Summary
explaining that when a law enforcement agency receives an ICE detainer, it is that agency's responsibility "to determine what authority or right they have to hold an individual" before complying
Summary of this case from N.S. v. HughesOpinion
Case No. 2:12-cv-09012-AB (FFMx)
2019-09-27
Gerardo GONZALEZ, et al., Plaintiffs, v. IMMIGRATION AND CUSTOMS ENFORCEMENT, an entity, et al. Defendants.
Mohammad K. Tajsar, Jennifer L. Pasquarella, Jordan Wells, Pro Hac Vice, Sylvia Torres-Guillen, Zoe Nicole McKinney, ACLU Foundation of Southern California, Los Angeles, CA, Cody H. Wofsy, ACLU Foundation Immigrants Rights Project, Mark M. Fleming, Pro Hac Vice, Ruben Loyo, Pro Hac Vice, National Immigrant Justice Center, Chicago, IL, Spencer E. W. Amdur, Pro Hac Vice, ACLU Foundation Immigrants Rights Project, New York, NY, for Plaintiffs. Ahilan T. Arulanantham, Jessica Karp Bansal, Peter J. Eliasberg, ACLU Foundation of Southern California, Chris Newman, National Day Laborer Organizing Network, Los Angeles, CA, Barrett S. Litt, Lindsay Battles, Kaye McLane Bednarski and Litt LLP, Pasadena, CA, Cecillia D. Wang, ACLU Foundation Immigrants Rights Project, San Francisco, CA, Omar C. Jadwat, Pro Hac Vice, ACLU Foundation Immigrants Rights Project, New York, NY, for Plaintiffs and Defendants. Andrew I. Baum, Glaser Weil Fink Howard Avchen and Shapiro LLP, Los Angeles, CA, Jin S. Choi, Oscar Andres Bustos, Paul B. Beach, Justin W. Clark, Lawrence Beach Allen and Choi PC, Glendale, CA, for Defendants. Anne Lai, UC Irvine School of Law, Irvine, CA, for Amicus.
Mohammad K. Tajsar, Jennifer L. Pasquarella, Jordan Wells, Pro Hac Vice, Sylvia Torres-Guillen, Zoe Nicole McKinney, ACLU Foundation of Southern California, Los Angeles, CA, Cody H. Wofsy, ACLU Foundation Immigrants Rights Project, Mark M. Fleming, Pro Hac Vice, Ruben Loyo, Pro Hac Vice, National Immigrant Justice Center, Chicago, IL, Spencer E. W. Amdur, Pro Hac Vice, ACLU Foundation Immigrants Rights Project, New York, NY, for Plaintiffs.
Ahilan T. Arulanantham, Jessica Karp Bansal, Peter J. Eliasberg, ACLU Foundation of Southern California, Chris Newman, National Day Laborer Organizing Network, Los Angeles, CA, Barrett S. Litt, Lindsay Battles, Kaye McLane Bednarski and Litt LLP, Pasadena, CA, Cecillia D. Wang, ACLU Foundation Immigrants Rights Project, San Francisco, CA, Omar C. Jadwat, Pro Hac Vice, ACLU Foundation Immigrants Rights Project, New York, NY, for Plaintiffs and Defendants.
Andrew I. Baum, Glaser Weil Fink Howard Avchen and Shapiro LLP, Los Angeles, CA, Jin S. Choi, Oscar Andres Bustos, Paul B. Beach, Justin W. Clark, Lawrence Beach Allen and Choi PC, Glendale, CA, for Defendants.
Anne Lai, UC Irvine School of Law, Irvine, CA, for Amicus.
FINDINGS OF FACT AND CONCLUSIONS OF LAW
ANDRÉ BIROTTE JR., UNITED STATES DISTRICT JUDGE I. INTRODUCTION
In an effort to regulate the flow of immigration, U.S. Immigration and Customs Enforcement ("ICE") collaborates with federal, state, and local law enforcement agencies ("LEAs") to issue immigration detainers. These detainers allow LEAs to keep an individual in custody up to 48 hours after their scheduled release date. The purpose of the immigration detainer is to give ICE time to become aware of the individual's immigration status, and, if necessary, take the detainee into custody for removal.
ICE has reworked its detainer form multiple times but currently uses the I-247A form for removal proceedings. The form provides four criteria for removability: (1) a final order of removal against the alien; (2) the pendency of ongoing removal proceedings against the alien; (3) biometric confirmation of the alien's identity and a records check of federal databases that affirmatively indicate, by themselves or in addition to other reliable information, that the alien either lacks immigration status or notwithstanding such status is removable under U.S. immigration law; and (4) statements made by the alien to an immigration officer and/or other reliable evidence that affirmatively indicate the alien either lacks immigration status or notwithstanding such status is removable under U.S. immigration law.
This lawsuit surrounds the third category of the I-247A, which authorizes ICE to remove an individual based solely on biometric confirmation of the individual's identity and a review of multiple federal databases which may provide information on a person's immigration status. The issues before the Court are: (1) whether the exclusive use of biometric confirmation and database checks violates the Fourth Amendment; and (2) whether the issuance of detainers to state and local law enforcement agencies that lack authority for civil immigration arrests violates the Fourth Amendment.
This matter was tried before the Court, sitting without a jury, from May 7, 2019 to May 16, 2019. Jennifer Pasquarella and Jessica Karp-Bansal of the ACLU appeared on behalf of Plaintiff Gerardo Gonzalez and all those similarly situated ("Plaintiff"). J. Max Weintraub and John J. W. Inkeles of the United States Department of Justice appeared for Defendant Immigration and Customs Enforcement ("Defendant" or "ICE").
The Court has heard the admissible evidence presented by the parties and the arguments of counsel. It has considered the credibility of the witnesses and all papers and exhibits presented by the parties for purposes of this trial, including admissions in the Final Pretrial Conference Order. The Court makes the following findings of fact and conclusions of law pursuant to Federal Rule of Civil Procedure 52.
II. FINDINGS OF FACT
A. Background
1. Plaintiff Gerardo Gonzalez is a natural born United States citizen. He was born in Pacoima, in the City of Los Angeles, California. Dkt. No. 484, Stipulated Fact ("SF") 1. Gonzalez has never been removable from the United States.
2. On December 31, 2012, ICE issued an immigration detainer to the Los Angeles County Sheriff's Department requesting that the Sheriff maintain custody of Plaintiff Gonzalez after he would otherwise have been released from custody. Dkt. No 484, SF 2.
3. Gonzalez's immigration detainer was based on information reviewed in electronic databases. See Dkt. No. 484 SF, 3-4.
4. No ICE agent ever interviewed Plaintiff Gonzalez prior to issuing a detainer. Dkt. No. 484, SF 8.
5. Gonzalez represents a class of individuals comprised of all current and future persons who are subject to an immigration detainer issued by an ICE agent located in the Central District of California, where the detainer is not based upon a final order of removal signed by an immigration judge or the individual is not subject to ongoing removal proceedings and the detainer was issued solely on the basis of electronic database checks. Dkt. No. 484, at 1-2.
There are three classes of individuals represented by Gonzalez in this action:
(1) The Judicial Determination Class is comprised of all current and future persons who are subject to an immigration detainer issued by an ICE agent located in the Central District of California, where the detainer is not based upon a final order of removal signed by an immigration judge or the individual is not subject to ongoing removal proceedings. This Class is limited to those detained for more than 48 hours.
(2) The Probable Cause Subclass is comprised of all current and future persons who are subject to an immigration detainer issued by an ICE agent located in the Central District of California, where the detainer is not based upon a final order of removal signed by an immigration judge or the individual is not subject to ongoing removal proceedings and the detainer was issued solely on the basis of electronic database checks.
(3) The Statutory Subclass is comprised of all current and future persons who are subject to an immigration detainer issued by an ICE agent located in the Central District of California, where the detainer is not based upon a final order of removal signed by an immigration judge or the individual is not subject to ongoing removal proceedings for whom ICE did not issue an administrative warrant of arrest at the time it issued an immigration detainer.
Plaintiff Simon Chinivizyan is a representative of the Judicial Determination Class and the Statutory Subclass.
6. Defendant ICE is a component of the Department of Homeland Security ("DHS"). Id. at 5.
B. ICE's Use of Immigration Detainers
7. An immigration detainer is a request by ICE to a federal, state, or local LEA that the agency hold an individual in custody for up to 48 hours after the person would otherwise be released so that ICE can take the person into custody. Dkt. No. 484, Admitted Fact ("AF") 3, SF 16, Trial Transcript ("Tr.") 1480:18-21 (Robbins).
