Algorithmic Jim Crow

Fordham Law Review, Nov 2017

This Article contends that current immigration- and security-related vetting protocols risk promulgating an algorithmically driven form of Jim Crow. Under the “separate but equal” discrimination of a historic Jim Crow regime, state laws required mandatory separation and discrimination on the front end, while purportedly establishing equality on the back end. In contrast, an Algorithmic Jim Crow regime allows for “equal but separate” discrimination. Under Algorithmic Jim Crow, equal vetting and database screening of all citizens and noncitizens will make it appear that fairness and equality principles are preserved on the front end. Algorithmic Jim Crow, however, will enable discrimination on the back end in the form of designing, interpreting, and acting upon vetting and screening systems in ways that result in a disparate impact.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

Algorithmic Jim Crow

Margaret Hu, Algorithmic Jim Crow Algorithmic Jim Crow Margaret Hu 0 1 0 Thi s Article is brought to you for free and open access by FLASH: The F ordham Law Archive of Scholarship and History. It has been accepted for inclusion in Fordham Law Review by an authorized editor of FLASH: The F ordham Law Archive of Scholarship and History. For more information , please contact 1 Washington and Lee University School of Law , USA Margaret Hu* This Article contends that current immigration- and security-related vetting protocols risk promulgating an algorithmically driven form of Jim Crow. Under the “separate but equal” discrimination of a historic Jim Crow regime, state laws required mandatory separation and discrimination on the front end, while purportedly establishing equality on the back end. In contrast, an Algorithmic Jim Crow regime allows for “equal but separate” discrimination. Under Algorithmic Jim Crow, equal vetting and database screening of all citizens and noncitizens will make it appear that fairness and equality principles are preserved on the front end. Algorithmic Jim Crow, however, will enable discrimination on the back end in the form of designing, interpreting, and acting upon vetting and screening systems in ways that result in a disparate impact. Currently, security-related vetting protocols often begin with an algorithm-anchored technique of biometric identification—for example, the collection and database screening of scanned fingerprints and irises, digital photographs for facial recognition technology, and DNA. Immigration reform efforts, however, call for the biometric data collection of the entire citizenry in the United States to enhance border security efforts and to increase the accuracy of the algorithmic screening process. Newly * Associate Professor of Law, Washington and Lee University School of Law. I would like to extend my deep gratitude to those who graciously offered comments on this draft or who offered perspectives and expertise on this research through our thoughtful discussions: Sahar Aziz, Jack Balkin, Kate Bartlett, Gerlinde Berger-Walliser, Joseph Blocher, danah boyd, Devon Carbado, Jennifer Chacón, Guy Charles, Andrew Christensen, Danielle Citron, Adam Cox, Deven Desai, Mark Drumbl, Michelle Drumbl, Josh Fairfield, Frank Pasquale, Jennifer Granick, Janine Hiller, Gordon Hull, Trina Jones, Stephen Lee, Charlton McIlwain, Steve Miskinis, Hiroshi Motomura, Kish Parella, Jeff Powell, Angie Raymond, Bertrall Ross, Victoria Sahani, Andrew Selbst, and Kevin Werbach. In addition, this research benefited greatly from the discussions generated from the Immigration Law Professors Workshop; the Law & Society Association Annual Meeting; Yale Law School’s Information Society Project’s Unlocking the Black Box Conference; the Privacy Law Scholars Conference; Data & Society Research Institute’s Eclectic Paper Workshop; Duke Law School’s Culp Colloquium; Texas A&M Faculty Workshop Series; UNC Charlotte Surveillance Colloquium; and the Wharton School of the University of Pennsylvania, Law & Ethics of Big Data Colloquium. Many thanks to the editorial care of Vincent Margiotta of the Fordham Law Review, and for the generous research assistance of Alexandra Klein, Kirby Kreider, Bo Mahr, and Carroll Neale. All errors and omissions are my own. developed big data vetting tools fuse biometric data with biographic data and internet and social media profiling to algorithmically assess risk. This Article concludes that those individuals and groups disparately impacted by mandatory vetting and screening protocols will largely fall within traditional classifications—race, color, ethnicity, national origin, gender, and religion. Disparate-impact consequences may survive judicial review if based upon threat risk assessments, terroristic classifications, datascreening results deemed suspect, and characteristics establishing anomalous data and perceived foreignness or dangerousness data— nonprotected categories that fall outside of the current equal protection framework. Thus, Algorithmic Jim Crow will require an evolution of equality law. INTRODUCTION.......................................................................................... 634  I. BIRTH OF ALGORITHMIC JIM CROW...................................................... 645 II. OVERVIEW OF JIM CROW: CLASSIFICATION AND SCREENING SYSTEMS ........................................................................................ 650 INTRODUCTION During the 2016 presidential campaign, then-candidate Donald J. Trump announced his intention to impose a “Muslim Ban,” which would prohibit Muslim entry into the United States1 as part of his counterterrorism strategy. 1. See, e.g., Jeremy Diamond, Donald Trump: Ban All Muslim Travel to U.S., CNN (Dec. 8, 2015, 4:18 AM), []. Immigration, constitutional, and national security experts have offered perspectives on the ongoing legal challenges surrounding the Travel Ban. See generally Margaret Hu, Crimmigration-Counterterrorism and the Travel Ban, 2017 WIS. L. REV. (forthcoming) (citing Adam Cox, Why a Muslim Ban Is Likely to Be Held Unconstitutional: The Myth of Unconstrained Immigration Power, JUSTSECURITY (Jan. 2017] Shortly before his election, Trump also announced a proposal for the “extreme vetting” of immigrants and refugees.2 Trump clarified that “[t]he Muslim ban is something that in some form has morphed into a[n] extreme vetting [protocol] from certain areas of the world.”3 On January 27, 2017, during his first week as president, Trump signed Executive Order 13,769, titled “Protecting the Nation from Foreign Terrorist Entry into the United States” (the “January 27, 2017, Order”) .4 Litigation concerning the constitutionality of this Executive Order5 focused on sections 3 and 5(c), provisions that relate to barring the entry of travelers and refugees from specific Muslim-majority countries into the United States.6 These controversial provisions were challenged as violating equal protection, due process, and the Establishment Clause of the First Amendment, among other constitutional and statutory claims.7 On March 6, 2017, President Trump issued a revised Executive Order (the “March 6 2017, Order”) , Executive Order 13,780. Issued under the same title as the January 27, 2017, Order, “Protecting the Nation from Foreign Terrorist Entry into the United States,” the March 6, 2017, Order superseded the 30, 2017, 10:21 AM), []; then citing Mark Tushnet, Mootness and the Travel Ban, BALKINIZATION (June 2, 2017, 1:18 AM) []; then citing Marty Lederman, Unlocking the Mysteries of the Supreme Court’s Entry Ban Case, JUSTSECURITY (June 27, 2017, 8:01 PM), 42577/mysteries-trump-v-irap/ []; and then citing Leah Litman & Steve Vladeck, How the President’s “Clarifying” Memorandum Destroys the Case for the Entry Ban, JUSTSECURITY (June 15, 2017, 8:01 AM), 42166/presidents-clarifying-memorandum-destroys-case-entry-ban/ []). 2. Gerhard Peters & John T. Wooley, Presidential Debate at Washington University in St. Louis, Missouri, AM. PRESIDENCY PROJECT (Oct. 9, 2016), []; see also Peter Margulies, Bans, Borders, and Justice: Judicial Review of Immigration Law in the Trump Administration at 35–48 (Roger Williams Univ. Sch. of Law, Research Paper No. 177, 2017) , [] (arguing for a more searching judicial review of “extreme vetting” and the need to recognize the significant long-term impact of “extreme vetting”). 3. Peters & Wooley, supra note 2. 4. Exec. Order No. 13,769, 82 Fed. Reg. 8977 (Feb. 1, 2017) [hereinafter January 27, 2017, Order]. 5. See, e.g., Washington v. Trump, 847 F.3d 1151, 1157 (9th Cir. 2017) (per curiam). 6. January 27, 2017, Order, supra note 4, §§ 3, 5(c). 7. See, e.g., Hawaii v. Trump, 859 F.3d 741, 760 (9th Cir.) (per curiam) (alleging violations of the Establishment Clause, the Equal Protection and Due Process Clauses of the Fifth Amendment (both procedural and substantive claims), the Immigration and Nationality Act, the Religious Freedom Restoration Act, and the Administrative Procedure Act), cert. granted, 137 S. Ct. 2080 (2017); Int’l Refugee Assistance Project v. Trump, 857 F.3d 554, 578–79 (4th Cir.) (claiming violations of the Equal Protection Clause of the Fifth Amendment, the Immigration and Nationality Act, the Religious Freedom Restoration Act, the Refugee Act, and the Administrative Procedure Act), cert. granted, 137 S. Ct. 2080 (2017); Washington, 847 F.3d at 1157, 1165, 1167 (alleging that the Executive Order violates that First Amendment’s Establishment Clause, due process, and equal protection); Darweesh v. Trump, 17 Civ. 480 (AMD), 2017 WL 388504 (E.D.N.Y. Jan. 28, 2017) (alleging that the Executive Order violates the Equal Protection and Due Process Clauses). January 27, 2017, Order.8 However, it left the extreme vetting provisions of the January 27, 2017, Order in place,9 and, in fact, expanded the vetting requirements in several respects.10 The extreme vetting requirements of the March 6, 2017, Order are now most fully articulated in section 5, titled “Implementing Uniform Screening and Vetting Standards for All Immigration Programs.”11 The travel restrictions and the vetting requirements were expanded yet again in a third iteration of the “Muslim Ban,” also referred to as the “Travel Ban” or the “Entry Ban.” On September 24, 2017, shortly before oral argument was scheduled for the U.S. Supreme Court on October 10, 2017, in the consolidated Travel Ban cases of Trump v. Hawaii and Trump v. International Refugee Assistance Project,12 President Trump signed a new Proclamation (the “September 24, 2017, Order”) .13 The September 24, 2017, Order is titled, “Enhancing Vetting Capabilities and Processes for Detecting Attempted Entry into the United States by Terrorists or Other Public-Safety Threats.”14 Thus, the most recent Order, as implied by the title, focuses more squarely on the extreme vetting provisions set forth by the prior Orders. More specifically, sections 1(a) through (h) of the September 24, 2017, Order focus on “identity-management and information-sharing capabilities, protocols, and practices” related to immigration screening and vetting.15 The next day, the Court ordered briefing as to whether the Travel Ban cases that had been scheduled for oral argument on October 10, 2017, were moot.16 At the time of publication, the litigation remains ongoing, including challenges to the September 24, 2017, Order.17 2017] Regardless of the final disposition of these litigation matters, it is significant to note that the extreme vetting provisions of the original and revised Executive Orders have received less judicial attention than the travel restrictions.18 The extreme vetting provisions do not appear to be dependent upon the authority of the Orders, and are presented in the Orders as an evolving and prospective administrative matter.19 Thus, the vetting provisions of the March 6, 2017, Order and the September 24, 2017, Order may not be fully challenged.20 This Article focuses on the long-term impact of modern vetting requirements, such as those prescribed in the Executive Orders referenced above,21 and other immigration-related screening protocols that are increasingly algorithmically anchored. It contends that the implementation of expanded vetting protocols22 risks implications that may be undertheorized due to an underappreciation of the mass cybersurveillance and disparate-impact consequences that surround current screening measures broadly promulgated by the U.S. Department of Homeland Security (DHS). Specifically, this Article advances the claim that DHS vetting and screening protocols risk introducing an algorithmically driven and technologically enhanced form of Jim Crow.23 Unlike the “separate but equal”24 de jure discrimination25 of a historic Jim Crow regime, Algorithmic Jim Crow risks imposing both de jure and de facto discrimination26 through an “equal but separate”27 regime. This Article explains how Algorithmic Jim Crow is an outgrowth of a digital era that 23. See generally JONATHAN SCOTT HOLLOWAY, JIM CROW WISDOM: MEMORY AND IDENTITY IN BLACK AMERICA SINCE 1940 (2013); JUMPIN’ JIM CROW: SOUTHERN POLITICS FROM CIVIL WAR TO CIVIL RIGHTS (Jane Dailey et al. eds., 2000); Mattias Smångs, Doing Violence, Making Race: Southern Lynching and White Racial Group Formation, 121 AM. J. SOC. 1329 (2016). For contemporary discussions on the complexity of what has been termed a “post-racial” America, see generally DEVON W. CARBADO & MITU GULATI, ACTING WHITE? RETHINKING RACE IN “POST-RACIAL” AMERICA (2013); THE NEW BLACK: WHAT HAS CHANGED—AND WHAT HAS NOT—WITH RACE IN AMERICA (Kenneth W. Mack & Guy-Uriel E. Charles eds., 2013) ; Charlton McIlwain, Racial Formation, Inequality and the Political Economy of Web Traffic, 20 INFO. COMM. & SOC’Y 1073 (2016); Angela Onwuachi-Willig, Policing the Boundaries of Whiteness: The Tragedy of Being “Out of Place” from Emmett Till to Trayvon Martin, 102 IOWA L. REV. 1113 (2017); Camille Gear Rich, Marginal Whiteness, 98 CALIF. L. REV. 1497 (2010); infra Part I.A. 24. Plessy v. Ferguson, 163 U.S. 537, 550–51 (1896), overruled by Brown v. Bd. of Educ., 347 U.S. 483 (1954) (“[W]e cannot say that a law which authorizes or even requires the separation of the two races in public conveyances is unreasonable . . . .”). 25. The Supreme Court has characterized de jure discrimination as encompassing statesanctioned or state-imposed discrimination under the law, prohibited under the Equal Protection Clause. See, e.g., Brown, 347 U.S. at 490 (explaining that the Fourteenth Amendment “proscrib[es] all state-imposed discriminations against the Negro race”). 26. Nonracial classifications that result in de facto discrimination or disparate-impact discrimination may not be found to violate the Equal Protection Clause. See, e.g., Village of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 270 (1977) (explaining that the party asserting an equal protection violation bears the burden to show that the governmental action was intended to discriminate against a suspect or protected class); Milliken v. Bradley, 418 U.S. 717, 745 (1974) (distinguishing de jure and de facto segregation with express and explicit policies that articulate race-based distinctions defined as de jure discrimination); see also Frank I. Goodman, De Facto School Segregation: A Constitutional and Empirical Analysis, 60 CALIF. L. REV. 275, 275 (1972); Richard A. Primus, Equal Protection and Disparate Impact: Round Three, 117 HARV. L. REV. 494, 496–97 (2003). 