INTRODUCTION
Until recently, Mutual Legal Assistance Treaties (“MLATs”) were dusty, specialists’ treaties used mostly to gather evidence during investigations of transnational crimes.[1] These treaties codify states’ mutual responsibilities to one another when one state seeks crime-related information held by an entity within the jurisdiction of another state, such as a foreigner’s e-mail stored on the domestic servers of a domestic e-mail provider.[2] Because of the increasingly slippery nature of data, evidence concerning quintessentially local events is decoupling from the locations and parties most responsible and is flowing abroad,[3] making MLAT requests necessary tools in even the most pedestrian of local prosecutions.
The legal system has a long history of developing tools for responding to evidence requests from foreign jurisdictions, but the current flood of requests looks different in kind, not just in volume, than what we have seen before.[4] With the rise of electronic data, decreasing “distance” between jurisdictions with different legal traditions and expectations has exponentially increased the number and complexity of cross-jurisdictional requests. This decreasing “distance” has not yet, however, lessened the need for domestic courts to ensure that each individual order they issue complies with the domestic rights that individuals accrue by touching their jurisdiction. Thus, the new wave of requests is exposing a fundamental tension behind the territorial jurisdiction regime.
A new generation of MLATs and MLAT substitutes will seek to address these new challenges. In doing so, they will likely adopt radically new structures. Existing proposals generally try to build various mechanisms to induce more trust between parties so that individual requests can undergo less review in receiving states. In other words, they propose forming a so-called “Trusted Nation Club.” But some citizens might reasonably not trust their state to police this boundary on their behalf absent judicial oversight. And even if they did, states must be able to verify that they can continue trusting other states.
So far, the only proposal for punishing states who abuse this increased trust is for a state or international organization to kick an entire state out of a theoretical Trusted Nation Club. For example, a state would lose the right to access the faster, reduced screening process if it routinely sought dissidents’ communications without sufficient suspicion of a crime covered by the agreement. While not purely binary (there is always public shaming), even a quasi-binary punishment regime is ill-suited to the kind of death-by-a thousand-cuts violations that citizens are concerned about.
Therefore, states should experiment with negotiating into the next round of MLAT treaties a limited private right of action against a foreign government in the case of a purposeful or negligent breach. This is a more democratized and gradated system of apportioning the cost of reviewing requests. Furthermore, the mere possibility of close scrutiny in anticipation of legal action will reduce the need for it, and will more fairly dole out non-binary punishments for abusing trust—inducing more compliance in whatever regime replaces the current MLAT system.
I. BACKGROUND ON MLATS AND MLAT REFORM
Providing a full review of the history of MLATs, their deficiencies, and currently-available substitutions is beyond the scope of this Recent Development. But by way of brief introduction: historically, MLATs were bilateral treaties that only covered requests by a prosecutor in a criminal matter.[5] States are currently exploring the benefits of multilateral treaties, and treaties that permit agencies with civil investigatory power to submit MLAT requests as well.[6] With few exceptions,[7] MLATs explicitly refuse to create private rights of action in the event of a treaty violation.[8]
In the United States, these MLATs transform foreign requests into domestically enforceable orders by routing requests through a review at the Office of International Affairs (OIA) at the United States Department of Justice (DOJ).[9] OIA confirms that the request is properly presented, and then sends it out to the proper federal prosecutor who in turn implements the discovery request by filing for an order before a federal judge.[10] The court then considers the motion on its regular calendar,[11] reviewing it mostly for Fourth or Fifth Amendment deficiencies and human rights risks.[12] Subsequent to a discovery order, the DOJ also reviews the company’s response to the order, taking the responsibility to remove excess data, and ensuring that the response respects users’ domestic data privacy rights and does not raise an undue risk of human rights violations.[13]
Litigants may alternatively seek letters rogatory, discretionary requests[14] that are slower[15] than MLATs but are available to private litigants and defendants.[16] They may also circumvent a foreign court entirely by convincing a domestic court to directly issue a subpoena/warrant based on a mostly-foreign entity’s alleged domestic responsibilities. This is despite the entities’ possible contradictory foreign responsibilities, such as those rising from: where they are incorporated, where their “nerve center” is located, where they store the data, or the nationality and location of the data owner.[17]
These doctrines are all partial solutions to a knotty legal problem introduced above: the inherent “un-territoriality” of data, which has torn at the logic of what we understand of as jurisdiction.[18] These problems are not necessarily new,[19] but the ease with which data can aggregate and flow electronically has exponentially eroded the traditional assumption that information about a crime will tend to be located physically close to the crime.
The few modern proposals that seek to address this development have mostly focused on accommodating the interests of the companies and government litigators. But the solutions proposed do not sufficiently consider the interests that private citizens have in a properly restrained international permissions process that subjects individual requests to review by an outside party.