8. Detainers are check-box forms developed by DHS and issued by ICE agents. See, e.g. , Trial Ex. 96. Agents rely on their research, in the field professional expertise, and, if necessary, legal counsel when issuing detainers. Trial Tr. 1441:1-25 (Robbins).
9. No judicial process is undertaken before or after a detainer is issued; however, detainers are issued along with an ICE arrest warrant. Dkt. No. 484, Admitted Fact ("AF") 5; Trial Tr. 1038:13-16 (Garibay).
The warrant accompanying the detainer is either the I-200: a warrant of arrest, or the I-205: a warrant of removal.
10. The detainer form indicates that a detainer is not valid unless served on the individual subject to that detainer. Trial Ex. 96; Trial Tr. 1422:7-1423:4 (Robbins).
11. ICE officers at ICE's Pacific Enforcement Response Center ("PERC") in Laguna Niguel, California issue detainers 24 hours a day to persons in custody in the Central District of California and after-hours to federal, state, and local LEAs in 42 states nationwide and two U.S. territories. Dkt. No. 484, SF 50; Trial Ex. 101.
ICE officers do not currently issue detainers in Alaska, Washington, Oregon, Arizona, New Mexico, Colorado, Oklahoma, and Florida.
12. PERC agents issue detainers based on electronic database searches and do not perform any other investigation outside database checks. Dkt. No 484, SF 51.
13. PERC agents do not conduct interviews of subjects before issuing a detainer. Dkt. No 484, SF 52.
14. In California, the Transparent Review of Unjust Transfers and Holds ("TRUTH") Act was passed to require LEAs to serve detainers. See Cal. Gov. Code § 7283
15. Since the inception of this lawsuit, ICE has used five different detainer forms: the December 2011 revision, the December 2012 revision, the June 2015 Form I-247D, the August 2015 Form I-247X, and the April 2017 Form I-247A. See Exs. 92-96.
16. The introduction of the detainer form marked a change in ICE's approach to removal.
C. The Shift to Automatic Detainers
17. In 1999, the United States introduced the Integrated Automated Fingerprint Identification System ("IAFIS"), a 10-rolled fingerprint identification system for use by federal, state, local, and international law enforcement and other authorized agencies. Trial Ex. 98.
18. The Automated Biometric Identification System ("IDENT") is managed by the DHS US-VISIT Program and stores and processes biometric and biographic information—including fingerprints and, where available, facial recognition data— for DHS national security, law enforcement, immigration, intelligence, and other DHS mission-related functions. Trial Ex. 47.
19. ICE connects directly to IDENT through its EAGLE interface. Dkt. No. 484, SF 91.
EID Arrest Guide for Law Enforcement.
20. IDENT contains over 237 million unique identities. Dkt. No. 484, SF 90.
21. IDENT contains fingerprint data for certain United States citizens, including those whom the FBI believes belong in the system, those who voluntarily enroll in certain "trusted traveler" programs, those who have applied to naturalize as United States citizens, and those who have filed applications for certificates of citizenship. Nemeth Dep. 26:4-27:6; 41:14-42:9; 59:15-19. 22. IDENT tracks information for each individual encounter with law enforcement. An encounter occurs when an individual's fingerprint is live captured by an ICE officer or agent. Trial Tr. 1161:13-24 (Nemeth).
23. IDENT automatically compares the biometrics from each new encounter with biometrics already contained within the database and collected from previous encounters by law enforcement officers and immigration officers. If the biometrics do not match what is already in the system, IDENT automatically assigns a new identity for the encounter and assigns the fingerprints a Fingerprint Identification Number ("FIN").
24. IDENT can determine an individual's identity even where there are typographical errors in previous encounter entries. Trial Tr. 1165:8-12 (Nemeth).
25. IDENT is a very accurate source of biometric matching. See Nemeth Dep. Tr. 67:13-20.
26. In 2006, the government began its effort to create interoperability between IAFIS and IDENT. The goal of the project was to enable the near real-time exchange of biometric and biographic information between agencies that is complete, accurate, and timely. This would allow fingerprints entered into one system to be automatically run against the other system in order to identify matches. The process enabled verification of a person's identity and facilitated the sharing of criminal and immigration information. Trial Ex. 98; Dkt. No. 484, SF 45.
27. By 2008, IDENT and IAFIS were interoperable. Dkt. No. 484, SF 102.
28. In 2008, ICE launched the Secure Communities program, which harnessed the interoperability technology as a new tool for immigration enforcement. Dkt. No. 484. SF 48.
29. The program aimed to increase the number of criminal aliens identified "[t]hrough the deployment and use of biometric-based identification systems" and anticipated that "all persons booked into custody [would] be automatically checked for their immigration status as well as prior criminal history". Trial Ex. 123 at 2-3.
30. Through Secure Communities, ICE sought to make the process of identifying aliens and issuing detainers more streamlined by requiring only "a single query by a participating local law enforcement agency (LEA)" to "confirm[ ] the identity and immigration status of a subject being processed during incarceration booking." Trial Ex. 99 at 3.
31. Secure Communities signaled a shift in ICE procedures. Prior to Secure Communities, ICE manually identified individuals for removal in jails and prisons nationwide through the Criminal Alien Program ("CAP"). Trial Ex. 123 at 4.
32. CAP placed ICE agents in all prisons and some jails in order to identify individuals potentially subject to removal through interviews and records review. Trial Tr. At 1445:8-1446:17 (Robbins); Trial Ex. 98 at ICE 220.003.
33. This process required ICE agents to interview a person to determine whether to issue a detainer. Trial Tr. 1373:11-1374:1 (Marin); ICE 30(b)(6) Philip T. Miller, Dep. Tr. 22:6-16.
34. In November 2014, Secure Communities was terminated. Dkt. No. 484, SF 20.
35. On February 20, 2017, Secure Communities was reinstated, and other field guidance regarding the enforcement of immigration laws was rescinded. See Trial Ex. 121.
36. The memorandum reintroducing Secure Communities also rescinded "all existing conflicting directives, memoranda, or field guidance regarding the enforcement of our immigration laws and priorities for removal" insofar as the materials conflicted with the United States' goal to "faithfully execut[e] immigration laws" and prioritize the removal of aliens who are convicted felons or who are involved in gang activity or drug trafficking. Id. at 2.
37. Since Secure Communities' reinstatement, several states have passed laws prohibiting compliance with detainers, while others have laws authorizing or mandating compliance. Dkt. No. 484, SF 35.
38. ICE officers continue to issue detainers to state and local law enforcement agencies, even if those agencies do not honor detainers. See Trial Tr. 1420:15-24 (Robbins).
D. Establishing Probable Cause for Removability
39. In November of 2014, ICE adopted a policy determining that it must have probable cause before requesting a person's arrest and detention on an immigration detainer. See Trial Ex. 97. The change was at least partially motivated by Fourth Amendment concerns regarding the use of immigration detainers. Id.
40. The Form I-247A, in use today, indicates that probable cause exists to support an individual's removability based on one of the following criteria: (1) a final order of removal against the alien; (2) the pendency of ongoing removal proceedings against the alien; (3) biometric confirmation of the alien's identity and a records check of federal databases that affirmatively indicate, by themselves or in addition to other reliable information, that the alien either lacks immigration status or notwithstanding such status is removable under U.S. immigration law; and (4) statements made by the alien to an immigration officer and/or other reliable evidence that affirmatively indicate the alien either lacks immigration status or notwithstanding such status is removable under U.S. immigration law. Trial Ex. 96.
41. The I-247A form retained the probable cause language of the I-247D and I-247X, and requests that the receiving law enforcement agency "maintain custody" of the individual for a "period NOT TO EXCEED 48 HOURS." See Trial Exs. 94-96.
The form reads: "DHS has determined that probable cause exists that the subject is a removable alien. This determination is based on: ... biometric confirmation of the alien's identity and a records check of federal databases that affirmatively indicate, by themselves or in addition to other reliable information, that the alien either lacks immigration status or notwithstanding such status is removable under U.S. immigration law ... [.]"
42. Paramount in determining probable cause of removability is identifying that a person is not a U.S. citizen. See, e.g. , Trial Tr. 217:13-16 (Stock) ("The most important thing to do before you decide whether someone's going to suffer immigration consequences ... is to figure out whether the person is a citizen first"); Dkt. No. 481-11 (ICE 30(b)(6) Schichel Dep. Tr. 164:14-18) (stating that confirming a person's alienage is "part of establishing probable cause" every time officers investigate a potential subject for a detainer).