27. See infra Part III.A. 2017] exploits cybersurveillance28 and dataveillance29 systems that are rapidly proliferating in both the public30 and private sectors.31 This Article demonstrates how immigration-related vetting and database screening protocols utilize newly developed big data32 screening, tracking, and profiling tools that attempt to verify identity and assess future risk.33 These tools are now actively deployed by DHS34 and utilize databases operated by the military35 and intelligence communities.36 Currently, vetting and screening protocols often begin with biometric identification37—for example, the digital collection and screening of scanned fingerprints through federal and state biometric databases in the United States and international biometric databases, such as those operated by ICPO-INTERPOL (Interpol).38 Biometric data currently collected by DHS include scanned fingerprints39 and irises,40 digital photos for facial recognition technology,41 and DNA.42 Consequently, implementation of extreme vetting protocols will likely include proposals for a tamper-resistant and fraud-proof biometric electronic identity card,43 such as a biometric ePassport.44 The Trump administration’s Executive Orders, for example, specifically mandate “Expedited Completion of the Biometric Entry-Exit Tracking System” by DHS.45 As part of new 2017] vetting protocols, DHS also seeks social media identification data46 and plans to seek social media user credentials,47 such as passwords to Facebook accounts of refugees and visa applicants.48 Newly developed “big data” cybersurveillance tools fuse biometric data with biographic data and internet and social media profiling to assess risk.49 This Article aims to explain how big data vetting is mistakenly presented as a procedure that is restricted to noncitizens: immigrants, refugee and asylum applicants, and visitors seeking a travel visa to the United States. Instead, such vetting is part of a web of technologies that DHS has termed “identity management.”50 The application of these technologies may eventually extend to the entire citizenry through a variety of policy proposals, including a biometric national identification system, and various mandatory vetting and database screening programs. Identity-management programs attempt to authenticate identity and assess the risk factors across entire populations, including the U.S. citizenry. Big data vetting, thus, is misunderstood as a protocol that is likely to be limited to immigration-related screening. More accurately, such vetting includes an evolving form of big data surveillance that attempts to assess criminal and terroristic risk across entire populations and subpopulations through mass data collection, database screening and data fusion, artificial intelligence, and algorithm-driven predictive analytics.51 The long-term consequences of modern big data surveillance can be better envisioned by anticipating how and why big data vetting protocols may be extended to the entire population. Eventually, all residents of the United States, both citizens and noncitizens, may face various stages of technological vetting and algorithmic screening as a part of a post-September 11, 2001, national security policy trajectory that embraces big data surveillance for its presumed efficacy. Importantly, in parallel with the extreme vetting protocols mandated by the Executive Orders, almost every immigration reform effort since 9/11 has called for biometric data collection from the entire citizenry in the United States to enhance border security efforts.52 At the same time, increasing concern regarding homegrown terrorism has resulted in a call to extend domestic surveillance and counterterrorism efforts to both citizens and noncitizens.53 The Snowden disclosures, for example, have further revealed how foreign-intelligencegathering tools, such as bulk metadata collection, can be indiscriminate in scope and impact both citizens and noncitizens.54 Identifying the vetting procedures embedded within Executive Order 13,769 and the constitutional challenges which followed its promulgation is particularly appropriate as 2017 marks the seventy-fifth anniversary of the signing of Executive Order 9066 by President Franklin Delano Roosevelt.55 That order, issued on February 19, 1942, and titled “Authorizing the Secretary of War to Prescribe Military Areas,” allowed for Japanese internment by delegating to the Secretary of War the authority “to take such other steps as he . . . may deem advisable to enforce compliance” with the exclusion of Japanese Americans and those of Japanese ancestry.56 The legal challenges mounted against Executive Order 9066 culminated in several U.S. Supreme Court cases, most notably, Korematsu v. United States.57 In this case, decided in 1944, the Court upheld the constitutionality to facilitate immigration officials in both discovering targets and then creating and administering cases against them”). 52. See Margaret Hu, Biometric ID Cybersurveillance, 88 IND. L.J. 1475, 1478–82 (2013). 53. See, e.g., Countering Violent Extremism, DEP’T HOMELAND SECURITY (Jan. 19, 2017), []; see also TREVOR AARONSON, THE TERROR FACTORY: INSIDE THE FBI’S MANUFACTURED WAR ON TERRORISM 19 (2013); Colin Moynihan, A New York City Settlement on the Surveillance of Muslims, NEW YORKER (Jan. 7, 2016), [] (“After the attacks of September 11, 2001, the New York Police Department began an intense surveillance operation that focused on Muslims in New York City . . . . They eavesdropped on conversations in restaurants and cafes, catalogued memberships in mosques and student organizations, and . . . tried to bait people into making inflammatory statements.”) . 54. See, e.g., Laura K. Donohue, Bulk Metadata Collection: Statutory and Constitutional Considerations, 37 HARV. J.L. & PUB. POL’Y 757, 863– 64 (2014 ); Laura K. Donohue, Section 702 and the Collection of International Telephone and Internet Content, 38 HARV. J.L. & PUB. POL’Y 117, 151–52, 157, 164 n.83, 202–19 (2015). 55. Exec. Order No. 9066, 3 C.F.R. §§ 1092–93 (1942). 56. Id. 57. 323 U.S. 214 (1944); see also Hirabayashi v. United States, 320 U.S. 81, 100 (1943) (“Distinctions between citizens solely because of their ancestry are by their very nature odious to a free people whose institutions are founded upon the doctrine of equality.”). 2017] of Executive Order 9066, reasoning in part: “[W]e are dealing specifically with nothing but an exclusion order. To cast this case into outlines of racial prejudice, without reference to the real military dangers which were presented, merely confuses the issue.”58 Drawing comparisons between Executive Order 13,769 and Executive Order 9066, and reviewing the original justification for Japanese internment, is critical here as President Trump and others have cited both FDR’s actions59 and Korematsu as precedent for the Muslim Ban and the development of a Muslim registry database.60 This Article proceeds in four parts. Part I describes how modern vetting procedures are intertwined with burgeoning identity-management systems. Based on a review of publicly available information, these vetting protocols are increasingly dependent upon the following: mass biometric data collection, automated or semiautomated biometric identification, and algorithm-dependent database screening programs. In a big data world, threat risk assessments and data-profiling tools do not necessarily begin with the identification of individuals on the basis of traditionally protected classifications, such as race, ethnicity, or national origin. Instead, because contemporary big data systems are data-classification oriented, vetting and screening protocols begin with the identification of individuals on the basis of numerical identification, such as passport numbers, and on the basis of biometric identification, such as the cataloguing of scanned fingerprints and irises. Part II describes how national security programs risk creating forms of discrimination similar to Jim Crow in that they are also based upon classification and screening protocols. Historic Jim Crow regimes started with a legal premise: that certain individuals could and should be classified on the basis of race. Next, Jim Crow laws utilized screening systems to enforce segregation based upon designated racial classification. This discussion explores why security threat assessments produced through algorithms and database screening may—similar to historic Jim Crow—also separate populations based upon particular classifications. New Algorithmic Jim Crow systems, like historic Jim Crow regimes, systems may present themselves as facially neutral. 58. Korematsu, 323 U.S. at 223. 59. “What I’m doing is no different than FDR,” Trump told ABC News during the presidential campaign. Meghan Keneally, Donald Trump Cites These FDR Policies to Defend Muslim Ban, ABC NEWS (Dec. 8, 2015, 1:01 PM), []. 60. Shortly after President Trump’s election, Carl Higbie, a former spokesman for the Great America Political Action Committee, stated on Fox News that a Muslim database registry would be legal and would “hold constitutional muster” under Korematsu, explaining, “We did it during World War II with the Japanese . . . .” Derek Hawkins, Japanese American Internment Is ‘Precedent’ for National Muslim Registry, Prominent Trump Backer Says, WASH. POST (Nov. 17, 2016), []. Part III explains how identity-management systems do not necessarily discriminate based on protected categories recognized under equal protection jurisprudence.61 Rather, newly emerging vetting systems are often centered on big data and generally driven by mass data collection and analysis. These systems, for instance, purport to be race neutral and not to target individuals based on a protected classification. Rather, it is often the case that results of data screening and vetting analytics deemed suspect and anomalous are isolated and targeted. Consequently, the “equal but separate” consequences of Algorithmic Jim Crow will allow for the “equal” vetting and screening of all citizens and noncitizens. At the same time, newly deployed vetting systems will allow federal and state governments to “separate” individuals based upon the vetting and screening actions mandated through security policy developments. Part IV further discusses why advocates of immigration federalism62 and national security federalism63—those seeking the expansion of state participation in the enforcement of federal immigration and national security law—have increasingly enacted biometric data harvesting and identitymanagement laws that mimic federal laws and programs. These state laws mandate the utilization of multiple dataveillance tools and biometric data screening devices, purportedly to further crime and immigration control and simultaneously support counterterrorism efforts. Yet, just as historic Jim Crow regimes delegated segregationist gatekeeping duties to state and private entities, contemporary immigration policy delegates restrictive immigration gatekeeping duties to state and private entities. Under Algorithmic Jim 61. Equal protection jurisprudence and the foundations for differing standards of judicial review based upon protected classification has yielded rich scholarship. See generally JOHN HART ELY, DEMOCRACY AND DISTRUST: A THEORY OF JUDICIAL REVIEW 105–81 (1980); Bruce A. Ackerman, Beyond Carolene Products, 98 HARV. L. REV. 713, 714–16 (1985); Mario L. Barnes et al., A Post-Race Equal Protection?, 98 GEO. L.J. 967 (2010); Katharine T. Bartlett, Tradition as Past and Present in Substantive Due Process Analysis, 62 DUKE L.J. 535, 540– 48 (2012); Robert M. Cover, The Origins of Judicial Activism in the Protection of Minorities, 91 YALE L.J. 1287, 1294–97 (1982); Trina Jones, Shades of Brown: The Law of Skin Color, 49 DUKE L.J. 1487 (2000); Michael Klarman, An Interpretive History of Modern Equal Protection, 90 MICH. L. REV. 213, 219 (1991); Melissa Murray, Equal Rites and Equal Rights, 96 CALIF. L. REV. 1395 (2008); Bertrall L. Ross II, The Representative Equality Principle: Disaggregating the Equal Protection Intent Standard, 81 FORDHAM L. REV. 175 (2012); Jed Rubenfeld, The Anti-Antidiscrimination Agenda, 111 YALE L.J. 1141, 1143 (2002); Kenji Yoshino, The New Equal Protection, 124 HARV. L. REV. 747, 748, 755–63 (2011). 62. Hiroshi Motomura is credited with introducing the term “immigration federalism” into academic discourse to describe state and local involvement in immigration. Peter J. Spiro, Learning to Live with Immigration Federalism, 29 CONN. L. REV. 1627, 1627 (1997); see also Pratheepan Gulasekaram & S. Karthick Ramakrishnan, Immigration Federalism: A Reappraisal, 88 N.Y.U. L. REV. 2074, 2096 (2013); Clare Huntington, The Constitutional Dimension of Immigration Federalism, 61 VAND. L. REV. 787, 788 n.6 (2008) (crediting Motomura with “defining immigration federalism as ‘states and localities play[ing a role] in making and implementing law and policy relating to immigration and immigrants’” (citing Hiroshi Motomura, Federalism, International Human Rights, and Immigration Exceptionalism, 70 U. COLO. L. REV. 1361, 1361 (1999))). 63. See, e.g., Matthew C. Waxman, National Security Federalism in the Age of Terror, 64 STAN. L. REV. 289, 289 (2012); see also Kris W. Kobach, Reinforcing the Rule of Law: What States Can and Should Do to Reduce Illegal Immigration, 22 GEO. IMMIGR. L.J. 459, 475 (2008). 2017] Crow, these technologically enabled gatekeeping duties involve race-neutral database screening of personally identifiable data and biometric data through federal vetting and screening protocols. The results, however, may not be race neutral, or may in fact have a disparate impact on traditionally protected classes. Part IV further explains how technological vetting protocols and algorithm-driven database screening systems may be insulated from successful legal challenges, as the law has not yet adapted to anticipate new forms of back-end discrimination facilitated by DHS’s rapid deployment of identity-management programs. The government, as in Korematsu, will likely defend any disparate-impact consequences as necessary and justified based upon threat risk assessments and nonracial classifications. Risk-based classifications and data characteristics deemed suspect fall outside of the protections recognized by current equal protection jurisprudence. This type of disparate impact, driven by database screening and technologically enhanced discrimination, may face limited or lenient review by a federal judiciary that generally grants broad deference to the government in matters of immigration and national security.64 Thus, the advent of Algorithmic Jim Crow will require an evolution of equality law. This Article concludes that current algorithm-driven vetting and biometricbiographic identification screening, especially once deployed across the entire citizenry, will likely lead to discriminatory profiling and surveillance on the basis of suspicious data as well as classification-based discrimination. These vetting and screening systems are likely to result in both direct and disparate discrimination, particularly based on race, color, ethnicity, national origin, and religion. In addition, recent immigration-control policy and programs demonstrate the government’s interest in delegating immigrationvetting duties to private actors, such as employers, and nonfederal actors, such as state and local law enforcement, which can exacerbate issues of racial profiling and discrimination. I. BIRTH OF ALGORITHMIC JIM CROW When President Trump signed Executive Order 13,769 on January 27, 2017, then-acting Attorney General Sally Yates was taken by surprise.65 Yates reviewed the Order as well as a number of briefs by individuals who sought to enjoin the Order in federal court and believed that it raised constitutional problems—namely Establishment Clause and due process concerns.66 Yates later explained to the New Yorker that, after reviewing arguments for and against the first Order, she thought that her two choices were either to resign or to refuse to defend the Order.67 After reviewing the evidence, Yates believed that the Order was ultimately based on religion and said, “I thought back to Jim Crow laws, or literacy tests. Those didn’t say that the purpose was to prevent African-Americans from voting. But that’s what the purpose was.”68 Yates drafted a letter to her colleagues at the U.S. Department of Justice, in which she stated: “At present, I am not convinced that the defense of the Executive Order is consistent with these responsibilities nor am I convinced that the Executive Order is lawful.”69 Yates directed Department of Justice attorneys not to defend the Order until she determined that it was appropriate to act.70 A few hours later, Yates received a letter from the White House that informed her that she had been fired.71 Yates’s invocation of Jim Crow deserves notice. The former acting Attorney General concluded