Opinion
22-cv-10209
09-27-2024
JOHN DOES et al., Plaintiffs, v. GRETCHEN WHITMER et al., Defendants.
OPINION AND ORDER (1) GRANTING IN PART PLAINTIFFS' MOTION FOR SUMMARY JUDGMENT (Dkt. 123) and (2) GRANTING IN PART DEFENDANTS' MOTION FOR SUMMARY JUDGMENT (Dkt. 129)
MARK A. GOLDSMITH, UNITED STATES DISTRICT JUDGE
TABLE OF CONTENTS
I. INTRODUCTION........................4
II. BACKGROUND........................5
A. History of SORA......................5
B. Related Litigation........................10
C. The Present Case........................14
III. ANALYSIS........................18
A. Retroactive Application of SORA 20211. ........................18
1.Controlling Precedent........................19
2. SORA 2021 Constitutes Punishment ........................23
a. Legislative Intent ........................24
b. SORA 2021's Actual Effects ........................26
i. History and Tradition ........................26
ii. Affirmative Disability or Restraint ........................30
iii. Traditional Aims of Punishment ........................31
iv. Rational Connection to a Non-Punitive Purpose ........................31
v. Excessive with Respect to Purpose........................37
B. Lack of Individualized Review........................38
1. Due Process ........................43
a. Procedural Due Process ........................43
b. Substantive Due Process........................46
i. Fundamental Rights ........................47
ii. Animus ........................53
iii. Rational Basis Review........................59
2. Equal Protection........................65
C. Unequal Opportunity to Petition for Removal........................65
D. Plea Agreements........................70
E. Non-Sex Offenses........................71
1. Mootness........................72
2. Class-Wide Relief........................74
3. Merits of Plaintiffs' Non-Sex Claim........................75
F. Non-Michigan Convictions........................77
G. Compelled Speech........................83
1. Compelled Speech Under the First Amendment........................84
2. SORA 2021 Reporting Requirements and Strict Scrutiny........................91
H. Forced Admission of Understanding........................95
1. Applicable Scrutiny........................96
2. Analysis of the Statement of Understanding Under Strict Scrutiny........................97
I. Reporting of Internet Identifiers........................99
J. Vagueness........................107
K. Statute of Limitations and Mootness Due to Federal SORNA........................110
L. Whether the Governor is a Proper Defendant........................114
IV. CONCLUSION........................114
I. INTRODUCTION
Plaintiffs filed this class action challenging the constitutionality of Michigan's Sex Offender Registration Act, Mich. Comp. L. § 28.723, et seq, as it was amended in 2021 (SORA 2021). Over the past decade, prior versions of SORA have faced similar challenges, and multiple courts within Michigan and the Sixth Circuit have found that predecessor versions of SORA were unconstitutional. This Court is not writing on a blank slate, but rather, in the wake of a long history of case law and legislative amendments.
Plaintiffs bring eleven claims, alleging that SORA 2021 violates constitutional protections regarding ex post facto, due process, equal protection, privileges and immunities, and the First Amendment. After extensive discovery, both parties have moved for summary judgment. Each side's motion is denied in part and granted in part.
The Court set a briefing schedule for the cross-motions for summary judgment (Dkt. 121). Plaintiffs filed a motion for summary judgment (Dkt. 123). Defendants filed a combined motion for summary judgment and response in opposition to Plaintiffs' motion (Dkt. 129). Plaintiffs filed a combined reply in support of their motion and response in opposition to Defendants' motion (Dkt. 131). Defendants filed a reply in support of their motion (Dkt. 132). The parties also filed a joint summary of arguments (Dkt. 133). The Court later directed supplemental briefing on the applicability of two cases, Doe v. Lee, 102 F.4th 330 (6th Cir. 2024) and People v. Lymon, __N.W.3d __, No. 164685, 2024 WL 3573528 (Mich. July 29, 2024). Plaintiffs filed a Lee brief (Dkt. 146) and a Lymon brief (Dkt. 155). Defendants filed a Lee brief (Dkt. 145) and a Lymon brief (Dkt. 154). " Because oral argument will not aid the Court's decisional process, the motions will be decided based on the parties' briefing. See E.D. Mich. LR 7.1(f)(2); Fed.R.Civ.P. 78(b).
As explained in detail below, the Court sustains some of the challenges and rejects others.
Plaintiffs are correct that making registrants subject to certain provisions that were not adopted at the time they committed their crimes violates the Constitution's protection against ex post facto laws. The Court also agrees that individuals who committed a crime that does not have a sexual element or circumstance cannot be labeled as sex offenders and saddled with registration requirements. However, the Court rejects the claim that all sex offenders are entitled to an individualized hearing, either before being placed on the registry or periodically thereafter, to determine if they are or remain dangerous. A right to a hearing will be recognized (beyond the limited circumstances currently allowed under the statute) for individuals who have been convicted of an offense without a sexual element that the state claims is a registrable offense based on a sexual circumstance. A right to a hearing will also be recognized where a nonMichigan offense triggers registration in Michigan based on the offense being substantially similar to a registerable Michigan offense. Regarding Plaintiffs' First Amendment challenges, the Court finds that some of SORA 2021's requirements violate free-speech rights, while others do not. The challenge regarding plea bargains is moot. The challenge based on various terms being vague will require further proceedings.
II. BACKGROUND
A. History of SORA
Michigan passed its first SORA in 1994, creating a non-public registry maintained solely for law enforcement use. See Mich. Pub. Act 295 (1994). Since then, the legislature has amended SORA multiple times, most notably in 2006, 2011, and 2021. A brief history of SORA provides context for Plaintiffs' challenge to SORA 2021.
• 1994 - Michigan created its first sex offender registry, effective October 1, 1995. The registry established a non-public law enforcement database. No regular verification or reporting was required; the only reporting obligation was to notify law enforcement within 10 days of a change of address, which did not need to be done in-person. Registry information was maintained for 25 years for people convicted of one offense and for life for those with multiple offenses. See Mich. Pub. Act 295 (1994).
• 1997 - A paper copy of registry information was made available to the public for inspection during regular business hours at local law enforcement agencies. See Mich.
Pub. Act 494 (1996).
• 1999 - The registry became public on the internet for the first time, with the names, addresses, physical descriptions, and birth dates of registered sex offenders available online. Registrants started being required to report in person quarterly or yearly, depending on their offense. The list of registrable offenses and categories of individuals required to register for life expanded. See Mich. Pub. Act 85 (1999).
• 2002 - Required registrants to report in person when they enrolled, disenrolled, worked, or volunteered at an institution of higher education. See Mich. Pub. Act. 542 (2002).
• 2004 - Photographs of registered sex offenders were published online; a fee was imposed on registrants, with failure to pay a crime. See Mich. Pub. Acts 237, 238 (2004).
• 2006 - Created “exclusion zones” which prohibited registrants from living, working, or loitering within 1,000 feet of a school, with criminal penalties for noncompliance; created a system allowing subscribers to receive electronic notifications when a person registers in or moves into a particular zip code. See Mich. Pub. Acts 121, 127, 132 (2005).
• 2011 - Divided registrants into three tiers, based on offense, that determine the length of registration and reporting frequency; expanded the number of registrants required to register for life; required all registrants to appear in person within three business days to update certain personal information. See Mich. Pub. Acts 17, 18 (2011).
• 2021 - Removed exclusion zones; created a new option for the Michigan State Police (MSP) to designate a method other than in-person reporting for certain information; eliminated tier information from the public SORA website. See Mich. Pub. Act 295 (2020).
The amendments most relevant to the present case are those enacted in 2006, 2011, and 2021. The 2006 amendment introduced exclusion zones (sometimes called “school zones” or “student-safety zones”), which barred registrants from living, working, or loitering within 1,000 feet of a school, with few exceptions. Mich. Pub. Acts 121, 127 (2005).
The 2011 amendments introduced a three-tier system that classifies registrants based on their offense. Mich. Pub. Acts 17, 18 (2011). The parties agree that the three-tier system remains largely unchanged in SORA 2021. Pls. Statement of Material Facts (SOMF) ¶ 139; Defs. Resp. to Pls. SOMF ¶ 139. A registrant's tier determines their length of registration, reporting requirements, inclusion in the public registry, and ability to petition for removal from the registry, as shown in the table below.
The Court notes that the section of SORA that defines Tier I, Tier II, and Tier III offenses, Mich. Comp. Law § 28.722, is difficult to follow. Section 28.722 cites to more than 20 sections of the Michigan penal code-many of which contain numerous subsections themselves-and adds various exclusions and limitations to those sections. Below is a non-exhaustive list of offenses generally corresponding to each tier. Plaintiffs submitted as an exhibit a chart listing crime codes, crime descriptions, and tiers, that appears to be a comprehensive list of registrable offenses. See Pls. SOMF ¶ 103 (citing Crime Codes Chart (Dkt. 127-23)). Plaintiffs do not describe, however, how this chart was created (whether by Plaintiffs or the MSP) or for what purpose it is used. Defendants did not voice any objections to the contents of the chart. Tier I offenses include possession of child sexually abusive material; indecent exposure involving fondling; unlawful imprisonment of a minor; criminal sexual conduct in the fourth degree; surveillance or photography of an unclothed person if the victim is a minor. Mich. Comp. L. § 28.722(r)(i)-(xi). Tier II offenses include soliciting a minor under 16 years of age for an immoral purpose; involving a child in sexually abusive activities or materials, or producing or financing such materials; using a computer to communicate with any person for the purpose of committing other specified offenses against minors; most crimes against nature or sodomy against a minor; most crimes of gross indecency against a minor 13 years of age or older; recruiting or providing a minor for commercial sexual activity; criminal sexual conduct in the fourth degree; criminal sexual conduct in the second and fourth degree committed against a minor between 13 and 18 years of age. Id. § 28.722(t)(i)-(xii). A Tier I offender who is subsequently convicted of a Tier I offense is also considered a Tier II offender. Id. § 28.722(s)(i). Tier III offenses include gross indecency against a minor less than 13 years of age; kidnapping of a minor; leading away of a child under 14 years of age; criminal sexual conduct in the first, second, or third degree against a minor less than 13 years of age; sexual contact with a dead human body; and assault with intent to commit criminal sexual conduct. Id. § 28.722(v)(i)-(vii). A Tier II offender who is subsequently convicted of a Tier I or Tier II offense is also considered a Tier III offender. Id. § 28.722(u)(i).
This table was created by the Court.
Tier I
7%
15 years
Once a year
Non-public
Yes, after 10 years
Tier II
20%
25 years
Twice a year
Public
No
Tier III
73%
Lifetime
Four times a year
Public
No
Pls. SOMF ¶ 76 (citing Class Data Report ¶¶ 5, 42 (Dkt. 123-6)). The Class Data Report notes that “130 people (0.3%) are not classified in one of the tiers, which appears to reflect that they have a special status due to court decisions, special conditions related to an out-of-state offense, or some other exception.” Class Data Report ¶ 42 n.11.
Mich. Comp. L § 28.725(11)-(13).
Mich. Comp. L § 28.725a(3)(a)-(c).
Mich. Comp. L § 28.728(4)(c).
Mich. Comp. L § 28.728c(1), (12). As discussed in more detail later in this opinion, certain Tier III juvenile offenders can also petition for removal after 25 years. Id. § (2), (13).
The parties agree that before the introduction of the three-tier system in 2011, almost three-quarters of Michigan's registrants were 25-year registrants. Pls. SOMF ¶ 103 (citing Answer to Am. Compl. ¶¶ 179-180) (Dkt. 111)). After the 2011 amendments, almost three-quarters of registrants were assigned lifetime registration terms. Id. Plaintiffs estimate that, after the enactment of SORA 2011, almost 17,000 people suddenly had to register for the rest of their lives, without any individualized assessment. Id.
Unless otherwise specified, if the Court cites to a party's statement of material facts for a proposition that the Court is accepting as an established fact, the Court does so because the opposing party either admits or does not dispute the fact.
SORA 2011 also imposed additional reporting requirements. In addition to requiring registrants to report in person at specified intervals, SORA 2011 required registrants to report in person within three business days whenever they changed their name, address, employment, schooling, travel plans, vehicle information, or internet identifiers.
As Plaintiffs point out, Am. Compl. ¶ 609, SORA 2011 required registrants to report this information “immediately,” which was defined as three business days. Mich. Comp. L. § 28.722(g) (2020). SORA 2021 ceased using the word “immediately” but maintains the same three business day requirement. Mich. Comp. L. § 28.725(1) (2021). Plaintiffs refer to this change as “cosmetic” rather than “substantive.” Am. Compl. ¶ 609.
“Internet identifier” means “all designations used for self-identification or routing in internet communications or posting.” Mich. Comp. L. § 28.722(g).
Following rulings in an earlier federal case (discussed further below), the legislature passed SORA 2021, which removed the exclusion zones so that there are no longer any restrictions on where a registrant can live or work. But SORA 2021 left intact many of the amendments introduced in SORA 2011, including the three-tier system that retroactively increased the registration length for many offenders. The only significant change with respect to the tier system in SORA 2021 is that a registrant's tier is no longer published on the public website. Mich. Comp. L. § 28.728(3)(e).
The parties agree that SORA 2021 maintains the bulk of the reporting requirements introduced in SORA 2011. Pls. SOMF ¶¶ 144, 146; Defs. Resp. to Pls. SOMF ¶¶ 144, 146. Registrants are still required to report changes in the personal information listed above to law enforcement within three business days, although the requirement to report internet identifiers no longer applies to pre-2011 registrants. Id. SORA 2021 allows for, but does not require, the MSP to establish methods other than in-person reporting for certain required updates. Mich. Comp. L. 28.725(1). In practice, the MSP requires registrants to report in person within three business days whenever they change their name, address, employment, or schooling. See Pls. SOMF ¶ 271 (citing Explanation of Duties Form at ¶¶ 6, 9 (Dkt. 126-17)); Defs. Resp. to Pls. SOMF ¶ 271. Only changes in travel plans, vehicle information, internet identifiers, and telephone numbers can be reported by mail, and the MSP could, in theory, return to requiring in-person updates at any time. MSP did not promulgate rules or seek public comment on the required manner of reporting, but Defendants note that the current requirements mirror those of the federal Sex Offender Registration and Notification Act (SORNA), 34 U.S.C. § 20901 et seq. Pls. SOMF ¶ 270; Defs. Resp. to Pls. SOMF ¶¶ 270-271.