43. Establishing probable cause of removability also requires a determination of a person's current immigration status. Trial Ex. 96 (noting, in pertinent part, that probable cause can be based on checks that indicate "that the alien either lacks immigration status or notwithstanding such status is removable under U.S. immigration law"); see also Dkt. No. 479-3 (Schichel Dep. Tr. 157:13-22).
44. ICE uses IDENT to satisfy the biometric confirmation described in the I-247A. Trial Tr. 1418:20-1419:2 (Robbins). ICE relies on an individual's fingerprints, which are taken at the time of arrest. Id.
45. In addition, law enforcement specialists at the Law Enforcement Support Center ("LESC") and analysts at the PERC search multiple databases to find "affirmative evidence of removability". Trial Tr. 1419:3-10 (Robbins).
E. Determining Citizenship
46. An individual's citizenship and immigration status is not static and may change multiple times over a lifetime. See, e.g. , Trial Tr. at 153:22-159:14 (Bacon).
47. Accordingly, a person's citizenship or immigration status at some fixed point in time may not be a reliable indicator of their current citizenship status. Trial Tr. 1092:6-17 (Garibay) (testifying that while a person may not have been a United States citizen in 2010, that has no bearing on whether they are a citizen presently).
48. A foreign-born person can become a United States citizen in one of three ways: (1) naturalization, (2) acquisition, or (3) derivation. Trial Tr. 162:22-164:21 (Bacon).
49. Generally, a person must be a lawful permanent resident ("LPR") to naturalize. Trial Ex. 131 at 1; Trial Tr. 512: 21-24 (Johansen-Mendez). People who serve or served honorably in the U.S. armed forces may be excused from this rule. See 8 U.S.C. §§ 1439 - 40.
50. In addition, a person may pass in and out of lawful or unlawful status; a common example of this is a child's derivation of citizenship. Trial Tr. 163:4-11 (Bacon).
51. Determining whether a person acquired or derived citizenship is a complex inquiry and illustrates the fluidity one may experience with respect to immigration status. See Trial Tr. 242:20-255:23 (Stock).
52. For derivative citizenship, the analysis can turn on, among other things: which parent is a citizen, when that parent became a citizen, whether the person's parents were married, whether and when the U.S. citizen parent lived in the United States and for how long, whether the father legitimated the child, whether the child lived in the custody of the U.S. citizen parent or parents, and at what point the child lived in the custody of the U.S. citizen parent(s). Id.
53. In some cases, a determination on derivative citizenship may depend on knowing whether a person's grandparent(s) were U.S. citizens or whether a person or their parent(s) served in the U.S. Armed Forces. Id.
F. Determining Immigration Status
54. Similarly, determining a person's immigration status may be a complex inquiry requiring a thorough examination of the individual's history.
55. Aside from foreign born-U.S. citizens, there are four major categories of foreign-born persons who lawfully reside within the United States: (1) immigrants, (2) nonimmigrants, (3) refugees/asylees, (4) and non-citizens who otherwise have permission to be present. Trial Tr. 148:17-162:19 (Bacon)
56. Immigrants are non-citizens with permission to reside permanently in the United States. Generally, these are lawful permanent residents ("LPRs"). Trial Tr. 149:9-18 (Bacon). United States Citizenship and Immigration Services ("USCIS") commonly grants LPR status, but the Department of State ("DOS") may also grant it. Trial Tr. 144:20-148:15, 149:19-150:1 (Bacon).
57. An LPR's status does not change unless an immigration judge determines that status should be taken away. Trial Tr. 531:14-16 (Johansen-Mendez).
58. Non-immigrants are non-citizens admitted to the United States on a visa for a specific purpose and/or period of time, such as for tourism, employment or study. Trial Tr. 151:14-154:20 (Bacon).
59. There are 23 statutory categories of nonimmigrant visas, which DOS generally grants. Trial Tr. 154:5 (Bacon); Trial Ex. 69. USCIS generally handles status changes after nonimmigrants enter the United States. Trial Tr. 155:2-156:3 (Bacon) (describing an example of a status change from F1 student visa to H1-B employment visa, as adjudicated by USCIS).
60. If a nonimmigrant violates the condition(s) of their status, such as by overstaying their authorized period of stay, their lawful status terminates. See generally 8 U.S.C. § 1227(a)(1)(C)(i) (rendering deportable any "nonimmigrant ... who has failed to maintain the nonimmigrant status in which the alien was admitted or to which it was changed ..., or to comply with the conditions of any such status.").
61. Refugees and asylees are non-citizens who receive permission to reside in the United States based on a well-founded fear of persecution in their native country. Trial Tr. 159:18-160:8 (Bacon).
62. Refugee status is granted outside the United States and involves different agencies, including the United Nations and ultimately the DOS and USCIS. Trial Tr. 160:9-16 (Bacon).
63. Asylum is granted inside the United States by either USCIS or, if the person is placed in removal proceedings, by the Executive Office for Immigration Review (EOIR). Trial Tr. 160:15-16 (Bacon); 587:4-20 (Johansen-Mendez).
64. Once granted, asylum is "essentially a permanent status unless something's done to change it, unless [the asylee] adjust[s] to lawful permanent residence or unless there's cause to terminate that asylee designation." Trial Tr. 516:2-7 (Johansen-Mendez).
65. Those permitted to be present in the United States who otherwise lack formal status are non-citizens granted special protected status. Trial Tr. 161:4-162:19 (Bacon).
66. The two largest protected groups of non-citizens are those who USCIS has granted Temporary Protected Status ("TPS") and Deferred Action for Childhood Arrivals ("DACA"). Id. ; Trial Ex. 72 at 1 (699,350 active DACA recipients); Trial Ex. 73 at 5 (approximately 437,000 TPS holders).
G. Two-Step Determination of Probable Cause
67. The process ICE undertakes to determine probable cause for removal requires analysts at the PERC and officers at LESC to take two steps.
68. First, officers and analysts rely on IDENT to match biometric information.
69. IDENT creates a new biometric input whenever a law enforcement officer has an encounter with an individual. Trial Tr. 1161:13-24 (Nemeth).
70. When an individual is arrested by an LEA on a criminal charge, the person's fingerprints are sent to the FBI and automatically run against IAFIS and IDENT. Dkt. No. 484 SF 45-46.
71. If there is a fingerprint match in IDENT, a notification—an Immigrant Alien Query ("IAQ")—is automatically generated and sent to LESC. Trial Tr. 1436:7-17 (Robbins). Accordingly, if an individual has numerous encounters with law enforcement, IDENT would capture all biometric and biographical data on that individual, regardless of typographical errors.
This would mean DHS has some record of a prior encounter with that person, including, among other things, an encounter as the person entered the country, or an application for an immigration benefit.
Thus, regardless of whether an individual has a common name spelling, or whether a field officer made errors when inputting biographical information, a person's FIN number will remain the same across multiple encounters and will be able to be located on IDENT through its search inquiry.
72. ACRIMe automatically searches the NCIC, NLETS, CIS, CLAIMS 3, CLAIMS 4, EID, IDENT, ADIS, SEVIS, and EOIR databases to match the FIN submitted with other encounters for that same FIN. Trial. Tr. 930:22-931:23; 933: 8-21 (Garibay).
ACRIMe (Alien Criminal Response Information Management System) is a web-based system that includes two primary applications:
(1) ACRIMe Mod: which is used by the LESC to process IAQs and transmit IARs back to the LEA and responsible ERO field office(s) or the PERC
(2) ACRIMe Field: which is used by the PERC and ERO Field Offices to see the results of the LESC's integrated records checks, process IARs, and initiate immigration enforcement, if applicable.
See Trial Ex. 78 at ICE029605.
73. ACRIMe then automatically generates an Immigrant Alien Response ("IAR") from information stored on the ten databases; the IAR is the automatically delivered to the PERC. Trial Tr. 930:3-5 (Garibay).
74. The IAR contains basic biographic information and criminal history, as well as a short statement about immigration status and removability.
The short statement provides only limited insight into an individual's immigration or citizenship status, and can often contain speculative language on a person's status. See e.g. Ex. 170 at ICE 0023957 ("D.H.S. records indicate that the subject was legally admitted to the U.S. as a non-immigrant until the date noted above. If this person has remained longer in the U.S. without D.H.S. permission, this person may be in violation of immigration laws. If the departure date is indicated above, no electronic record of a new entry has been found. Subject appears to be a non immigrant overstay.") (emphasis added).
75. Next, an analyst conducts a first level review of the IAR and makes a recommendation to an ICE deportation officer about whether a detainer should be issued. Dkt. No. 484 SF 54, 58.