that the Executive Order might lead to a disparate impact on the basis of protected classifications such as national origin and religion. At the same time, she also recognized that the Executive Order presented itself as facially neutral, much like the facially neutral literacy tests promulgated under Jim Crow laws that disproportionately burdened the voting rights of minority communities. Under historic Jim Crow, literacy tests, poll taxes, and other obstacles to voting rights were equally applied to all voters.72 Although these obstacles did not explicitly inquire into voters’ race, they nonetheless significantly disenfranchised minority communities. Therefore, they served discriminatory ends even though the race of the voter was never technically a basis for denying access to the ballot.73 Much like literacy tests and poll taxes, post-9/11 security initiatives may disproportionately impact minority communities even though they do not explicitly effectuate decisions based on protected attributes. An inquiry into modern-day screening and vetting systems depends upon in an understanding of myriad post-9/11 national security programs and policy initiatives. Contemporary screening and vetting systems utilize algorithms to determine a wide range of questions, including identity and associational assessments, to gauge risk. For example, extreme vetting systems like the one promulgated by the Executive Orders may bring about disproportionate 2017] burdens on minority communities. Potential discrimination facilitated or exacerbated by technological means appearing to be facially neutral may evade legal challenge requiring careful inquiry. Because big data screening and tracking systems unfold in ways that are difficult to see—for example, through algorithm-driven determinations, internet-based database screening programs, and social media monitoring— it is critical to explore how modern vetting protocols may be linked to preexisting post-9/11 identity-management systems that are dependent upon cybersurveillance and dataveillance tools. To grasp how extreme vetting can be extended to the entire citizenry, it is helpful to compare No Fly database screening systems with potential extreme vetting database screening systems. For example, based on what is known of both programs, it appears that many of the database screening protocols overlap.74 Part I, therefore, explains how vetting systems will increasingly rely upon database screening, including universal biometric databases, to sweep entire populations and subsets within populations to assess terroristic and criminal risk. To better understand the Trump administration’s policy on extreme vetting, it is important to reconstruct the justification for such a policy based upon historical background and prior policy developments implemented during the Obama administration. Many of the policies advanced by the Executive Order are not only an outgrowth of 9/11, but they are specifically reactive to the Paris attacks in November 2015 and the San Bernardino attack in December 2015. Both terrorist events led to multiple immigration policy proposals and adjustments to current vetting procedures. On the evening of November 13, 2015, coordinated terrorist attacks were staged in Paris, France, which included mass shootings, suicide bombings, and hostage takings.75 According to news reports, the terrorist attacks left 129 dead and 352 wounded, including ninety-nine in serious condition.76 The terrorist group Islamic State in Iraq and Syria (ISIS) immediately claimed responsibility.77 According to news accounts, ISIS announced immediately after the Paris attacks that three teams of eight terrorists had carried them out.78 Seven of the terrorists were reportedly killed through self-detonated suicide bombs.79 In the days following the attacks, intelligence reports indicated that at least three of the eight terrorists had used falsified passports.80 A passport found on the body of a terrorist who had died at the Stade de France (“National Stadium”) was reported to be an illegitimate Syrian passport and it was reported that the terrorist had allegedly claimed Syrian refugee status in France.81 Two other terrorists killed at the National Stadium allegedly carried false Turkish passports.82 By November 19, 2015, just six days after the attacks, the governors of thirty-one states had announced their refusal to admit or resettle Syrian refugees in their respective states.83 One week after the Paris attacks, Michael Ignatieff, formerly the Edward R. Murrow professor of public policy at the Harvard Kennedy School and currently the rector and president of Central European University in Budapest, expressed a position widely held by many experts: an international biometric identification system would help to address the refugee crisis in Europe and simultaneously serve national security interests.84 He stated that “[t]he world badly needs a new migratory regime—based around an internationally authorized biometric ID card, with a date of permitted entry and a mandatory exit.”85 On November 24, 2015, while standing next to then-President François Hollande of France less than two weeks after the Paris attacks, then-President Barack Obama announced that “the [U.S.] government was developing ‘biometric information and other technologies that can make [refugee identification] more accurate.’”86 On December 2, 2015, fourteen people were killed and twenty-one were seriously injured in a terrorist attack in San Bernardino, California.87 The attack consisted of a mass shooting and an attempted bombing.88 The perpetrators, Syed Rizwan Farook and Tashfeen Malik, a married couple residing in California, targeted a San Bernardino County Department of 2017] Public Health holiday party.89 Farook, a Pakistani American U.S. citizen born in Illinois, was employed by the Department of Public Health.90 Malik was a Pakistani-born lawful permanent resident who had recently migrated to the United States.91 According to media accounts, the Muslim couple had been self-radicalized, inspired by ISIS.92 On the same day as the San Bernardino attack, Paul Ryan, Speaker of the U.S. House of Representatives, explained that lawmakers were considering various legislative reforms to increase security vetting of refugees and immigrants in response to the threat of ISIS, including requiring countries “to issue smart e-passports with biometric chips.”93 Less than one week later, in the immediate aftermath of the Paris and San Bernardino attacks, then-presidential candidate Trump called for what has been referred to as a “Muslim Ban” or “Travel Ban”: an executive action that would prohibit entry of any Muslim into the United States.94 Trump did not provide specifics on how he would prohibit Muslims from entering the United States; however, he later clarified that the ban would be temporary to allow the government to assess its current immigration procedures and “suspend immigration from regions linked with terrorism.”95 Trump’s call for a “Muslim Ban” was followed by calls for surveillance of Muslim communities and mosques.96 Six months later, on June 12, 2016, Omar Mateen, an Afghani American born in the United States, killed forty-nine people and wounded fifty-three in a shooting rampage at the Pulse nightclub in Orlando, Florida.97 According to news reports, Mateen had proclaimed allegiance to ISIS shortly before committing “the worst mass shooting in United States history.”98 In the wake of the Orlando attack, then-candidate Trump explained that the profiling of Muslims in the United States was necessary as a preemptive measure to prevent future attacks.99 On August 16, 2016, Trump announced a proposal for the “extreme vetting” of immigrants and refugees in a campaign speech on ISIS.100 He explained: “In the Cold War, we had an ideological screening test. The time is long overdue to develop a new screening test for the threats we face today. I call it extreme vetting. I call it extreme, extreme vetting.”101 Then, on August 31, 2016, Trump delivered an address on his proposed immigration policy.102 In addition to his promise to build a wall on the southern border of the United States and his reassertion that “Mexico will pay for the wall,” Trump explained that he would also implement a “biometric entry-exit system for tracking visa-holders.”103 These promises of enhanced national security and border security systems, as well as the various Executive Orders restricting travel, inherently represent technological developments in the promulgation of emerging cybersurveillance technologies and algorithmicdriven screening and vetting protocols. To draw parallels between historic Jim Crow and Algorithmic Jim Crow, this Article turns to an overview of Jim Crow. II. OVERVIEW OF JIM CROW: CLASSIFICATION AND SCREENING SYSTEMS Artificial intelligence and algorithms are not usually perceived as resulting in discrimination. In fact, they may appear to be equality-compliant or even equality-enhancing in that algorithmic screening and vetting can be applied equally across entire populations and subpopulations. Screening and classification systems, however, even when facially neutral and algorithmically based, can lead to profound constitutional challenges. The historical framing in this section is necessary to assist in the interrogation of new classification and screening systems that are flourishing under security rationales and presented as technologically objective and colorblind. Therefore, to better grasp why the framing of Algorithmic Jim Crow is now needed, Part II lays a factual predicate to explain the foundational legal premises for historic Jim Crow regimes. Traditional Jim Crow laws first required the government—often state and county governments—to engage in formal and standardized protocols to assign racial classification to citizens of that state or county. Once a racial classification system was determined under the law, screening protocols, also established under Jim Crow laws, enabled the separation of populations on the basis of that racial classification. Consequently, understanding the basic mechanics of how separation was enforced under the law begins with an understanding of historic Jim Crow classification and screening systems. 2017] A. Historical Framing of Jim Crow This overview is not intended to be an exhaustive history of the legal issues and the nature of Jim Crow—other scholarship has addressed that subject in thoughtful detail.104 Instead, this background intends to sketch out an understanding of the scope and context of the Jim Crow era and to further clarify that its laws and policies were not confined merely to segregating locational sites and imposing restrictions on movement. Rather, Jim Crow was: [A] structure of exclusion and discrimination devised by white Americans to be employed principally against black Americans . . . . Its central purpose was to maintain a second-class social and economic status for blacks while upholding a first-class social and economic status for whites. . . . In the South, Jim Crow discrimination at its height existed not only by statute but by custom and racial ‘etiquette,’ and it was rigidly enforced by both the law enforcement agencies and courts as well as by ordinary white citizens who were neither policemen nor judges but who often took the law into their own hands as though they were.105 104. See generally 1 RACE, LAW, AND AMERICAN HISTORY 1700–1990: AFRICAN AMERICANS AND THE LAW (Paul Finkelman ed., 1992); FRANK J. SCATURRO, THE SUPREME COURT’S RETREAT FROM RECONSTRUCTION: A DISTORTION OF CONSTITUTIONAL JURISPRUDENCE (2000); STEPHEN L. WASBY ET AL., DESEGREGATION FROM BROWN TO ALEXANDER (1977); C. VANN WOODWARD, THE STRANGE CAREER OF JIM CROW (3d rev. ed. 2002); Gabriel J. Chin, Jim Crow’s Long Goodbye, 21 CONST. COMMENT. 107 (2004); Gabriel J. Chin & Randy Wagner, The Tyranny of the Minority: Jim Crow and the CounterMajoritarian Difficulty, 43 HARV. C.R.-C.L.L. REV. 65 (2008); James W. Fox, Jr., Doctrinal Myths and the Management of Cognitive Dissonance: Race, Law, and the Supreme Court’s Doctrinal Support of Jim Crow, 34 STETSON L. REV. 293 (2005); James W. Fox, Jr., Intimations of Citizenship: Repressions and Expressions of Equal Citizenship in the Era of Jim Crow, 50 HOW. L.J. 113 (2006); Rachel D. Godsil, Race Nuisance: The Politics of Law in the Jim Crow Era, 105 MICH. L. REV. 505 (2006); Ariela J. Gross, Litigating Whiteness: Trials of Racial Determination in the Nineteenth-Century South, 108 YALE L.J. 109 (1998); Trina Jones, Brown II: A Case of Missed Opportunity?, 24 L. & INEQ. 9 (2006); José Roberto Juárez, Jr., Recovering Texas History: Tejanos, Jim Crow, Lynchings & the University of Texas School of Law, 52 S. TEX. L. REV. 85 (2010); Kenneth W. Mack, Foreword: A Short Biography of the Civil Rights Act of 1964, 67 SMU L. REV. 229 (2014); Kenneth W. Mack, Law, Society, Identity, and the Making of the Jim Crow South: Travel and Segregation on Tennessee Railroads, 1875–1905, 24 L. & SOC. INQUIRY 377 (1999) [hereinafter Mack, Law, Society, Identity]; Kenneth W. Mack, Rethinking Civil Rights Lawyering and Politics in the Era Before Brown, 115 YALE L.J. 256 (2005); David Martin, The Birth of Jim Crow in Alabama 1865–1896, 13 NAT’L BLACK L.J. 184 (1993); Jennifer Roback, Southern Labor Law in the Jim Crow Era: Exploitative or Competitive, 51 U. CHI. L. REV. 1161 (1984); Benno C. Schmidt, Jr., Principle and Prejudice: The Supreme Court and Race in the Progressive Era. Part 1: The Heyday of Jim Crow, 82 COLUM. L. REV. 444 (1982); Barbara Y. Welke, Beyond Plessy: Space, Status, and Race in the Era of Jim Crow, 2000 UTAH L. REV. 267; John W. Wertheimer et al., “The Law Recognizes Racial Instinct”: Tucker v. Blease and the BlackWhite Paradigm in the Jim Crow South, 29 L. & HIST. REV. 471 (2011); Joseph R. Palmore, Note, The Not-So-Strange Career of Interstate Jim Crow: Race, Transportation, and the Dormant Commerce Clause, 1878–1946, 83 VA. L. REV. 1773 (1997); Anders Walker, Jim Crow’s Unwritten Code, JOTWELL (Jan. 16, 2017 ), []. 