Plaintiffs argue that “SORA 2021 makes minimal changes to the structure, substance, and overall requirements of [SORA 2011], and is, if anything, more complex, vaguer, and harder to understand than the version that preceded it.” Am. Compl. ¶ 213 (Dkt. 108). They note that SORA 2021 “retains the identical tier system, the identical lengthy/lifetime registration periods, and the virtually identical onerous reporting requirements and online public registry, all without any individual assessment of risk . . . .” Id. ¶ 214. Defendants, on the other hand, argue that “[t]he new SORA amended and repealed the provisions of the old SORA that were found to be unconstitutional.” Defs. Br. Supp. Mot. for Summ. J. at 4.
Because SORA 2021 retains features of its predecessor statutes, the Court begins by summarizing prior cases that addressed constitutional challenges to prior iterations of SORA.
B. Related Litigation
In an earlier action, six registered sex offenders challenged the constitutionality of SORA 2011. See John Does 1-4 v. Snyder, 932 F.Supp.2d 803 (E.D. Mich. 2013), rev'd and remanded sub nom. Does #1-5 v. Snyder, 834 F.3d 696 (6th Cir. 2016) (Does I). The plaintiffs sued Michigan's Governor and the director of the MSP, alleging violations of: (i) the Ex Post Facto Clause; (ii) the plaintiffs' fundamental rights to travel; (iii) the plaintiffs' fundamental rights to engage in common occupations of life; (iv) the plaintiffs' rights to direct the education and upbringing of their children; (v) the First Amendment; (vi) the Due Process Clause because of the retroactive and oppressive nature of the act; (vii) the Due Process Clause because certain provisions were vague and impossible to comply with; and (viii) the Headlee Amendment to the Michigan Constitution. Does I, 932 F.Supp.2d at 808.
The Court refers to all district and circuit court opinions from the earlier case as Does I, with reporter information listed to identify the specific opinion.
In a series of three opinions, the district court rejected most of the plaintiffs' constitutional claims, including the ex post facto challenge. See Does I, 932 F.Supp.2d 803 (E.D. Mich. 2014); Does I, 101 F.Supp.3d 672 (E.D. Mich. 2015); Does I, 101 F.Supp.3d 722 (E.D. Mich. 2015). The court ruled in plaintiffs' favor as to some claims, finding that certain provisions of SORA 2011 were unconstitutionally vague; that imposing strict liability for compliance violations violated due process; and that the retroactive application of certain internet-reporting requirements violated the First Amendment.
On appeal, the Sixth Circuit reversed. See Does I, 834 F.3d at 706. The court addressed the plaintiffs' ex post facto challenge first and found that SORA 2011 imposed punishment within the meaning of that doctrine. Id. at 699-705. Accordingly, the court held that the “retroactive application of SORA's 2006 and 2011 amendments to [p]laintiffs is unconstitutional, and it must therefore cease.” Id. at 706. The court declined to opine on the plaintiffs' other constitutional arguments because its ex post facto decision meant that “none of the contested provisions [could] be applied to the plaintiffs in [the] lawsuit . . . .” Id. But the court did note that the “[plaintiffs' arguments on these other issues are far from frivolous and involve matters of great public importance.” Id. The Sixth Circuit remanded to the district court for entry of a judgment. Id.
In January 2018, the district court entered a stipulated final judgment. See Stipulated Final Judgment, Does I, No. 12-11194 (E.D. Mich. Jan. 26, 2018) (Dkt. 153). Among other things, the judgment declared that “retroactive application of [SORA's] 2006 and 2011 amendments violates the Ex Post Facto Clause of the U.S. Constitution” and enjoined “defendants, their officers, agents[,] servants, employees, and attorneys, and other persons in active concert or participation with them . . . from enforcing the 2006 and 2011 SORA amendments against the plaintiffs.” Id. at 2. The judgment also ordered that “if the Michigan legislature amends or replaces SORA to implement the Sixth Circuit's holding in [Does I], this injunction shall terminate on the effective date of any such amendments or new statute.” Id. at 4.
To the Court's knowledge, the stipulated final judgment does not appear on Westlaw.
A class action was filed just days after the Sixth Circuit's decision in Does I in 2016, bringing largely the same constitutional challenges to SORA 2011 as were brought in Does I, but this time seeking relief on behalf of all class members, not just the Does I plaintiffs. See Doe v. Snyder, 449 F.Supp.3d 719 (E.D. Mich. 2020) (Does II). But after filing Does II, “the parties agreed to focus their resources on legislative reform,” and the briefing schedule was repeatedly delayed. Does II, 449 F.Supp.3d at 726. In May 2019, the district court entered a stipulated order in Does II, granting declaratory relief for the plaintiffs, holding that “the 2006 and 2011 amendments were unconstitutional as applied to the ex post facto subclass.” Id. Eventually, after legislative relief failed to materialize, plaintiffs filed renewed motions for partial summary judgment. Id.
In February 2020, the court in Does II granted plaintiffs' motions for summary judgment, finding that “[f]or several years, registrants have been forced to comply with unconstitutional provisions of SORA.” Id. at 737. The court entered permanent injunctive relief and stated that, upon the entry of final judgment, Defendants would be “permanently enjoined from enforcing any provision of SORA against members of the ex post facto subclasses” and would be permanently enjoined from enforcing certain unconstitutional provisions against any registrant. Id. The court gave the parties 60 days to formulate a joint, proposed form of judgment. Id. Before those 60 days passed, however, the COVID-19 pandemic swept the nation, delaying the entry of the final judgment for more than a year, until August 2021. Am. Compl. ¶¶ 203-205. During that time, the legislature passed SORA 2021, which took effect on March 24, 2021. Mich. Pub. Act 295 (2020).
As Does I and Does II were being litigated, a criminal case concerning the constitutionality of SORA was proceeding through the Michigan state courts. See People v. Betts, 968 N.W.2d 497 (Mich. 2021). In Betts, the defendant had pled guilty to second-degree criminal sexual conduct in 1993, two years before SORA took effect. Id. at 501. In 2012, after he had successfully completed parole, Betts failed to report his change of residence, e-mail address, and purchase of a new vehicle within three business days, as required by SORA 2011. Id. Charged with violating SORA 2011's registration requirements, he sought dismissal of the criminal proceeding, arguing that retroactive application of SORA 2011 violated ex post facto protections. Id. at 501-502.
The Michigan Supreme Court ultimately agreed with Betts, holding that SORA 2011, “when applied to registrants whose criminal acts predated the enactment of the 2011 SORA amendments, violates the constitutional prohibition on ex post facto laws.” Id. at 521. The court ordered that the defendant's conviction of failure to register as a sex offender be vacated. Id.
In the wake of that decision, the Michigan Supreme Court held that SORA 2021 constitutes punishment and violates the Michigan Constitution's prohibition of cruel or unusual punishment insofar as it applies to non-sexual offenders. See People v. Lymon, __N.W.3d __, No. 164685, 2024 WL 3573528 (Mich. July 29, 2024). In Lymon, the defendant was convicted of torture, unlawful imprisonment, felonious assault, and possession of a firearm during the commission of a felony. Id. at *3. These convictions stemmed from an incident where the defendant held his wife and children at gunpoint overnight and repeatedly threatened to kill them. Id. at *2-*3. Although Lymon's offenses did not include a sexual component, he was placed on the Michigan sex offender registry as a Tier I offender because two of his three unlawful-imprisonment convictions involved minors. Id. at *3. The court held that the defendant was entitled to removal from the sex-offender registry. Id. at *17.
C. The Present Case
The ten named Plaintiffs in this case-John Does A, B, C, D, E, F, G, H, Mary Doe, and Mary Roe-filed this action seeking declaratory and injunctive relief on behalf of themselves and all others similarly situated. All named Plaintiffs are or have been required to register as sex offenders. The backgrounds of each named Plaintiff are summarized below.
• John Doe A (John Doe #1 in Does I) - In 1990, John Doe A committed an armed robbery, during which he forced a mother and her teenage son into a building. He pled guilty to armed robbery and weapons charges and no contest to kidnapping. He was sentenced to 20 to 40 years in prison. At the time of John Doe A's arrest, Michigan did not have a sex offender registry. When he was released from prison, he was retroactively required to register as a sex offender, because kidnapping of a minor triggers registration. After the 2011 SORA amendments, he was retroactively classified as a Tier III offender, and his registration period was extended from 25 years to life. Pls. SOMF ¶¶ 3-8.
• John Doe B (John Doe #3 in Does I) - In 1998, when John Doe B was 19 years old, he had a sexual relationship with a 14-year-old girl. He pled guilty to attempted criminal sexual conduct in the third degree for having sex with a minor. He was sentenced to four years of probation under the Holmes Youthful Trainee Act (HYTA), a record sealing statute for young offenders. His HYTA status was later revoked because he missed a
reporting deadline. After the 2011 SORA amendments, he was retroactively classified as a Tier III offender, and his registration period was extended from 25 years to life. Id. ¶¶ 9-17.
• John Doe C (John Doe #4 in Does I) - In 2005, when John Doe C was 23 years old, he had a sexual relationship with a 15-year-old girl. He pled guilty to attempted criminal sexual conduct in the third degree. John Doe C was required to register for 25 years. After the 2011 SORA amendments, he was retroactively classified as a Tier III offender, and his registration period was extended from 25 years to life. Id. ¶¶ 18-23.
• John Doe D (John Doe #1 in Does II) - In 2000, when John Doe D was 19 years old, he had a sexual relationship with a 14-year-old girl. He pled guilty to attempted criminal sexual conduct in the third degree under the HYTA. He served about a month in jail and was sentenced to three years of probation. After his release, his HYTA status was revoked because he violated the terms of his probation. After the 2011 SORA amendments, he was retroactively classified as a Tier III offender, and his registration period was extended from 25 years to life. Id. ¶¶ 41-48.
• John Doe E (John Doe #3 in Does II) - In 1994, when John Doe E was 21 years old, he pled no contest to three counts of criminal sexual conduct in the second degree after being accused of engaging in inappropriate touching of his six-year-old nephew. At the time, Michigan did not have a sex offender registry. He was sentenced to 90 days in custody and five years of probation. When SORA was first enacted in 1995, he was subject to registration for 25 years. As a result of the 2011 amendments, he was retroactively reclassified as a Tier III offender, and his registration period was extended from 25 years to life. Id. ¶¶ 49-56.
• John Doe F (John Doe #4 in Does II) - In 2013, John Doe F pled guilty to a two-year misdemeanor of sexual misconduct in the fourth degree. He was classified as a Tier II registrant, which remains the case under SORA 2021. Id. ¶¶ 57-63.
• John Doe G (John Doe #5 in Does II) - In 2006, John Doe G was convicted of third-degree sexual assault of a child in Nebraska. He was required to register for 10 years. In 2010, he moved to Michigan to be closer to his family. In Michigan, he was required to register for 25 years. As a result of the 2011 amendments, he was retroactively reclassified as a Tier III offender, and his registration period was extended from 25 years to life. Id. ¶¶ 64-68.
• John Doe H (John Doe #6 in Does II) - In 2015, John Doe H pled no contest to criminal sexual conduct in the fourth degree for sexually touching a woman. He was sentenced to
five years of probation. He was classified as a Tier I offender and required to register for 15 years. Under SORA 2021, this remains the same. Id. ¶¶ 69-71.
• Mary Doe (Mary Doe in Does I) - In 2003, while living in Ohio, Mary Doe pled no contest to unlawful conduct with a minor for having a sexual relationship with a 15-year-old male. She was sentenced to three years in prison and required to register for 10 years. After serving fewer than eight months, Mary Doe was granted judicial release. The terms of her probation required her to move to Michigan to live with her parents. Under Michigan law at the time, she was required to register for 25 years. As a result of the 2011 amendments, she was retroactively reclassified as a Tier III offender, and her registration period was extended from 25 years to life. Id. ¶¶ 24-32.
• Mary Roe (Mary Roe in Does I) - In 2002, at the age of 19, Mary Roe had sex with a 14-year-old boy. She pled guilty to criminal sexual conduct in the third degree. She served about 2.5 years in prison. At the time of her conviction, Mary Roe was required to register for 25 years. As a result of the 2011 amendments, she was retroactively reclassified as a Tier III offender, and her registration period was extended from 25 years to life. In 2016, she was threatened with prosecution because her job was within 1,000 feet of a school, as prohibited by the 2016 amendments to SORA. She filed a suit challenging SORA's constitutionality. See Roe v. Synder, 240 F.Supp.3d 697 (E.D. Mich. 2017). Pursuant to a settlement in that case, her registration term was reduced from life to 25 years and she was removed from the public registry. But under SORA 2021, she is required to register and comply for life again. Id. ¶¶ 33-40.
Plaintiffs filed a notice that John Doe F “has expunged his conviction and has been removed from the sex offender registry.” See Pl's 8/6/24 Notice at 1 (Dkt. 156). “The class and subclasses for which Mr. Doe F is a named representative all are represented by other class and subclass representatives.” Id.
Plaintiffs filed a motion for class certification, which the Court granted in a stipulated order. See Mot. to Certify Class (Dkt. 5); 5/18/22 Order (Dkt. 35). The Court certified a “primary class,” defined as “people who are or will be subject to registration under Michigan's [SORA],” as well as seven subclasses. 5/18/22 Order; 5/9/23 Order (Dkt. 109). Plaintiffs John Does A, B, C, D, E, F, G, H, Mary Doe, and Mary Roe are named as representatives of the primary class. 5/18/22 Order at 1-2. As of January 24, 2023, the primary class consisted of 45,145 people subject to SORA. Pls. SOMF ¶ 72 (citing Class Data Report ¶¶ 1, 24, 127-128 (Dkt. 123-6)). A description of the seven subclasses is below:
• Pre-2011 ex post facto subclass (31,249 people, 69% of the primary class) - Defined as “members of the primary class who committed the offense(s) requiring registration before July 1, 2011.” Plaintiffs John Does A, B, C, D, E, F, G, Mary Doe, and Mary Roe are class representatives.