II. CURRENT MLAT REFORM PROPOSALS
Recent academic analyses of MLAT reform agree on broad goals of reform: speeding up requests, respecting the right of all governments to enforce domestic law to the limits of their jurisdiction, and ensuring respect for human rights.[20] The United States has unique interests in this debate because it places a comparatively high due process barrier in front of government demands for electronic data, and because it receives a comparatively large share of MLAT requests for data stored domestically.[21] Additionally, the United States has an interest in the Internet’s potential to expand respect for human rights, privacy, and free expression, and this would be particularly upended if an MLAT breakdown results in greater jurisdictional expansion,[22] or the balkanization of the Internet through data localization.[23] There are numerous proposals that seek to streamline and bolster non-MLAT processes, but this Recent Development focuses on those that seek to salvage something from the MLAT paradigm.
A. Reducing Inefficient MLAT Request Processes at DOJ
There are some uncontroversial proposals in these recommendations that are generally agreed upon. These include: better funding DOJ and OIA,[24] permitting electronic handling of requests,[25] and providing translation and better training to foreign lawyers on how to comply with domestic constitutional and statutory requirements for subpoenas/warrants.[26] They are unlikely to garner much opposition if their proponents can persuade skeptics that the added resources will materially reduce delay and confusion.
However, there is good reason to challenge some plans claiming to streamline (what critics characterize as) the overly bureaucratic process of routing requests through centralized agencies.[27] The specific remedies proposed include introducing a 30-day response time limit,[28] and removing “redundant” steps.[29] But at least in theory these steps have the benefit of introducing careful review and valuable procedural safeguards.
-
“Rocket Docket”/Specialized Court Proposal
One recommendation would create a “rocket docket” that specializes in quickly filing MLAT requests.[30] Although many requests get routed to the Northern District of California or the District of Columbia (based on where the companies are located), OIA does not currently have an explicit policy to route all MLATs to a “rocket docket” that has developed an expertise in processing these kinds of motions.[31]
The proponents of creating an exclusive “rocket docket” in these district courts claim that the expertise such offices will develop will better ensure compliance and consistency.[32] This may be, but—as with any specialty court proposal—this creates the possibility of developing bad law without the chance for circuit/district courts to tee up a split.[33]
Although district courts have historically taken a relatively perfunctory review of MLAT requests, this is an important safeguard to retain. This is especially true because there is no reason to assume that the future of MLAT abuse will resemble the past. Because of the profound uncertainty of what form the future process will take, Article III judges should still be responsible for deciding that a subpoena does not violate the constitution by, for example, creating an “egregious violation of human rights.”[34]
In addition to the value of requiring different district court judges to review these questions redundantly, critics should not discount the potential value in having the DOJ’s review include both specialists and non-specialists. The proposal for a “rocket docket” glosses over the fact that DOJ already has a single point-of-review for MLATs: OIA. OIA reviewers can already develop “expertise”, so sending the requests out to federal prosecutors’ offices for secondary review is valuable because it forces non-specialist lawyers to review the decisions of OIA. Hopefully they can help prevent OIA from getting stuck in an echo chamber. It is not clear how seriously these various U.S. Attorney’s Offices take this responsibility or how reformers might encourage them to take it more seriously, and this might influence whether these extra procedures are worth protecting or expanding.[35]
B. Bypassing Domestic Prosecutors Entirely
There has also been much discussion on how to create new paths requesting states can use to lessen the burden on recipient states’ governments. In the United States, this would mean permitting other states to issue demands directly to companies without DOJ involvement.[36]
These reform proposals tend to agree that implementing the oversight common to them cannot succeed without increasing transparency. Under the existing procedure, United States MLAT requests are sent to companies in the form of a domestic court order under 28 U.S.C. §1782 or 18 U.S.C. §3512,[37] and often those companies may not know that a particular request comes from a foreign government or which foreign government it is.[38]
There are two kinds of possible rights violations these proposals try to avoid through transparency: 1) those that are directed, or at least permitted, by an official foreign government policy; and 2) those that are the result of a single overzealous individual or group within government that is pursuing a particular investigation.
The new proposals address the first kind of violation (those officially tolerated or encouraged) well. The reforms include “red flag triggers” that aggregated requests can use when they exhibit suspicious tendencies, and transparency reports that permit some sort of ex post review.[39] These reforms generally trust states to analyze whether their own direct government-company requests are appropriately “proportional and necessary.”[40] The proposals justify this by assuming that the new transparency required will disclose information about individual requests to other governments and international law enforcement organizations, and that these entities will have the ability and desire to ensure that neither type of abuse is occurring.
On the other hand, these proposals are not well designed to discover or dissuade individual overzealous prosecutors from abusing this trust. The next section more fully discusses why any trust created is ripe for abuse unless individuals have the right to discover and seek recourse for abuses of trust by individual state officials.
III. THE TRUSTED NATION CLUB AND ITS VERIFICATION REGIME
One of the most popular proposals tying this all together is to create some sort of “Trusted Nation Club”: a group of states that are entitled to a special, expedited process.[41] The Club members would screen other states’ entry based on an analysis of their commitment to the rule of law and human rights.[42] The Club would initially enable the easy transfer of subscriber information only when investigating “serious crimes, organized crime, terrorism, and cyber crime.”[43] The proposals contemplate eventually expanding these procedures to cover other types of crime, requests sent directly to companies, and loosening the—already partially waived[44]—requirement of “dual criminality”.[45] Increased transparency would induce good behavior, and would permit discovery of those states abusing the trust.[46] This allows states in the Club to minimize process for other states that they believe are least likely to abuse the legal system to infringe on domestic rights or to facilitate violating human rights.
