel, Judge William Pryor explained that, contrary to the conclusion reached by the district court, forensic searches of electronic property at the border “do[] not require any suspicion.”25×25. Id. Judge Pryor pointed to two customs statutes passed by the First Congress, which also proposed the Fourth Amendment.26×26. See id. at 1232. Both laws permitted customs officials to enter and search vessels without a warrant, and one permitted officials to do so in international waters.27×27. Id. Judge Pryor also asserted that other constitutional provisions — such as Congress’s Article I, Section 8 powers to lay and collect taxes, regulate foreign commerce, and establish a uniform rule for naturalization — buttress the government’s authority to conduct suspicionless searches at the border.28×28. Id.

Turning to caselaw, Judge Pryor observed that neither the Supreme Court nor the Eleventh Circuit has ever required reasonable suspicion for a border search of property.29×29. Id. at 1232–33. In one case, United States v. Montoya de Hernandez,30×30. 473 U.S. 531 (1985). the Supreme Court required that the nonroutine detention of a person at the border be supported by reasonable suspicion.31×31. Id. at 541. Later, in United States v. Flores-Montano,32×32. 541 U.S. 149 (2004). the Supreme Court declined to extend that requirement to a border search involving disassembly of a vehicle’s fuel tank.33×33. Id. at 155. The panel noted that Eleventh Circuit “precedent considers only the personal indignity of a search, not its extensiveness.”34×34. Touset, 890 F.3d at 1234 (internal quotation marks omitted) (quoting United States v. Vega-Barvo, 729 F.2d 1341, 1346 (11th Cir. 1984)). Moreover, Eleventh Circuit precedent permits suspicionless border searches of a ship crew member’s living quarters, al-though “a home ‘receives the greatest Fourth Amendment protection.’”35×35. Id. at 1233 (quoting United States v. Alfaro-Moncada, 607 F.3d 720, 729 (11th Cir. 2010)).

Judge Pryor acknowledged that Touset created a circuit split with the Fourth and Ninth Circuits,36×36. See United States v. Kolsuz, 890 F.3d 133, 137 (4th Cir. 2018); United States v. Cotterman, 709 F.3d 952, 957 (9th Cir. 2013) (en banc). which “have concluded . . . that the Fourth Amendment requires at least reasonable suspicion for forensic searches of electronic devices at the border.”37×37. Touset, 890 F.3d at 1234. The Touset panel was “unpersuaded” by the other circuits’ reasoning, which underscored the invasive nature of forensic searches of digital property.38×38. Id. Instead, the panel emphasized that “a traveler’s ‘expectation of privacy is less at the border’”39×39. Id. at 1235 (quoting United States v. Flores-Montano, 541 U.S. 149, 154 (2004)). and “[t]he [g]overnment’s interest in preventing unwanted persons and effects is at its zenith at the international border.”40×40. Id. (alterations in original) (quoting Flores-Montano, 541 U.S. at 152). The panel further observed that “child pornography offenses overwhelmingly involve the use of electronic devices for the receipt, storage, and distribution of unlawful images,” expressing reluctance to “create special protection” for such property.41×41. Id. at 1235–36. Urging judicial restraint, the panel cited the “longstanding historical practice” of deferring to the policymaking branches in the border search context42×42. Id. at 1237 (quoting United States v. Kolsuz, 890 F.3d 133, 153 (4th Cir. 2018) (Wilkinson, J., concurring in the judgment)). and argued that Congress may, if it wishes, afford individual privacy more than constitutionally minimal protection.43×43. Id. at 1236–37.

Alternatively, the panel held that the government was reasonably suspicious of Touset,44×44. Id. at 1237. asserting that the evidence obtained from Yahoo and Xoom, and during the pre-search investigation, provided a “‘particularized and objective basis for suspecting’ that Touset possessed child pornography on his electronic devices.”45×45. Id. (quoting Denson v. United States, 574 F.3d 1318, 1341 (11th Cir. 2009)).

Judge Corrigan concurred. He noted that the government had changed its position on appeal.46×46. Id. at 1238–39 (Corrigan, J., concurring in part and concurring in the judgment). Because the “new-found government position present[ed] a different and difficult question,” Judge Corrigan joined only the panel’s alternative holding.47×47. Id. at 1239.