• Retroactive extension of registration subclass (16,723 people, 37% of the primary class) - Defined as “members of the primary class who were retroactively required to register for life as a result of amendments to SORA.” Plaintiffs John Does A, B, C, D, E, G, Mary Doe, and Mary Roe are class representatives.
• Barred from petitioning subclass (size unknown) - Defined as “members of the primary class who are ineligible to petition for removal from the registry and for whom ten or more years will have elapsed since the date of their conviction for the registrable offense(s) or from their release from any period of confinement for that offense(s), whichever occurred last, and who (a) have not been convicted of any felony or any registrable offense since; (b) have successfully completed their assigned periods of supervised release, probation, or parole without revocation at any time of that supervised release, probation, or parole; and (c) have successfully completed an appropriate sex offender treatment program, if successful completion of a sex offender treatment program was a condition of the registrant's confinement, release, probation, or parole.” Plaintiffs John Does A, C, E, F, G, Mary Doe, and Mary Roe are class representatives.
• Non-sex-offense subclass (298 people, 0.7% of the primary class) - Defined as “members of the primary class who are or will be subject to registration for an offense without a sexual component including convictions for violating M.C.L. § 750.349 (other than convictions for violating M.C.L. § 750.349(1)(c) or M.C.L. § 750.349(1)(f)), § 750.349b, § 750.350, or a substantially similar offense in another jurisdiction.” Plaintiff John Doe A is the class representative.
• Non-Michigan offense subclass (3,100 people, 7% of the primary class) - Defined as “members of the primary class who are or will be subject to sex offender registration under Mich. Comp. Laws 28.722(r)(x); (t)(xiii); (v)(viii); or 28.723(1)(d), for a conviction or adjudication from a jurisdiction other than Michigan.” Plaintiffs John Doe G and Mary Doe are class representatives.
• Plea bargain subclass (size unknown) - Defined as “members of the primary class who gave up their right to trial and pled guilty to a registrable offense in Michigan and who, as a result of retroactive amendments to SORA, (a) were retroactively subjected to SORA even though there was no registration requirement at the time of their plea; or (b) had their registration terms retroactively extended beyond that in effect at the time of their plea.” Plaintiffs John Does A, B, C, D, E, and Mary Roe are class representatives.
• Post-2011 subclass (13,848 people, 31% of the primary class) - Defined as “members of the primary class who committed the offense(s) requiring registration on or after July 1, 2011.” Plaintiff John Doe H is the class representative.5/18/22 Order; 5/9/23 Order; Pls. SOMF ¶¶ 80-85 (citing Class Data Report ¶¶ 24, 129-149); Defs. Resp. to Pls. SOMF ¶¶ 80-85.
Plaintiffs previously filed a motion for a preliminary injunction (Dkt. 7), which the Court denied without prejudice. See 9/15/22 Order (Dkt. 54). In the same order, the Court also denied Defendants' motion to dismiss Plaintiffs' complaint (Dkt. 41). Id. The Court explained that it would be premature to consider either motion before the parties had the opportunity to develop a factual record. Id. Since then, the parties have engaged in significant discovery, and a thorough factual record is now before the Court. The Court will address each of Plaintiffs' claims in turn.
The Court applies the traditional summary judgment standard as articulated in Scott v. Harris, 550 U.S. 372, 380 (2007). The movant is entitled to summary judgment if that party shows that there is no genuine dispute as to any material fact and that the movant is entitled to judgment as a matter of law. Fed.R.Civ.P. 56(a). If the movant makes an initial showing that there is an absence of evidence to support the nonmoving party's case, the nonmovant can only survive summary judgment by coming forward with evidence showing there is a genuine issue for trial. Celotex Corp. v. Catrett, 477 U.S. 317, 324-325 (1986).
A. Retroactive Application of SORA 2021
Plaintiffs bring two counts regarding the retroactive application of SORA 2021 to registrants. In Count I, Plaintiffs assert that retroactively applying SORA 2021 to John Does A, B, C, D, E, F, G, Mary Doe, Mary Roe, and the pre-2011 ex post facto subclass violates the Ex Post Facto clause of the Constitution. Am. Compl. ¶¶ 727-736. In Count II, Plaintiffs contend that retroactively requiring lifetime registration for John Does A, B, C, D, E, G, Mary Doe, Mary Roe, and the retroactive extension subclass violates the Ex Post Facto Clause. Id. ¶¶ 737-745.The Court awards summary judgment to Plaintiffs on both counts. Specifically, the Court finds that retroactively increasing reporting requirements and extending registration terms violates the Ex Post Facto Clause.
In Count II, Plaintiffs also allege that retroactively requiring lifetime registration violates due process. Am. Compl. ¶¶ 742-743. Because the Court finds that retroactive application of SORA 2021 violates the Ex Post Facto Clause, the Court declines to address Plaintiffs' alternate due process argument. Granting Plaintiffs' requested relief on more than one basis would be “unnecessary and inefficient . . . especially [as to] the resolution of any constitutional claims, which should be avoided whenever possible . . . .” Doe #1 v. Lee, 518 F.Supp.3d 1157, 1206 (M.D. Tenn. 2021) (collecting cases).
The Ex Post Facto Clause provides that “No State shall . . . pass any . . . ex post facto Law.” U.S. Const. art. I, § 10, cl. 1. The clause proscribes retroactive punishment. Does I, 834 F.3d at 699. “A law qualifies as ex post facto if it changes the legal consequences of acts committed before its effective date.” Doe v. Lee, 102 F.4th 330, 336 (6th Cir. 2024) (punctuation modified).
Plaintiffs offer two arguments in support of their ex post facto claims. First, Plaintiffs argue that Does I-in which the Sixth Circuit held that certain provisions from SORA 2011 violate the Ex Post Facto Clause-requires a finding that SORA 2021 also violates the Ex Post Facto Clause. Second, Plaintiffs argue that even if Does I does not control, SORA 2021 constitutes punishment, rendering it an ex post facto law. The Court addresses both arguments in turn.
1. Controlling Precedent
The parties disagree about which case controls this Court's decision. Plaintiffs argue that Does I controls. Pls. Br. Supp. Mot. for Summ. J. at 5-9. Defendants, on the other hand, argue that Willman v. Att'y Gen. of U.S., 972 F.3d 819 (6th Cir. 2020), a case addressing federal SORNA, controls. Br. Supp. Def. Mot. Summ. J. at 3. Neither party is fully correct.
Congress passed SORNA in 2006. Willman, 972 F.3d at 822 (citing Pub. L. No. 109-248; 120 Stat. 590; 34 U.S.C. § 20901 et seq). “The statute aimed to make more uniform what had remained a patchwork of federal and 50 individual state registration systems, with loopholes and deficiencies that had resulted in an estimated 100,000 sex offenders becoming missing or lost.” Id. (punctuation modified). SORNA does not create its own registration system; instead, SORNA states that “[a] sex offender shall register, and keep the registration current, in each jurisdiction where the offender resides, where the offender is an employee, and where the offender is a student.” Id. (citing 34 U.S.C. § 20913(a)). “Instead of imposing requirements on states, SORNA conditions federal funds on states' voluntary compliance with a federal sex offender registration regime.” Id. at 824 n.2.
Defendants argue that because SORA 2021 “is almost identical to federal SORNA in all material respects, the Willman decision binds this Court and requires it to dismiss the Ex Post Facto challenge.” Id. at 3. Defendants err because the factual context and legal issues in Willman differ from those present in the instant case, and the terse discussion of the Ex Post Facto Clause in that case offers no explanation as to how it might be applicable here.
The plaintiff in Willman was convicted in a Michigan court in 1993 of assault with intent to commit criminal sexual conduct involving sexual penetration. Willman, 972 F.3d at 822. He was required to register under Michigan SORA when it became effective in 1995. Id. More than two decades later, Congress passed federal SORNA in 2006, which required him to register as a matter of federal law. Id. In 2019, Willman filed a federal court action challenging his registration requirements under Michigan state law and federal SORNA. Id. The district court entered a stipulated order regarding Willman and the state defendants, declaring that Willman no longer had to register under Michigan SORA. Id. The U.S. Attorney General, the only remaining defendant, filed a motion to dismiss, arguing that the stipulated order had “no bearing on Willman's obligations under federal law (i.e., SORNA).” Id. The district court granted the motion to dismiss. Id.
On appeal, the Sixth Circuit described the “principal issue” as “whether the registration and notification obligations set forth in [federal SORNA] apply to sex offenders who are convicted under state law but are not subject to that state's sex offender registration and notification requirements.” Id. at 821. The Sixth Circuit held that “federal SORNA obligations are independent of state-law sex offender duties,” requiring Willman to register with Michigan as a matter of federal law. Id. at 824.
In addition to his statutory claim that he was not required to register, Willman brought several constitutional challenges to SORNA, including an ex post facto challenge, all of which were rejected. But Willman's analysis was terse, offering no guidance here. The specific ex post facto argument was not even set forth in the opinion. Without any description of the claim, the court simply rejected it, citing a prior decision, United States v. Felts, 674 F.3d 599, 605-606 (6th Cir. 2012). Id. at 824. By contrast, Felts engaged in a broader discussion of that plaintiff's ex post facto claim. However, a review of the Felts opinion shows why the case is not applicable here.
Willman's constitutional challenges were brought on the following grounds: (i) the Ex Post Facto Clause, (ii) the Fifth Amendment Double Jeopardy Clause, (iii) the Eighth Amendment prohibition of cruel and unusual punishment, (iv) the First Amendment's right to privacy, (v) the Fourteenth Amendment Privileges Or Immunities Clause, (vi) Article IV's Privileges And Immunities Clause, (vii) the Fourth Amendment prohibition of unreasonable seizures, (viii) the First Amendment overbreadth doctrine, and (ix) the Fourteenth Amendment Due Process Clause. Willman, 972 F.3d at 824-827.
In Felts, the plaintiff was convicted in Tennessee of rape of a child and aggravated sexual battery. Felts, 674 F.3d at 702. After his release from prison, Felts moved to different states without notifying the Tennessee registration authorities. Id. Felts was then convicted of failing to register under federal SORNA, even though Tennessee had not completely implemented the act. Id. Felts argued that his new conviction violated the Ex Post Facto Clause because it increased the level of punishment for his crimes. Id. at 605-606. The Sixth Circuit rejected this argument, explaining that “SORNA provides for a conviction for failing to register; it does not increase the punishment for the past conviction. Felts's crime of failing to update his sex offender registry after the enactment of SORNA was entirely separate from his crime of rape of a child and aggravated sexual battery.” Id. at 606.
The context, issues, and holding in Felts are entirely distinguishable from those present here. Plaintiffs here argue that SORA 2021 imposes retroactive punishment solely for their original convictions; Felts addressed whether the later failure-to-register conviction was vulnerable on ex post facto grounds. Neither Felts nor Willman controls here.
Does I, on the other hand, is controlling, in part. In Does I, the Sixth Circuit held that two provisions from the 2006 and 2011 amendments to SORA violated the Ex Post Facto Clause: (i) the exclusion zones introduced in 2006 and (ii) the requirement that registrants report in person within three business days to update certain personal information introduced in 2011. Does I, 834 F.3d at 706. SORA 2021 no longer contains exclusion zones, but it does contain the same onerous reporting requirements. Does I, therefore, requires the finding that the in-person reporting requirement violates the Ex Post Facto Clause.
As discussed previously, SORA 2021 allows for, but does not require, the MSP to establish methods other than in-person reporting for certain required updates. Because this is optional, it does not provide meaningful relief to Plaintiffs. The Court views the 2021 reporting requirement as substantially similar to the 2011 requirement.
Plaintiffs argue that Does I went even further, purportedly holding that all amendments introduced in 2006 and 2011 violate the Ex Post Facto Clause, including the retroactive application of increased registration length for certain offenders, which was introduced along with the three-tier system in 2011 and remains in SORA 2021. Pls. Br. Supp. Mot. for Summ. J. at 6-9. Plaintiffs rely on the Sixth Circuit's pronouncement that “[t]he retroactive application of SORA's 2006 and 2011 amendments to [p]laintiffs is unconstitutional, and it must therefore cease.” Id. at 6 (citing Does I, 834 F.3d at 706). But Does I focused only on exclusion zones and reporting requirements; the court did not even discuss the retroactive extension of registration terms.
The Sixth Circuit's recent opinion in Lee, 102 F.4th 330, a case challenging Tennessee's sex offender registration act, confirms this interpretation. In describing the court's holding in Does I, the court stated that, “[i]n 2016, we enjoined enforcement of two amendments to Michigan's sex offender registry law, known as SORA . . . . In [Does I], two sex offenders challenged the 2006 and 2011 amendments to SORA which prohibited registrants from living, working, or loitering within 1,000 feet of a school and required all registrants to appear in person immediately to update information such as new vehicles or internet identifiers . . . .” Lee, 102 F.4th at 337. This confirms that the “amendments” referenced in the Does I holding were the exclusion zones and in-person reporting requirements specifically discussed by the court.
Does I is also controlling in that it describes the framework used to analyze ex post facto challenges to SORA. In the next section, the Court applies this framework to SORA 2021 and finds that the retroactive extension of registration terms also violates the Ex Post Facto Clause.
2. SORA 2021 Constitutes Punishment
The Supreme Court has laid out a two-part framework for determining whether a law constitutes retroactive punishment:
The determinative question is whether the legislature meant to establish civil proceedings. If the intention was to impose punishment, that ends the inquiry. If, however, the intention was to enact a regulatory scheme that is civil and nonpunitive, the [c]ourt must further examine whether the statutory scheme is so punitive either in purpose or effect as to negate the State's intention to deem it civil.Smith v. Doe, 538 U.S 84, 85 (2003). This framework was adopted by the Sixth Circuit in Does I. Does I, 834 F.3d at 700. Thus, the Court's first question is whether the Michigan legislature intended to impose punishment in passing SORA 2021. Finding that the legislature did not intend to impose punishment, the Court then asks whether SORA 2021 is so punitive either in purpose or effect as to negate the legislature's intention. The Count finds that it is.