Although proponents present minor differences in what this system would look like, there are so far only two criticisms that categorically question whether this is a good idea.
First, this would raise the politically fraught decision of who to admit.[47] This a non-trivial problem, but does not appear to be a persuasive reason to avoid trying to create a Trusted Nation Club altogether. It is worth testing whether a rough consensus, non-corner equilibrium can be reached.
Second, this would not actually address the MLAT requests that are the most problematic: those issued by states with little familiarity with our legal system, and who might respond to continued MLAT failure by forcing data relocation to their countries.[48] While true, this ignores the benefit of easing any pressure on the understaffed offices that have to deal with the more challenging requests. It also fails to recognize the (admittedly marginal) value of the additional enticement this offers to bad actors to adopt and respect international norms.
In addition to the two critiques above, this Recent Development raises, for the first time, a third concern: that the currently contemplated punishments for abusing trust are essentially binary, in-or-out decisions made by a state concerning another state. They depend on states discovering and publically outing other states for violating this trust. Although the question of how to punish is, in many ways, similar to the question of who to admit into a potential Club, it raises a novel threat. Namely, that any Trusted Nation Club that gets off the ground is likely to result instead in a Trusted Nation Cartel that effectively colludes to undermine the rights of its citizens.
But that is only true because, when drafting our MLATs, we created an artificial shortage of punishment options by leaving finer-grained punishments off the table.
A. Currently Proposed Punishments
Accepting the principle of a Trusted Nation Club, any such Club would require robust checks preventing other Club members from seeking an order that violates the rules of the Club (which in turn violates individual states’ domestic law). But what sanction is available if a state catches another state breaking the rules? And how likely are different kinds of violations to be discovered?
The current proposals rely on transparency to scare bad state-actors into behaving, and to allow states to discover violators that should be kicked out of the Club.[49] This decision of whether to kick someone out of the Club would turn on a complex balance. This balance has not yet been concretely formulated by Club proponents, but in practice it would likely encompass something like the following: 1) whether the cumulative, aggregated foreign violations have undermined domestic individuals’ constitutional and statutory protections to such an extent that further participation in the scheme would be an endorsement of that unconstitutional process, and 2) whether this harm is sufficient to outweigh the benefits the state is seeking by remaining a Club member.
The complex process of tallying and weighing these interests would, ironically, be an example of “simple outcasting”—a straightforward invocation of Article 60 of the Vienna Convention on the Law of Treaties.[50] It would get more abstract if the scheme contemplates a role for Interpol (or another international organization) in this decision.[51] This then would require states and citizens to place their faith in a non-judicial, international agency to correctly strike this complex balance on their behalf.[52]
B. Limitations on Punishing Individual Violations
Therefore, the currently proposed sanctions are comparatively discrete punishments that are invoked once a state reaches a tipping point. They are reasonably well matched to catching and punishing the first kind of abuse individuals can be harmed by: aggregate, centralized abuse. They are not, however, well matched to the second kind of abuse: those by individual functionaries in far flung local offices overseas that are hard to attribute to a centralized policy purposefully circumventing the protections of a foreign constitution.
There is no domestic solution to this second problem. Take the OIA in the United States as an example: like other centralized MLAT offices, the OIA is purely administrative and does not have the ability to “compel or instruct” other states when their actions violate the treaty.[53] Given the reciprocity required by these treaties, this is also true of centralized offices in other states dealing with United States requests proceeding through their courts. The current treaties permit states to deny requests, but the clamor for reform shows that states consider even this resource drain too burdensome. Governments have shown no appetite for undertaking the more intensive responsibility of pursuing remedies for these individual violations.
Were a domestic solution to appear, it could conceivably take the form of a state earmarking the bureaucratic resources saved from membership in the Trusted Nation Club and reinvesting them for the purpose of discovering manipulations of that trust. But even assuming a full rededication of resources, this is a step backwards. These resources will be spread over more requests and will be less efficient at discovering nonconforming filings. The benefit of the Trusted Nation Club is entirely based on the bureaucratic savings officials accrue when they no longer must bear the cost of having to transpose a domestic legal matter into the language (both institutional and linguistic) of the foreign legal regime. Therefore, inside the Trusted Nation Club, officials reviewing a random subset of incoming requests would be required to know as much—if not more—about how criminal prosecutions are built in the requesting state than they do when reviewing requests now. Additionally, they would be working off of documents that—if they are in fact less of a burden on foreign government employees to produce—include less information than current MLAT requests provide.[54]
Nor is there a real international solution. No reformer has yet proposed a sufficient alternate method of policing whether individual foreign law enforcement officials are violating the trust this scheme relies on. Such an adequate solution would smooth out the binary nature of the in-or-out punishment discussed supra in Section III.A. It would enforce good behavior and encourage states to police their officials sufficiently closely. For example, this might take the form of public shaming or of developing informal backchannels to flag likely abusers’ requests.