For three reasons, the Eleventh Circuit in Touset correctly refrained from requiring individualized suspicion at the border. First, two Founding-era customs statutes provide ample support for the conclusion that the Fourth Amendment was not originally understood to require suspicion for border searches, even when technological changes are accounted for. Second, the Touset panel properly conducted its balancing analysis, emphasizing the strength of the governmental interest at the border as well as the express limits of precedent ascribing significant weight to privacy interests in digital property. Third, both longstanding historical practice and respect for congressional competence counsel for judicial restraint. In rejecting the conclusion reached by other circuits, the Eleventh Circuit gave due weight to all three considerations.

Fourth Amendment analysis proceeds in two steps. First, courts examine the Amendment’s meaning in light of how it was understood at the time of its adoption.48×48. See Wilson v. Arkansas, 514 U.S. 927, 931 (1995). Relevant evidence of meaning includes late-eighteenth-century common law rules49×49. See, e.g., California v. Hodari D., 499 U.S. 621, 624 (1991). and statutes passed by early Congresses.50×50. See, e.g., United States v. Ramsey, 431 U.S. 606, 616–17 (1977); United States v. Watson, 423 U.S. 411, 420–21 (1976). Second, absent adequate guidance from the Founding era, courts conduct a balancing test, weighing the individual and governmental interests at stake.51×51. Riley v. California, 134 S. Ct. 2473, 2484 (2014) (citing Wyoming v. Houghton, 526 U.S. 295, 300 (1999)). Both steps strongly favor the government at the border — neither a warrant nor probable cause is ever necessary.52×52. Ramsey, 431 U.S. at 616–19; see also United States v. Montoya de Hernandez, 473 U.S. 531, 537–38 (1985); Anthony G. Amsterdam, Perspectives on the Fourth Amendment, 58 Minn. L. Rev. 349, 359 (1974) (suggesting that border searches are permissible “without either a warrant or any individuating judgment”). The most demanding constraint the government might face at the border is individualized suspicion, though it is rarely required.53×53. See, e.g., Montoya de Hernandez, 473 U.S. at 541–42; United States v. Brignoni-Ponce, 422 U.S. 873, 881–82 (1975).

The Touset panel correctly invoked two Founding-era customs statutes in support of its conclusion that border searches of property do not require suspicion. The First Congress — the same Congress that proposed the Fourth Amendment — passed both statutes, laws whose “[manifest] historical importance” the Supreme Court has affirmed.54×54. Ramsey, 431 U.S. at 616–17. The panel cited the first statute in discussing Congress’s authority to regulate the entry of persons and effects and referenced the second law because it permitted “customs officials to board vessels even before they reached the United States.”55×55. Touset, 890 F.3d at 1232. Interestingly, though, the court probably underestimated the import of these laws. The first statute required individualized suspicion,56×56. Act of July 31, 1789, ch. 5, § 24, 1 Stat. 29, 43 (repealed 1790) (“[E]very [designated customs official] . . . shall have full power and authority, to enter any ship or vessel, in which they shall have reason to suspect any goods, wares or merchandise subject to duty shall be concealed; and therein to search for, seize, and secure any such goods, wares or merchandise . . . .” (emphasis added)). while the second, passed a year later, did not require any level of suspicion.57×57. Act of Aug. 4, 1790, ch. 35, § 31, 1 Stat. 145, 164–65 (repealed 1799) (“[I]t shall be lawful for [designated customs officials] to go on board of ships or vessels in any part of the United States, or within four leagues of the coast thereof, if bound to the United States . . . for the purposes of . . . examining and searching the said ships or vessels; and the said officers respectively shall have free access to the cabin, and every other part of a ship or vessel . . . .”). Taken together, the laws demonstrate that the First Congress contemplated and understood how to require reasonable suspicion — and how to withhold that requirement — when it wanted. They also show that the First Congress did not understand the Fourth Amendment to require individualized suspicion to search property at the border — otherwise, it would not have passed a law permitting suspicionless searches. In any event, the customs statutes passed by the First Congress provide a solid foundation for the Touset court’s holding.