105. JERROLD M. PACKARD, AMERICAN NIGHTMARE: THE HISTORY OF JIM CROW vii–viii (2002). For other sources on the history and impact of Jim Crow, see generally F. MICHAEL HIGGINBOTHAM, GHOSTS OF JIM CROW: ENDING RACISM IN POST-RACIAL AMERICA (2013); 2017] identities and ‘ensuring the security” of the Jet Propulsion Laboratory at NASA.270 In a footnote, the Court noted a compelling question raised by the scientists that had been dismissed by the Ninth Circuit as unripe and had not been made the subject of a cross-petition: a question of the so-called “suitability” criteria that the government used to determine employment eligibility at Jet Propulsion Laboratory.271 These factors include consideration of a candidate’s financial and emotional health as well as things like “carnal knowledge.”272 The “suitability” criteria were derived from a ninety-fourpage government vetting protocol document titled, “NASA Desk Guide for Suitability and Security Clearance Processing, Version 2.”273 Specifically, understanding the scientists’ constitutional informational privacy claim requires an understanding of the morality- and charactertesting criteria that were open for questioning and evaluation during the background-check process required under NASA’s implementation of HSPD-12. On page sixty-five of the desk guide, NASA includes an “Issue Characterization Chart” that allows NASA to assess individuals’ character and “suitability” based on more than 100 itemized characteristics.274 These items appeared to assess good moral character and trustworthiness. Characteristics on the evaluation include: “[d]runk”; “[b]ad check”; “[p]attern of irresponsibility as reflected in credit history”; “[c]arnal knowledge”; “indecent proposal”; “sodomy”; “voyeurism [or] peeping tom”; “[m]ailing, selling, or displaying obscene material”; “[b]eastiality”; “[p]attern of excessive [substance abuse] as reflected in inability to function responsibly [and] medical treatment or poor health”; “[d]isorderly conduct”; “[a]ttitude [and] [p]ersonality [c]onflict”; “[t]respassing”; and “[m]inor traffic violation.”275 Upon successful completion of the NASA “Suitability and Security Clearance Processing” protocol, the desk guide authorizes the agency to issue the NASA employee or private contractor a biometric ID card in accordance with HSPD-12.276 Failure to pass this newly implemented clearance process results in the termination of employment.277 While the “suitability” criteria were not before the Court, the acting solicitor general nevertheless felt compelled to assert at oral argument that “NASA will not and does not use” such objectionable criteria “to make contractor credentialing decisions.”278 The need for such assurance indicates the scientists did indeed have real privacy concerns, even if they did not crystallize into part of a live claim before the Court. Thus, lost in this case was a real concern about what information NASA, or any other government agency, is allowed to seek under the identityverification procedures imposed by HSPD-12 and whether the constitution, under a privacy right, imposes any fundamental limiting principles on that identity-verification process. The acting solicitor general’s assurance that intimate personal details regarding credit card debt and carnal knowledge, for example, will not be considered by NASA is nothing more than that—just an assurance. Meanwhile the “suitability” criteria that could be used to determine the denial of the issuance of a biometric ID card under HSPD-12 makes clear that fears that government identity-management programs may become overbroad and overintrusive are not paranoid or baseless. Nelson, by affirming HSPD-12, may now pave the way for the implementation of a biometric credentialing program and uniform biometricbased dataveillance program on a national scale. Nelson also demonstrates how suitable character testing or morality testing can be built into modern vetting protocols in civilian background checks, as the facts of the case demonstrated that NASA employees and contractors were required to demonstrate trustworthiness and good character before receipt of the biometric identification card. Under a universal biometric identification system, however, suitability testing or character-vetting protocols could be embedded within the database screening system itself. Thus, the morality testing would not necessarily arrive at the front end of the vetting process, as was seen in Nelson. Rather, the accumulation of biometric and biographic data enables both biometric and suitability testing. Rather than clearing a suitability assessment in order to qualify for a biometric ID card, a biometric-anchored database screening system could allow for moral- and suitability-criteria testing on the back end of the vetting process. Recent disclosures by Edward Snowden, for example, explain how biometric data can be fused with biographic data to assess risk.279 This development in Supreme Court jurisprudence is, thus, significant because the original announcement of Trump’s “Muslim Ban” indicated that the proposal was inclusive of U.S. citizens. Specifically, on December 8, 2015, shortly after then-candidate Trump announced plans for the Muslim 278. Nelson, 562 U.S. at 143 n.5. 279. See James Risen & Laura Poitras, N.S.A. Collecting Millions of Faces from Web Images, N.Y. TIMES (May 31, 2014), [] (explaining that biometric data can be combined with “two dozen data points” that include DHS databases and other federal databases, such as “Transportation Security Administration No Fly List, [a person’s] passport and visa status, known associates or suspected terrorist ties, and comments made about [an individual] by informants to American intelligence agencies”). 2017] travel ban, he suggested in a nationally televised interview280 that the ban could possibly extend to Muslim Americans.281 Trump invoked former President Franklin D. Roosevelt’s World War II proclamation that U.S. citizens who were potentially “enemy aliens” could be detained.282 On January 28, 2017, one day after President Trump signed the Executive Order imposing restrictions on the travel and immigration of citizens of Muslim-majority nations, Fox News asked former New York City Mayor Rudy Giuliani whether the Executive Order was, in fact, a Muslim ban.283 Guiliani explained: “[W]hen [Trump] first announced it [during the campaign], he said, ‘Muslim ban.’ He called me up. He said, ‘Put a commission together. Show me the right way to do it legally.’”284 Giuliani elaborated further: “And what we did was, we focused on, instead of religion, danger . . . . Perfectly legal, perfectly sensible. And that’s what the ban is based on. It’s not based on religion. It’s based on places where there are [sic] substantial evidence that people are sending terrorists into our country.”285 This suggests that discriminatory vetting and screening protocols can evade judicial review if a protected class is targeted indirectly through “race-neutral” criteria, such as threat risk assessments. In Washington v. Trump,286 litigation that addressed the first Executive Order, the Ninth Circuit disagreed that the government had established sufficient evidence of an impending national security threat.287 On February 9, 2017, in upholding the Western District of Washington’s grant of a temporary restraining order, halting the implementation of the Executive Order, the Ninth Circuit concluded: “[T]he Government has not offered any evidence or even an explanation of how the national security concerns that justified those designations [of travel and immigration restrictions], which triggered visa requirements, can be extrapolated to justify an urgent need for the Executive Order to be immediately reinstated.”288 In subsequent litigation, the Fourth and Ninth Circuits agreed that the Government had failed to show a sufficient justification for the Executive Order. In Hawaii v. Trump,289 the Ninth Circuit panel explained that the president had not made a “sufficient finding . . . that entry of the excluded classes would be detrimental to . . . the United States.”290 In International Refugee Assistance Project, the Fourth Circuit noted that while the government argued that it had a national security purpose in issuing the Order, evidence supporting such a purpose was “comparably weak[er]” than then-candidate Trump’s statements about a Muslim ban, subsequent statements on the issues, statements made by his advisors, as well as the issuance of and statements made by President Trump and his advisors regarding the second Executive Order.291 At the time this Article was written, the Supreme Court had granted certiorari in both cases and consolidated them for argument.292 On February 24, 2017, the Associated Press reported that a leaked memo drafted at the request of the DHS acting Under Secretary for Intelligence and Analysis concluded “citizenship is an unlikely indicator of terrorism threats to the United States.”293 The memo states that the analysis undertaken by DHS specifically analyzed the threat of the “seven countries [that were] impacted by [section 3 of Executive Order] 13769.”294 The DHS memo states that “of [eighty-two] people the government determined were inspired by a foreign terrorist group to carry out or try to carry out an attack in the United States [since the Syrian conflict commenced in March 2011], just over half were U.S. citizens born in the United States.”295 The DHS memo further states that the terrorists were from “[twenty-six] countries, led by Pakistan, Somalia, Bangladesh, Cuba, Ethiopia, Iraq and Uzbekistan. Of these, only Somalia and Iraq were among the seven nations included in the ban.”296 Both the Ninth and Fourth Circuits discussed this memorandum and relied on it in their rulings.297 Importantly, the original January 27, 2017, Order states that vetting policy should include a test to assess fidelity to founding principles and the Constitution.298 Statements by Trump suggest that such vetting should 2017] include a test to assess loyalty to the United States and whether an individual will “support our country, and love deeply our people.”299 He further promoted, as a candidate, the implementation of profiling and preventative measures, such as mass surveillance, to assess terroristic risk.300 In Nelson, twin innovations in national security policy and biometric surveillance policy included a machine-readable biometric ID card encoded with digitalized biometric data and other personally identifiable data, as was required by the HSPD-12 program. In the suitability criteria developed by NASA, a version of extreme vetting emerged. The January 27, 2017, Order discusses the need to implement loyalty tests that demonstrate “proAmerican” values. Thus, extreme vetting may be expanded to encompass similar abstract assessments of character and morality as a part of threat risk assessments. By affirming the credentialing protocol surrounding HSPD-12 and sanctioning an identity-management technology, Nelson opens the door to profound questions of constitutional law, electronic privacy law and policy, and surveillance policy that have yet to be resolved. These questions include the role of biometric technology and dataveillance in national security policy and immigration-control policy. It now remains to be seen whether HSPD-12 will eventually serve as a programmatic and technological prototype for a national biometric ID system, such as a biometric social security card or biometric ePassport, in the future. B. Delegating Vetting and Database Screening Protocols to States and Private Entities In addition to de facto discrimination, Algorithmic Jim Crow regimes can promote de jure discrimination or discrimination as a matter of law.301 Under historic Jim Crow regimes, enforcement of segregationist laws was delegated to both public and private entities.302 Those who participated in segregation gatekeeping often did so under the threat of legally imposed sanctions.303 Resistance to the mandate to segregate train service, for instance, led to the initiation of a legal challenge to Louisiana’s Jim Crow laws in Plessy, which required the cooperation of railway companies that were frustrated with their gatekeeping duties under the Separate Car Act.304 Railway companies opposed the segregation law on the ground that running two train cars—one for white passengers and one for black passengers—was economically costly, especially for train routes on which ridership had proven to be light.305 The petitioners also argued that it was unlawful to delegate segregationist gatekeeping to the private companies who would be fined for allowing black passengers to ride white railcars.306 Homer Adolph Plessy had been selected to violate the Jim Crow law specifically because he was a light-skinned black man who could “pass” as a white man.307 Yet, a similar de jure discrimination scheme may be emerging in the modern era. How do private and state immigration gatekeepers determine whether an individual is lawfully present in the United States? Under the Immigration Reform and Control Act of 1986 (IRCA),308 the federal government delegated immigration enforcement authority to all employers, public and private, to assist in immigration gatekeeping duties through the examination of paper-based documents that purport to establish identity and citizenship status.309 Under IRCA, employers faced civil and criminal fines for failure to participate in sorting out undocumented immigrants from the workforce.310 In 1990, the Wall Street Journal editorial pages compared federal employer-sanctioning policies required under federal immigration law to historic Jim Crow regimes. The publication explained that private entities were once again asked to engage in discrimination311 under the law by effectually being deputized as immigration gatekeepers. Specifically, the Wall Street Journal described IRCA as “the first legislation since Jim Crow where the government is so closely aligned with a process that produces discrimination.”312 From the 1970s to the present, immigration laws at the federal and state level have attempted to restrict immigrant access to transportation and travel, employment, education, housing, and benefits.313 In contrast to historic Jim 2017] Crow regimes, however, the targeting of the undocumented immigrant population does not need to proceed under a façade of “equality” because such discrimination is often construed as legally permissible. Undocumented immigrants, with important exceptions, do not enjoy the broad civil rights protections and constitutional rights afforded to U.S. citizens.314 Even lawful immigrants may face more restricted rights than U.S. citizens.315 Those arguing in favor of tough immigration actions, including those defending the Executive Orders, have explained this position as a legal defense of such actions.316 Yet, for decades, lawful immigrants and those perceived to be foreign have alleged that they suffer from a form of collateral discrimination: an assumption of undocumented status and accidental targeting that stems from restrictive immigration laws.317 Studies have consistently shown that vetting and screening protocols required by immigration gatekeeping—sometimes referred to as “show me your papers” laws—incentivize racial profiling.318 In other words, mandatory document checks often target those perceived to be foreign: those who may be isolated on the basis of race, color, ethnicity, national origin, religion, and “foreignness” characteristics,319 such as accent, clothing, and a failure to present “whiteness” characteristics.320 In response to growing empirical evidence that immigration-related screening and delegated gatekeeping duties by the government reliably led to discrimination, Congress increasingly looked to technological screening methods as “race-neutral” tools to achieve the same means.321 Throughout the 1990s until the present, immigration reform legislation has proposed database-driven methods to implement screening and gatekeeping functions.322 Many of these database screening methods are experimental and still under testing.323 Nonetheless, the 9/11 terror attacks accelerated the rollout of these experimental vetting and screening systems.324 The immigrant status of the 9/11 hijackers led to calls for policy initiatives that could facilitate the identification and more efficient tracking of immigrants and potential terrorists through cybersurveillance and dataveillance technologies.325 Many of these technologies were dependent upon biometric data monitoring and database-facilitated algorithmic sorting tools.326 The impetus was not so much to avoid bias in screening but to harness the supposed efficiencies and reliability of a database-centered means of screening. Since 9/11, immigration policy and national security policy have increasingly converged. At the federal level, this convergence could be seen in the increasing adoption of big data identity-management systems aimed to screen the population to determine who could receive rights and benefits, such as the No Fly List,327 the No Work List (“E-Verify”),328 and the No 322. See id. 323. Id. § 3(b)(b)(1) (“[E]valuating whether the problems identified by the report submitted under subsection (a) have been substantially resolved . . . .”); id. § 3(b)(b)(2) (“[D]escribing what actions the Secretary of Homeland Security shall take before undertaking the expansion of the basic pilot program to all 50 States in accordance with section 401(c)(1), in order to resolve any outstanding problems raised in the report filed under subsection (a).”). 324. See Press Release, Transp. Sec. Admin., TSA to Test New Passenger Pre-Screening System (Aug. 26, 2004), [] (describing the implementation of a post-9/11 passenger prescreening program that checks passengers’ names against terrorist watchlists in an effort to improve the use of “no fly” lists). 325. The 9/11 Commission Report, for example, emphasized the need to incorporate biometric data into identity-management tools and systems in order to augment border security and national security objectives. NAT’L COMM’N ON TERRORIST ATTACKS UPON THE U.S., THE 9/11 COMMISSION REPORT 385–92 (2004), 911Report.pdf [] (“Linking biometric passports to good data systems and decision-making is a fundamental goal.”). 