The Court notes that there is an inherent contradiction imbedded in this test, as the “intent” inquiry of the first prong seems similar to the “purpose” inquiry of the second prong. The Court's understanding is that the first prong focuses on legislative intent, whereas the second prong focuses on the effect of the statute in practice.
a. Legislative Intent
To determine legislative intent, “courts must first ask whether the legislature, in establishing the penalizing mechanism, indicated either expressly or impliedly a preference for one label or the other.” Id. at 93. The Supreme Court warned in Smith that “because the Court ordinarily defers to the legislature's stated intent . . . only the clearest proof will suffice to override legislative intent and transform what has been denominated a civil remedy into a criminal penalty.” Id. (punctuation modified).
The Michigan legislature included the following statement of purpose in SORA 2021, unchanged from SORA 2011:
The legislature declares that the sex offenders registration act was enacted pursuant to the legislature's exercise of the police power of the state with the intent to better assist law enforcement officers and the people of this state in preventing and protecting against the commission of future criminal sexual acts by convicted sex offenders. The legislature has determined that a person who has been convicted of committing an offense covered by this act poses a potential serious menace and danger to the health, safety, morals, and welfare of the people, and particularly the children, of this state. The registration requirements of this act are intended to provide law enforcement and the people of this state with an appropriate, comprehensive, and effective means to monitor those persons who pose such a potential danger.Mich. Comp. L. § 28.721a.
The Sixth Circuit analyzed this statement of purpose in Does I and found that it “evinces no punitive intent.” Does I, 834 F.3d at 700. The court admitted that the plaintiffs pointed to “some features [of SORA 2011] that might suggest a punitive aim-e.g., SORA is triggered solely by criminal offenses and the registration requirement is recorded on the judgment; registration is handled by criminal justice agencies like the police; SORA imposes criminal sanctions; and it is codified in Chapter 28 of the Michigan Code, a chapter that deals with police-related laws ....”
Id. But the court noted that similar arguments had been previously raised and rejected by the Supreme Court, id. (citing Smith, 538 U.S. at 95), so the court found “no warrant for concluding that SORA's intent is punitive,” id. at 701.
In the present case, Plaintiffs argue that new facts warrant a finding that SORA 2021 is intended to punish. Pls. Br. Supp. Mot. for Summ. J. at 9-10. Plaintiffs highlight that the legislature chose to maintain some of the 2011 amendments that Does I criticized and adopted SORA 2021 “after being presented with uncontroverted evidence that registries don't work.” Id. at 9. In sum, Plaintiffs argue that “[t]he facts in the record on legislative animus coupled with the legislature's decision to ignore both Does I and evidence-based information support a finding that the new law is intended to punish.” Id. at 9-10.
In response, Defendants argue that “it is part of the political process to disagree over how to address contentious issues” and take issue with Plaintiffs' claim that the legislature was presented with “uncontroverted evidence” that registries do not work. Defs. Br. Supp. Mot. for Summ. J. at 5. Defendants emphasize that it is difficult to measure the effect of a law designed to prevent future potential crimes. Id. at 5-6.
The Court agrees with Defendants and finds insufficient evidence to declare that the legislature's intent was punitive. True, the legislature maintained one provision that Does I criticized-the onerous in-person reporting requirements. But it also removed the amendment Does I criticized most strongly-the exclusionary zones. And as discussed later in this opinion, Plaintiffs' evidence does not conclusively establish that registries like SORA 2021 do not work. Such evidence does not constitute “the clearest proof” that the legislature adopted SORA 2021 with a punitive intent.
The Michigan Supreme Court reached the same conclusion regarding the legislature's non-punitive intent in its recent analysis of SORA 2021. Lymon, 2024 WL 3573528, at *5-*6. The court noted that the legislature did not amend the statement of intent in SORA 2021, “nor did it amend the statute in any other way that would imply its intent in enacting the statute had changed.” Id. at *6. “To the contrary, the Legislature's choice to remove provisions [the Michigan Supreme Court] identified in Betts as especially punitive in effect supports a theory that the Legislature's recent amendments were made to support a characterization of SORA as civil.” Id.
In harmony with these cases, the Court concludes that the legislature's intent in adopting SORA 2021 was not punitive.
b. SORA 2021's Actual Effects
The second step is to ask whether SORA 2021's actual effects are punitive. The Supreme Court has identified five factors that are particularly relevant in ex post facto challenges to registration statutes. See Smith, 538 U.S. at 97 (citing Kennedy v. Mendoza-Martinez, 372 U.S. 144, 168-169 (1963)). These include whether the challenged law: (i) has been regarded in our history and traditions as a punishment, (ii) imposes an affirmative disability or restraint, (iii) promotes the traditional aims of punishment, (iv) has a rational connection to a nonpunitive purpose, or (v) is excessive with respect to this purpose. Id. These factors are “neither exhaustive nor dispositive, . . . but useful guideposts.” Id. (punctuation modified). The Court considers each factor in turn.
i. History and Tradition
As to the first factor, the relevant question is whether the regulatory scheme has been regarded in the nation's history and traditions as punishment. Id. Courts typically consider whether a statute's restrictions resemble the traditional punishments of banishment, shaming, and parole/probation. See, e.g., Smith, 538 U.S. at 97-99; Does I, 834 F.3d at 701-703; Lymon, 2024 WL 3573528, at *7; Betts, 968 N.W.2d at 508-510.
In Does I, the Sixth Circuit held that SORA 2011 resembled traditional punishment. Does I, 834 F.3d at 701-703. With respect to banishment, the court particularly criticized exclusionary zones, explaining that exclusion zones “resemble[s], in some respects at least, the ancient punishment of banishment.” Id. at 701-702. The court also found that SORA 2011's requirement for public disclosure of information resembled traditional shaming punishments. Id. at 702-703. The court noted that SORA 2011 went beyond republishing information that was already publicly available, like criminal history, and published public “tier classifications corresponding to the state's estimation of present dangerousness without providing for any individualized assessment.” Id. at 702. In addition, the court noted that SORA 2011 sometimes disclosed “otherwise nonpublic information,” such as sealed criminal records. Id. at 703. Finding also that SORA 2011 resembled parole/probation, the court referenced the statute's exclusion zones, in-person reporting requirements, and imposition of imprisonment for failure to comply. Id. For all these reasons, the court held that the first factor weighed in Plaintiffs' favor. Id.
Defendants argue that SORA 2021 does not resemble banishment because it no longer contains exclusion zones. Defs. Br. Supp. Mot. for Summ. J. at 6-7. In Lymon, the Michigan Supreme Court came to the same conclusion: “[A]lthough we conclude that [SORA 2021] resembles shaming and parole, we conclude that [SORA 2021] no longer resembles banishment because its present iteration no longer includes a prohibition on registrants living, working, or loitering within 1,000 feet of school property.” Lymon, 2024 WL 3573528, at *7 n.11.
The Court agrees with Defendants that SORA 2021 no longer resembles banishment. But as the Michigan Supreme Court held, the Court still finds that SORA 2021 resembles shaming and parole/probation.
Plaintiffs criticize the registry's public website as “a world-wide wall of shame where registrants are ‘branded [as] a potentially violent menace by the state.'” Pls. Br. Supp. Mot. for Summ. J. at 12 (citing Betts, 968 N.W.2d at 514). In response, Defendants argue that the website does not expressly state that any registrant is currently dangerous. Defs. Resp. to Pls. SOMF ¶ 242. Plaintiffs do not dispute this, but they contend that the website implies that registrants are currently dangerous, nonetheless. As explained by Plaintiffs, “[t]he initial search page signals dangerousness, stating: ‘This registry is made available through the Internet with the intent to better assist the public in preventing and protecting against the commission of future criminal sexual acts by convicted sex offenders.'” Pls. SOMF ¶ 242 (citing Lageson Report ¶¶ 32, 43 (Dkt. 123-14); Registry Screenshots at 1-2 (Dkt. 127-24)). Dr. Sarah Lageson, an expert for Plaintiffs whose research focuses on the impact of digital technologies on criminal punishment, believes that “[t]his messaging signals a highly dangerous type of criminal who requires constant public monitoring and scrutiny.” Lageson Report ¶¶ 3, 43. The Court agrees. There is no disputing that the public website implies that the registrants listed might be dangerous.
Defendants also argue that the public website does not constitute shaming because most of the information it includes is already public. Defs. Br. Supp. Mot. for Summ. J. at 7. But not all public information is made equal; re-packaging information and providing it to the public in a different form can in and of itself increase shaming. The electronic notification system is particularly troublesome. As the Michigan Supreme Court explained in Betts, the notification system reduces the effort required to obtain registry information, thereby “increase[ing] the likelihood of social ostracism based on registration.” Betts, 968 N.W.2d at 551-552.
For these reasons, the Court finds that SORA 2021 resembles shaming.
Plaintiffs argue that “SORA 2021 resembles the traditional punishment of parole/probation.” Pls. Br. Supp. Mot. for Summ. J. at 10 (punctuation modified). As Plaintiffs put it, “registrants are under significant state supervision, and a failure to comply, like the failure to comply with parole conditions, potentially subjects the offender to imprisonment.” Id. at 1011 (punctuation modified). Plaintiffs go so far as to contend that the “range of affirmative requirements and adverse consequences experienced as a result of registration well exceed those associated with customary probation and parole.” Id. at 11 (citing Wayne Logan, Knowledge as Power: Criminal Registration and Community Notification Laws in America 138 (1st ed. 2009)).
In response, Defendants contend that SORA is less severe than probation/parole because it is notification based, not permission based. Defs. Br. Supp. Mot. for Summ. J. at 9-12. The crux of Defendants' argument is that, under the probation/parole system, authorities can dictate certain choices for probationers/parolees, such as requiring individuals to work, dictating where a person can or cannot be, and prohibiting association with anyone under the age of 18. Id. at 10. They contend that SORA, on the other hand, allows registrants to make their own choices and merely requires them to notify law enforcement of certain changes. Id. According to Defendants,
Probation or parole dictates whom someone can date, requires that they must work, mandates disclosure of internet passwords, requires installation and payment for monitoring systems, dictates where a person can be or cannot be, mandates agreement to warrantless searches, prohibits possession of children's toys, and prohibits association with anyone under the age of 18 . . . . Those types of conditions pale in comparison to notifying police of changes to personal information and periodic reporting.Defs. Br. Supp. Mot. for Summ. J. at 10. Plaintiffs disagree, arguing that “[s]ome probationers/parolees will have permission-based restrictions” but “[o]thers will not.” Pls. Reply at 11 (citing Sample Probation Order (Dkt. 128-18)).
The Court finds that Plaintiffs have the better of the argument. With its significant restraints on registrants, including requiring regular in-person reporting and the threat of imprisonment for failure to comply, SORA 2021 resembles the traditional punishments of parole/probation.
In light of all these considerations, SORA 2021 resembles traditional forms of punishment.
ii. Affirmative Disability or Restraint
The second factor requires an inquiry into “how the effects of the [act] are felt by those subject to it.” Smith, 538 U.S. at 99-100. “If the disability or restraint is minor and indirect, its effects are unlikely to be punitive.” Id.
The Sixth Circuit found that SORA 2011 imposed significant restraints. Does I, 834 F.3d at 703-704. The court called the exclusion zones the statute's “most significant” restraint, but also cited in-person registration and the threat of imprisonment for failure to comply as important restraints. Id. at 703.
SORA 2021 removed the exclusion zones, allowing registrants to live and work anywhere. But as previously discussed, SORA 2021 maintained the bulk of SORA 2011's in-person reporting requirements. Registrants are still required to report in person at regular intervals throughout the year, depending on tier, and within three days whenever certain personal information changes. As the Michigan Supreme Court explained in Lymon,
Overall, [SORA 2021] continues to impose significant obligations on registrants by requiring the immediate disclosure of extensive personal information, annual (or more-frequent) in-person visits to law enforcement, and the payment of fees. [SORA 2021] ensures compliance with these requirements through the potential for
imprisonment, which is the paradigmatic affirmative restraint .... Accordingly, this factor weighs toward a finding that [SORA 2021] constitutes punishment- although it weighs less heavily in that regard than [SORA 2011], which additionally restricted registrants' choices and actions through student-safety zones and imposed more-frequent in-person reporting requirements.Lymon, 2024 WL 3573528, at *10.
The Court agrees with Lymon's characterization of SORA 2021. Although SORA 2011 imposed more significant restraints than SORA 2021, the Court still finds that SORA 2021 imposes significant restraints on registrants.
iii. Traditional Aims of Punishment
In Does I, the Sixth Circuit found that “[SORA 2011] advances all the traditional aims of punishment: incapacitation, retribution, and specific and general deterrence.” Does I, 834 F.3d at 704. As the court explained, “[SORA 2011's] very goal is incapacitation insofar as it seeks to keep sex offenders away from opportunities to reoffend. It is retributive in that it looks back at the offense (and nothing else) in imposing its restrictions ....Finally, its professed purpose is to deter recidivism . . . and it doubtless serves the purpose of general deterrence.” Id. None of this changed with SORA 2021. See Lymon, 2024 WL 3573528, at *10-*11 (noting that “the changes to [SORA 2021] have not materially affected [the court's] analysis of this factor” and holding that SORA 2021 still supports the aims of general deterrence, specific deterrence, and retribution). This factor weighs in Plaintiffs' favor.
iv. Rational Connection to a Non-Punitive Purpose
The statute's rational connection to a nonpunitive purpose is “‘a most significant' factor in [the Court's] determination that the statute's effects are not punitive.” Does I, 834 F.3d at 704 (quoting Smith, 538 U.S. at 102). The parties seem to agree that the stated goal of SORA 2021, which is to promote public safety, constitutes a valid, non-punitive purpose. The parties strongly disagree, however, regarding the efficacy of SORA in achieving that purpose.
In evaluating whether sex offender registration systems bear a rational relationship to a non-punitive purpose, courts have considered social-science evidence. Plaintiffs urge that this Court do the same here, contending that it shows that SORA 2021 fails to reduce sexual recidivism, and may possibly increase it. See Pl. SOMF ¶ 153; Pl. Br. Supp. Mot. for Summ. J. at 16-17. Plaintiffs point to a body of scholarship, including reports of their experts, supporting that view. Pl. SOMF ¶ 153. However, as discussed below, courts reviewing the scientific literature have expressed the view that there is no universally accepted scientific view about the efficacy of registration systems. And the parties' submissions confirm that there is no unanimity of scientific opinion on the subject. While the Court concludes that there are strong, science-based opinions challenging the impact of registration systems like SORA 2021 on recidivism, it cannot conclude that the state of scientific opinion mandates a finding that SORA 2021 bears no rational connection to a non-punitive purpose.