But these attempts to change to a more continuous punishment regime are likely not enough because they merely reduce the severity of the sanction for abusing trust while not manifestly increasing the likelihood of discovering abuse. They still rely on someone knowing that a particular individual has abused, or is likely to abuse, the system.
Taking public shaming first, if we assume arguendo that public shaming is a useful tool, that is only so because states take allegations of duplicity seriously. Therefore, before a state would risk publicly calling out their fellow Club member, the same type of evidence proving an individual official’s abuses is likely necessary. And as is discussed infra in Section III.D, the current proposals do not contemplate the kind of transparency necessary for such a discovery to be likely.
As for informal backchannels, a state official would have to be relatively confident before willingly disclosing to a foreign state official that they suspect a compatriot of abusing the system. It is possible to build into a bureaucracy the kind of transparency and accountability necessary to produce this confidence. But, taking for example the amount of high profile Brady violations in the United States that are unearthed only during litigation, it is clear that individual prosecutors operating in adversarial judicial systems have a remarkable ability to obfuscate their dubious ethical litigation decisions from their superiors.[55] Many of these violations are surely prevented or discovered by superiors before they reach court, but there is no shortage of cases that slip through—despite the embarrassment and sanctions available for such a violation.[56]
Therefore, this Recent Development proposes that we can more properly induce good behavior through a loosening of the complete refusal to grant private rights of action in the event of an MLAT breach. At the very least this democratizes the burden of deciding to closely review a particular individual MLAT request, and provides a real threat of discovering and punishing bad individual government actors after they try to abuse the trust their government has earned.
C. Limited Private Right of Action as a Verification Regime
Unless the United States is planning on abusing the trust of other states or on permitting others to do so with impunity, the creation of a private right of action need not subject the United States to a serious increase in process/risk when it requests data from overseas.
Existing MLATs almost uniformly exclude a private right of action against a foreign government in domestic courts for violating the treaty.[57] MLAT requests have been refused for not complying with the terms of the treaty and domestic law,[58] but if courts discover after the fact that a request is granted erroneously, they tend to treat it as an act that was merely not controlled by the MLAT, not as a violation of the MLAT.[59]
In the United States, this refusal to recognize an enforceable private benefit has a large impact on whether evidence is subject to the Fourth Amendment’s exclusionary rule in domestic prosecutions,[60] but it would also be relevant when considering a possible civil action against a foreign entity.[61]
Policy makers might logically wonder why they would create any new privately enforceable right of action against themselves in foreign courts by non-domestic-persons. But this ignores the fact that the right can initially be very narrowly defined, and that it is reciprocal.
The underlying rationale supporting this partial waiver of sovereign immunity is the same as the rationale underlying the decision states make to put any process between the receipt of an MLAT request and a response: we, as states, are concerned about foreign governments infringing on rights/interests protected by domestic law—constitutional, statutory, or international human rights. Essentially, if both states are reciprocally bound, then that is a fair price to pay for the benefit one expects to receive from binding the other party. If the United States joins a Trusted Nation Club, it provides a trusted foreign government with some new benefit (by reducing checks and providing a less onerous process). We should consider imposing a new mandatory condition of joining the Club: permitting a targeted, private right of action for abusing this trust.
This change would require renegotiating MLAT language—a process already required for a Trusted Nation Club anyway. But in the United States, at least, it would not necessarily require additional legislation, because it would be relatively simple to tie to the Foreign Sovereign Immunities Act or the Alien Tort Statute’s grant of jurisdiction for treaty violations between the United States and foreign governments.[62]
It is beyond this Recent Development’s scope to propose exactly how “large” of a crack this would be, and who could raise which causes of action in which forums. This can, and should, be negotiated while states find the proper balance by trial and error. At a minimum, it should be relatively uncontroversial to permit a right of action for some purposeful violations of the treaty—like filing court documents with the intent to deceive or obfuscate the improper motive of the request. Only states that are considering taking advantage of the trust would be disadvantaged, and they can hardly complain. Especially if the remedy available is initially limited to monetary damages.
There is always a worry of nuisance suits, but the briefing required to dismiss a case for failure to state a claim would largely turn on clear-cut, treaty interpretation questions concerning the narrow right created. This is admittedly more time consuming than submitting an off-the-shelf brief that can invoke an ironclad bar to suit. But if a government is already presenting an argument to their courts that a plaintiff failed to state a claim, then the marginal effort required would manifest as merely making sure that the government adequately presents their alternate theories for why the claim must fail. Importantly, states can experiment with different shapes and sizes of waiver, slowly widening them until the costs outweigh the benefits.
D. The Transparency Required Is Feasible
Will this recommendation escape the shoals that threaten all major MLAT reform proposals: states’ reluctance to submit to the kind of transparency regime necessary to actually implement the plan? There is admittedly no perfect solution. However, this solution evades more of this hazard than any other current proposal by creating ex post remedies for as much abuse as is discoverable under any conceivable regime.