326. See SIMSON GARFINKEL, DATABASE NATION: THE DEATH OF PRIVACY IN THE 21ST CENTURY 37–67 (2000); KELLY A. GATES, OUR BIOMETRIC FUTURE: FACIAL RECOGNITION TECHNOLOGY AND THE CULTURE OF SURVEILLANCE 1–2 (2011) (“The suggestion that an automated facial recognition system may have helped avert the September 11 terrorist attacks was perhaps the most ambitious claim circulating about biometric identification technologies in the aftermath of the catastrophe.”); ANIL K. JAIN ET AL., INTRODUCTION TO BIOMETRICS vii (2011) (“[T]he deployment of biometric systems has been gaining momentum over the last two decades in both public and private sectors. These developments have been fueled in part by recent [post-9/11] government mandates stipulating the use of biometrics for ensuring reliable delivery of various services.”). See generally JENNIFER LYNCH, FROM FINGERPRINTS TO DNA: BIOMETRIC DATA COLLECTION IN U.S. IMMIGRANT COMMUNITIES AND BEYOND (2012); Laura K. Donohue, Technological Leap, Statutory Gap, and Constitutional Abyss: Remote Biometric Identification Comes of Age, 97 MINN. L. REV. 407 (2012); Hu, supra note 52. 327. 49 U.S.C. § 44903 (2012); 49 C.F.R. pts. 1540, 1544, 1560 (2016). 328. E-Verify is a “test pilot” program jointly operated by DHS and the Social Security Administration that enables employers to screen employees’ personally identifiable data (e.g., name, birth date, and Social Security number) through government databases over the internet in order to “verify” the identity of the employee. U.S. CITIZENSHIP & IMMIGR. SERVS., DEP’T OF HOMELAND SEC., E-VERIFY USER MANUAL FOR EMPLOYERS 1 (2014), 2017] Citizenship List (managed by Secure Communities329 and the DHS’s Prioritized Enforcement Program330). These database screening and digital watchlisting systems purport to further crime control, immigration control, and counterterrorism objectives. The E-Verify database has not only been used to restrict employment opportunities, but it is alleged that landlords have used the database to screen tenants and that school officials have used the database to screen students.331 Similarly, the No Vote List (“SAVE”332 and “HAVA”333) has been used for voter purges and to restrict driver’s licenses as well as access to welfare benefits.334 After 9/11, the federal government sought a sharp increase in personnel who could conduct vetting and implement screening protocols to increase the effectiveness of immigration gatekeeping.335 Thus, the federal government also increasingly invited state and local law enforcement to participate in the enforcement of federal immigration law under a “force multiplier” theory336 of delegation of immigration gatekeeping.337 Under the expanded immigration gatekeeping mandates of DHS, state and local governments were granted access to DHS database screening systems and invited to screen arrestees through these systems.338 After 9/11, the federal government also experimented with the merging of database screening protocols to eliminate the separation between civil and criminal immigration database screening protocols.339 Under immigration federalism and national security federalism, state laws have increasingly captured post-9/11 identity-management technologies introduced by DHS. Consequently, comparisons between historic Jim Crow regimes and contemporary immigration enforcement regimes have intensified in recent years. De jure discrimination has been alleged in state laws that mandate vetting protocols and the delegation of vetting and The Hunt for Noncitizen Voters, 65 STAN. L. REV. ONLINE 66, 66 (2012). For more information on the SAVE database screening program, see DEP’T OF HOMELAND SEC., PRIVACY IMPACT ASSESSMENT FOR THE SYSTEMATIC ALIEN VERIFICATION FOR ENTITLEMENTS (SAVE) PROGRAM 12 (Aug. 26, 2011), []. 333. 52 U.S.C. § 21083 (2012) (originally enacted as Help America Vote Act of 2002 (HAVA), Pub. L. No. 107-252, 116 Stat. 1666 (2002)). HAVA requires each state to implement and maintain an electronic database of all registered voters. Id. HAVA also requires states to verify the identity of the voter registration application through crosschecking the applicant’s driver’s license or last four digits of the applicant’s Social Security number. Id. § 21083(a)(5)(A)(i). If the individual has neither number, the state is required to assign a voter identification number to the applicant. Id. § 21083(a)(5)(A)(ii). Each state election office is tasked with overseeing election rules and procedures for that state in the implementation of HAVA. President Signs H.R. 3295, “Help America Vote Act of 2002,” SOC. SEC. ADMIN. (Nov. 7, 2002), []. 334. See SAVE Database—Issues with Obtaining SSN and Driver’s Licenses, NAFSA, [] (last visited Oct. 16, 2017 ). 335. See Jennifer M. Chacón, The Transformation of Immigration Federalism, 21 WM. & MARY BILL RTS. J. 577, 589 (2012); Kobach, supra note 63, at 545; Christopher N. Lasch, Preempting Immigration Detainer Enforcement Under Arizona v. United States, 3 WAKE FOREST J.L. & POL’Y 281, 328 (2013). 336. See Memorandum from John Kelly, Sec’y of Homeland Sec., to Kevin McAleenan, Acting Comm’r, U.S. Customs and Border Prot. et al., Implementing the President’s Border Security and Immigration Enforcement Improvements Policies 4 (Feb. 20, 2017), []. 337. Id. 338. See supra note 329 and accompanying text. 339. S-COMM involves both civil DHS immigration database screening and FBI criminal record database screening simultaneously. See supra note 329 and accompanying text. 2017] screening responsibilities. Similar to the segregationist gatekeeping duties that were delegated under historic Jim Crow regimes, restrictionist immigration gatekeeping protocols under Algorithmic Jim Crow have been criticized as promoting both de jure and de facto discrimination against citizens and lawful immigrants that may be perceived to be foreign.340 Specifically, state laws have proposed that legal penalties and liabilities could be incurred by employers, police, landlords, doctors, school officials, and state benefits administrators who fail to conduct vetting and screening protocols. For instance, the Los Angeles Times invoked Jim Crow comparisons after passage of Alabama House Bill 56, a state law that attempted to control unwanted migration in part by delegating immigration screening to both private and public entities.341 Wade Henderson, president of the Leadership Conference on Civil and Human Rights, also drew a comparison. On June 9, 2011, the day Alabama House Bill 56 was signed into law, he remarked: “[E]ven Bull Connor himself would be impressed,” referring to the famed segregationist who served as Birmingham’s public safety commissioner tasked with enforcing the city’s Jim Crow laws during the 1950s.342 Similarly, the governor of Arizona signed SB 1070 into law in 2013, referred to as the “racial profiling” law and the “show me your papers law.”343 Shortly thereafter, Reverend Lennox Yearwood of the Hip Hop Caucus, wearing a “Boycott Arizona” cap, declared during an interview that the Arizona immigration law is “our Jim Crow moment for the 21st century.”344 He further stated, “We can’t have anyone being checked based on the hue of their color . . . . We need to put our lives on the line [in protest] . . . . We need to stand up.”345 Members of Congress likened the Arizona law to Jim Crow and historic apartheid systems.346 In future comprehensive immigration reform proposals and in the implementation of extreme vetting protocols—especially with the Supreme Court in both Whiting and Arizona apparently endorsing a potential merger between criminal and civil database screening protocols—it is possible that other public and private actors could be delegated counterterrorism intelligence-gathering duties pursuant to immigration gatekeeping duties under the “force multiplier” approach. In Whiting, the Supreme Court upheld the constitutionality of delegating immigration database screening to private employers through the passage of the Legal Arizona Workers Act (LAWA).347 Under LAWA, employers in Arizona were not only documentinspecting immigration gatekeepers.348 LAWA also transformed public and private employers into database screening gatekeepers through a legal requirement that they run all new hires through the E-Verify identitymanagement system, which allows employers to screen employees through various federal agency databases.349 Similarly, in Arizona, the Supreme Court upheld the constitutionality of delegating immigration-database screening to state and local law enforcement after the passage of Arizona’s SB 1070.350 Similar to LAWA, state and local law enforcement were transformed into database screening gatekeepers through a legal requirement that they run all arrestees and those suspected of unlawful presence through S-COMM, an internet-based database screening system that checks biometric data (scanned fingerprints) against DHS and FBI databases. For example, in addition to Arizona’s establishment of an employersanctioning regime in LAWA (holding private employers responsible), and a police-sanctioning regime in SB 1070 (holding state and local law enforcement responsible), the state proposed a landlord-sanctioning regime in SB 1611,351 enacted a state-worker-sanctioning regime in HB 2008,352 proposed a hospital-worker-sanctioning regime in SB 1405,353 and proposed and enacted a public-school-worker- or teacher-sanctioning regime in SB 1407354 and SB 1141.355 Each of Arizona’s proposed sanctioning regimes requires a screening or vetting system because otherwise, the Arizona legislature has contended, it 2017] could not verify who among its residents is an unauthorized immigrant and who is not. The identity-management technology often relied upon by the state gatekeeping law involved an algorithm-based database screening protocol, often supplemented by a paper-based inspection, to log personally identifiable data into a preexisting vetting and database screening system operated by DHS. Although not all of these measures passed, the gatekeeping and screening aspects of Arizona’s proposed comprehensive immigration reform strategy, as well as other state and local laws passed or considered in recent years, has resulted in an unprecedented expansion of document inspection and database-driven screening protocols.356 Arizona’s aggressive stance on restrictive immigration gatekeeping is enabled by the introduction of big data vetting analytics and database screening potentialities, often conducted through internet-based screening of DHS and other federal agency databases. It shows the multifold opportunities for electronic vetting in daily life. And while it is a system to screen out immigrants, at the national level and in the national security context, newly emerging digital watchlisting and database screening programs such as the No Fly List make clear that database screening is easily adaptable to other kinds of vetting for various purposes. Those screened or vetted will individually encounter a purportedly neutral and colorblind process but with the result that they fall into groups that can start to look much like the kinds of classifications that would normally offend the Constitution’s equal protection guarantees. Consequently, the extreme vetting protocols and the implementation of a Muslim registry should be understood within the context of delegated database screening protocols, such as those proposed and passed by Arizona. As can be seen in the discussion above, the efforts by the federal and state government to collect and screen data under the auspices of immigrationcontrol policy now extend to a wide range of contexts, including employment screening and day-to-day policing. This appears to be consistent with datacollection and screening policies under the Trump administration. “Asked where [Muslims] would be registered, [Trump] said Muslims would be signed up at ‘different places . . . . [I]t’s all about [data] management.’”357 Candidate Trump specifically referred to the need to deploy DHS identitymanagement technologies: “Trump tied his reasoning for the database to the need to identify who is in the country legally. ‘It would stop people from coming in illegally,’ Trump said. ‘We have to stop people from coming into our country illegally.’”358 C. Litigating Algorithmic Jim Crow The number of individuals and private entities affected by government identity-management programs is growing as rapidly as the programs themselves, and future attempts to seek judicially imposed limits on such programs appear inevitable. The broader question, thus, is not whether the scientists in Nelson were denied a constitutional right to privacy but whether any limiting principle can be articulated to curtail the government’s attempt to engage in post-9/11, semiuniversal vetting and screening systems, such as the No Fly List, and biometric dataveillance credentialing, such as HSPD-12, in the name of furthering national security, crime control, and immigration policy. Under an “equal but separate” regime, identity-management systems that purport to collect and sort data of individuals equally, however, may impose disparate consequences through colorblind vetting protocols and “raceneutral” database screening systems. Yet, whether biometric-based identification systems can be presented as colorblind and “race-neutral” is in doubt. Racial characteristics are among the data collected in biometric databases. Soft biometric data, for instance, includes digital analysis or automated determination of age, height, weight, race or ethnicity, color of skin and color of hair, scars and birthmarks, and tattoos.359 Further, newly developed big data vetting tools fuse biometric data with biographic data and internet and social media profiling to algorithmically assess risk. Data fusion techniques are not race neutral, as recent reports have exposed how data analytics can result in pinpointing racial, ethnic, and socioeconomic characteristics through big data analysis tools.360 As plaintiffs in the No Fly List litigation allege, those disparately impacted by mandatory vetting and screening protocols will largely fall within traditional classifications—race, color, national origin, ethnicity, and religion—depending on what may be determined to be suspect criteria.361 Recent immigration-control policy and programs demonstrate the government’s interest in delegating immigration-vetting duties to private actors,362 such as employers, and nonfederal actors, such as state and local law enforcement363 or their privatized subdelegates,364 which can exacerbate issues of racial profiling and discrimination. For instance, LAWA and Arizona’s SB 1070 are examples of immigration federalism and national security federalism. Immigration federalism traditionally has denoted state 2017] and local efforts to control or mitigate the impact of unwanted migration or to regulate the admission of noncitizens across state borders.365 In the past few years, several thousand state and local immigration-related laws have been considered in almost every state.366 This level of immigration federalism activity is unprecedented in U.S. history.367 Also unprecedented, however, is the manner in which immigration federalism is intersecting with two other critical movements in U.S. history: (1) an increasing reliance on database-sorting technology and dataveillance in federal immigration policy to facilitate state-federal partnerships in the control of unwanted migration; and (2) a post-9/11 national security policy of national security federalism that encourages a state-federal partnership in the furtherance of intelligence-gathering and homeland security objectives under a “force multiplier” theory. In the DHS memos released by then-DHS Secretary John Kelly on February 21, 2017, which implemented the Executive Orders on immigration signed by President Trump on January 25, 2017, DHS stated that the executive branch would apply a “force multiplier” approach to the immigration-control and national security effort.368 Identity-management technologies that rely upon the internet and digital databases to verify identity have been developed to help execute these goals. As a presidential candidate, Trump explained that