Plaintiffs point to expert reports of Elizabeth Letourneau (Dkt. 123-9), J.J. Prescott (Dkt. 12310), Kelly Socia (Dkts. 123-11, 123-12, 123-13), Kristen Zgoba (Dkt. 123-15), and Karl Hanson (Dkt. 123-7, 123-8).
The Michigan Supreme Court concluded that the efficacy of sex offender registry regimes, like SORA 2021, is the subject of “robust scientific debate [with] no universally accepted conclusion on the matter[]” Lymon, 2024 WL 3573528, at *12 n.18. The court looked back at its earlier decision in Betts, 968 N.W.2d 497, where it had found that certain “studies demonstrate that, at a minimum, [SORA 2011's] efficacy is unclear.” Id. at 12. It confirmed in Lymon that its prior conclusion remains true with respect to SORA 2021. See id. (“The same and additional studies continue to support the uncertainty of SORA's general efficacy.”).
The same conclusion was reached in a 2017 U.S. Department of Justice report, which analyzed studies about sex offender registration and notification (SORN) systems and found that “research on the effectiveness of [SORN] remains relatively limited and findings from the studies are somewhat inconclusive.” U.S. Department of Justice, Sex Offender Management Assessment and Planning Initiative at 202 (March 2017). As the study explains:
Defendants cite to the DOJ report in their reply. See Defs. Reply at 4-5. The report can be found at: https://smart.ojp.gov/sites/g/files/xyckuh231/files/media/document/somapifullreport.pdf.
Studies based on a comparison of outcomes for sex offenders subject and not subject to SORN also produced mixed findings. An arguable lack of sufficient scientific rigor may further cloud the import of studies in this area. Therefore, the results of SORN research undertaken to date continue to leave open questions about the effects of registration and community notification requirements.Id.
One potential shortcoming of studies measuring the efficacy of registries on recidivism is the difficulty in measuring the prevalence of sexual offending. As the study points out, “[e]ven with the best sources of data, it is extremely difficult to estimate the actual number of sex crimes committed because of low levels of reporting.” Id. at 3. Moreover, “conclusions about the extent of sex offender recidivism and the propensity of sex offenders to reoffend over the life course inherently involve some uncertainty.” Id. at 121. In part, because of these difficulties in measuring offenses and, more specifically, recidivism rates, “problems found in sex offender recidivism research no doubt have contributed to a lack of consensus among researchers regarding the proper interpretation of some research findings and the validity of certain conclusions.” Id. at 107.
Similar conclusions can be found in a 2022 report prepared by the Federal Research Division of the Library of Congress under an interagency agreement with the U.S. Department of Justice's Office of Sex Offender Sentencing, Monitoring, Apprehending, Registering, and Tracking. See Office of Justice Programs, Sex Offender Registration and Notification Act- Summary and Assessment of Research (April 2022). The report finds that “research is not conclusive about whether SORN laws have mitigated sex offender recidivism.” Id. at 18-19. The study adds that “[a]s a whole, literature on the impacts of registration on [registered sex offenders] was indeterminate in its findings” with limitations “caused by methodological shortcomings restricting reliability, validity, and/or applicability of findings to only those individuals in the samples.” Id. at 2. Ultimately, the report concludes that “drawing sound conclusions about policy decisions” regarding SORN laws “is difficult due to the current cohort of limited research and lack of convincing data.” Id. at 23.
The report can be found at: https://www.ojp.gov/pdffiles1/smart/305231.pdf.
In opposing this conclusion, Plaintiffs present significant expert opinions that the registries lack efficacy, arguing that they demonstrate a “broad scientific consensus” that registries do not reduce-and may increase-sexual recidivism. Pls. SOMF ¶ 153. While the expert reports do reach the ultimate conclusion that public notification registration systems, like SORA 2021, do not significantly reduce recidivism and may increase it, there are concessions within the reports acknowledging that some studies do show some public benefit through reduction of recidivism. Further, some studies show that registration systems can have the separate public benefit of deterring first-time offenders.
For example, Elizabeth Letourneau, a public health professor at Johns Hopkins University, reported on a number of studies bearing on these issues. While many supported Plaintiffs' position, some diverged. See Letourneau Report ¶ 7 (Dkt. 123-9) (acknowledging that of “14 studies [that] specifically examined policy effects on sexual re-offense rates .... [a]ll but two fail[ed] to find any effect on sexual or violent re-offense rates”). One divergent analysis was a 2009 New Jersey study, which detected declining rates of sex crimes during and after implementation of that state's registration law in 1994. Id. This finding was tempered by a 2012 follow-up study, which found that “registration had no effect on sexual or violent re-offense rates.” Id. Similarly, a 2010 South Carolina study found a “small but significant deterrent effect on first-time sex crimes following implementation of the registration in 1995,” but “no effect on reoffense rates.” Id. A Minnesota study found “lower recidivism risk . . . relative to what was predicted based on their risk scores” following implementation of a registration law. Id. A Washington study found that adoption of a revision in the registration law “was associated with a significant reduction in recidivism.” Id.
Citing Veysey, B. M., Zgoba, K., & Dalessandro, M., A preliminary step towards evaluating the impact of Megan's Law: A trend analysis of sexual offenses in New Jersey from Justice 1985 to 2005, Justice Research and Policy, 10, 1-18 (2008).
Citing Tewksbury, R., Jennings, W. G., & Zgoba, K. M., A longitudinal examination of sex offender recidivism prior to and following the implementation of SORN, Behavioral Sciences and the Law, 30, 308-328.
Citing Letourneau, E. J., Levenson, J. S., Bandyopadhyay, D., Armstrong, K. S., & Sinha, D., Effects of South Carolina's sex offender registration and notification policy on deterrence of adult sex crimes, Criminal Justice Review, 35, 295-317 (2010).
Citing Duwe, G., & Donnay, W., The impact of Megan's Law on sex offender recidivism: The Minnesota experience, Criminology, 46(2), 411-446 (2008).
Citing Barnoski, R., Sex offender sentencing in Washington state: Did community notification influence recidivism?, Olympia: Washington State Institute for Public Policy (2005).
Another expert for Plaintiffs, Michigan Law Professor J.J. Prescott, also opined that community notification laws, like SORA, reduce offenses by non-registrants. See Prescott Report ¶ 11 (Dkt. 123-10) (“The threat of becoming subject to a community notification regime-and the shame and collateral effects that accompany being publicly identified as a ‘sex offender'-appear to have a measurable deterrent effect (i.e., it reduces offenses) by nonregistrants.”).
Prescott also concluded that “it is very unlikely that these laws are reducing recidivism by registrants; instead, it is probable that these laws are actually increasing recidivism . . . .” (emphasis in original). Prescott Report ¶ 11.
To be sure, these divergent studies were accompanied by critiques and limitations. But there is at least some evidence supporting the view that registration systems, like SORA 2021, promote some non-punitive goal. It is not fair to say that scientific studies mandate a finding of no public benefit at all.
For example, the Minnesota study was questioned because it involved registrants who were subject to “intensive supervised release conditions.” Letourneau Report ¶ 7. The Washington study was questioned by Letourneau because it did not appear in a peer-reviewed journal. Id.
While Plaintiffs focus primarily on the impact registration systems may or may not have on recidivism, they fail to adequately address other potential benefits of SORA. As just discussed, some studies cited by Plaintiffs show that registration systems can have a deterrent effect on firsttime offenders, which is undoubtedly a benefit to public safety. In addition, SORA provides members of the public with valuable information that they can use to protect themselves and their families. As explained by Defendants, “SORA is a tool for the people to use to potentially prevent criminal sexual activity from happening in the first instance.” Defs. Reply at 2. “It provides factual information to the public so residents may decide how to interact with individuals [who] are registered sex offenders[] (e.g., a member of the public may decide not to date someone on the registry; or a member of the public may decide not to let their children be cared for [by] a neighbor that is a registered sex offender.)” Id. at 2-3.
Plaintiffs' singular focus on recidivism might have been prompted by the stated goal of preventing harm from repeat offenders found in SORA's legislative declaration, Mich. Comp. L. § 27.721a. But “[i]n our review of governmental purposes . . . we need not rely only upon those purposes the legislature, litigants, or [other courts] have espoused, but may also consider any other rational purposes possibly motivating enactment of the challenged statute.” Mountain Water Co. v. Mont. Dep't of Pub. Serv. Regul., 919 F.2d 593, 597 (9th Cir. 1990).
Here, there is apparently undisputed evidence showing that SORN laws promote the goal of public safety, at least by deterring first-time offenders, although there is also mixed evidence regarding whether they reduce recidivism. SORN laws also serve the goal of providing the public with information they can use to protect themselves. For these reasons, the Court finds that SORA is rationally connected to a non-punitive purpose.
v. Excessive with Respect to Purpose
The final factor-whether the statute is excessive with respect to its purpose-weighs in Plaintiffs' favor. It is clear that SORA 2021 exacts a heavy toll on registrants. This includes requiring frequent in-person reporting, publishing personal information online, and often requiring registration for life. Yet, as discussed above, there is a spirited debate regarding its effectiveness, particularly as it relates to reducing recidivism. While it is true that SORA contributes to public safety by deterring first-time offenders and providing members of the public with information they can use to protect themselves, it is not clear that these purposes alone justify the heavy toll on registrants. Notably, this weighing of factors takes place in the “ex post facto” context, where the state has changed the “rules of the game” after registrants have committed their offenses-a context in which the Constitution has provided express protection. This leads the Court to conclude that SORA 2021 is excessive for purposes of its ex post facto analysis. This is in harmony with other courts addressing the same issue. See Does I, 834 F.3d at 705; Betts, 968 N.W.2d. at 513-515.
Taking all the Mendoza-Martinez factors into account supports a finding that SORA 2021 constitutes punishment. The Court holds that the 2006 and 2011 amendments to SORA violate the Ex Post Facto Clause.
To be clear, this holding means that the in-person reporting requirements and retroactive extension of registration terms originally introduced in SORA 2011 cannot be applied to a registrant who committed an offense before SORA 2011 was enacted.
B. Lack of Individualized Review
Plaintiffs allege that SORA 2021 violates equal protection and due process by imposing lengthy and lifetime registration without individualized consideration of risk. Am. Compl. ¶¶ 746756. According to Plaintiffs, SORA is a “one-size-fits-all regime designed for the highest risk offenders, without any individualized assessment of current risk. It presumes that all those convicted of a sex offense pose the same high risk to public safety . . . .” Id. ¶ 250.
Plaintiffs are correct that SORA 2021, like many other sex offender registration acts across the nation, is an offense-based statute that does not offer individualized review in most circumstances. But SORA 2021 is not a “one-size-fits-all” regime as Plaintiffs contend. As discussed already, SORA categorizes registrants into three tiers based on their offense. A registrant's tier determines their length of registration, frequency of reporting, inclusion in the public registry, and ability to petition for removal from the registry. Thus, not all registrants are treated the same; the severity of a registrant's offense determines their registration requirements. Regardless, Plaintiffs argue that SORA 2021's lack of individualized review “restricts the liberty of thousands of people who pose no appreciable risk with no public benefit.” Pls. Br. Supp. Mot. for Summ. J. at 28-29.
As discussed later in the opinion, SORA 2021 provides individual review for certain individuals convicted of offenses without a sex element.
Plaintiffs do not specify exactly what they mean by “individualized review.” In their amended complaint, Plaintiffs argue that “[a]ctuarial risk assessment instruments-which are used to determine the statistical likelihood that an individual will reoffend based on known diagnostic indicators-are far better at predicting recidivism risk than the fact of a conviction.” Am. Compl. ¶ 283 (citing Hanson Report ¶ 27-32 (Dkt. 1-4); Zgoba Report ¶ 36-37 (Dkt. 1-8)). Plaintiffs discuss two actuarial instruments, the Static-99R and its predecessor the Static-99, in detail in both their amended complaint and statement of material facts. But Plaintiffs do not mention either in their brief supporting their motion for summary judgment, and they do not explicitly endorse using Static-99R to determine registration requirements. Further, they do not specify when and how often they think an individualized risk assessment should occur. For instance, one can imagine a system that conducts an initial individualized review when determining a registrant's reporting requirements. Alternatively, one can imagine an individualized review that occurs later, to determine if a registrant's risk-level has changed. Plaintiffs also do not say who should perform this review-whether a court, the MSP, or some other decisionmaker.
Because both parties submitted evidence regarding the Static-99R, the Court will briefly address its feasibility as a tool for individualized review.
Plaintiffs describe Static-99R as “[t]he most widely used and well-researched sex offense risk assessment instrument[] in the world.” Pls. SOMF ¶ 232. Plaintiffs submitted an expert report by Dr. Karl Hanson, who, along with a colleague, created the Static-99 in 1999 and the Static-99R in 2009. See Hanson Report (Dkt. 123-7). As explained by Dr. Hanson, the Static-99 “assess the recidivism risk of adult male with a history of sexual crime . . . .” Id. ¶ 29. The Static-99 groups individuals into five levels of risk for sexual offending, Level I (the lowest risk category) through Level V (the highest risk category). Id. at ¶ 36. The assessment is based on ten factors, including the nature of the sex-related offense, demographics (including age at release and relationship history), sexual criminal history, and general criminal history. Id. at 29. The static-99R is truly “static” in that it estimates risk at a specific point in time based on factors present at that time. Id. at 34.
According to Defendants, Static-99R is a flawed measure because it predicts recidivism- i.e., the chances of someone being convicted for a subsequent offense-and not re-offenses, i.e., the chances of someone committing another offense. Defs. Br. Supp. Mot. for Summ. J. at 19. Defendants argue that Static-99 scores do not account for offenses that were never reported by the victim; offenses that were reported by the victim, but not reported to law enforcement; offenses reported to law enforcement but not resulting in charges or arrests; arrests not resulting in convictions; cases of criminal sexual conduct that were pled down to non-sexual offenses; or cases that do not end in a conviction for criminal sexual conduct. Id. at 18-19. Defendants also submit that the five risk levels used by Static-99R are arbitrary. Id. at 18.