It is important to note that analysts are not federal employees nor immigration agents, rather they are contractors.
76. The analysts need not search each database independently. Trial Tr. 933:5-9 (Garibay).
77. However, the analyst may, at their discretion, run independent database checks. See Trial Tr. 1438:19-1439:11 (Robbins) ("... on complex cases [analysts] have access to go through multiple databases to make sure they get it right, and they also have the ability to talk to legal counsel, their supervision, the [deportation officers], or the officer's supervisors"); see also Trial Tr. 1437:19-1438:2 (Robbins) ("The analyst at the PERC will take the case. They will use [ACRIMe] and other databases to make a recommendation on that particular case—an enforcement recommendation or a no action recommendation, to the deportation officer.")
H. The Databases
78. ICE does not rely on one single database to determine citizenship and immigration status.
79. When Plaintiff Gonzalez filed this case, agents at the PERC issued detainers based on searches of IDENT and four other databases: CIS, CLAIMS 3, TECS, and ENFORCE/EARM. Ex. 102 at ICE 652.
80. In December 2017, the PERC adopted ACRIMe Field, which searches IDENT and nine other databases: CIS, CLAIMS 3, CLAIMS 4, ADIS, SEVIS, EOIR, EID, NCIC, and NLETs. Dkt. No. 484, SF 56; Trial Tr. 933:8-21 (Garibay).
81. CIS—the Central Index System, is a database created by USCIS and contains biographical information on individuals. Trial Tr. 175:3-14 (Bacon).
82. CLAIMS 3 is a database ICE relies on for information on immigration benefits applications. CLAIMS 3 allows an ICE officer or analyst to review a person's applications for, among other things, lawful permanent residency, temporary protected status, and work authorization. Trial Tr. 510:16-21 (Johansen-Mendez).
83. CLAIMS 3 also provides information on how an application was adjudicated (grant or deny), and whether a credential (e.g., employment authorization card) was generated a result of the adjudication. Trial Tr. 510:22-511:2 (Johansen-Mendez).
84. CLAIMS 4 exclusively holds information about naturalization applications and provides information on how the application was adjudicated. Trial Tr. 511: 16-19 (Johansen-Mendez).
85. ADIS provides arrival and departure information for non-immigrant visa holders. ADIS pulls its arrival and departure information from TECS and IDENT. Dkt. No. 484, SF 77; Trial Ex. 54 at 12-13; Trial Tr. 513:1-514:17 (Johansen-Mendez).
Arrival and Departure Information System.
86. ICE relies on other databases to supplement information it receives from CIS, CLAIMS 3, CLAIMS 4, and ADIS.
Coupled with those discussed in substance, ICE also reviews:
PCQS (or "Person Centric Query Search" which provides a "snapshot" of results from CIS and CLAIMS but with a "little less detail") see Trial Tr. 967:10-21 (Garibay); Trial Tr. 1515:9-17 (Johansen-Mendez);
SQ11 (a lookout alert system, e.g., for individual's previously deported) (Trial Tr. 969:4-16 (Garibay);
SQ94 (provides information on individuals that were issued an I-94 when admitted into the country) Trial Tr. 969:17-25 (Garibay);
ELIS 1 & 2 (provide extremely limited information on immigration benefits) Trial Tr. 1512:8-1513: 4 (Johansen-Mendez);
TECS (a portal to NCIC—a database related to criminal history—and ADIS);
California Birth Index (database containing information (i.e., date of birth, full name, country of birth, gender, and mother's maiden name) on U.S. citizens born in California; however, contains no information on births after 1995. See Trial Ex. 228; and
CCD (contains information about visas that noncitizens have applied for; may contain information about an individual's passport) Trial Tr. 1532: 9-18 (Johansen-Mendez).
87. SEVIS , a database maintained by ICE, contains records of non-citizens who enter the United States as students and exchange visitors (F, M, and J visas), as well as their dependents. Dkt. No. 484, SF 65.
Student and Exchange Visitor Information System.
88. EOIR provides basic information about immigration court proceedings. EOIR does not reflect how the proceeding was adjudicated, simply that the case was terminated. Dkt. No. 484, SF 67; Dkt. No. 481-2 (Cano Dep. Tr. 66:6-12).
89. EID contains information about previous encounters with ICE officers. Trial Tr. 951:1-10 (Garibay). Specifically, EID provides information on whether a person is currently in removal proceedings and whether a person has been previously removed or deported. Id.
90. EID is a repository for all records created, updated, and accessed by other software applications, including ENFORCE and EAGLE. Trial Tr. 951:4-952:21 (Garibay).
91. NCIC and NLETS are criminal databases and do not provide any information for establishing a person's citizenship or immigration status. These databases are relevant for removability purposes. See e.g. , 8 U.S.C. §§ 1182(a)(2)(A)-(I) ; Trial Tr. 607:7-608:10 (Johansen-Mendez).
National Crime Information Center.
National Law Enforcement Telecommunications System.
I. Shortcomings in the Databases
92. The databases on which ICE relies for information on citizenship and immigration status often contain incomplete data, significant errors, or were not designed to provide information that would be used to determine a person's removability.
93. The CIS database is the "spine" of USCIS's functions. Trial Tr. 137:6 (Bacon).
94. CIS was first established in 1985 and it provides basic biographical information on an individual. Trial Tr. 175:3-14 (Bacon).
95. CIS contains five pieces of data that analysts or officers may use in determining citizenship or immigration status: (1) first name, (2) last name, (3) date of birth, (4) country of birth, (5) and classification when the individual entered the United States. Id.
96. The five pieces of biographical information recorded on CIS are not independently verified. Trial Tr. 176:3-12 (Bacon).
97. Indeed, data entered into CIS is often erroneous. Dkt. No. 481-5 (ICE 30(b)(6) Dep. Hamm Tr. 219:17-220:2); see also Trial Tr. 521:8-524:1, 527:3-7, 523:13-19 (Johansen-Mendez).
98. In 1995, the Government Accountability Office ("GAO") described CIS as "incomplete and inaccurate," noting that 22 percent of records "contained either misspelled names, incorrect name order, or incorrect nationality." Dkt. No. 484, SF 104; Trial Ex. 62 at 3.
99. Moreover, a 2012 DHS study reported that CIS's class of admission field was wrong for 12 percent of individuals it studied. Trial Ex. 58 at 1.
100. Many of the errors in the class of admission field result from USCIS's failure to update CIS when a person's immigration status changes. The class of admission field is "derived from DHS's administrative action related to a person's initial application/petition or apprehension." Trial Ex. 43 at 7.
101. Updates to the class of admissions field are meant to occur automatically when other databases that feed information to CIS are updated, but there are often problems with these uploads. Id. ; Trial Ex. 251 at 30 ("CIS does not always contain the latest status updates from USCIS benefits case tracking systems ... In these cases, uploads to the CIS may be delayed by data input and uploading processes, or the transfer of data between systems may fail initially."); id. at xi ("Among other issues, when decisions concerning noncitizens' statuses, such as changes affecting employment status and length of authorized stay in the United States, are updated in one system, one or more uploads from a case tracking system are required for them to appear in the Central Index System (CIS). This process delays their availability in CIS and introduces opportunity for error , especially if there are problems with uploads.") (emphasis added); Trial Tr. 522:1-9 (Johansen-Mendez) (CIS often "would have an old class of admission that was no longer updated.").
102. In an April 10, 2012 email to ICE Headquarters, Senior ICE official David Marin wrote: "For the past couple of months, AFOD Martinez has noticed an increase in the volume of biometric Immigration Alien Responses (IAR) for individuals that are determined, via agency database checks, to be removable, only to later discover that the person is a United States citizen." Dkt. No. 484, SF 107; Trial Ex. 14 at 1. Mr. Marin then asked for a "mechanism in which we can modify the CIS database to indicate the individual's true nationality and citizenship " because "[w]e have found that it is not easy to get USCIS to change someone's status in CIS. " Id. (emphasis added).
103. In a follow-up email on May 9, 2012, Marin wrote that "[US]CIS is not updating its databases, so when we run our queries it shows the subject is an [Lawful Permanent Resident], we take appropriate action. This case appears to be exactly that; CIS databases show the subject as a LPR, so the hold was properly placed based on the information we had at the time. Now the subject claims to be a [United States Citizen] and has a [naturalization certificate]; if this is true and the Central Index System was timely updated, we may not have run into this issue. Unfortunately these types of cases occur frequently. " Dkt. No. 484., SF 108; Trial Ex. 15] (emphasis added).