immigration-control and counterterrorism efforts required “a lot of systems, beyond databases.”369 Because these statutes are perceived as targeting primarily those born in foreign countries but residing here, the questions of government intrusion and disparate impact are obscured. However, while the state and federal laws at issue may be immigration laws first, they are still identity-screening laws, and the entire population—citizens and noncitizens—is subject to their vetting and screening protocols. Consequently, immigration federalism, when combined with national security federalism, is driving the exponential expansion of identity-management programs and biometric-database screening. Given the historical connection between mass data collection and mass discrimination,370 federal courts may require an inquiry into the discriminatory animus in the design of the vetting protocols and the database screening systems. That inquiry will likely start from an assessment of the disparate impact of identity-management technologies. Relatedly, challengers and the courts must contemplate the disparate impact of the algorithms themselves, the screening inputs that produce the results, and the discriminatory result of other data-driven decision-making tools, rather than in the personal animus of the screener. Judicial review should evolve to question analytical assumptions and to develop evaluative methods to interrogate the underlying algorithms informing the screening and vetting systems. Challengers and federal courts must also become more aware of other types of discrimination that can be facilitated by digitalized vetting and screening protocols. Although “data driven discrimination” is not currently recognized, we can begin challenging the collection of data under privacy theories and attempt to limit the ways in which judgments can be made based on the analysis of such data. Nelson, for example, challenged data-driven decisions that imposed arbitrary definitions of suitability on moral proclivities rather than decision-making founded on a secure rational basis.371 Further, identity-management technologies and Algorithmic Jim Crow may force innovations in constitutional data-privacy theories. These may include, for example, asserting a reasonable expectation of privacy under the Fourth Amendment’s prohibition of the search and seizure of data. The original complaints filed to enjoin LAWA’s mandatory E-Verify database screening alleged a Fourth Amendment violation.372 Among other challenges, the Chamber of Commerce argued that E-Verify required an unconstitutional search and seizure of an employee’s personally identifiable information by Arizona employers.373 The Fourth Amendment challenge, however, was not the driving force behind the litigation, was withdrawn by stipulation before the district court,374 and was not before the Supreme Court in Whiting, which focused on the question of preemption.375 Although the United States only challenged section 2(B) of Arizona’s SB 1070 on preemption grounds, the American Civil Liberties Union (ACLU) had originally raised other constitutional claims, including a Fourth Amendment claim, to the implementation of section 2(B), which required mandatory biometric-database screening of those suspected of unlawful presence.376 Papers Show Census Role in WWII Camps, USA TODAY (Mar. 30, 2007), []; see also supra notes 132, 134 and accompanying text. 371. See supra notes 253–78 and accompanying text. 372. See Ariz. Contractors Ass’n v. Candelaria, 534 F. Supp. 2d 1036, 1061 (D. Ariz. 2008). 373. Id. 374. Id. 375. Chamber of Commerce v. Whiting, 563 U.S. 582, 594 (2011). 376. Valle del Sol v. Whiting, No. CV 10-1061-PHX-SRB, 2012 WL 8021265, at *2 (D. Ariz. Sept. 5, 2012). 2017] Once again, the Fourth Amendment claim was not before the Supreme Court in Arizona, which also focused on preemption.377 It is important to note as well that, similar to Korematsu, the No Fly List challengers have relied on due process rather than equal protection. As in Korematsu, the government has defended the No Fly List as a national security program that does not target classifications of individuals, but, rather, targets risk. In his dissent in Korematsu, Justice Frank Murphy stated that “the order deprives all those within its scope of the equal protection of the laws as guaranteed by the Fifth Amendment.”378 Of the relationship between due process and equal protection, William Eskridge has observed that “[t]he Due Process Clause announces a procedural norm.”379 To the extent that the Due Process Clause is recognized to carry a substantive element, Eskridge explains the courts demand “a fit between the reasonableness of the deprivation (whatever the process) and the ‘law of the land.’ The Equal Protection Clause requires the state to justify any difference in procedural or substantive treatment of one person vis-à-vis another.”380 Consequently, the Equal Protection Clause may be less useful than other constitutional options that can force political change, such as the Due Process Clause and the First Amendment.381 Eskridge suggests, however, that the Due Process Clause can secure important rights at the individual level.382 He notes that our conception of due process is more elastic and can track changing standards of social progress.383 Further, Perhaps the most fundamental value found in the Due Process Clause is the idea that the state is obligated to treat every person as a presumptively worthwhile human being who is entitled to respect and humane treatment. This principle is the key reason Buck v. Bell and [Korematsu] were wrongly decided.”384 Eskridge signals that the time might be right to view equal protection and due process as “interchangeable and interdependent” in the vindication of individual rights.385 There are benefits to prevailing under an equal protection claim,386 namely, “the Equal Protection Clause alone offers a minority group a 377. See generally Arizona v. United States, 567 U.S. 387 (2012). 378. Korematsu v. United States, 323 U.S. 214, 234–35 (1944) (Murphy, J., dissenting) (“In excommunicating them without benefit of hearings, this order also deprives them of all their constitutional rights to procedural due process. Yet no reasonable relation to an ‘immediate, imminent, and impending’ public danger is evident to support this racial restriction, which is one of the most sweeping and complete deprivations of constitutional rights in the history of this nation in the absence of martial law.”). 379. William N. Eskridge, Jr., Destabilizing Due Process and Evolutive Equal Protection, 47 UCLA L. REV. 1183, 1187 (2000). 380. Id. at 1187–88. 381. Id. at 1214. 382. Id. at 1183. 383. Id. at 1210. 384. Id. 385. Id. at 1216. 386. Id. potential constitutional jackpot at the wholesale level, that is, in challenges to an array of interconnected discriminations in state benefits as well as burdens.”387 Eskridge posits, however, that due process can yield similar wholesale benefits to protection as the Equal Protection Clause; under the Constitution, there is no theoretical or historical limit to extending wholesale rights to classes of individuals under a due process theory.388 In fact, a “destabilizing due process” that offers multiple opportunities to challenge discrimination can result in an “evolutive equal protection.”389 Thus, the equal protection process may need to be pushed to evolve to realize new forms of discrimination once the Due Process Clause forces the federal courts to confront the unreasonableness of wholesale deprivations and the need to grant wholesale benefits to challengers alleging the infringement. In the context of Algorithmic Jim Crow, however, the interrelationship between due process and equal protection is more pragmatic. Challenging algorithm-driven vetting and screening protocols under due process claims means demanding answers about the “black box” processes that may flag individuals as potential risks or threats.390 As the algorithms and dataanalytic processes become more transparent, equal protection violations can no longer be as easily disguised. This destabilization or disruption of government deprivations made possible by due process challenges can give new evaluative impetus to the evolution of the types of protections offered under the Equal Protection Clause. Arguably, this process of destabilizing algorithmic due process is already occurring under due process and equal protection challenges, among others. The No Work List has been implicated in an equal protection challenge.391 The No Vote List has been challenged392 under section 2 of the Voting Rights 387. Id. (“[T]he Court’s apparent classification-based approach offers a tremendous reward for groups that can persuade judges that the classification legally defining their group is suspect.”). 388. Id. at 1216; id. at 1219 (“There may be no deep theoretical or even historical reason why the Due Process Clause’s principles of fairness, antiarbitrariness, and dignity could not be applied on the wholesale level.”). 389. Id. at 1186. 390. See PASQUALE, supra note 191, at 101–03; Danielle Keats Citron, Technological Due Process, 85 WASH. U. L. REV. 1249, 1260 (2008); Citron & Pasquale, supra note 196, at 3–4; Kate Crawford & Jason Schultz, Big Data and Due Process: Toward a Framework to Redress Predictive Privacy Harms, 55 B.C. L. REV. 93, 122 (2014); Hu, supra note 160, at 1759; Daniel J. Steinbock, Data Matching, Data Mining, and Due Process, 40 GA. L. REV. 1, 3 (2005). 391. See Puente Ariz. v. Arpaio, 76 F. Supp. 3d 833, 842 (D. Ariz. 2015), rev’d in part, vacated in part, 821 F.3d 1098 (9th Cir. 2016). In Puente, the plaintiffs challenged the constitutionality of state statutes “that criminalize the act of identity theft done with the intent to obtain or continue employment.” Id. The state statutes in question required employers to use E-Verify and included provisions to ensure employers’ participation in the E-Verify program: the “Legal Arizona Workers Act” and “Employment of Unauthorized Aliens.” Id. at 844. The district court preliminarily enjoined the enforcement of the statutes, finding that the plaintiffs had demonstrated a likelihood of success on the merits of their equal protection claim. Id. at 854–56. 392. See Arcia v. Fla. Sec’y. of State, 772 F.3d 1335, 1348 (11th Cir. 2014) (reversing and remanding the district court’s grant of judgment as a matter of law to the Secretary of State of 2017] Act393 and the “90 Day Provision” of the National Voter Registration Act.394 The No Citizenship List has been challenged under both procedural due process and equal protection claims under the Fifth and Fourteenth Amendments,395 as well as under the Fourth Amendment.396 In addition, the subsequent litigation of the No Citizenship List was found to implicate the Tenth Amendment under the anticommandeering doctrine.397 The No Fly List and Terrorist Watchlist have been challenged under multiple legal claims Florida and declaring that the SAVE database screening program for voter purges were in violation of the 90-day provision of the National Voter Registration Act) . 393. Id. The original complaint alleged that the SAVE database screening program aimed at removing noncitizens from voter registration rolls violated section 2 of the Voting Rights Act, asserting protection for “citizens . . . having ‘less opportunity than other members of the electorate to participate in the political process and to elect the representatives of their choice.’” Complaint for Declaratory and Injunctive Relief at 18, Arcia v. Detzner, 908 F. Supp. 2d 1276 (S.D. Fla. 2012) , vacated, 2015 WL 11198230 (S.D. Fla. Feb. 12, 2015) (No. 1222282-CIV), 2012 WL 2308560 (quoting 42 U.S.C. § 1973 (2012)). 394. Complaint for Declaratory and Injunctive Relief, supra note 393, at 2. The original complaint also alleged that the SAVE database screening program aimed at removing noncitizens from voter registration rolls violated section 8(b)(1) of the National Voter Registration Act, also known as the “90 Day Provision,” with plaintiffs asserting that the statute “prohibits the systematic purging of eligible voters from the official voter list for the State of Florida, within 90 days before the date of a primary or general election for Federal office.” Id. 395. See, e.g., Complaint at 1, Galarza v. Szalczyk, No. 10-cv-06815, 2012 WL 1080020 (E.D. Pa. Mar. 30, 2012) , 2010 WL 4822758. In the original complaint, the plaintiff brought an “action under the Fourth, Fifth and Fourteenth Amendments to the United States Constitution, the Civil Rights Act of 1964 and the authority of Bivens v. Six Unknown Named Agents of Federal Bureau of Narcotics, 403 U.S. 388 (1971).” Id. Under the Fifth Amendment, the plaintiff alleged, “Issuance of an immigration detainer against Plaintiff based on his Hispanic ethnicity violated his right to be free from discrimination on the basis of ethnicity under the equal protection clause of the Fifth Amendment.” Id. ¶ 94. Under the Fourteenth Amendment, the plaintiff alleged, “Treating Plaintiff as presumptively subject to detention and removal as an ‘alien’ on the basis of his Hispanic identity violated his rights under the equal protection clause of the Fourteenth Amendment.” Id. ¶ 104. The plaintiff also alleged due process claims under the Fourteenth and Fifth Amendments. Under the Fourteenth Amendment, the complaint alleges that the defendants violated Plaintiffs right to due process of law guaranteed by the Fourteenth Amendment of the United States Constitution [by]: a) [i]mprisoning Plaintiff pursuant to a detainer issued on less than probable cause [and]; b) [f]ailing to give Plaintiff notice of and an opportunity to be heard regarding the grounds for the detainer before imprisoning Plaintiff pursuant to it. Id. ¶ 114. Under the Fifth Amendment, the complaint alleged that the defendants “violated the Fifth Amendment by acting in the following ways: a) [v]iolating the terms of 8 U.S.C. § 1357, as interpreted by the courts, by issuing detainers on less than probable cause; b) [m]isrepresenting immigration detainers as orders for mandatory detention contrary to 8 C.F.R. § 287.7(a).” Id. ¶ 100. 396. Id. ¶ 90. In the original complaint, the plaintiff also brought an action under the Fourth Amendment, alleging that “[t]he issuance of the detainer against Plaintiff occurred without probable cause to believe that he was an ‘alien’ subject to detention and removal. That issuance constituted an unreasonable seizure in violation of Plaintiff’s rights under the Fourth Amendment.” Id. 397. See Galarza v. Szalczyk, 745 F.3d 634, 636 (3d Cir. 2014). In Galarza, the Third Circuit concluded that immigration holds are not mandatory commands, but rather—per the Tenth Amendment—discretionary for state agencies. Id. Therefore, the court reasoned, a previously dismissed § 1983 claim against the county that allegedly held the plaintiff was erroneously dismissed and remanded. Id. at 645. and constitutional theories,398 including procedural due process and substantive due process under the Fifth Amendment.399 Algorithmic Jim Crow may not be challenged successfully on equal protection grounds, but, rather, on other legal grounds, such as informational privacy grounds, the legal claim in Nelson. Nelson is especially useful to the analysis here as it involves a challenge to both a mandatory biometric ID program and the vetting protocols associated with the program.400 Like the loyalty requirements imposed by the January 27, 2017, Order,401 the vetting protocols challenged in Nelson also involved screening criteria to determine trustworthiness, morality, and suitability—criteria subsequently criticized by the Court as subjective and objectionable.402 Consequently, legal responses to Algorithmic Jim Crow may require preconceiving identity-management harms to encompass a broad range of legal theories. As in the No Fly List litigation, the government will likely defend disparate-impact consequences as justified based upon risk assessments, terroristic classifications, data-screening results deemed suspect, and characteristics establishing unsuitability.403 These are classifications and characteristics not protected by equal protection jurisprudence. To acknowledge the harms emerging from Algorithmic Jim Crow, equality law should be broadened to recognize data-driven discrimination and recognition of algorithm- and big data-derived disparate impact, rather than limiting protection to only animus-based, classificationdriven discrimination. CONCLUSION Algorithmic Jim Crow regimes are distinguished from historic Jim Crow regimes in several significant respects. Algorithmic Jim Crow is cybersurveillance driven and dataveillance dependent, built around the transparency of biometric identity and other technologies of identity management, monitoring internet and social media activity and contact lists through telephony databases, database screening and digital watchlisting enforcement, and other emerging big data surveillance techniques. In contrast, traditional Jim Crow is law driven, built around the transparency of racial identity; monitoring economic, educational, political, and social 398. See Ibrahim v. Dep’t of Homeland Sec., 62 F. Supp. 3d 909, 914 (N.D. Cal. 2014) (challenging Ibrahim’s inclusion on the No Fly List under the First Amendment (freedom of association and freedom of religion), Fourth Amendment (freedom from unreasonable search and seizure), Fifth Amendment (procedural due process and substantive due process), and Fourteenth Amendment (equal protection)). 