Defendants note that Dr. Hanson testified that he was not aware of any state that uses Static-99 as a basis for determining how long someone needs to register as a sex offender. Id. at 22 (citing Hanson Dep. Tr. at 64 (Dkt. 125-3)). Dr. Hanson testified that he is aware of various states that utilize Static-99R “within their correctional systems,” but he did not elaborate how. Hanson Dep. Tr. at 64. He also stated that his recommendation “would be not to write in a specific measure [into the law] but to write in a risk level, which is . . . associated with the appropriate levels of intervention.” Id. at 63.
In addition to debating the utility of Static-99R, the parties also dispute whether performing individualized review of any sort would be economically feasible. Defendants argue that “the legislature opted to use a simple conviction-based statute rather than a complicated, time consuming and expensive individualized risk assessment.” Defs. Resp. to Pls. SOMF ¶ 64. But the parties dispute exactly how expensive and time-consuming individualized review might be.
Defendants cite the report submitted by Dr. Darrel Turner, which estimates that an individualized risk assessment costs between $8,000 to $20,000 and takes between eight and 26 hours to complete. Defs. Resp. to Pls. SOMF ¶ 151 (citing Turner Decl. at PageID.7091 (Dkt. 128-22)). Based on this estimate, conducting an individualized risk assessment for all current SORA registrants-roughly 44,000 people-would cost between $352 million and $880 million and take between 352,000 and 1.1 million hours. Id. Defendants also cite the declaration of Dr. Anna Salter, who stated that she spends at least 15 hours on each sex offender evaluation. Id. (citing Salter Decl. at 16). Based on that estimate, it would take 660,000 hours to conduct evaluations for all registrants. Id.
Plaintiffs argue that these estimates are based on “comprehensive evaluations . . . performed for civil commitment proceedings by psychologists,” and that “routine risk assessments” would be “far less complicated and much less expensive.” Pls. Resp. to Defs. SOMF ¶ 83 (punctuation modified) (citing Hanson Rebuttal Report at 24-25 (Dkt. 123-8)). Plaintiffs cite the testimony of James Kissinger, the State Administrative Manager for Sexual Abuse Prevention Services for the Michigan Department of Corrections (MDOC), who estimated that calculating the Static-99R for an individual can take as little as 15 minutes. Id. (citing Kissinger Dep. Tr. at 62-63 (Dkt. 126-5)). But as Defendants point out, Kissinger stated that it can take anywhere from 15 minutes to an hour to calculate a Static-99R. Defs. Resp. to Pls. SOMF ¶ 151 (citing Kissinger Dep. Tr. at 62-63). Using this estimate, it would take 11,000 to 44,000 hours to calculate a Static-99R for all registrants. Id. Plaintiffs do not provide a cost estimate for calculating a Static-99R for all registrants, but they contend that many registrants have already been assessed by MDOC. Pls. Resp. to Defs. SOMF ¶ 83 (citing Kissinger Dep. Tr. at 29-35).
Aside from potentially towering costs, an individualized assessment scheme carries another risk: the risk of making poor predictions of dangerousness. Whether the assessment is made at the outset of registration, or periodically throughout the term, there is an unknown risk that the evaluators will make a mistake about who should be required to register and for how long. A legislature might conclude that an offense-based system at least gives some certainty that those who commit serious offenses are monitored for significant periods, perhaps for life, and thereby mitigate the risk of an erroneous prediction. Making choices about whether to adopt bright-line rules or customized assessments of people, entities, or transactions subject to government regulation is the hallmark of the legislative function. “Under the system of government created by our Constitution, it is up to legislatures, not courts, to decide on the wisdom and utility of legislation.” Ferguson v. Skrupa, 372 U.S. 726, 729 (1963). There is nothing irrational in choosing a bright-line rule for a sex offender registration system.
In sum, it is not clear what form of individualized review Plaintiffs envision, or what the associated costs would be. What is clear is that the legislature could reasonably conclude that individualized review carried risks not deemed worth taking.
In any case, Plaintiffs' claim is vulnerable for another reason: Any potential right to review would require a showing of entitlement to avoid or remedy a constitutional violation. As discussed below, Plaintiffs' two claimed bases for constitutional violations-due process and equal protection-come up short.
1. Due Process
The Due Process Clause of the Fourteenth Amendment protects citizens from government deprivation of “life, liberty, or property, without due process of law.” U.S. Const. amend. XIV, § 1. Due process has two components: procedural and substantive. Bambach v. Moegle, 92 F.4th 615, 624 (6th Cir. 2024). Procedural due process rights “protect individuals from deficient procedures that lead to the deprivation of cognizable liberty interests.” Id. (punctuation modified). Substantive due process rights “ensure that-regardless of the procedural protections available- the government may not deprive individuals of fundamental rights unless the action is necessary and animated by a compelling purpose.” Id. (punctuation modified).
Plaintiffs do not specify whether they bring a procedural or substantive due process claim, but it appears they bring only a substantive claim. Because arguing that SORA lacks individualized review is akin to arguing that SORA's procedures are inadequate, the Court analyzes Plaintiffs' claim under both due process components.
In their motion for summary judgment, Defendants state that “[i]t appears that Plaintiffs assert only a substantive Due Process claim ....” Defs. Br. Supp. Mot. for Summ. J. at 28. Plaintiffs do not refute this in their response. Pls. Resp. at 14-15.
a. Procedural Due Process
To the extent Plaintiffs argue that SORA 2021's lack of individualized review violates procedural due process, this claim fails. The Supreme Court has held that registrants are not entitled to a hearing to determine risk where that fact is irrelevant to a state's statutory scheme, as is the case here. See Conn. Dep't of Pub. Safety v. Doe, 538 U.S. 1, 8 (2003) (DPS).
In DPS, the Supreme Court addressed a procedural due process challenge to Connecticut's sex offender registry. Like SORA 2021, the Connecticut statute at issue was conviction-based; “the law's requirements turn[ed] on an offender's conviction alone.” DPS, 538 U.S. at 2. The plaintiff argued that the Connecticut law “deprive[d] him of a liberty interest . . . without notice or a meaningful opportunity to be heard.” Id. at 6. Without deciding whether the plaintiff properly alleged deprivation of a liberty interest, the court held that there was no procedural due process violation. Id. at 8.
The Supreme Court explained that the plaintiff was not entitled to a hearing to determine whether he was currently dangerous because that fact was irrelevant to Connecticut's statute, which was strictly conviction-based. Id. at 7-8. The court explained that “even if respondent could prove that he is not likely to be currently dangerous, Connecticut has decided that the registry information of all sex offenders-currently dangerous or not-must be publicly disclosed.” Id. at 7 (emphasis in original). In enacting an offense-based statute, Michigan has made the same decision here. Because “[p]laintiffs who assert a right to a hearing under the Due Process Clause must show that the facts they seek to establish in that hearing are relevant under the statutory scheme,” id. at 8, Plaintiffs have not established a procedural due process violation.
The Sixth Circuit has previously rejected an attempt to distinguish SORA from DPS. See Fullmer v. Mich. Dep't of State Police, 360 F.3d 579 (6th Cir. 2004). In Fullmer, the plaintiff argued that the registration and public disclosure aspects of SORA deprived him of a constitutionally-protected liberty interest without “giving him notice and an opportunity to be heard on whether he is a threat to the public safety.” Fullmer, 360 F.3d at 581 (punctuation modified). The plaintiff attempted to distinguish DPS because a provision in Michigan's SORA states that “[t]he legislature has determined that a person who has been convicted of committing an offense covered by this act poses a potential serious menace and danger to the health, safety, morals, and welfare of the people” and that “[t]he registration requirements of this act are intended to provide law enforcement and the people . . . an . . . effective means to monitor those persons who pose such a potential danger.” Id. at 582 (citing Mich. Comp. L. § 28.721a). The plaintiff contended that, unlike in DPS, dangerousness was a component of the registration scheme under SORA.
The court rejected this argument, explaining that “[r]egardless of the language in the statute, the information on the registry's website makes it clear to anyone accessing the registry that all sex offenders convicted after a certain date are listed, without exception. Moreover, there is nothing on the website to indicate that the state has made an individual determination as to a registrant's dangerousness.” Id. at 582. Relying on DPS, the court found that a determination regarding dangerousness was irrelevant to Michigan's statutory scheme, which was also conviction-based. Id. at 581-583. The court held that there was no procedural due process violation and noted that the plaintiff had not brought a substantive due process claim. Id. at 582.
Nothing of significance to the analysis of this claim has changed since Fullmer was decided. SORA 2021 contains the same provision regarding the legislature's determination, and the registry's current website does not indicate that the state has made an individual determination as to a registrants' dangerousness. Therefore, any procedural due process claim brought by Plaintiffs fails under DPS and Fullmer.
b. Substantive Due Process
Substantive due process “ensure[s] that-regardless of the procedural protections available-the government may not deprive individuals of fundamental rights unless the action is necessary and animated by a compelling purpose.” Bambach, 92 F.4th at 624. Rights protected by the Due Process Clause must be “deeply rooted in this Nation's history and tradition and implicit in the concept of ordered liberty, such that neither liberty nor justice would exist if they were sacrificed.” Gore v. Lee, 107 F.4th 548, 562 (6th Cir. 2024) (quoting Washington v. Glucksberg, 521 U.S. 702, 720-721 (1997)). If legislation infringes on a fundamental right, strict scrutiny applies. Does I, 932 F.Supp.2d at 814. Where no fundamental right is implicated, the statute need only survive rational basis review. Id. (citing City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432 (1985)).
Plaintiffs argue that the aspects of SORA 2021 that restrict their fundamental rights to work and travel should be analyzed under strict scrutiny. Pls. Br. Supp. Mot. for Summ. J. at 29. “With respect to the rest of SORA [2021],” id., Plaintiffs argue that an “exacting rational basis scrutiny” applies because the law was motivated by animus, id. at 29-32. But even if standard rational basis applies, Plaintiffs contend that SORA 2021 is unconstitutional because it irrationally imposes extensive burdens with no public safety benefit. Id. at 33-35. The Court disagrees. As explained below, Plaintiffs have not shown that SORA 2021 significantly burdens any fundamental right. Further, Plaintiffs have failed to show that SORA 2021 was motivated by animus. Finally, as explained below, SORA 2021 passes constitutional muster under the rational basis standard.
Plaintiffs also contend that SORA impacts their fundamental right to speak. Pls. Br. Supp. Mot. for Summ. J. at 29. Plaintiffs' right to speak is already protected by the First Amendment, and the Court will address the First Amendment implications of SORA later in this opinion.
i. Fundamental Rights
Plaintiffs argue that SORA restricts their fundamental rights to work and travel. The Court begins by addressing the fundamental right to work, then turns to travel. With respect to work, Plaintiffs argue that SORA impacts them in the following ways:
• “Registrants' work addresses are posted online, meaning that employers who hire registrants will have their business address show up in registry searches.” Pls. SOMF ¶ 339 (citing Mich. Comp. L. § 28.728(2)(d)).
• While Michigan's unemployment rate was 4.3% in January 2023, 45% of class members who are not incarcerated report no current employment. Id. ¶ 340 (citing Class Data Report ¶¶ 20, 109-110 (Dkt. 123-6)).
• According to Plaintiffs, “research demonstrates that being on a sex offender registry dramatically reduces employment options for registrants, and frequently results in job loss.” Id. ¶ 339. Plaintiffs cite two expert reports in support of this claim. Id. (citing Socia Report ¶¶ 22-23 (Dkt. 123-11); Zgoba Report ¶¶ 25-30 (Dkt. 123-15)). Both reports cite surveys where roughly 40% of respondents reported losing a job because of their registration status. Socia Report ¶ 27; Zgoba Report ¶ 25.
Plaintiffs do not explain whether and why it would violate the law for employers to choose not to hire convicted sex offenders.
The Court notes that this is not an apples-to-apples comparison. Plaintiffs' 45% statistic does not account for whether individuals are actively looking for work, which is the definition used by Michigan (as well as the federal government) to track unemployment.
Neither expert report explains how the surveys controlled for the impact of registrants' criminal record, separate from their registry status.
In response, Defendants admit that registrants may have more difficulty finding employment but note that this is also true for individuals with criminal histories. Defs. Resp. to Pls. SOMF ¶ 339. Defendants argue that the HR policies of certain employers may prohibit employing individuals with criminal histories, and that some individuals with criminal histories may “lack the motivation and hard or soft skills to obtain[] employment.” Id.
According to Plaintiffs, “[Registrants report that it is appearing on the online registry, not their conviction, that turns employers away.” Pls. SOMF ¶ 354. Plaintiffs provide various examples of registrants who claim that their registry status has impeded their job opportunities, but many of their specific facts are contested by defense.
For example, Plaintiffs make the following claims, which Defendants contest:
• Plaintiffs claim that John Doe B worked at a family business for many years because he “learned from experience that he would be unable to find work elsewhere.” Pls. SOMF ¶ 342 (citing Doe B Dep. Tr. at 8, 60-61 (Dkt. 125-5)). He now has his real estate license but he has had sales fall through when a buyer or seller learned he was on the registry. Id. In response, Defendants point out that John Doe B has earned a good livelihood, making $70,000 a year in the family business and approximately $100,000 a year as a real estate agent. Defs. Resp. to Pls. SOMF ¶ 342 (citing Doe B Dep. Tr. at 8-10).
• Plaintiffs claim that John Doe C has been “repeatedly terminated because he is on the registry.” Pls. SOMF ¶ 342 (citing Doe C Dep. Tr. at 9, 17-18, 54-55 (Dkt. 125-6)). He was fired from one job because “an anonymous caller exposed his status.” Id. He was fired from another job after a copy of a newspaper publishing registry listings, including his, appeared in the break room at work. Id. In response, Defendants note that John Doe C has been at the same employer for a decade. Defs. Resp. to Pls. SOMF ¶ 343 (citing John Doe C Dep. Tr. at 10-11).