104. DHS is aware of the errors in the class of admission field in CIS. See Trial Ex. 43 at 7 (if law enforcement seeks to detain an individual based on the class of admission code in CIS, they would request the physical A-File for further investigation); Trial Tr. 186:2-187:5 (Bacon) (same); Trial Ex. 244 at 6 (advising SAVE agencies not to rely on the class of admission code alone because it can be wrong, but to view it together with other information, such as immigration documents).
105. Further, CIS's primary function is not to provide indicia of citizenship or to establish probable cause for removal; CIS is meant to direct users to a person's A-File. Trial Tr. 140: 23-24 (Bacon) ("CIS has only one real function and that's to direct people to where the A-File is."); Trial Tr. 420:25 – 421:4 (Meinhardt) ("the main reason for accessing [CIS] is to pin down not only just the bare essentials information, but really the A-File number"); see also Trial Ex. 42 at 4 ("The purpose of CIS is to provide a searchable central index of A-Files and to support the location and transfer of A-Files among DHS personnel and offices as needed in support of immigration benefits and enforcement actions.").
106. CLAIMS 3 and CLAIMS 4 are USCIS case management systems designed to track and process applications for certain immigration benefits and naturalization. Trial Ex. 41 at 2; Trial Ex. 42 at 1.
107. CLAIMS 3 is an "old, legacy, mainframe system[ ] that do[es] not have the capability to interface in real-time with other systems." Id.
108. CLAIMS 3, like CIS, contains information that is often incorrect. See Dkt. No. 481-5 (ICE 30(b)(6) Hamm Dep. Tr. 220:14-19) ("[t]he information is outdated. It's antiquated. It's not updated appropriately.").
109. Further, individuals familiar with CLAIMS 3 consider the database's error rate to be close to 30 percent. See id. ; see also Trial Tr. 791:7-792:17 (Corrales); Trial Ex. 237 at 12 ("USCIS' CLAIMS 3, used to store immigration benefits application data, was not updated in a timely manner").
110. Both CLAIMS 3 and CLAIMS 4 destroy information after 15 years. Trial Tr. 789:16-18 (Corrales); Trial Tr. 773: 13-15 (Corrales).
This means that neither CLAIMS database contains any information from before September 27, 2004.
111. As a result, any individuals who obtained LPR status before 2004 and have not had any further interaction with USCIS would not be located in a search of CLAIMS 3. This information would only be located in CIS, which has already been discussed as an error-laden source of information.
112. ADIS is intended as a system for generating leads on visa overstays to be further investigated including through non-database sources. Trial Ex. 54 at 2 (purpose of ADIS is "[f]acilitating the identification and investigation of individuals who may have violated their terms of admission.)
113. ADIS has its limitations as a source of information in that no comprehensive system to record entry and exit information has been adopted and there is limited information on individuals who enter at land borders. Trial Ex. 237 at 7, 18-21; see also Trial Ex. 236 at 3, 12-13.
114. Indeed, Customs and Border Protection ("CBP") policy is not to take any enforcement action "based solely on ADIS (or any other system-generated) alerts," and to mitigate risk of error by "corroborat[ing] all information [in ADIS] before taking an adverse action" against a person. Trial Ex. 54 at 14.
115. As recently as 2017, the DHS OIG found that ADIS incorrectly identified visa overstays more than 42 percent of the time. Trial Ex. 237 at 24. Twenty-five percent of people ADIS and SEVIS identified as visa overstays had actually applied for and/or received permission to extend their stay from USCIS. Id. Seventeen percent had in fact departed the country as required, but may have returned on a subsequent visit with permission on a new, non-immigrant visa. Id.
116. In contrast, ICE relies solely on ADIS for information about non-immigrants entering and exiting the country on visas without corroborating the information stored on ADIS.
117. Without a comprehensive collection of information about both exits and entries, ADIS has severe limitations as a source of information about who has entered the country on a visa and whether they have overstayed that visa.
118. SEVIS contains information on three types of visa holders, and shares the same entry/exit information in ADIS. Dkt. No. 484, SF 65; Trial Ex. 54 at 3.
119. TECS, the predecessor to ADIS, also has limitations on its information. See Trial Ex. 251 at A-17 ("TECS is the most error prone database accessed by E-Verify, which particularly affects the ability to accurately verify nonimmigrants"); Trial Ex. 236 at 12 ("Because of concerns about the reliability of the department's overstay data, neither DHS nor its predecessor has regularly reported annual overstay rates to Congress since 1994.").
120. In addition to the numerous errors in the databases discussed above, a number of immigration and citizenship statuses are either not captured, or captured on databases with dubious reliability. 121. As an initial matter, there is no national database of all U.S.-born citizens.
While CCD may contain some passport information, it is not a broad-reaching database that captures all U.S.-born citizens.
122. Similarly, ICE has never had access to any database of derivative or acquired citizens, because none exists. Dkt. No. 484, SF 115.
123. With respect to asylees and refugees, ICE could potentially derive information in CIS's class of admissions, but only to the extent the information captured in the class of admissions field is either properly updated or completely filled out.
124. ICE is not required to search the Refugees, Asylum and Parole System ("RAPS") database—the database containing information about refugee and asylum applications. Trial Ex. 53 at 66.
125. ICE relies upon CLAIMS 3 for information on TPS and DACA. Dkt. No. 484, SF 74.
126. All told, the collection of datapoints ICE gathers from the various databases does not provide affirmative indicia of removability to satisfy probable cause determination because the aggregation of information ICE receives from the databases is largely erroneous and fails to capture certain complexities and nuances of immigration law.
127. ICE does not look at databases for a single record that affirmatively indicates probable cause for removal, but instead weaves together a patchwork web of databases owned and maintained by separate and distinct agencies to make complex determinations about citizenship, immigration status, and removability. No single database contains all the information needed to make a probable cause determination; no one piece of information provides ICE probable cause to issue a detainer to members of the Probable Cause Subclass, and the supplementation of additional databases—each with its own deficiencies—does not provide sufficient indicia of probable cause.
J. The Effect of ICE's Reliance on the Databases for Probable Cause Determinations
128. Errors in the databases ICE reviews to make its probable cause determinations for removal have, on multiple occasions, led to arrests of U.S. citizens and lawfully-present non-citizens.
129. Data produced by ICE during the period of May 2015 to February 2016 reveals that of the 12,797 detainers issued during that time frame, 771 were lifted because the individuals were either U.S. citizens or otherwise not subject to removal. See Trial Tr. 812:12-817:25 (Corrales); Trial Ex. 125.
130. Of those 771 detainers lifted, 42 explicitly provide that the detainer was lifted because the individual was a U.S. citizen. Trial Ex. 124.
131. Further, evidence introduced in this case includes dozens of additional examples of U.S. citizens wrongly subject to detainers because of database errors. See Trial Ex. 163. These examples illustrate the real-life impact of the relying on a makeshift set of databases to make citizenship and immigration status determinations.
The narratives resolving these improperly issued detainers include language highlighting the fallible nature of the databases. See e.g. , Trial Ex. 175 (detainer lifted because CIS wrongly showed lifted LPR status) ("Upon review of database information it was determined that subject was amenable to DHS/ICE enforcement action with the following status: LPR-SA1. Based on this information an ICE Immigration Detainer was lodged. On 12/05/2015, the detainer was lifted ... [.] Subject appears to have derived citizenship in the United States."); Trial Ex. 187 (detainer lifted two weeks later after second level review of subject's file at PERC because subject derived citizenship from naturalized parent) ("Upon review of database information it was determined that subject was amenable to DHS/ICE enforcement action as a LPR. Based on this information an ICE Immigration Detainer was lodged .... [Second level review] performed by PERC, subject had perceedings [sic] terminated in 2013 due to his probative claim to [citizenship] ... father [naturalized] on 10/11/2000 when subject was 16, subject autoderived on 02/27/2001 when [the Child Citizenship Act of 2000] went into effect."); see also Trial Ex. 202 (detainer lifted because subject was naturalized in 1983 and no database had information on subject's naturalization application); Trial Ex. 164 (detainer issued to U.S. born citizen); Trial Ex. 165 (detainer issued to U.S. born citizen based on database checks determining subject had entered without inspection, two months later, second level review determined subject's U.S citizenship)
132. The cited instances of citizens being misclassified are not exhaustive. Indeed, practitioners in immigration law echo the frequency with which ICE issues detainers to U.S. citizens and lawfully present non-citizens. See Trial Tr. at 230:7-232:22, 240:17-242:12 (Stock) (describing routine database error that leads to derivative citizens being wrongly identified as lawful permanent residents); Trial Ex. 36 (describing same); Trial Tr. 916:7-20 (Corrales) (testifying that while counsel at ICE, she was tasked with reviewing 50 to 200 memos regarding potential claims to U.S. citizenship arising solely out of the Los Angeles area in a two year period.).