Although some recent Supreme Court precedent has suggested that Founding-era guidance is not dispositive when considering privacy concerns related to digital property, the Touset panel was justified in relying on such guidance. In Riley v. California,58×58. 134 S. Ct. 2473 (2014). for instance, the Supreme Court held that reasoning applying the Fourth Amendment to digital property “has to rest on its own bottom.”59×59. Id. at 2489. One reason that the original understanding of the Fourth Amendment might not resolve questions about the constitutionality of digital data searches is the uniquely personal, sensitive nature of information stored on digital devices.60×60. See Carpenter v. United States, 138 S. Ct. 2206, 2217 (2018). Another reason is the immense storage capacity of digital devices — for computers, the equivalent of “the amount of information contained in the books on one floor of a typical academic library.”61×61. Orin S. Kerr, Searches and Seizures in a Digital World, 119 Harv. L. Rev. 531, 542 (2005). The Framers surely could not have contemplated the ubiquity of handheld personal libraries, the underregulated search of which raises the specter of the principal evil the Fourth Amendment was designed to confront: general warrants.62×62. Groh v. Ramirez, 540 U.S. 551, 572 (2004) (Thomas, J., dissenting); Thomas K. Clancy, The Role of Individualized Suspicion in Assessing the Reasonableness of Searches and Seizures, 25 U. Mem. L. Rev. 483, 527–28 (1995); Kerr, supra note 61, at 568–71. Of course, even permissible suspicionless searches cannot exceed their constitutionally justifiable scope. See United States v. Molina-Isidoro, 884 F.3d 287, 295–97 (5th Cir. 2018) (Costa, J., specially concurring) (suggesting that the border exception permits searches for “contraband” but not for “evidence of crimes”). Nonetheless, the distinctiveness of the border search context militates for permitting suspicionless searches — even intrusive ones — of digital property. The First Congress explicitly authorized customs officials to board and search vessels extensively and without suspicion, affording officials “free access to the cabin, and every other part of . . . [the] vessel.”63×63. Act of Aug. 4, 1790, ch. 35, § 31, 1 Stat. 145, 164 (repealed 1799). Because the “boarding of a private vessel [is] similar to entry of a private house”64×64. United States v. Villamonte-Marquez, 462 U.S. 579, 605 (1983) (Brennan, J., dissenting). and “a home ‘receives the greatest Fourth Amendment protection,’”65×65. United States v. Alfaro-Moncada, 607 F.3d 720, 729 (11th Cir. 2010) (emphasis added) (quoting United States v. McGough, 412 F.3d 1232, 1236 (11th Cir. 2005)). deeming the customs statutes insufficiently precise guidance is difficult to justify.

The Touset panel also conducted a proper balancing analysis for two reasons.66×66. But see United States v. Kolsuz, 890 F.3d 133, 146 (4th Cir. 2018) (requiring individualized suspicion for a nonroutine search). Kolsuz notwithstanding, Flores-Montano suggests that added protection for nonroutine searches attaches only to “highly intrusive searches of the person.” 541 U.S. 149, 152 (2004) (emphasis added). In only two instances has the Supreme Court required individualized suspicion to conduct a border search. One such instance concerned the prolonged detention of a person suspected of smuggling drugs in her alimentary canal. United States v. Montoya de Hernandez, 473 U.S. 531, 541 (1985). The other concerned a roving patrol stop away from border checkpoints and tainted by racial discrimination. United States v. Brignoni-Ponce, 422 U.S. 873, 885–87 (1975). The first is the narrow holdings of Riley and Carpenter v. United States.67×67. 138 S. Ct. 2206 (2018). In Carpenter, which was decided a month after Touset, the Supreme Court held that the government’s obtaining cell-site location information about an individual from a third party was unconstitutional without a warrant, limiting the third-party exception to the general warrant requirement. Id. at 2217. In Riley, the Court held that an officer must generally obtain a warrant before conducting a search of cell phone data incident to arrest. 134 S. Ct. 2473, 2485 (2014). Both decisions extended Fourth Amendment protections to contexts in which such protections otherwise do not attach, ascribing dispositive weight to privacy interests in digital property in doing so.68×68. See Carpenter, 138 S. Ct. at 2219–20; Riley, 134 S. Ct. at 2484–85. For a discussion of other exceptions to the general warrant requirement, see Groh v. Ramirez, 540 U.S. 551, 572 (2004) (Thomas, J., dissenting) (listing the exceptions to the general warrant requirement). But both cases also noted the possibility of “case-specific exceptions.”69×69. Carpenter, 138 S. Ct. at 2222; Riley, 134 S. Ct. at 2494. And Carpenter, the later decision, expressly declined to extend its holding to “collection techniques involving foreign affairs or national security,”70×70. Carpenter, 138 S. Ct. at 2220. strongly suggesting that the border exception was not at issue.