• Plaintiffs claim that Mary Doe received a certificate in medical billing near the top of her class, but her externship placement “was unwilling to hire her because its name and address would be posted on the registry.” Pls. SOMF ¶ 348 (citing Mary Doe Dep. Tr. at 27-40 (Dkt. 125-10). She submitted over 100 resumes for other jobs before finding one. Id. Her current employer is unaware of her registry status, but Mary Doe expects she would be fired if it were discovered. Id. In response, Defendants submit that Mary Doe was at one employer from 2017 through 2022, and that she left that position voluntarily for a new position with a “huge increase in salary and benefits” that she is still in. Defs. Resp to Pls. SOMF ¶ 348 (citing Mary Doe Dep. Tr. at 38-41).
Regardless of whether there are actual employment difficulties, no court decision of which the Court is aware has found that employment in the private sector is a fundamental right. In fact, the district court in Does I considered and rejected such an argument. Does I, 932 F.Supp.2d 803. As that court explained, “[w]hile the freedom to choose and pursue a career, to engage in any of the common occupations of life, qualifies as a liberty interest . . . there is no ‘general right to private employment.'” Id. at 817 (punctuation modified) (citing Cutshall v. Sundquist, 193 F.3d 466, 479 (6th Cir. 1999)). Stated differently, legislation that bars individuals from pursuing particular careers may infringe on a fundamental right, but legislation that incidentally makes obtaining private employment more difficult does not. Finding that SORA 2011 did not bar the plaintiffs from freely electing a particular career, the court rejected their due process argument on that ground. Id. at 817-818.
SORA 2021-like SORA 2011-does not prohibit registrants from pursuing any particular career. SORA 2021 may make it more difficult for registrants to find employment in general, but as the Does I court noted, there is no general right to private employment. And SORA 2021 imposes less of a burden on finding work than SORA 2011, which prohibited registrants from working within 1,000 feet of a school, thereby eliminating certain job opportunities. “A charge that merely makes a plaintiff less attractive to other employers but leaves open a definite range of opportunity does not constitute a liberty deprivation.” Cutshall, 193 F.3d at 478 (punctuation modified) (holding that Tennessee sex registration statute did not deprive registrant of procedural due process based on impact on his job possibilities, because any past judicial recognition of a constitutionally-protected interest related to employment involved termination of government employment, not private employment).
While Plaintiffs undoubtedly face more difficulty finding employment than those who have not been convicted of a sex offense, they have not shown a fundamental right has been infringed.
With respect to travel, Plaintiffs argue that SORA impacts them in the following ways:
• For any travel of more than seven days, registrants must provide advance notice to the police. Registrants must state where they are going, where they will stay, and when they
will return. Pls. SOMF ¶ 360 (citing Mich. Comp. L. §§ 28.725(1)(e); 28.727(1)(e). Plaintiffs contend that the practical effect of this requirement is that “[a]lmost all Plaintiffs limit travel to no more than six days . . . .” Id. ¶ 366.
• For international travel of more than seven days, registrants must report in person at least 21 days in advance. Id. ¶ 360 (citing Mich. Comp. L. §§ 28.725(8)).
• Registrants “must plan to travel so that they are able to register in person at specified intervals.” Id ¶ 361 (citing Mich. Comp. L. §§ 28.725a(3)).
• “Because registration in one state generally triggers registration in other states, if registrants travel, they must comply with all applicable registration laws in other jurisdictions. Any travel out-of-state requires extensive research to determine what the registry requirements are in the states through which and to which registrants travel.” According to Plaintiffs, this “significantly restricts registrants' ability to associate with family or friends out of state, or indeed to leave the state for any purpose.” Id. ¶ 363.
• Registrants convicted of an offense against a minor can only get passports that identify them as sex offenders. Id. ¶ 365 (citing 22 U.S.C. § 212b(c)(I); 34 U.S.C § 21503(f)).
Plaintiffs cite Mich. Comp. L. § 28.725(1)(e), but this appears to be a typo. The correct provision should be § 28.725(2)(b).
Plaintiffs do not cite to the record in support of this contention. In response, Defendants admit that “several plaintiffs, who are only a few of the approximately 44,000 registrants, made the statement.” Defs. Resp. to Pls. SOMF ¶ 366.
Plaintiffs do not explain why Michigan should be faulted for aspects of other states' registration systems that may be triggered by virtue of an individual being required to register as a sex offender in Michigan.
Plaintiffs provide numerous examples of individual plaintiffs who would like to travel but do not do so, or who have experienced difficulty while traveling, because of their registration status. Id. ¶¶ 367-378. For example:
SORA's three-week notice requirement for international travel limits Doe G's career advancement, as he has had to tell his employer he cannot travel internationally. He was forced to cancel a business trip to China because he would have been away for 30 days during his reporting period. He doesn't visit his father in Florida or sister in Illinois for fear of violating those states' laws or ending up on those states' registries. He missed his niece's graduation for similar reasons.Id. ¶ 373 (citing Doe G Dep. Tr. at 20-22 (Dkt. 125-9).
The right to travel has been recognized as fundamental in certain contexts. As explained in Does I, “[n]either the Supreme Court nor the Sixth Circuit has recognized [] a generically defined fundamental ‘right [to travel].' Instead the Supreme Court has found a fundamental right to interstate travel that is comprised of three distinct components: (1) ‘the right of a citizen of one State to enter and to leave another state'; (2) ‘the right to be treated as a welcome visitor rather than an unfriendly alien when temporarily present in the second State'; and (3) ‘for those travelers who elect to become permanent residents, the right to be treated like other citizens of that State.'” Does I, 932 F.Supp.2d. at 814-815 (emphasis in original) (citing Saenz v. Roe, 526 U.S. 489, 500 (1999)). The court further explained that “the Sixth Circuit, in the absence of any express statement by the Supreme Court, has also recognized a fundamental right to intrastate travel, which it found to include the ‘right to travel locally through public spaces and roadways.'” Id. at 15 (citing Johnson v. City of Cincinnati, 310 F.3d 484, 498 (6th Cir. 2002)).
In Johnson, the Sixth Circuit struck down a city's drug-exclusion ordinance, which banned an individual for up to ninety days from public streets and sidewalks in all drug-exclusion zones if the individual is arrested or taken into custody within any drug-exclusion zone for one of several enumerated drug offenses. Johnson, 310 F.3d at 487. The ordinance in Johnson is distinguishable from SORA 2021, which does not ban registrants from entering any particular areas within the state.
But the Does I court also noted that SORA 2011 did not create any actual barriers preventing registrants from entering or exiting the state. Id. at 815. The court explained that the requirement that registrants report travel of greater than seven days “does not . . . make out-ofstate travel impossible.” Id. “It merely requires more advanced planning and less spontaneity on the part of a traveling registrant.” Id. The court held that “mere burdens on a person's ability to travel from state to state are not necessarily a violation of their right to travel” and that although SORA 2011 “imposes some burdens on registrants who wish to travel for extended period of times, the burdens cannot be fairly characterized as substantial or unreasonable.” Id. The same is true with respect to SORA 2021, which contains the same requirement that registrants report travel of greater than seven days and adds no new restrictions with respect to travel.
The court next addressed the quarterly in-person reporting requirement and held that “[a]t most, the provision imposes only a de minimis, incidental burden on registrants who wish to travel during these periods.” Id. at 816. Again, the same is true of SORA 2021, which requires registrants to report in person either once, twice, or four times a year, depending on their tier. SORA 2021 assigns required reporting months, but registrants can choose any day within those months to report in person. At most, this requirement prevents certain registrants from traveling on four days within the year, and registrants have a fair degree of flexibility in choosing those days. The Does I court's description of this burden as “de minimis” is accurate.
Notably, SORA 2021 imposes significantly less of a burden on travel than SORA 2011, which contained exclusion zones. Yet the Does I court found that SORA 2011, even with exclusion zones, did not substantially burden Plaintiffs' rights to interstate or intrastate travel because the exclusion zones “[did] not restrict registrants from entering or traveling through” exclusion zones. Id. at 816. Rather, the court explained that the exclusion zones “regulat[ed] the type of activity registrants [could] engage in when they are within 1,000 feet of school property.” Id. The court held that the exclusionary zones were “tailored to prohibit only the type of conduct the state has reasonably identified as creating potential harm to children.” Id. at 817. In the present case, SORA 2021 imposes no restrictions on the activities registrants can engage in when they are within 1,000 feet of a school.
Because SORA 2021 only imposes incidental burdens on a registrant's ability to travel, strict scrutiny does not apply.
Other courts in this circuit have reached the same conclusion with respect to Tennessee's sex offender registration statute, which is similar to SORA 2021. See, e.g., Doe v. Rausch, 648 F.Supp.3d 925, 949 (W.D. Tenn. 2023); Jefferies v. Lee, No. 22-cv-02258, 2023 WL 2724249, at *8 (W.D. Tenn. Mar. 30, 2023).
ii. Animus
Even if SORA 2021 does not violate a fundamental right, Plaintiffs argue that an “exacting rational relationship standard” must apply because SORA 2021 was motivated by animus. Pls. Br. Supp. Mot. for Summ. J. at 29-32. Plaintiffs contend that “[t]his more searching form of rational basis review applies when a law exhibits a desire to harm a politically unpopular group.” Id. at 29 (citing Lawrence v. Texas, 539 U.S. 558, 580 (2003) (O'Connor, J., concurring). This argument is echoed in the amicus curie brief filed by law professors William Araiza, Eric Janus, and Sandra Mayson in support of Plaintiffs' equal protection and substantive due process claims. See Araiza Br. (Dkt. 136).
The amicus brief refers to a line of United States Supreme Court cases-which the amici refer to as the “animus cases”-where the court applied the exacting rational relationship standard to laws that “might reflect prejudice or stereotype.” Id. at 7-9. According to the brief, these cases “illustrate the objective features of legislation that suggest potential stereotype or prejudice.” Id. at 7. These features include: (i) “laws that target a politically disfavored group,” (ii) laws that “impose[] broad or severe burdens on the disadvantaged group,” and (iii) “rarity,” or laws that impose “discriminations of an unusual character.” Id. at 7-8 (punctuation modified). Plaintiffs cite similar factors to consider when determining whether a statute was motivated by animus. Pls. Br. Supp. Mot. for Summ. J. at 30-31. In addition to the amicus brief factors, Plaintiffs contend that courts should consider legislative history and background. Id.
The Court notes that Plaintiffs and the amicus brief provide no authority showing that a court has adopted these factors as a test.
A consideration of these factors, however, does not establish that SORA 2021 was motivated by animus. As an initial matter, the present case is distinguishable from the “animus cases” cited by the amicus brief because those cases addressed the regulation of politically disfavored groups who did not do anything wrongful or objectionable. See Romer v. Evans, 517 U.S. 620 (1996) (LGBTQ people); Cleburne, 473 U.S. 432 (1985) (severely intellectually disabled); Plyler v. Doe, 457 U.S. 202 (1982) (children of undocumented immigrants); U.S. Dep't of Agric. v. Moreno, 413 U.S. 528 (1973) (hippies). Unlike these groups, sex offenders did do something wrongful by committing dangerous acts that are often violent in nature. And, in many cases, they did so in extremely concerning ways. If they are “politically disfavored,” that is explainable based on conduct that is understandably troubling to law-abiding citizens. None of the “animus cases” addressed an unpopular group whose members had been convicted of similarly dangerous crimes.
Plaintiffs rely heavily on Bannum, Inc. v. City of Louisville, 958 F.2d 1354 (6th Cir. 1992). In Bannum, the plaintiff argued that a city zoning ordinance violated equal protection because it required community treatment centers for federal offenders (CTCs) to obtain a special permit to operate but did not require operators of other group residences to do the same. Bannum, 958 F.2d at 1355-1356. The CTC program was intended to facilitate the reintegration of federal offenders into society. Id. at 1355. The court applied a “rational relationship level of review,” which asked “whether the classifications drawn by the zoning ordinance are rationally related to a legitimate state interest.” Id. at 1360. The court described the standard as “highly deferential” but also warned that rational relationship review can be “exacting” in that “the desire to impede a politically unpopular group is not a legitimate state interest.” Id. (punctuation modified).
The city's main justification for the ordinance was that the occupants of a CTC are more likely to commit crimes than a person never having been convicted of a crime. Id. The Sixth Circuit held that “[i]f the city's goal was to protect its residents from recidivists, then some data reflecting the extent of the danger must exist in order to render the different treatment of CTCs rationally related to that goal.” Id. at 1360-1361. Explaining that the city's expert witness found that literature on the topic was inconclusive, the court rejected the city's proffered purpose. Id. Instead, the court concluded that “the purpose behind different treatment of CTCs . . . is to assure residents . . . that they would not find themselves with a CTC as a neighbor.” Id. at 1361. The court held that this justification was invalid. Id. Bannum does not provide clear guidance here. Its holding seems to be that the special permit requirement reflected a naked fear of felons generally, without any evidence that they would likely re-offend. Id. at 1360 (“The city was able to present the district court with no evidence supporting its contention that CTCs present a danger to the community.”). That is different from the present case because, as discussed above, there is at least some evidence of recidivism being thwarted by the registration system. It is also different from the present case because, as discussed above, there is some evidence that registration systems deter first-time offenders, and registration systems serve the goal of empowering individuals to pursue measures to promote their own safety.
Further, Bannum appeared to qualify its pronouncement that naked fear of federal offenders justified heightened scrutiny when it stated that “[i]t is important to note that the CTCs . . . do not house felons convicted of crimes of violence involving firearms, or sexual offenders.” Id. at 1361 n.3 (emphasis added). Similarly, it also stated that “the city may well have a legitimate interest in requiring a conditional use permit of an institution housing those with a prior record of violent felony offenses.” Id. at 1361. Thus, Bannum does not suggest that a regulation impacting convicted sex offenders must pass exacting scrutiny on the theory that the law operates against a politically unpopular group. Violent offenders and sex offenders, in particular, gain no right to invoke special constitutional protection under that decision. The “politically unpopular” factor does not weigh in favor of Plaintiffs.