133. ICE endeavors to reduce its error rates when issuing detainers. See Trial Tr. 1469:19-20 (Robbins).
K. Alternate Sources of Information
134. Biometric verification and database checks are not the only methods for ICE to gather information that could support probable cause for removal.
135. As discussed above, while CIS provides cursory biographical information, its primary function is to direct users to a person's A-File. Trial Tr. 420:25-421:4 (Meinhardt) ("the main reason for accessing [CIS] is to pin down not only just the bare essentials information, but really the A-File number"); see also Trial Ex. 42 at 4 ("The purpose of CIS is to provide a searchable central index of A-Files and to support the location and transfer of A-Files among DHS personnel and offices as needed in support of immigration benefits and enforcement actions.").
136. An A-File is a physical file that contains hard copy documents on a person's immigration status. Trial Tr. 169:18-170:12 (Bacon). The A-File is capable of painting a more robust picture of a person's current immigration status.
137. USCIS maintains millions of A-Files and has millions more in digitized format.
138. Further, physical A-Files may be delivered by expedited mail without serious difficulty.
139. Moreover, the most reliable source of information on a subject's immigration or citizenship status is the individual. Trial Tr. 369:9-13 (Stock) ("I think you need to interview the person .... Because often, when you interview the person, you discover facts that would cause you to realize that they are a United States citizen."); this is because some questions regarding an individual's status can only be answered during the interview process. Trial Tr. 1364:4-13 (Marin) (testifying that "in some cases" there is information that would only be available to ICE through an interview of a person). 140. The interview process is not new to ICE; face-to-face interviews and jail checks were ICE policy prior to the introduction of Secure Communities. See Dkt. No. 481-7 (ICE 30(b)(6) Philip T. Miller, Dep. Tr. 22:6-16; Trial Tr. 1373:11-1374:1 (Marin)
I. The SAVE and E-Verify Programs
141. SAVE and E-Verify are two long-standing DHS programs designed to verify immigration status and employment status, respectively, using electronic databases. Trial Tr. 624:11-625:11 (Winkler). The programs use the same core databases ICE relies upon, as well as several others. See Trial Exs. 240 at 4; 241 at 8; 242 at 20-24; 243 at 4.
142. SAVE and E-Verify provide an instructive comparison to ICE's use of databases to issue detainers. Both programs take into account the significant challenges to using databases that are incomplete and error-prone when verifying immigration and employment status. As a result, as Rachel Winkler—who served in USCIS as a subject matter expert on verification programs—explained, neither program uses databases on their own to determine a person's immigration status. Trial Tr. 646:15-648:7 (Winkler).
143. To account for systemic errors in the databases, both SAVE and E-Verify have built in layers of review. The programs incorporate manual database searches and paper records reviews to resolve disparities between records provided by an individual and records in the databases. Trial Ex. 244 at 2; Trial Tr. 657:3-660:4 (Winkler).
144. Both SAVE and E-Verify have been subject to significant government oversight, including studies to evaluate the reliability of the databases used. Trial Tr. 663:22-664:16, 672:3-14 (Winkler).
145. One study commissioned by USCIS in 2012 and conducted by Westat, examined E-Verify's accuracy rate at each of the different levels of review. Trial Ex. 251.
146. The study found that, of 8.2 million cases submitted in fiscal year 2009, only 58 percent of eligible workers could be accurately verified during E-Verify's automated review level. Id. at xi.
During automatic searches, E-Verify "checks some systems to determine if employment authorization can be confirmed." Trial Ex. 251. The databases reviewed at the automatic review level include, among others: CIS, TECS and CLAIMS 4. Id.
147. The study also found that 90 percent of eligible workers could be accurately verified during second level review, at which stage USCIS employees take 24 hours to run manual checks of the database. Id. at xiii; id. at 6.
148. At the final stage of review, which involves the review of paper immigration documentation, E-Verify could accurately verify 94 percent of eligible workers. Id. at x ("It is estimated that 94 percent of FNCs were accurately issued to unauthorized workers and 6 percent were inaccurately issued to employment-authorized workers.").
149. The highest percentage (35 percent) of employment-authorized workers who could not be verified at all were foreign-born U.S. citizens. Trial Ex. 251 at 48.
150. The report cited two reasons for this: (1) the absence of most naturalization data in DHS systems from before the mid-1990s (which is consistent with CLAIMS 4's 15-year retention schedule, given that 15 years before 2012 was 1997) and (2) the unavailability of information about derivative citizens. Id. at 76 ("Most USCIS naturalization data before the mid-1990s are still not reflected in USCIS systems. Data are also frequently not available for workers who derived U.S. citizenship status."); see also id. at 51 ("Retaining accurate information on noncitizens who have become U.S. citizens remains a major challenge to E-Verify ...."); id. at A-8 (derivative citizenship information "not normally entered into an automated system").
III. CONCLUSIONS OF LAW
1. This Court has jurisdiction under 28 U.S.C. § 1331.
2. Venue is proper in this district as a significant portion of the events underlying the Complaint occurred within the Central District of California, and Plaintiffs are part of a class of individuals who have been or will be subject to an immigration detainer issued by an ICE agent located in this District.
Establishing Fourth Amendment Probable Cause
3. The Fourth Amendment protects against "unreasonable searches and seizures." U.S. Const. amend. IV.
4. To prevail on their unlawful seizure claim, Plaintiffs must prove that class members have been or will be seized based solely on their ICE detainers without lawful authority or probable cause. See, e.g. , Terry v. Ohio , 392 U.S. 1, 38, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968) ("The infringement on personal liberty of any ‘seizure’ of a person can only be ‘reasonable’ under the Fourth Amendment if we require the police to possess ‘probable cause’ before they seize him."); see also Velazquez v. City of Long Beach , 793 F.3d 1010, 1018 (9th Cir. 2015) (a "claim for unlawful arrest is cognizable ... as a violation of the Fourth Amendment, provided the arrest was without probable cause or other justification") (internal quotation marks omitted)
5. The probable cause standard is "incapable of precise definition or quantification into percentages because it deals with probabilities and depends on the totality of the circumstances." Maryland v. Pringle , 540 U.S. 366, 371, 124 S.Ct. 795, 157 L.Ed.2d 769 (2003).
6. Thus, "[t]o determine whether an officer had probable cause to arrest an individual, we examine the events leading up to the arrest, and then decide ‘whether these historical facts, viewed from the standpoint of an objectively reasonable ... officer, amount to’ probable cause." Id. ; see also Gerstein v. Pugh , 420 U.S. 103, 111, 95 S.Ct. 854, 43 L.Ed.2d 54 (1975) (probable cause requires "facts and circumstances ‘sufficient to warrant a prudent man in believing that the (suspect) had committed or was committing an offense.’ ")
Computer Databases May be Used to Establish Probable Cause
7. Law enforcement agencies may rely on computer databases to establish probable cause if it is reasonable for them to do so. See Herring v. United States 555 U.S. 135, 146-47, 129 S.Ct. 695, 172 L.Ed.2d 496 (2009) (declining to apply exclusionary rule because an officer's reliance on a computer database is reasonable where no evidence of routine or widespread errors on the database exists); id. at 146, 129 S.Ct. 695 ("In a case where systematic errors were demonstrated, it might be reckless for officers to rely on an unreliable warrant system."); Arizona v. Evans , 514 U.S. 1, 17, 115 S.Ct. 1185, 131 L.Ed.2d 34 (1995) (O'Connor, J., concurring) ("Surely it would not be reasonable for the police to rely ... on a recordkeeping system ... that routinely leads to false arrests.") (emphasis in original). Probable Cause is Required to Issue Immigration Detainers
8. The Fourth Amendment applies to immigration arrests generally, or arrests on detainers specifically. See e.g. , Tejeda-Mata v. INS , 626 F.2d 721, 724-25 (9th Cir. 1980) (applying "the constitutional requirement of probable cause" to immigration arrests); Morales v. Chadbourne , 793 F.3d 208, 211 (1st Cir. 2015) (holding an ICE agent must have probable cause to issue an immigration detainer).
9. The Fourth Amendment's protection against unreasonable seizure in the context of a warrantless detainer serves an exceedingly important function in the immigration context because many of the backstops that exist in the criminal justice system are absent in the immigration system. See e.g. , 8 U.S.C. § 1226(c) (no right to pre-trial release on bail or bond); 8 U.S.C. § 1229a(b)(4)(A) (no right to an attorney at the government's expense in removal proceedings); Dkt. No. 484, AF 5-7; Dkt. No. 264 at 2-3 (no judge or neutral arbiter ever reviews an ICE official's probable cause determination to issue a detainer or make an arrest).