Transparency is not at the top of everyone’s wish list. Law enforcement often must rely on sealed requests in order to not tip off the subjects to their investigations.[63] They can be expected to oppose any disclosure that they believe will give the public information about their priorities, techniques, or procedures.[64] This reluctance was recently demonstrated in the relatively weak, permissive transparency requirements of the Privacy Shield Framework Principles negotiated between the European Union and United States, which replaced the International Safe Harbor Privacy Principles, previously struck down by the European Court of Human Rights.[65]
These kinds of transparency proposals fail to address the concern that they may well result in regimes that disclose the size and shape of the haystacks, but that forbid people to search for needles.[66] On the contrary, the waiver proposed here gives defendants (or their counsel) the power to probe and bring to light any particular needle that was used to facilitate a specific violation of their rights. This vests the person with the best chance of discovering abuse with the power and the incentive to challenge any improper redactions or ex post justifications covering up violations of the MLAT.
This is undoubtedly not a solution for all concerns. Specifically, the proposal provides no remedy when individuals have no way to discover these violations, i.e. when they are not permitted to review or challenge evidence used to prosecute or to violate their rights. But if the aggregate screens work as advertised, then that will have to be the mechanism for discovering these kind of large, centrally endorsed, extra-judicial programs that prevent individuals from knowing how evidence against them was gathered.[67] The private right of action will be a parallel treatment for the second, separate ailment.
CONCLUSION
The more ambitious reform proposals assume that there is no need for individualized scrutiny of each separate MLAT request. They build robust systems for catching large systemic violations of the MLAT trust, but they rely on states and international organizations to police these boundaries. This Recent Development notes that this may turn the Trusted Nation Club into a Trusted Nation Cartel, because these proposals have failed to subcontract to defendants the cost and benefit of reviewing one-by-one the legality of these requests. If no one else is picking up this burden, individual defenders will be happy to at least have the opportunity to pick it up themselves, and are the parties best suited for seeking individual remedies for those harmed by the violations they find.
*Former Stanford Law School Fellow at the Electronic Frontier Foundation. In addition to the editors, I would like to thank Lee Tien, Mark Jaycox, Wafa Ben Hassine, Michael Frenkel, Andrew Crocker, Professor Beth Van Schaack, and Professor Alan Weiner for their advice and feedback. The analysis and views presented in this Recent Development are those of the author alone, and do not reflect the views of the Electronic Frontier Foundation or Stanford Law School.
[1] See Virginia M. Kendall & T. Markus Funk, The Role of Mutual Legal Assistance Treaties in Obtaining Foreign Evidence, 40 LITIG. 59, 59 (2014).
[2] MLATs do cover other types of cooperation (for example, extradition), but this Recent Development is limited to those areas of MLATs that concern electronic discovery and warrants beyond national jurisdictions’ boundaries.
[3] See generally Jennifer Daskal, The Un-territoriality of Data, 125 YALE L.J. 326 (2015).
[4] See Tim Wu, Is Internet Exceptionalism Dead?, in THE NEXT DIGITAL DECADE: ESSAYS ON THE FUTURE OF THE INTERNET 179, 187 (Berin Szoka & Adam Marcus eds., 2010) (explaining that the Internet has revealed some latent assumptions that we no longer have the luxury of relying on).
[5] The United States had signed MLATs with sixty-four countries, fifty-eight MLATs were as of 2013. 7 FOREIGN AFFAIRS MANUAL § 962.1(d), https://fam.state.gov/searchapps/viewer?format=html&query=MLAT&links=ML… M/07FAM0960.html#M962_1. It also has two Mutual Legal Assistance Agreements (“MLAAs”)—one between the United States and China, and one between the American Institute in Taiwan and the Taipei Economic and Cultural Representative Office. Bureau of Int’l Narcotics & Law Enforcement Affairs, 2012 International Narcotics Control Strategy Report (INCSR): Treaties and Agreements, U.S. DEP’T STATE (Mar. 7, 2012), http://www.state.gov/j/inl/rls/nrcrpt/2012/vol2/184110.htm.
[6] The United States also signed a superseding multilateral MLAT with the members of the European Union as a whole. T. Markus Funk, Mutual Legal Assistance Treaties and Letters Rogatory: A Guide Judges, FED. JUDICIAL CENT. INT’L LITIG. GUIDE 6 (2014), http://www.fjc.gov/public/pdf.nsf/lookup/mlat-lr-guide-funk-fjc-2014.pd… 2014.pdf; see also 7 FOREIGN AFFAIRS MANUAL § 962.1(d); Andrew K. Woods, Data Beyond Borders: Mutual Legal Assistance in the Internet Age, GLOBAL NETWORK INITIATIVE 15-16 (Jan. 2015), https://globalnetworkinitiative.org/sites/default/files/GNI%20MLAT%20Re….
[7] Those without explicit limitations (see, e.g., Extradition and Mutual Assistance in Criminal Matters, U.S.-Turk., Jan. 1, 1981, 32 U.S.T. 3111; Mutual Legal Assistance Treaty, U.S. Switz. May 25, 1973, 27 U.S.T. 2019) are silent on the issue, but reviewing courts will likely deny an attempt to bring a private right of action based on their violation because of the general assumption that private rights of action to enforce international treaties must be explicitly granted in those document. Cf. Goldstar (Panama) S.A. v. United States, 967 F.2d 965, 968 (4th Cir. 1992).