The second reason Touset got the balancing equation correct is the combination of the border search’s sui generis character71×71. Clancy, supra note 62, at 557 n.329; see also Montoya de Hernandez, 473 U.S. at 538. and the scale of interests tilted strongly in the government’s favor.72×72. United States v. Flores-Montano, 541 U.S. 149, 152–54 (2004). If an individual’s “expectation of privacy is less at the border”73×73. Id. at 154. and “[t]he [g]overnment’s interest in preventing the entry of unwanted persons and effects is at its zenith at the international border,”74×74. Id. at 152. the argument for extending the logic of Riley and Carpenter to the border must do more than merely note the existence of weighty privacy considerations: it must explain why the privacy interests in digital information are sufficient to overcome both the government’s interest at its height and the individual’s diminished interest.75×75. But see United States v. Kolsuz, 890 F.3d 133, 144–46 (4th Cir. 2018) (omitting discussion of a heightened government interest). Also important, the enhanced governmental interest at the border affords the government more leeway to deter the entry of illegal persons and effects. Indeed, the Supreme Court has held the deterrent impact of a search-and-seizure regime relevant to the regime’s reasonableness.76×76. See United States v. Martinez-Fuerte, 428 U.S. 543, 557 (1976) (“[Requiring reasonable suspicion] would largely eliminate any deterrent to the conduct of well-disguised smuggling operations, even though smugglers are known to use these highways regularly.”). Likewise, the Touset panel underscored the lack of alternatives available to stem the inflow of child pornography, which Congress has declared contraband and which is usually stored in digital files.77×77. See Touset, 890 F.3d at 1234–35; see also United States v. Brignoni-Ponce, 422 U.S. 873, 881 (1975) (holding that “the absence of practical alternatives for policing the border” weighs in the government’s favor). Thus, as the panel reasoned, a departure from the default rule that no individualized suspicion is required at the border would “create special protection” for a particularly harmful form of contraband.78×78. Touset, 890 F.3d at 1235.

Both historical practice and respect for legislative competence also support the Touset panel’s exercise of judicial restraint. Because “[e]mpirical questions lie at the heart of the tension between privacy and security interests at the border,”79×79. Kolsuz, 890 F.3d at 150 (Wilkinson, J., concurring in the judgment). judicial articulation of a constitutional minimum would amount to “a hugely consequential policy judgment.”80×80. Id. at 151. As the Touset panel recognized, such policy judgments are the kind that Congress is better suited to and routinely does make.81×81. Touset, 890 F.3d at 1237. And Congress, like many state legislatures, is often vigilantly protective of privacy in digital data — a modern example is the USA Freedom Act, passed in 2015.82×82. See Orin S. Kerr, The Effect of Legislation on Fourth Amendment Protection, 115 Mich. L. Rev. 1117, 1120 (2017); see also Touset, 890 F.3d at 1236–37. In this light, the First Congress’s regulation of customs officials’ conduct at the border stands as a first example in a history of careful congressional standard setting that is as old as the Fourth Amendment itself. Imposing a Fourth Amendment floor at the border without congressional input would amount to an inflexible, “hugely consequential policy judgment” that would lack the benefits of consultation with national security officials and privacy advocacy groups, as well as the constraining influence of legislative consensus-building.

Recent Supreme Court precedents make clear that the Fourth Amendment treats digital property differently. But so too does the Constitution treat the law of border search differently — and it has done so since the nation’s Founding. In light of this longstanding, “manifest” historical authority,83×83. United States v. Ramsey, 431 U.S. 606, 617 (1977). courts would be imprudent to “charg[e] unnecessarily ahead”84×84. Kolsuz, 890 F.3d at 150 (Wilkinson, J., concurring in the judgment). in a manner inconsistent with their Article III function.85×85. See id. at 148. In Touset, the Eleventh Circuit reaffirmed three fundamental Fourth Amendment principles: the importance of the Fourth Amendment’s original public meaning, the distinctiveness of border search doctrine in constitutional balancing, and the prudence of judicial restraint in view of congressional silence. Fidelity to these principles duly respects the justifications of the border search exception and the Founding generation’s codification of border search standards.