10. Given the lack of additional safeguards in the immigration context, the initial process ICE officials utilize to make arrests and issue detainers must be sufficiently supported by lawful authority and probable cause.
ICE Violates the Fourth Amendment by Issuing Detainers through State and Local Officers Who Lack Authority to Make Civil Immigration Arrests
11. The Court first considers whether ICE violates the Fourth Amendment by issuing detainers through state and local law enforcement agencies that independently lack the authority to issue immigration detainers.
12. State and local law enforcement "generally lack authority to arrest individuals suspected of civil immigration violations." Santos v. Frederick Cnty. Bd. of Com'rs , 725 F.3d 451, 465 (4th Cir. 2013)
13. While "federal law does not preclude local enforcement of the criminal provisions" of federal immigration law, Gonzales v. City of Peoria , 722 F.2d 468, 475 (9th Cir. 1983), overruled on other grounds by Hodgers-Durgin v. de la Vina , 199 F.3d 1037 (9th Cir. 1999), the Immigration and Nationality Act ("INA") expressly entrusts the enforcement of civil immigration law to federal agents and officers. Arizona v. United States , 567 U.S. 387, 408, 132 S.Ct. 2492, 183 L.Ed.2d 351 (2012).
14. Thus, where state officers may perform the functions of immigration law, federal law makes clear those circumstances. Id. ("Federal law specifies limited circumstances in which state officers may perform the functions of an immigration officer.").
15. A number of statutes highlight the scope of authority federal law generally grants to state and local governments. See 8 U.S.C. § 1357(g)(1) (granting immigration enforcement authority to state or local government officials in a formal agreement with a state or local government.); 8 U.S.C. § 1103(a)(10) (authority may be extended in the event of an "imminent mass influx of aliens arriving off the coast of the United States).
16. Case law also suggests that state and local law enforcement agencies must consent to the delegation of federal immigration functions. See Michigan v. DeFillippo , 443 U.S. 31, 36, 99 S.Ct. 2627, 61 L.Ed.2d 343 (1979) ("Whether an officer is authorized to make an arrest ordinarily depends, in the first instance, on state law.").
17. This is true even when it comes to arrests for federal offenses. See Ker v. State of Cal. , 374 U.S. 23, 37, 83 S.Ct. 1623, 10 L.Ed.2d 726 (1963) ("This Court, in cases under the Fourth Amendment, [has] long recognized that the lawfulness of arrests for federal offenses is to be determined by reference to state law insofar as it is not violative of the Federal Constitution."); Gonzales v. City of Peoria , 722 F.2d at 475 (assessing lawfulness of arrest by local police for unlawful entry in violation of 8 U.S.C. § 1325 by first concluding federal law permitted the arrest and then asking "whether state law grants Peoria police the affirmative authority to make [the] arrest[ ]"); see also Santos v. Frederick Cnty. Bd. of Com'rs , 725 F.3d 451, 464 (4th Cir. 2013) ("[l]ocal law enforcement officials may detain or arrest an individual for criminal violations of federal immigration law without running afoul of the Fourth Amendment, so long as the seizure is supported by reasonable suspicion or probable cause and is authorized by state law.")
18. A fundamental tenet of federalism requires states to determine the powers and responsibilities of their own officers and any attempt to subvert states' control over their law enforcement runs afoul of the Tenth Amendment. Cf Printz v. United States , 521 U.S. 898, 922, 928-32, 117 S.Ct. 2365, 138 L.Ed.2d 914 (1997) (holding the Tenth Amendment prevents the federal government from "impress[ing] into its service—at no cost to itself—the police officers of the 50 States").
19. Thus, even where federal law permits state or local officers to make civil immigration arrests, the authority for such arrests must come from state law. See 8 U.S.C. § 1357(g)(1) (permitting federal-state agreements authorizing state and local officials to perform immigration enforcement functions, but only "to the extent consistent with State and local law."); 8 U.S.C. § 1252c(a) (permitting arrest "to the extent permitted by relevant State and local law").
20. ICE concedes that a detainer itself does not provide the legal authority for a state or local officer to make a civil immigration arrest. See Dkt. No. 297-1 ¶ 162 ("[T]he detainer does not authorize, but merely requests, action by the LEA); see also Dkt. No. 481-9 (ICE 30(b)(6) Rapp Dep. Tr. 134:23-135:24) (explaining that it "would be up to the individual LEA to determine what authority or right they have to hold an individual [on an immigration detainer] beyond the completion of their [ ] criminal process.").
21. ICE is liable for the arrests other law enforcement agencies make on immigration detainers, including arrests made by state and local entities that have no authority to make arrests on detainers. Dkt. No. 42 at 8.
22. For the foregoing reasons, the Court determines that ICE violates the Fourth Amendment by issuing detainers to state and local law enforcement agencies in states that do not expressly authorize civil immigration arrests in state statute. Accordingly, the Court enters judgment for Plaintiffs on that claim.
ICE's Issuance of Detainers Through the Reliance on Inaccurate, Incomplete, and Error-Filled Databases Violates the Fourth Amendment
1. Next, the Court considers whether the databases ICE relies on to issue its detainers are reliable sources of information for establishing probable cause.
2. This claim is brought on behalf of the Probable Cause Subclass, which is comprised of all current and future persons who are subject to an immigration detainer issued by an ICE agent located in the Central District of California, where the detainer is not based upon a final order of removal signed by an immigration judge, or the individual is not subject to ongoing removal proceedings and the detainer was issued solely on the basis of electronic database checks. See Dkt. No. 484 at 1-2.
3. As discussed above, database sources must be reliable to satisfy the Fourth Amendment. See e.g. , Arizona v. Evans , 514 U.S. 1, 17, 115 S.Ct. 1185, 131 L.Ed.2d 34 (1995) (O'Connor, J. concurring); see also United States v. Esquivel-Rios , 725 F.3d 1231, 1238 (10th Cir. 2013) (Gorsuch, J.) (holding, "in a case that hinges entirely on the reliability of a computer database, the district court overlooked" testimony "casting doubt on its reliability").
4. Courts consider a number of factors when determining whether a database is reliable.
5. First, courts look to whether a database contains the complete set of pertinent information; that is, whether it lacks certain material information. A database that is incomplete in a manner that necessarily produces erroneous arrests is not a database that can be reasonably relied upon by officers. See Smith v. Oklahoma City , 696 F.2d 784, 787 (10th Cir. 1983) (finding a violation of the Fourth Amendment because an arrest warrant relied on a computer database that did not provide the necessary information to issue such warrants); see also Morales v. Chadbourne , 235 F. Supp. 3d 388, 401 (D.R.I. 2017), appeal dismissed , No. 17-1300, 2017 WL 4574440 (1st Cir. May 24, 2017) ("A database search is only successful and its results are only reliable under a probable cause analysis if the information contained in the database is complete and if the search is thorough and based on available identifies.")
6. In a case addressing the issue, the Ninth Circuit determined that the Immigrant Index System database, which included records limited in time, did not contain records predating 1983, and excluded "millions of people who are legitimately present in the United States," did not provide sufficient cause for the search. Orhorhaghe v. I.N.S. , 38 F.3d 488, 493 (9th Cir. 1994).
7. Courts also consider the error rates and inaccuracies within a database to determine their reliability. See People v. Jones , 110 Misc.2d 875, 443 N.Y.S.2d 298, 304 (N.Y. City Crim. Ct. 1981) (granting suppression in case where officers estimated an inaccuracy rate of a computer generated alarm system "as high as twenty percent, an inaccuracy rate that is unquestionably substantial.").
8. For example, courts have identified Fourth Amendment concerns where underlying information was not up-to-date or the database provided static information regarding a fact that can change over time. See e.g. , United States v. Beltran Vivanco , No. CR 14-840 RB, 2014 WL 12789017, at *5 (D.N.M. Aug. 13, 2014) (finding Fourth Amendment violation where officer "relied on a database that suffered from a fatal flaw," specifically that the database "utilized static insurance information to make the dynamic assessment of whether a vehicle is currently insured").