[8] See Michael A. Rosenhouse, Annotation, Validity, Construction, and Application of Mutual Legal Assistance Treaties (MLATs), 70 A.L.R. FED. 2D 375, at §§ 6-7.
[9] Peter Swire & Justin D. Hemmings, Re-Engineering the Mutual Legal Assistance Treaty Process, N.Y.U. ANN. SURV. AM. L. (forthcoming 2016) (manuscript at 9-12).
[10] Id. at 10.
[11] Id.
[12] Funk, supra note 6, at 6-7, 10-11.
[13] Swire & Hemmings, supra note 9, at 9-12.
[14] Funk, supra note 6, at 5, 17.
[15] Funk, supra note 6, at 18; Swire & Hemmings, supra note 9, at 11 (citing REP AND RECOMMENDATIONS OF THE PRESIDENT’S REVIEW GROUP ON INTELLIGENCE AND COMMUNICATIONS TECHNOLOGIES (2013), https://www.whitehouse.gov/sites/default/files/docs/2013-12- 12_rg_final_report.pdf).
[16] The first three MLATs the United States signed (Switzerland, Turkey, and the Netherlands) do permit some defense discovery through the MLAT process, but the more recent do not. Funk, supra note 6, at 12; see also Mutual Assistance in Criminal Matters, U.S.-Neth., Sept. 15, 1983, T.I.A.S. 10734; Extradition and Mutual Assistance in Criminal Matters, U.S.-Turk., Jan. 1, 1981, 32 U.S.T. 3111; Mutual Legal Assistance Treaty, U.S.-Switz. May 25, 1973, 27 U.S.T. 2019.
[17] See, e.g., In re Warrant to Search a Certain Email Account Controlled & Maintained by Microsoft Corp., 15 F. Supp. 3d 466, 474-75 (S.D.N.Y. 2014), (interpreting the Stored Communications Act to require United States Internet Service Providers to produce information “within [their] control” even when the United States declined to seek an MLAT request in Ireland against Microsoft’s Irish subsidiary).
[18] See generally Daskal, supra note 3, at 330 (“Put bluntly, data is destabilizing territoriality doctrine.”).
[19] Id. at 329-30.
[20] With the possible exception of those analyses that propose an ultimate goal of changing domestic law to focus on the nationality of the person who creates or owns the data and not on ‘territoriality’—defined as where the user resides and where the “content is produced.” See Reema Shah, Comment, Law Enforcement and Data Privacy: A Forward-Looking Approach, 126 YALE L.J. 543, 550 (2015); see also Orin S. Kerr, The Next Generation Communications Privacy Act, 162 U. PA. L. REV. 373, 418 (2014) (arguing that United States domestic law should be amended to determine which legal standards apply based on user location).
[21] Gail Kent, Sharing Investigation-Specific Data with Law Enforcement—An International Approach 8 (Feb. 14, 2014) (Stanford Public Law Working Paper), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2472413.
[22] Jonah Force Hill, Problematic Alternatives: MLAT Reform for the Digital Age, NAT’L. SEC. J. HARV. L. ONLINE http://harvardnsj.org/2015/01/problematic-alternatives-mlat-reform-for-…; Woods, supra note 6, at 2.
[23] Hill, supra note 22; Kent, supra note 21, at 7-8; Swire & Hemmings, supra note 9, at 2; Woods, supra note 6, at 2.
[24] Hill, supra note 22; Kent, supra note 21, at 13; Swire & Hemmings, supra note 9, at 2; Woods, supra note 6, at 10; Liberty and Security in a Changing World, PRESIDENT’S REVIEW GROUP ON INTELLIGENCE AND COMMUNICATIONS TECHNOLOGIES 227-28 (Dec. 12, 2013), https://www.whitehouse.gov/sites/default/files/docs/2013-12-12_rg_final… [hereinafter Review Group Report.
[25] Hill, supra note 22; Kent, supra note 21, at 12-13; Review Group Report, supra note 24, at 228; Woods, supra note 6, at 2.
[26] Hill, supra note 22; Kent, supra note 21, at 12; Swire & Hemmings, supra note 9, at 2; Woods, supra note 6, at 26.
[27] Hill, supra note 22; Kent, supra note 21, at 13; Swire & Hemmings, supra note 9, at 27- 30; Review Group Report, supra note 24, at 228.
[28] Woods, supra note 6, at 11.
[29] Hill, supra note 22; Kent, supra note 21, at 13.
[30] Swire & Hemmings, supra note 9, at 27-30.
[31] Id. at 27.
[32] Id.
[33] See Diane P. Woods, Is It Time to Abolish the Federal Circuit’s Exclusive Jurisdiction in Patent Cases? 13 CHI.-KENT J. INTELL. PROP. 1, 3 (noting that uniformity “says nothing about quality or accuracy” in patent decisions (alteration in original)); cf. Janet Cooper Alexander, Jurisdiction-Stripping in a Time of Terror, 95 CALIF. L. REV. 1193, 1201-03 (discussing the exclusive jurisdiction of the D.C. Circuit over Guantanamo Bay challenges under the Detainee Treatment Act).