399. See First Amended Complaint paras. 52–72, Ibrahim, 62 F. Supp. 3d 909 (N.D. Cal. 2014) (No. C06-0545 WHA), 2006 WL 2330786; Complaint for Injunctive and Declaratory Relief paras. 216–255, Latif v. Holder, 28 F. Supp. 3d 1134 (D. Or. 2014) (No. 10-CV-750BR). 400. NASA v. Nelson, 562 U.S. 134, 140–42 (2011). 401. See supra notes 298–99 and accompanying text. 402. Nelson, 562 U.S. at 143 n.5. 403. See Defendants’ Cross-Motion for Summary Judgment, Latif, 28 F. Supp. 3d 1134 (D. Or. 2014) (No. 3:10-cv-00750-BR), 2015 WL 11347548. 2017] activity; and utilizing traditional criminal enforcement and detention tools as well as small data surveillance techniques. Algorithmic Jim Crow describes an “equal but separate” system of de jure and de facto discrimination rather than the “separate but equal” discrimination of historic Jim Crow. The goal of Algorithmic Jim Crow is not physical separation per se. Rather, all individuals subjected to an Algorithmic Jim Crow regime may be equally vetted through database screening and digital watchlisting systems. The separation, however, is achieved through data discrimination applied on the back end of screening and vetting protocols rather than overt social and economic discrimination and legal apartheid applied on the front end of segregationist regimes. The “equal but separate” impact of Algorithmic Jim Crow will likely manifest itself in the big data assessment of risk factors that purport to predict terroristic and criminal threat rather than segregation systems of racial or ethnic classification. In other words, individuals will be at risk of disparate treatment on the basis of suspicious algorithmic results and anomalous data, or “foreignness” characteristics. Thus, disparate treatment stemming from cybersurveillance and dataveillance may not be characterized as traditional discrimination: discrimination on the basis of a historically protected class, for instance, race, color, ethnicity, national origin, and sex. This type of identity-management, technology-based discrimination may, therefore, fall outside current interpretations of the scope of the Fourteenth Amendment’s Equal Protection Clause and outside the reach of the protection of civil rights statutes. Algorithmic vetting and biometric identification, especially once deployed across an entire citizenry, will likely lead to discriminatory profiling and surveillance on the basis of suspicious digital data and internet and social media activity deemed “suspect,” as well as classification-based discrimination, such as the isolation of those emigrating from Muslimmajority nations. These systems are likely to result in both direct and collateral discrimination on the basis of citizenship status, national origin, and religion, in particular. In addition, recent immigration-control policies and programs demonstrate the government’s willingness to delegate screening and vetting duties to private actors, such as employers and local law enforcement, which can exacerbate issues of racial profiling and discrimination. This discrimination may face limited or lenient review by a federal judiciary that generally grants broad deference in matters of immigration and national security. Because Algorithmic Jim Crow may appear to offer equality in theory, it may not be challenged successfully on equal protection grounds under the current equal protection framework. Thus, the jurisprudence must evolve to encompass new harms and recognize the disparate-impact harms of Algorithmic Jim Crow regimes. At the same time, Algorithmic Jim Crow must be challenged under other legal theories, including search and seizure of data under the Fourth Amendment, procedural due process and informational privacy rights under substantive due process guarantees of the Fifth and Fourteenth Amendments, First Amendment theories, and other statutory and constitutional theories. Wholesale disruptions to Algorithmic A. Historical Framing of Jim Crow .............................................. 651 B. Classification and Screening .................................................... 654  C. Cyberarchitecture of Algorithmic Jim Crow ............................ 658  III . THEORETICAL EQUALITY UNDER ALGORITHMIC JIM CROW ............. 663  A. Limitations of Equal Protection as a Legal Response to Algorithmic Jim Crow............................................................. 663  B. No Fly List and Discrimination on the Back End of Vetting and Database Screening Protocols......................................... 668  IV. FUTURE OF ALGORITHMIC JIM CROW . ............................................... 671  A. Biometric Credentialing and Vetting Protocols: NASA v . Nelson ..................................................................................... 672  B . Delegating Vetting and Database Screening Protocols to States and Private Entities ...................................................... 679 C. Litigating Algorithmic Jim Crow.............................................. 688  CONCLUSION ............................................................................................. 694 8. Exec . Order No. 13 , 780 , 82 Fed. Reg. 13 , 209 ( Mar . 6, 2017 ) [ hereinafter March 6 , 2017 , Order]. 9. Id. §§ 1 - 2 ; see also id. § 5 (“Implementing Uniform Screening and Vetting Standards for All Immigration Programs” ). 10. Compare id., with January 27 , 2017 , Order, supra note 4, § 4. 11. March 6, 2017 , Order, supra note 8, § 5. 12. 137 S. Ct . 2080 ( 2017 ). 13. Presidential Proclamation , Enhancing Vetting Capabilities and Processes for Detecting Attempted Entry into the United States by Terrorists or Other Public-Safety Threats (Sept . 24, 2017 ) [hereinafter “September 24, 2017 , Order”], 2017 /09/24/enhancing-vetting -capabilities-and-processes-detecting-attemptedentry [https://perma .cc/R678-KL5F]. 14. Id . 15. Id . ; see also infra note 50 and accompanying text (discussing U.S. Department of Homeland Security definition of “identity management”). 16. Trump v. Int'l Refugee Assistance Project, Nos. 16-1436 , 16 - 1540 , slip op. (U.S. Sept. 25 , 2017 ), [] (ordering parties to file letter briefs addressing whether, or to what extent , the Proclamation issued on September 24 , 2017 , may render the consolidated cases moot ). 17. See , e.g., Complaint for Declaratory and Injunctive Relief at 26-27 , Iranian Alliances Across Borders, Univ. of Md. Coll. Park Chapter v. Trump, No. 8 : 17 -cv-02921 -GJH (D. Md . Oct. 2 , 2017 ) (seeking declaratory and injunctive relief from the September 24, 2017, Order and alleging that the Order violates the antidiscrimination provision of the Immigration and Nationality Act, 8 U .S.C. § 1152(a)(1)(A) ( 2012 )) ; Letter from ACLU to Hon . Theodore D. Chuang , U.S. Dist . Ct. for the Dist . of Md. (Sept. 29 , 2017 ), - pmc-letter [] (seeking to amend the complaint in International Refugee Assistance Project in light of the September 24, 2017 , Order); see also 35. Current refugee vetting procedures include database screening through the U.S. Department of Defense's Defense Forensics and Biometrics Agency's (DFBA) Automated Biometric Identification System (ABIS). Id. (“A biometric record check of the Department of Defense's (DOD) records collected in areas of conflict (predominantly Iraq and Afghanistan) . DOD screening began in 2007 for Iraqi applicants and has now been expanded to all nationalities .”). 36. Current refugee vetting procedures include database screening through the “National Counterterrorism Center/Terrorist Screening Center (terrorist watch lists)” and the “FBI Fingerprint Check through Next Generation Identification (NGI) .” Id. 37. Biometrics is “[t] he science of automatic identification or identity verification of individuals using physiological or behavioral characteristics . ” JOHN VACCA, BIOMETRIC TECHNOLOGIES AND VERIFICATION SYSTEMS 589 ( 2007 ). 38. See Databases , INTERPOL, Databases [] (last visited Oct . 16 , 2017 ). 39. See , e.g., DEP'T OF HOMELAND SEC., PRIVACY IMPACT ASSESSMENT FOR THE AUTOMATED BIOMETRIC IDENTIFICATION SYSTEM (IDENT) 15 ( 2012 ), sites/default/files/publications/privacy/PIAs/privacy_pia_usvisit_ident_appendixj_jan2013.p df []; Office of Biometric Identity Management Identification Services , DEP'T HOMELAND SECURITY , [] (last visited Oct . 16 , 2017 ). 40. See , e.g., DEP'T OF HOMELAND SEC., PRIVACY IMPACT ASSESSMENT FOR THE IRIS AND FACE TECHNOLOGY DEMONSTRATION AND EVALUATION (IFTDE) 2 ( 2010 ), []. 41. See , e.g., DEP'T OF HOMELAND SEC., PRIVACY IMPACT ASSESSMENT UPDATE FOR THE STANDOFF TECHNOLOGY INTEGRATION AND DEMONSTRATION PROGRAM: BIOMETRIC OPTICAL SURVEILLANCE SYSTEM TESTS 2 ( 2012 ), privacy_pia_st_stidpboss_dec2012.pdf []. 42. See , e.g., DEP'T OF HOMELAND SEC., PRIVACY IMPACT ASSESSMENT FOR THE RAPID DNA SYSTEM 2 ( 2013 ), 20130208 .pdf []. 43. See , e.g., Charles E. Schumer & Lindsey O. Graham , The Right Way to Mend Immigration, WASH . POST (Mar. 19 , 2010 ), [] (“We would require all U.S. citizens and legal immigrants who want jobs to obtain a hightech, fraud-proof Social Security card . Each card's unique biometric identifier would be stored only on the card . . . .”). 44. See , e.g., Eric Markowitz , Retina Scanners and Biometric Passports: A Look at the Futuristic Tech That Could Scan Refugees , INT'L BUS. TIMES (Nov. 25, 2015 , 11 :29 AM), http://www.ibtimes. com/retina-scanners-biometric-passports-look-futuristic-tech- could- scanrefugees- 2199960 []. 45. March 6, 2017 , Order, supra note 8, § 8; January 27 , 2017 , Order, supra note 4, § 7. 46. John Burnett, Former Immigration Director Defends U.S. Record on Refugee Vetting, NPR (Feb. 3 , 2017 , 4 :35 PM), 2017 /02/03/513311323/formerimmigration-director -defends-u-s-record-on-refugee-vetting [https://perma .cc/U99Q-RWT5] (noting that the former director of the Office of U.S. Citizenship and Immigration Services of the U.S. Department of Homeland Security under the Obama administration “point[ed] out that his office had been checking Facebook, Twitter and Instagram accounts of prospective refugees from Syria and Iraq since 2015 ”). 47. Alexander Smith , U.S. Visitors May Have to Hand over Social Media Passwords: DHS , NBC NEWS ( Feb . 8, 2017 , 7 :51 AM), -have-hand-over-social-media-passwords-kelly-n718216 [https://perma .cc/7WK4-FDKB]. 48. See id. ; see also Notice of Modified Privacy Act System of Records , 82 Fed. Reg. 179 ( Sept . 18, 2017 ). 49. See , e.g., DEP'T OF HOMELAND SEC ., supra note 33; see also infra Part I.A. 50. DHS offers this definition of identity management: Identity Management (IdM) deals with identifying and managing individuals within a government, state, local, public, or private sector network or enterprise. In addition, authentication and authorization to access resources such as facilities or, sensitive data within that system are managed by associating user rights, entitlements, and privileges with the established identity . Cyber Security Division Identity Management Program Video , DEP'T HOMELAND SECURITY , -technology/cyber-security-division-identity-managementprogram-video [https://perma .cc/9NGG-8G28] (last visited Oct . 16 , 2016 ). 51. See , e.g., STEVEN FINLAY , PREDICTIVE ANALYTICS , DATA MINING AND BIG DATA: MYTHS, MISCONCEPTIONS AND METHODS 3 ( 2014 ) ; ERIC SIEGEL, PREDICTIVE ANALYTICS: THE POWER TO PREDICT WHO WILL CLICK, BUY , LIE , OR DIE 59-60 ( 2013 ) ; NATE SILVER, THE SIGNAL AND THE NOISE: WHY SO MANY PREDICTIONS FAIL-BUT SOME DON'T 417 - 18 ( 2012 ) ; see also Spencer Woodman, Palantir Provides the Engine for Donald Trump's Deportation Machine , INTERCEPT (Mar. 2 , 2017 , 1 :18 PM), 2017 /03/02/ palantirprovides-the-engine-for-donald-trumps-deportation-machine/ [] (reporting that the DHS awarded a private contractor a $41 million contract to build an “Investigative Case Management” system to allow DHS to “access a vast 'ecosystem' of data 64. See , e.g., Shirin Sinnar , Rule of Law Tropes in National Security , 129 HARV. L. REV. 1566 , 1569 ( 2016 ) ; see also Sahar F. Aziz, Policing Terrorists in the Community, 5 HARV . NAT'L SECURITY J. 147 , 222 ( 2014 ) (discussing counterterrism law enforcement). 65. See Ryan Lizza, Why Sally Yates Stood Up to Trump, NEW YORKER (May 29 , 2017 ), -yates-stood-up-to-trump [https://perma .cc/JHF9-73DM] (explaining that Yates learned of the Order upon being notified by a deputy who read the news online ). 66. Id . (“Yates read through the briefs, and thought that two arguments against the order were particularly strong . . . . [ The order] arguably violated the Establishment Clause of the First Amendment . And . . . there seemed to be serious due-process questions .”). 67. Id . 68. Id . 69. Letter from Sally Yates, Acting Attorney Gen., Dep't of Justice , to Dep' t of Justice (Jan. 30 , 2017 ), []. 70. Id . 71. Lizza , supra note 65 (“ The statement was sent to thousands of department employees around the country . About four hours later, at around 9 P.M., McGahn's office asked the senior Trump appointee to deliver a letter to Yates, notifying her that she had been fired .”). 72. See , e.g., Guy-Uriel E. Charles & Luis Fuentes-Rohwer , State's Rights, Last Rites , and Voting Rights, 47 CONN. L. REV. 481 , 486 nn. 23 - 24 ( 2014 ); Atiba R.Ellis, The Cost of the Vote: Poll Taxes, Voter Identification Laws, and the Price of Democracy, 86 DENV . U. L. REV. 1023 , 1024 n. 7 ( 2009 ). 73. Ellis , supra note 72, at 1040 n. 79 , 1041 - 50 . 74. See , e.g., Ibrahim v. Dep't of Homeland Sec ., 62 F. Supp . 3d 909 , 929 (N.D. Cal . 2014 ) (“By this order, all defendants shall specifically and thoroughly query the databases maintained by them, such as the TSDB, TIDE, CLASS, KSTF, TECS, IBIS, TUSCAN, TACTICS, and the no-fly and selectee lists . . . .”). 75. Adam Nossiter et al., Three Teams of Coordinated Attackers Carried Out Assault on Paris, Officials Say; Hollande Blames ISIS , N.Y. TIMES (Nov. 14, 2015 ), 2015 /11/15/world/europe/paris-terrorist-attacks.html []. 76. Id . 77. Id . 78. Id . 79. Steve Almasy et al., Paris Massacre: At Least 128 Killed in Gunfire and Blasts, French Officials Say, CNN (Nov. 14 , 2015 , 9 :48 AM), 2015 /11/13/ world/paris-shooting/ []. 80. Christiane Amanpour & Thom Patterson, Passport Linked to Terrorist Complicates Syrian Refugee Crisis , CNN (Nov. 15 , 2015 , 12 :24 PM), 2015 /11/15/ europe/paris-attacks-passports/index.html []. 81. Id . 82. Id . 83. Ashley Fantz & Ben Brumfield, More Than Half the Nation's Governors Say Syrian Refugees Not Welcome , CNN (Nov. 19 , 2015 , 3 :20 PM), 2015 /11/16/ world/paris-attacks-syrian - refugees-backlash/ [] (reporting that the many states refused to accept Syrian refugees for resettlement , including: Alabama, Arizona, Arkansas, Florida, Georgia, Idaho, Illinois, Indiana, Iowa, Kansas, Louisiana, Maine, Maryland, Massachusetts, Michigan, Mississippi, Nebraska, Nevada, New Hampshire, New Jersey, New Mexico, North Carolina, North Dakota, Ohio, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Wisconsin, and Wyoming). 