9. Finally, in determining whether a database is reliable for the purposes of probable cause, courts consider whether the databases were intended to supply probable cause for an arrest and whether attendant safeguards are in place to ensure accuracy and completeness. See Millender v. Cnty. of Los Angeles , 620 F.3d 1016, 1029 n.7 (9th Cir. 2010) (disregarding information gathered on CALGANG database for purpose of determining probable cause for search warrant for firearms because the database "is not designed to provide users with information upon which official action may be taken" and "cannot be used to provide probable cause for an arrest"), rev'd on other grounds sub nom . Messerschmidt v. Millender , 565 U.S. 535, 132 S.Ct. 1235, 182 L.Ed.2d 47 (2012).
10. The evidence presented at trial establishes that ICE violates the Fourth Amendment by relying on an unreliable set of databases to make probable cause determinations for its detainers. The flaws that exist in the databases are precisely those flaws courts have deemed fatal: (1) the databases provide static, often outdated, information about dynamic facts; (2) the databases are incomplete, often missing crucial pieces of information otherwise necessary for making probable cause determinations; and (3) the databases were never intended to be used to make probable cause determinations in the immigration context.
11. While ICE relies on several different databases in an attempt to compile enough information on a subject and make an adequate probable cause determination, the databases used by ICE, which have their limitations detailed herein—standing alone without any additional checks—do not sufficiently establish probable cause of removal. A number of factors guide the Court's reasoning.
12. First, information stored in the multiple databases ICE searches is often static and outdated. Citizenship and immigration status are complex inquiries that sometimes require information that cannot be obtained through databases that do not receive constant updates or real-time information about a subject. See People v. Joseph , 128 Ill.App.3d 668, 83 Ill.Dec. 883, 470 N.E.2d 1303, 1306 (1984) ("When ... computerized records are not kept up to date, a citizen may be subject to a deprivation of his liberty without any legal basis.").
13. ICE's reliance on static databases to assess citizenship and immigration status and determine whether an individual should remain in the custody of law enforcement agencies beyond their sentence is similarly unreasonable.
14. None of the databases on which ICE depends necessarily reflect a person's immigration status at the time when a detainer is set to be issued. This is because those databases reflect a person's immigration status at a particular point in time, but fail to reliably show how or whether that status has changed over time.
15. For example, the databases do not interact with one another in an effort to cross-reference or check the accuracy of the information entered.
16. CIS—the central database ICE relies upon in its determination of immigration status—provides no information on derivative citizenship. CIS regularly reports incorrect information about derivative citizens. See Trial Tr. 258:4-14, 261:12-15 (Stock). The result is a potential misclassification of "[t]housands and thousands of people, potentially even millions" given the various requirements for derivation of citizenship, their dynamism, and the failure of immigration officials to adequately or properly track those changes in status. Trial Tr. 274:19-275:10 (Stock). The result, of course, is that many U.S. citizens become exposed to possible false arrest when ICE relies solely on deficient databases.
17. Further, CIS "frequently" shows naturalized citizens as green card holders or fails to reflect an extension of a non-immigrant's period of stay. Trial Exs. 14-15.
18. The other databases on which ICE relies suffer similar flaws. At best, the databases reflect a person's immigration status at a particular point in the past, but—independently and as a set—the databases fail to reliably show how or whether that status has changed over time.
19. Further, the databases suffer from serious incompleteness, which impacts ICE's ability to rely on the databases in making probable cause determinations.
20. In Orhorhaghe , the Ninth Circuit considered a warrantless immigration arrest, which relied upon a database that did not contain any information about a subject's entry into the United States. INS relied on the absence of information in the database as an indication that the subject had illegitimately entered the United States. The Ninth Circuit found that "the absence of any record of Orhorhaghe's entry into the United States from the INS computer system did not provide any additional basis for suspecting that he was an illegal alien rather than a legal alien or American citizen. INS agents would be unable to locate any entry records for literally millions of people who are legitimately present in the United States. Most obviously, there would be no occasion to record the entry of native-born American citizens ... [.]" 38 F.3d at 498 ; see also Duckett v. United States , 886 A.2d 548, 549-550, 552 n.8 (D.C. 2005) (finding that the reliance on a lack of vehicle registration information in a law enforcement database that downloaded DMV records on a weekly basis did not provide reasonable grounds for a stop, since a lack of registration information could mean either that the car was not registered or that it had been registered less than a week previously.)
21. These cases demonstrate that relying on the absence of information in a database known for being incomplete is unreasonable. Whether the database lacks historical records, Orhorhaghe , 38 F.3d at 498-99 ; categorically excludes certain kinds of information, Esquivel-Rios , 725 F.3d at 1236 ; or is outdated or has delays in its updating of information, Duckett , 886 A.2d at 552 n.8 ; databases in such a state cannot be relied upon for probable cause determinations.
22. In addition, the core databases ICE relies on suffer from structural flaws, incompleteness, and pervasive errors that render the databases unreliable.
23. Evidence before the Court has shown that the set of databases ICE checks, and the information stored therein, contain serious errors.
24. Testimony before the Court, from practitioners familiar with each of the databases suggests errors were common, and often anticipated in each of the databases. Communications between an ICE senior officer and ICE Headquarters indicates that the misclassification of U.S. citizens as LPRs was a frequent occurrence. Dkt. No. 484, SF 108; Trial Ex. 15.
25. Finally, and perhaps most tellingly, the databases ICE uses are unreliable because no single database used was intended to provide any indication of probable cause of removability.
26. Immigration and citizenship law are complex and require a taxing examination of a person's history—the databases ICE uses were not created to track those complexities. The evidence before the Court shows ICE relies on the databases to cobble together information from disparate systems that are not at all intended to establish probable cause of removal.
27. The use of information stored on the databases here deviates from the purpose for which those databases were designed—accordingly, the use must be examined with careful scrutiny. See, e.g. , Beltran-Vivanco , 2014 WL 12789017 at *7 (noting that the database relied on by the officer "was not designed to provide a dynamic record of a vehicle's current insurance status" and finding a Fourth Amendment violation).
Plaintiffs Meet the Requirements for Injunctive Relief
28. "[A] plaintiff seeking a permanent injunction ... must demonstrate: (1) that it has suffered an irreparable injury; (2) that remedies available at law, such as monetary damages, are inadequate to compensate for that injury; (3) that, considering the balance of hardships between the plaintiff and defendant, a remedy in equity is warranted; and (4) that the public interest would not be disserved by a permanent injunction." Monsanto Co. v. Geertson Seed Farms , 561 U.S. 139, 156-57, 130 S.Ct. 2743, 177 L.Ed.2d 461 (2010).
29. "It is well established that the deprivation of constitutional rights unquestionably constitutes irreparable injury.’ " Melendres v. Arpaio (9th Cir. 2012), 695 F.3d 990, 1002 (quoting Elrod v. Burns , 427 U.S. 347, 373, 96 S.Ct. 2673, 49 L.Ed.2d 547 (1976) ). The Ninth Circuit has repeatedly "upheld injunctions against pervasive violations of the Fourth Amendment" like those present here. See e.g. , Easyriders Freedom F.I.G.H.T. v. Hannigan , 92 F.3d 1486, 1501 (9th Cir. 1996).
30. The class members here face days of unconstitutional imprisonment as a result of immigration detainers. Such injury is not adequately compensated by monetary damages. Nelson v. Nat'l Aeronautics & Space Admin. , 530 F.3d 865, 882 (9th Cir. 2008) ("Unlike monetary injuries, constitutional violations cannot be adequately remedied through damages and therefore generally constitute irreparable harm."), rev'd and remanded on other grounds , 562 U.S. 134, 131 S.Ct. 746, 178 L.Ed.2d 667 (2011).
31. The balance of hardships favors Plaintiffs as Defendants "cannot reasonably assert that [they are] harmed in any legally cognizable sense by being enjoined from constitutional violations." Zepeda v. INS , 753 F.2d 719, 727 (9th Cir. 1983). The injunction Plaintiffs request encompasses only what is necessary to remedy the Fourth Amendment violations. It leaves numerous feasible avenues for ICE to investigate and effectively enforce immigration law.
32. Finally, "it is always in the public interest to prevent the violation of a party's constitutional rights." Melendres , 695 F.3d at 1002. Plaintiffs only seek for Defendants to adhere to their constitutional mandate and cease issuing unlawful detainers that result in unlawful imprisonment.
33. For the foregoing reasons, concurrently with adopting the above findings of facts and conclusions of law, the Court shall issue a permanent injunction enjoining ICE from issuing detainers to state and local law enforcement agencies in states where there is no explicit state statute authorizing civil immigration arrests on detainers and enjoining ICE from issuing detainers to Probable Cause Subclass members based solely on database searches that rely upon information from sources that lack sufficient indicia of reliability for a probable cause determination for removal.
Plaintiffs shall submit a Proposed Judgment within 14 days of the issuance of this order.