[34] See Funk, supra note 6, at 10-11 (citing In re Search of Premises Located, 634 F.3d 557 (9th Cir. 2011)).
[35] Cf. Matthews v. Eldrige, 424 U.S. 319 (1976) (balancing the interests at issue, the risk of error, and the cost/potential corrective value of additional procedures). In addition to Swire, at least one other academic believes that there are far too many levels of review and that they “are at best redundant and at worst unnecessarily bureaucratic,” “do[ing] no more to protect the privacy of the user, and instead frustrating the investigative or judicial process in the [requesting] country.” Kent, supra note 21, at 6-7.
[36] Mailyn Fidler, MLAT Reform: Some Thoughts from Civil Society, LAWFARE (Sept. 11, 2015, 12:22 PM), https://www.lawfareblog.com/mlat-reform-some-thoughts-civil-society; see also Kent, supra note 21, at 24 (describing “controversial” proposals that would permit direct responses from domestic companies). But see Woods supra note 6, at 15 (describing the challenge a company faces when it is responsible for complying with only valid foreign request for production).
[37] Kendall & Funk, supra note 1, at 2.
[38] Swire & Hemmings, supra note 9, at 32-33 (citing Woods, supra note 6, at 11).
[39] Kent, supra note 21, at 23-24.
[40] Id. at 19.
[41] Swire & Hemmings, supra note 9, at 33-41; Woods, supra note 6, at 16; see also Fidler, supra note 36 (describing such a proposal as “practical”); Kent, supra note 21, at 14-15 (encouraging easier sharing of subscriber information with limited number of states); cf. Kerr, supra note 20, at 418 (proposing a United States domestic regime in which certain countries are deemed “sufficiently protective and democratic” to honor their legal process).
[42] Kent recommends using the following to determine membership: transparency, right of appeal, accountability, and evidence of past behavior. Kent, supra note 21, at 15 (citing Ian Brown & Douwe Korff, Digital Freedoms in International Law: Practical Steps to Protect Human Rights Online, GLOBAL NETWORK INITIATIVE 2012). Kent also proposes those members of the ‘five eyes’ or those in the G8 24/7 Network or the Council of Europe Convention on Cybercrime and MLATs. Kent, supra note 21, at 15.
[43] Kent, supra note 21, at 15.
[44] Kendall & Funk, supra note 1, at 2-3.
[45] Kent, supra note 21, at 16.
[46] Id. at 22; Mark Jaycox & Lee Tien, Reforms Abound for Cross-Border Data Requests, ELECTRONIC FRONTIER FOUND. (Dec. 27, 2015), https://www.eff.org/deeplinks/2015/12/reformsabound-cross-border-data-r….
[47] Woods, supra note 6, at 16.
[48] Id.
[49] Kent, supra note 21, at 22; Woods, supra note 6, at 7.
[50] Oona A. Hathaway & Scott J. Shapiro, Outcasting: Enforcement in Domestic and International Law, 121 Yale L.J. 252, 326 (2011) (citing Vienna Convention on the Law of Treaties art. 60, May 23, 1969, 1155 U.N.T.S. 331).
[51] See, e.g., Kent, supra note 21, at 21.
[52] This would, perhaps, look similar to the International Coffee Organization, which permits centralized International Coffee Council to exclude members from the benefits of membership See Hathaway & Shapiro, supra note 50 at 330.
[53] Kent, supra note 21, at 7.
[54] Both of these points remain relevant even if the Trusted Nation Club permits domestic officials to have easy unlimited access to the entire dossier of the foreign government’s case. As a trivial example of the difficulty that remains, consider the burden that even fluent United States lawyers and law enforcement individuals face when deciphering raw domestic criminal history reports created by an unfamiliar database. See, e.g., BUREAU OF JUSTICE STATISTICS, U.S. DEP’T OF JUSTICE, USE AND MANAGEMENT OF CRIMINAL HISTORY RECORD INFORMATION: A COMPHREHENSIVE REPORT 33 (1993), http://www.bjs.gov/content/pub/pdf/CCHUSE.PDF (“The formats in use vary so greatly that it is probably true that no two State criminal history record formats are identical and many of them are not even similar… . As a review of these samples will show, the formats vary from columnar designs with titles over values to various forms of linear designs utilizing indentations or upper- and lower-case type to distinguish titles from values. Some of the records may leave some data fields blank while others display “unknown” in all spaces where information is not provided.”).
[55] Brady violations occur when a prosecutor withholds “material evidence that has a significant adverse impact on the accuracy of the guilt/innocence determination.” See Cynthia E. Jones, A Reason to Doubt: The Suppression of Evidence and the Inference of Innocence, 100 J. CRIM. L. & CRIMINOLOGY 415, 436 (2010).