84. See Michael Ignatieff , The Refugees and the New War, N.Y. REV. BOOKS (Dec. 17 , 2015 ), - new-war/ []. 85. Id . ; see also Katie Worth, Can Biometrics Solve the Refugee Debate? , PBS (Dec. 2 , 2015 ), -biometrics-solve-the-refugee-debate/ []. 86. Markowitz , supra note 44. 87. Michael S. Schmidt & Richard Pérez-Peña , F.B.I. Treating San Bernardino Attack as Terrorism Case , N.Y. TIMES ( Dec . 4, 2015 ), 2015 /12/05/us/ tashfeen-malik - islamic-state.html []. 88. Id . 89. Id . 90. Id . 91. Id . 92. Id . 93. Lisa Lambert et al., House to Consider Changes to Visa Waiver Program, Including 'Smart' Passports, REUTERS (Dec. 2 , 2015 , 8 :16 AM), 12/02/us-usa -congress-visas-idUSKBN0TL1CV20151202 [https://perma .cc/W32F-SLZR]. 94. See Diamond, supra note 1. 95. Associated Press, How Donald Trump's Plan to Ban Muslims Has Evolved , FORTUNE (June 28 , 2016 ), 2016 /06/28/donald-trump - muslim-ban/ []. 96. See David Mark & Jeremy Diamond, Trump: 'I Want Surveillance of Certain Mosques,' CNN (Nov . 21, 2015 ), 2015 /11/21/politics/trump-muslimssurveillance/ []. 97. See Lizette Alvarez et al., Orlando Gunman Was 'Cool and Calm' After Massacre , Police Say, N.Y. TIMES (June 13, 2016 ), 2016 /06/14/us/orlandoshooting.html []. 98. Lizette Alvarez & Richard Pérez-Peña , Orlando Gunman Attacks Gay Nightclub, Leaving 50 Dead, N.Y. TIMES (June 12, 2016 ), 2016 /06/ 13/us/orlando-nightclub-shooting.html []. 99. See Emily Schultheis, Donald Trump: U.S. Must “Start Thinking About” Racial Profiling, CBS NEWS ( June 19 , 2016 ), -afterorlando-racial-profiling-not-the-worst- thing- to-do/ []. 100. See , e.g., Diamond, supra note 1. 101. Diamond , supra note 22. 102. Emily Schultheis , Donald Trump Doubles Down in Immigration Speech: “Mexico Will Pay for the Wall,” CBS NEWS (Aug . 31, 2016 ), -immigration- speech- in-phoenix/ []. 103. Id . 270. Nelson , 562 U.S. at 143. 271. Id . at 143 n.5. 272. Id . ; see also AGENCY HUM . RES . DIV., NAT'L AERONOTICS & SPACE ADMIN., SREF30000-0003, NASA DESK GUIDE FOR SUITABILITY AND SECURITY CLEARANCE PROCESSING VERSION 2 , at 51 ( 2008 ) [hereinafter NASA DESK GUIDE] , SecurityDeskGuide.pdf []. 273. NASA DESK GUIDE, supra note 272 , at 51. 274. Id . at 65. 275. See Nelson, 562 U.S. at 143 n. 5; NASA DESK GUIDE , supra note 272, at 65- 67 . The NASA Desk Guide provides this caveat: “[T]raffic violations not required to be admitted on OF306 or other application material/QSP will not be considered issues . ” NASA DESK GUIDE, supra note 272 , at 67. 276. NASA DESK GUIDE, supra note 272 , at 71-93. 277. Id . 280. Christopher Snyder , Trump Doubles Down on Vow to Bar Muslims, FOX NEWS (Dec. 8 , 2015 ), -for-complete-shutdownon-muslims-entering-us .html []. 281. See Ali Vitali, At South Carolina Rally, Donald Trump Defiant on Muslim Ban , NBC NEWS (Dec. 7 , 2015 ), -rallytrump-defiant-steadfast-muslim-ban-n475951 [https://perma .cc/D3WC-5S56]. 282. Snyder , supra note 280. 283. Amy B. Wang , Trump Asked for a 'Muslim Ban,' Giuliani Says-and Ordered a Commission to Do It 'Legally,' WASH . POST (Jan. 29 , 2017 ), -for-a-muslimban-giuliani-says-and-ordered-a-commission-to- do- it-legally/ []. 284. Id . 285. Id . 286. 847 F.3d 1151 ( 9th Cir . 2017 ) (per curiam). 287. Id . at 1168. 288. Id . at 1168 n.7. 289. 859 F.3d 741 ( 9th Cir .) (per curiam), cert . granted, 137 S. Ct . 2080 ( 2017 ). 290. Id . at 770 , 775 . 291. Int'l Refugee Assistance Project v . Trump , 857 F.3d 554 , 591 - 92 (4th Cir.), cert. granted, 137 S. Ct . 2080 ( 2017 ) (“Plaintiffs also point to the comparably weak evidence that EO-2 is meant to address national security interests, including the exclusion of national security agencies from the decision-making process, the post hoc nature of the national security rationale, and evidence from DHS that EO-2 would not operate to diminish the threat of potential terrorist activity .”). 292. Trump v. Int'l Refugee Assistance Project , 137 S. Ct . 2080 , 2086 ( 2017 ). The author reserves for future scholarship further inquiry into and analysis of the Supreme Court's final disposition and resolution of these matters . 293. Vivian Salama & Alicia A . Caldwell, DHS Report Disputes Threat from Banned Nations, ASSOCIATED PRESS (Feb. 24 , 2017 ), a30a4570f693291c866/ dhs-intel-report-disputes-threat-posed-travel-ban-nations [https://perma .cc/Z8NL-8VW6]. 294. Memorandum from the Dep't of Homeland Sec., Office of Intelligence and Analysis, Citizenship Likely an Unreliable Indicator of Terrorist Threat to the United States ( 2017 ), []. 295. Salama & Caldwell, supra note 293. 296. Id . 297. Hawaii v. Trump , 859 F.3d 741 , 759 , 784 n. 23 ( 9th Cir .) (per curiam), cert . granted, 137 S. Ct . 2080 ( 2017 ) ; Int'l Refugee Assistance Project v . Trump , 857 F.3d 554 , 575 , 591 - 92 , 596 (4th Cir.), cert. granted sub nom . 137 S. Ct . 2080 ( 2017 ). 298. January 27, 2017 , Order, supra note 4, § 1. 299. Michael D. Shear & Helene Cooper, Trump Bars Refugees and Citizens of 7 Muslim Countries , N.Y. TIMES (Jan. 27, 2017 ), 2017 /01/27/us/politics/trumpsyrian-refugees.html []. 300. See Schultheis, supra note 99. 301. See Mack , Law, Society, Identity, supra note 104, at 394-95 ( observing that de jure discrimination reflected both a shift in law and social rhetoric). 302. See generally Sunstein, supra note 232. 303. See David Benjamin Oppenheimer, Martin Luther King, Walker v. City of Birmingham, and the Letter from Birmingham Jail , 26 U.C. DAVIS L. REV. 791 , 796 ( 1993 ) (“[S]hortly before King's arrival the bus station manager had been jailed for permitting African American passengers to use the white waiting room .”); see also TOM R. TYLER, WHY PEOPLE OBEY THE LAW 21 ( 1990 ) (“Social control refers specifically to altering citizens' behavior by manipulating access to valued social resources or by delivering or threatening to deliver sanctions .”). 304. Harris , supra note 105, at 187 , 207 (citing 1890 La. Acts 111). 305. HIGGINBOTHAM, supra note 105, at 88. 306. Harris , supra note 105, at 212-13. 307. Id . at 212. 308. Immigration Reform and Control Act of 1986 , Pub. L. No. 99 - 603 , 100 Stat. 3359 (codified in scattered sections of 8 U .S.C.). 309. See 8 U.S.C. §§ 1324a ( 2012 ). 310. Id . 311. Studies by the U.S. Government Accountability Office found that the employer sanctions provision of the Immigration Reform and Control Act of 1986 had resulted in widespread discrimination . See, e.g., U.S. GOV'T ACCOUNTABILITY OFFICE , GAO/T-GGD- 92 - 21, IRCA-RELATED DISCRIMINATION : ACTIONS HAVE BEEN TAKEN TO ADDRESS IRCARELATED DISCRIMINATION, BUT MORE IS NEEDED 2 ( 1992 ), []; U.S. GOV'T ACCOUNTABILITY OFFICE , GAO/TGGD-90-51, IRCA ANTI -DISCRIMINATION AMENDMENTS OF 1990 , at 3 ( 1990 ), 90 -51 []. 312. Editorial , Clocking Immigration Sanctions, WALL ST. J., Apr . 16 , 1990 , at A12. 313. See Motomura, supra note 62 , at 1361; Michael J. Wishnie , Laboratories of Bigotry?: Devolution of the Immigration Power , Equal Protection, and Federalism, 76 N.Y.U. L. REV. 493 , 513 - 14 & nn . 106 - 10 ( 2001 ) ; see also Stephen Lee, Private Immigration Screening in the Workplace, 61 STAN . L. REV. 1103 , 1130 ( 2009 ); Huyen Pham, The Private Enforcement of Immigration Laws , 96 GEO. L.J. 777 , 780 - 81 ( 2008 ). 314. See Hiroshi Motomura , The Curious Evolution of Immigration Law: Procedural Surrogates for Substantive Constitutional Rights , 92 COLUM. L. REV. 1625 , 1632 ( 1992 ). The Supreme Court has recognized potential due process claims of persons in the United States who are noncitizens including those present unlawfully . See Kerry v. Din , 135 S. Ct . 2128 , 2133 - 34 ( 2015 ); Zadvydas v . Davis , 533 U.S. 678 , 692 - 95 ( 2001 ); Landon v . Plasencia , 459 U.S. 21 , 33 - 34 ( 1982 ); Kleindienst v . Mandel , 408 U.S. 753 , 762 - 65 ( 1972 ). 315. See 8 U.S.C. § 1324b(a)(3)(B ) ( 2012 ) (restricting certain protections under the antidiscrimination provision of the Immigration and Nationality Act to those immigrants with lawful permanent residence status). 316. Reply in Support of Emergency Motion for Stay Pending Appeal at 5 , Washington v. Trump , 847 F.3d 1151 ( 9th Cir . 2017 ) (per curiam) (No . 17 - 35105 ), 2017 WL 492504. 317. See U.S. GOV'T ACCOUNTABILITY OFFICE , GAO/T-GGD- 90 -51, IRCA ANTIDISCRIMINATION AMENDMENTS OF 1990 , supra note 311, at 3. 318. Id . 319. Id . 320. Id .; see also Cheryl I. Harris, Whiteness as Property , 106 HARV. L. REV. 1707 , 1725 ( 1993 ). 321. See Basic Pilot Program Extension and Expansion Act of 2003, Pub . L. No. 108 - 156 , 117 Stat. 1944 (codified at 8 U.S.C. §§ 1101 , 1324a ( 2012 )) (authorizing the development of the “basic pilot program for employment eligibility verification” to implement a technologically improved method to screen immigrants) . []. E-Verify is referred to as the “Basic Pilot Program” in the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA) and in subsequent congressional action extending its funding . Id. at 77-78; see also Basic Pilot Program Extension and Expansion Act of 2003 , Pub. L. No. 108 - 156 , 117 Stat. 1944 (codified at 8 U.S.C. §§ 1101 , 1324a ( 2012 )) ; Basic Pilot Extension Act of 2001, Pub . L. No. 107 - 128 , 115 Stat. 2407 ( 2002 ) (codified at 8 U .S.C. §§ 1101 , 1324a ). For a thorough discussion of EVerify and its legal implications, see generally Juliet P . Stumpf, Getting to Work: Why Nobody Cares About E-Verify (And Why They Should) , 2 U.C. IRVINE L. REV. 381 ( 2012 ). 330. The DHS Prioritized Enforcement Program (PEP) was announced by DHS Secretary Jeh Johnson on November 20, 2014, to replace the S-COMM program; however, it appears that the database screening protocols of S-COMM will remain intact under PEP. See Memorandum from Jeh Charles Johnson, Sec'y, Dep't of Homeland Sec ., to Thomas S. Winkowski, Acting Director, Immigr. & Customs Enf't 2 (Nov. 20 , 2014 ), []. On February 20 , 2017 , former DHS Secretary John Kelly signed an implementation memo announcing that S-COMM would be reinstated and PEP would be rescinded . See Memorandum from John Kelly , Sec'y of Homeland Sec., to Kevin McAleenan, Acting Comm' r, U.S. Customs & Border Prot., Enforcement of the Immigration Laws to Serve the National Interest (Feb . 20, 2017 ), files/publications/17_0220_S1_ Enforcement-of-the-Immigration-Laws-to- Serve- theNational-Interest.pdf []. 331. Stumpf , supra note 328, at 400 n. 87 ( citing MARC R. ROSENBLUM , EVERIFY: STRENGTHS, WEAKNESSES, AND PROPOSALS FOR REFORM , MIGRATION POL'Y INST . 5 , 7 ( 2011 ); Mary D. Fan, Post-Racial Proxies : Resurgent State and Local Anti-“ Alien ” Laws and UnityRebuilding Frames for Antidiscrimination Values, 32 CARDOZO L . REV. 905 , 923 - 24 , 935 - 36 ( 2011 ); Kati L. Griffith, Discovering 'Immployment' Law: The Constitutionality of Subfederal Immigration Regulation at Work , 29 YALE L. & POL'Y REV . 389 , 417 , 424 - 26 ( 2011 ) ; Rigel C. Oliveri, Between a Rock and a Hard Place: Landlords, Latinos, Anti-Illegal Immigrant Ordinances , and Housing Discrimination, 62 VAND. L. REV. 55 , 116 ( 2009 )). 332. In recent years, state election officials have used the Systematic Alien Verification for Entitlements (SAVE) database screening protocol to conduct voter purges . See Fatma Marouf, 340. See , e.g., Kevin R. Johnson , A Case Study of Color-Blindness: The Racially Disparate Impacts of Arizona's S.B. 1070 and the Failure of Comprehensive Immigration Reform , 2 U.C. IRVINE L. REV. 313 , 319 - 20 ( 2012 ) ; Karla Mari McKanders , Immigration Enforcement and the Fugitive Slave Acts: Exploring Their Similarities , 61 CATH. U. L. REV. 921 , 947 ( 2012 ). 341. Beason-Hammon Taxpayer and Citizen Protection Act, No. 2011 - 535 ( 2011 ) (codified at ALA . CODE §§ 31 - 13 -1 to 31 -13- 30 , 32 -6- 9 ( 2017 )). 342. Richard Fausset, Alabama Enacts Anti- Illegal-Immigration Law Described As Nation's Strictest , L.A. TIMES (June 10, 2011 ), nation/la-na - alabama-immigration- 20110610 [] (“'This draconian initiative signed into law this morning by Gov. Robert Bentley is so oppressive that even Bull Connor himself would be impressed,' said Wade Henderson , head of the Leadership Conference on Civil and Human Rights . . . . ' HB 56 is designed to do nothing more than terrorize the state's Latino community .'”). 343. Kasie Hunt , Dems: Ariz Law Like Jim Crow, Apartheid, POLITICO (Apr. 28 , 2010 ), []. 344. Michael McIntee , AZ Is Our Jim Crow Moment of 21st Century, YOUTUBE (July 24 , 2010 ), []. 345. Id . 346. Hunt , supra note 343. 347. See Legal Arizona Workers Act, ch. 279 , 2007 Ariz. Sess. Laws 1312 (codified at ARIZ . REV. STAT . ANN. §§ 13 - 2009 , 23 -211 to 23 - 214 ( 2008 )). 348. Immigration Reform and Control Act of 1986 , Pub. L. No. 99 - 603 , 100 Stat. 3359 (codified in scattered sections of 8 U .S.C.). 349. See supra note 328 and accompanying text . 350. Arizona v. United States , 567 U.S. 387 , 416 ( 2012 ). Section 2(B) of SB 1070, which was slightly modified and later codified in the Arizona Revised Statutes, provides: For any lawful stop, detention or arrest made by [an Arizona] law enforcement official or a law enforcement agency . . . in the enforcement of any other law or ordinance of a county, city or town or this state where reasonable suspicion exists that the person is an alien and is unlawfully present in the United States, a reasonable attempt shall be made, when practicable, to determine the immigration status of the person, except if the determination may hinder or obstruct an investigation. Any person who is arrested shall have the person's immigration status determined before the person is released . ARIZ. REV. STAT . ANN. § 11 - 1051 ( 2012 ). 351. S.B. 1611 , 50th Leg., 1st Reg. Sess. (Ariz . 2011 ). 352. ARIZ. REV. STAT. ANN. §§ 1 - 501 , 1 - 502 (Supp. 2011 ). 353. S.B. 1405 , 50th Leg., 1st Reg. Sess. (Ariz . 2011 ). 354. S.B. 1407 , 50th Leg., 1st Reg. Sess. (Ariz . 2011 ). 355. ARIZ. REV. STAT. ANN. § 15 - 802 (Supp. 2011 ). 356. See Fred H. Cate , Government Data Mining: The Need for a Legal Framework , 43 HARV. C.R-C.L.L. REV . 435 , 439 - 40 ( 2008 ). 357. Vaughn Hillyard , Donald Trump's Plan for a Muslim Database Draws Comparison to Nazi Germany , NBC NEWS (Nov. 20 , 2015 ), election /trump-says-he-would-certainly-implement-muslim-database-n466716 [https://perma .cc/RML4-KTZK]. 358. Id . 359. ENCYCLOPEDIA OF BIOMETRICS 1235 (Stan Z. Li & Anil K . Jain eds., 2009 ). 360. See FED. TRADE COMM'N , DATA BROKERS : A CALL FOR TRANSPARENCY AND ACCOUNTABILITY 56 ( 2014 ), -transparency-accountability-report-federal-trade-commission-may2014/140527databrokerreport .pdf []. 361. See Complaint at 8-9, Latif v. Holder , 28 F. Supp . 3d 1134 (D. Or . 2014 ) (No. 10 - CV- 750-BR). 362. See Lee , supra note 313 , at 1130; Pham, supra note 313, at 780-81; Stumpf, supra note 328, at 382. 363. See generally Cox & Miles, supra note 329; Lasch, supra note 329. 364. See Jennifer M. Chacón , Privatized Immigration Enforcement, 52 HARV. C.R-C.L.L. REV . 1 , 8 ( 2017 ). 365. See Hiroshi Motomura , The Rights of Others: Legal Claims and Immigration Outside the Law , 59 DUKE L.J. 1723 , 1729 ( 2010 ) (“Only after the Civil War did today's prevailing view of immigration federalism-that federal immigration regulation displaces any state laws on the admission and expulsion of noncitizens-begin to emerge .”). 366. The National Conference of State Legislatures compiles annual reports on state legislative activity regarding immigration as part of the Immigration Policy Project . See generally State Laws Related to Immigration and Immigrants , NAT'L CONF. ST . LEGISLATURES (Aug. 6 , 2017 ), -related-to-immigrationand-immigrants .aspx []. 367. See id. 368. Memorandum from John Kelly, supra note 330, at 3. 369. Hillyard , supra note 357. 370. See EDWIN BLACK , IBM AND THE HOLOCAUST: THE STRATEGIC ALLIANCE BETWEEN NAZI GERMANY AND AMERICA'S MOST POWERFUL CORPORATION 21 ( 2001 ) ; Aebra Coe, ExAmbassador Wants Ford , IBM Apartheid Liability Reviewed, LAW360 (Mar. 21 , 2016 ), [] ; Haya El Nasser,

This is a preview of a remote PDF:

Margaret Hu. Algorithmic Jim Crow, Fordham Law Review, 2017,