[56] See id. (concluding that Brady violations should never occur, and therefore that it is problematic that we are still stuck in the debate over whether they are “epidemic” or “episodic”). But this is a not just a problem for the United States. Other systems have also had trouble enforcing their individual employees’ obligations to disclose exculpatory evidence. See, e.g., Rob Evans & Paul Lewis, CPS in crisis as allegations of suppressed evidence wreck trials, THE GUARDIAN (July 18, 2011), http://www.theguardian.com/environment/2011/jul/18/cps-mark-kennedy-tri… (reporting that “senior officials” in the prosecution office “repeatedly fail[ed]” to meet their disclosure obligations); see also Stanley Z. Fisher, The Prosecutor’s Ethical Duty to Seek Exculpatory Evidence in Police Hands: Lessons from England, 68 FORDHAM L. REV. 1379, 1410-1413 (2000) (describing English prosecutors’ difficulty in complying with law requiring them to disclose exculpatory evidence);
[57] See Rosenhouse, supra note 8.
[58] Dera J. Nevin & Marc Jenkins, Information, Knowledge, and the Pursuit of Privacy, 38 AM. J. TRIAL ADVOC. 485, 507 (2015) (citing United States v. Schneider, 2202 BCSC 104 (July 5, 2002)).
[59] United States v. Rommy, 506 F.3d 108, 129-30 (2d Cir. 2007) (“[Defendant] cannot demonstrate that the [U.S.-Netherlands] treaty creates any judicially enforceable individual right that could be implicated by the government’s conduct here.”); United States v. Davis, 767 F.2d 1025, 1030 (2d Cir. 1985) (noting that the U.S.-Switzerland MLAT conferred only specific “judicially enforceable rights on individuals,” not implicated in that case); see In re Request from the United Kingdom, 685 F.3d 1, 9 (1st Cir. 2013).
[60] See, e.g., United States v. Castrillon, 2007 WL 2398810 (S.D.N.Y. 2007) (basing exclusionary rule analysis on whether foreign law enforcement were under instruction of domestic law enforcement).
[61] See, e.g., Foreign Sovereign Immunities Act 28 U.S.C. § 1605(a)(1) (permitting jurisdiction in the event of explicit waiver of sovereign immunity), Alien Tort Statute, 28 U.S.C. § 1350.
[62] See 28 U.S.C. § 1350. Alternately, states could affirmatively create a “private right of action” for these particular treaties by passing implementing legislation. Oona A. Hathaway, Sabria McElroy, & Sara Aronchick Solow, International Law at Home: Enforcing Treaties in U.S. Courts, 37 YALE J. INT’L L. 51, 56 (2012).
[63] Microsoft recently revealed the extent of this desire for secrecy, claiming that law enforcement’s use of the gag order provision of the Electronic Communications Privacy Act was unconstitutionally overzealous. Todd Bishop, Interview: Why Microsoft is suing the U.S. Justice Department over cloud security, GEEKWIRE (Apr. 14, 2016), http://www.geekwire.com/2016/interviewmicrosoft-suing-u-s-justice-depar… (“I think that people probably should be surprised that the government is now so routinely seeking secrecy orders that have no end date at all. I don’t know that people have thought about this deeply in the past but the practice in this country is that secrecy’s been the exception not the norm. It required a standard of proof by the government and it was understood to last for only as long as needed. When you get over 1,700 secrecy orders that are permanent, it really means that they will last forever.”); Steve Lohr, Microsoft Sues Justice Department to Protest Electronic Gag Order Statute, N.Y. TIMES (Apr. 14, 2016), http://www.nytimes.com/2016/04/15/technology/microsoft-sues-us-over-ord….
[64] See, e.g., 5 U.S.C. 552(b)(7)(E); Exemption 7(E), FREEDOM OF INFORMATION GUIDE, DEP’T OF JUSTICE (May 2004), http://www.justice.gov/oip/foia-guide-2004-edition-exemption-7e. The current proposals could easily result in a hazy-disclosure regime based on a similar model to DOJ’s practice under 18 U.S.C. § 2809. See Letter from James M. Cole, U.S. Deputy Att’y Gen., to Tech Company Gen. Counsels (Jan. 27, 2014), http://www.justice.gov/iso/opa/resources/366201412716018407143.pdf (permitting recipients of government to disclose publicly only bands of how many requests they received).
[65] See EU-U.S. Privacy Shield Principles, U.S. DEP’T OF COMMERCE 29-30 (Feb. 2016), https://www.commerce.gov/sites/commerce.gov/files/media/files/2016/eu_u… df.pdf. (“Privacy Shield organizations may voluntarily issue periodic transparency reports on the number of requests for personal information they receive by public authorities for law enforcement or national security reasons, to the extent such disclosures are permissible under applicable law.”)
[66] See, e.g., Jennifer Daskal & Andrew K. Woods, Cross-Border Data Requests: A Proposed Framework, LAWFARE (Nov. 24, 2015), https://lawfareblog.com/cross-border-data-requests-proposedframework (discussing a proposed transparency regime).
[67] Official endorsement of parallel construction (where evidence collected using particular means is not admitted in court but can lead to admissible evidence) is, unfortunately, not subject to review at all in individual cases, but must rather be challenged in cases that appear to U.S. lawyers to be seeking advisory opinions without a live case or controversy.