As to the second factor, it is clear that SORA imposes burdens on registrants. Plaintiffs discuss these burdens in detail, particularly as they relate to SORA's impact on access to housing, employment, education, and travel. Defendants do not seem to dispute that SORA imposes such burdens on registrants. This factor weighs in Plaintiffs' favor.
As to the third factor, Plaintiffs argue that SORA is a structural aberration because it is “unlike other laws.” Pls. Br. Supp. Mot. for Summ. J. at 32. According to Plaintiffs, SORA “criminalizes ordinary behavior (that is not criminal for non-registrants) and imposes extensive supervision for decades/life on a disfavored group without any individual assessment.” Id. Plaintiffs contend that other “[n]on-punitive systems that impose significant restrictions on liberty (e.g., child protective services, guardianships, medication compliance for mentally ill), all turn on individual assessments and periodic review to ensure these restrictions are warranted.” Id.
The only case Plaintiffs cite in support of this claim is Bassett v. Snyder, 59 F.Supp.3d 837 (E.D. Mich. 2014). In Basset, five same-sex couples, each with one partner employed by a local municipality, challenged the constitutionality of a Michigan law that prohibited public employers from providing health care and other fringe benefits to the domestic partners of their employees. Id. at 839. The plaintiffs argued that the law was “nothing more than a mean-spirited attempt to deny health care benefits to the same-sex domestic partners of public employees on the basis of their sexual preference, . . . .” in violation of equal protection. Id.
In determining whether the law was motivated by animus, the court asked, in part, whether it was a structural aberration. Id. at 847, 855. The court explained that “[structural aberration occurs when [a] law (1) ‘impose[s] wide-ranging and novel deprivations upon the disfavored group;' or (2) ‘ stray [s] from the historical territory of the lawmaking sovereign to eliminate privileges that a group would otherwise receive.'” Id. at 847 (citing Bishop v. Smith, 760 F.3d 1070 at 1100 (10th Cir. 2014) (Holmes, J., concurring)). The court found that the law in question was a structural aberration because “the legislature took the unusual step of telling local governments that they could not offer employee fringe benefits to a specific class of people who had been receiving them up to then.” Id. at 855.
Basset is distinguishable from the present case. For one, Basset is more similar to the traditional “animus cases” discussed above because it dealt with a politically unpopular group whose members had not done something violent or dangerous. In addition, the law at issue in Basset deprived individuals of benefits they had previously been receiving, due to no fault of the individual. To the extent SORA 2021 deprives registrants of any benefits, it does so because the registrant has been convicted of a dangerous crime.
Further, Plaintiffs admitted in the ex post facto context that “SORA 2021 resembles the traditional punishment of parole/probation.” Pls. Br. Supp. Mot. for Summ. J. at 10 (punctuation modified). In an about-face, Plaintiffs now try to point out various ways in which SORA 2021 differs from parole and probation. For the reasons already discussed in the ex post facto section, the Court finds that SORA 2021 resembles probation and parole. This and Plaintiffs' lack of case law supporting the rarity of SORA 2021 weigh in favor of finding that SORA is not a structural aberration.
Plaintiffs note the following distinctions between SORA 2021 and probation/parole:
• Probation and parole are individually determined, while sex offender registration requirements are not. Pls. Resp. to Defs. SOMF ¶ 44 (citing SORA, Probation & Parole Comparison Chart (Dkt. 131-10)).
• Probation and parole usually only last for two to four years, while registration requirements can last for decades or life. Id. (citing SORA, Probation & Parole Comparison Chart). But the Court notes that, depending on the severity of the crime, parole-like systems like supervised release can last for life. See United States v. Zabel, 35 F.4th 493, 508 (6th Cir. 2022) (“There is no question that the district court had authority to impose a life term of supervised release. Congress insists that lifetime supervision be available to courts in sentencing sexual offenders ....”) (punctuation modified).
• Most probationers and parolees can petition to be discharged early, whereas most sex offender registrants cannot. Id. (citing SORA, Probation & Parole Comparison Chart).
Lastly, Plaintiffs argue that SORA's legislative history indicates that it was motivated by animus. Pls. Br. Supp. Mot. for Summ. J. at 31-32. Plaintiffs cite to legislators who have called registrants ‘beasts,” “monsters,” “animals,” and “the human equivalent of toxic waste.” Id. at 31 (citing Wayne Logan, Knowledge as Power: Criminal Registration and Community Notification Laws in America 95 (1st ed. 2009)). But the Michigan legislature consists of 38 senators and 110 representatives-statements from a handful of members do not represent the views of the legislature at large. See Mich. Const. art. IV, §§ 2-3.
Moreover, courts have recognized the combative tenor of political discourse, instructing that the First Amendment reflects “a profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open.” Snyder v. Phelps, 562 U.S. 443, 452 (2011). For better or worse, the use of inflammatory language is fairly widespread throughout the continuum of political issues. But “[a]s a Nation we have chosen . . . to protect even hurtful speech on public issues to ensure that we do not stifle public debate.” Id. at 461. There is nothing unique in that regard pertaining to the sex offender registration issue. While demeaning comments are certainly troubling, the use of strong language in debates about sex offenders should not be surprising. All sex offenses are condemnable and many involve particularly heinous conduct.
Plaintiffs also emphasize that, for recent enactments of SORA, legislators ignored a “scientific consensus that registries don't work.” Pls. Br. Supp. Mot. for Summ. J. at 31. But as already discussed, the literature regarding the effect of registries on recidivism is not conclusive. These arguments, therefore, fail to establish that the legislative history demonstrates that SORA was motivated by animus.
In sum, considering all of the factors that courts have considered in the animus cases, the present case hardly seems like it falls within that category. Further, the Court is not aware of any case holding that sex offenders are subject to exacting scrutiny or otherwise comparing sex offenders to the groups at issue in the cases cited by Plaintiffs and amici.
iii. Rational Basis Review
Plaintiffs argue that SORA 2021 is unconstitutional no matter what standard of review is applied because it cannot survive even rational basis review. Pls. Br. Supp. Mot. for Summ. J. at 33-35. Under rational basis review, the relevant question is whether the statute at issue is “rationally related to legitimate government interests.” Doe v. Mich. Dep't of State Police, 490 F.3d 491, 501 (6th Cir. 2007) (citing Washington v. Glucksberg, 521 U.S. 702, 728 (1997)). “That deferential standard does not require States to show that a classification is the only way, the best way, or even the most defensible way to achieve their interests.” Gore v. Lee, 107 F.4th 548, 561 (6th Cir. 2024).
Plaintiffs argue that SORA 2021 does not meet this standard because “imposing extensive burdens that have no public safety benefit on thousands of people who present no appreciable risk is irrational.” Pls. Br. Supp. Mot. for Summ. J. at 33 (punctuation modified). According to Plaintiffs, “unrebutted evidence” shows SORA does not promote public safety and instead shows that SORA undermines the key factors for reentry (housing, employment, and social connections) and harms survivors. Id. at 33. Further, Plaintiffs contend that SORA wastes millions of dollars each year on “a system that undermines its very purpose.” Id. In response, Defendants argue that the state has a compelling interest in protecting public safety. Defs. Br. Supp. Mot. for Summ. J. at 31.
As already discussed, the social science regarding the efficacy of SORA in reducing recidivism is mixed. And Plaintiffs' theory that SORA is irrational because it promotes recidivism by impeding “key factors for reentry” to society such as housing and employment, Pls. Br. Supp. Mot. for Summ. J. at 33, has been questioned by the Office of Justice Programs report cited above. It noted that studies examining the impacts of registration on offenders' “employment and finances, housing, and physical and psychological well-being” were limited in that the studies were largely based on “self-reported data provided by [registrants], their family members, and treatment providers in surveys and interviews with researchers.” Office of Justice Programs, Sex Offender Registration and Notification Act-Summary and Assessment of Research, at 9 (April 2022). Given the apparent indeterminate state of the scientific record, it cannot be said that Michigan has selected an irrational registration scheme on the theory that it promotes recidivism.
The indeterminate state of scientific opinion also serves to temper the assertion of Plaintiffs that lengthy registration periods-and in particular lifetime registration-is irrational. This is predicated on the contention that after several years-perhaps as few as 10-the recidivism rate for registrants with an average risk level who remain offense-free equals the risk of offending for all males in the general population. Pls. SOMF ¶ 187.
However, determining who is offense-free is a fraught undertaking, given the severe problem of underreporting of sex offenses. As Rachel Lovell, director of the Criminology Research Center at Cleveland State University, stated in her declaration: “[s]exual recidivism research based on official, court/administrative records from criminal justice agencies provides biased and unrepresentative estimates of repeat sexual offending.” Lovell Decl. ¶ 6 (Dkt. 128-19). According to Lovell, sexual assault “is the most underreported violent crime” in the country. Id. Only about a third of such offenses are reported to law enforcement and five percent or less result in a conviction. Id. Thus “sexual recidivism cannot be used interchangeably with repeat sexual offending.” Id. ¶ 4. Lovell opines that one meta-analysis of 808 empirical studies reported sexual recidivism rates that varied from 0 to 68%. Id. ¶ 7. Lovell asserts that sexual recidivism estimates cannot be reliably used to determine if sexual offenders live offense-free. Id. at ¶ 9.
Similar conclusions are reached by Rachel Goodman-Williams, an assistant professor of psychology at Wichita State University. She opined:
Recidivism studies based solely on criminal history data are therefore not measuring sexual offenders' behaviors so much as they are measuring the behavior of the criminal legal system; they are not effectively measuring multiple instances of sexual offending so much as they are measuring multiple instances of being caught and held accountable by a system with a uniquely poor track record of doing so. Measures of serial sexual offending that include data less vulnerable to case attrition suggest that approximately 40% of sexual offenders are serial sexual offenders.Goodman-Williams Decl. ¶¶ 1, 51 (Dkt. 128-20) (emphasis omitted). Thus, there is nothing irrational in maintaining a registration system with lengthy periods of registration.
Nor is it irrational for the system to cover people who have a low risk of reoffending. As an example, even accepting one of Plaintiffs' expert's statistics that 80% to 90% of convicted male offenders will never be reconvicted of a new sex crime, Letourneau Report ¶ 12, that still leaves 10% to 20% who well may recidivate. Rationality does not demand a perfect or even a “good” success rate to pass constitutional muster. See Gore, 107 F.4th at 561 (6th Cir. 2024).
For the same reason, Plaintiffs' contention that most sex offenses are not committed by strangers, but by persons known to their victims, is of no constitutional moment. See Pls. SOMF ¶ 159 (citing Socia Report ¶ 4 (Dkt. 123-11); Prescott Report ¶ 32 (Dkt. 123-10); Lovell Decl. ¶ 10). Their theory appears to be that registration serves no purpose as to victims who know their perpetrator because they already are aware of their perpetrator's past. Id. ¶ 160. But other victims do not know of that past and may want to avoid becoming a fresh victim. And even according to Plaintiffs' estimates, 13-15% of sex crimes reported to the police are committed by strangers. Id. ¶ 159. Providing information that will help the public protect themselves from even a portion of reported sex crimes is a legitimate goal. Again, the Constitution does not demand legislative perfection.
As discussed previously, rationality is further established by recognizing that SORA serves at least two other legitimate purposes completely unrelated to recidivism: deterring firsttime offenders and providing information that individuals can use to protect themselves and their families. These purposes alone are each legitimate government purposes sufficient to justify SORA under rational basis review.
Although there are no federal cases adopting Plaintiffs' argument, Plaintiffs cite to certain state cases that they contend “have held that non-reviewable lifetime registration without any opportunity for judicial review violates due process ....” Pls. Br. Supp. Mot. for Summ. J. at 34. All are distinguishable.
In Powell v. Keel, 860 S.E.2d 344 (S.C. 2021), the respondent was convicted of a sex offense and required to register under South Carolina's SORA for life. Powell, 860 S.E.2d at 345 346. As explained by that state's supreme court, South Carolina's SORA “generally mandate[d] that a person required to register as a sex offender must do so biannually for life.” Id. at 347. The respondent challenged the constitutionality of this lifetime requirement, arguing that it violated various constitutional protections, including due process. Id. at 346.
The court agreed with the respondent and held that “SORA's lifetime registration requirement without judicial review violated due process.” Id. at 348. But the court specifically held that “the initial mandatory imposition of sex offender registration” was constitutional. Id. Powell only found that “SORA's lifetime registration requirement without any opportunity for judicial review to assess the risk of re-offending” violated due process. Id.
But Michigan SORA 2021, unlike South Carolina SORA, only imposes lifetime registration on Tier III offenders-there is a degree of tailoring imbedded in the statute. This ameliorates the Powell court's concern that “there is no evidence in the record that current statistics indicate all sex offenders generally pose a high risk of re-offending.” Id. at 349. Under Michigan SORA, not all sex offenders are required to register for life. And as already discussed, the social science regarding the risk of recidivism is mixed. The Michigan legislature has determined that individuals who committed certain offenses should be required to register for life without any chance for reassessment. This determination is far more reasonable than requiring all registrants to register for life. Powell, therefore, is not directly applicable to the present case.
The court in Powell described the State's purpose in enacting SORA as “protecting the public from a high risk of re-offending.” Powell, 433 S.C. at 466. The “high risk of re-offending” language most likely came from the “purpose” provision of South Carolina's SORA, which states that “[statistics show that sex offenders often pose a high risk of re-offending.” Id. at 348 (citing S.C. Code Ann. § 23-3-400 (2007 & Supp. 2020)). This Court notes that Michigan SORA 2021 does not include the same language in its statement of purpose. Instead, Michigan SORA 2021 states that “[t]he legislature has determined that a person who has been convicted of committing an offense covered by this act poses a potential serious menace and danger to the health, safety, morals, and welfare of the people, and particularly the children, of this state.” Mich. Comp. L. § 28.721a. The legislature could reasonably consider an individual with even a low risk of reoffending to be a “potential serious menace and danger.”
Plaintiffs also cite to Doe v. Dep't of Pub. Safety, 444 P.3d 116 (Alaska 2019). In that case, the plaintiff argued that the Alaska Sexual Offender Registration Act (ASORA) violated the due process clause of the Alaska Constitution by requiring all sex offenders to register without providing a procedure for them to establish that they do not represent a threat to the public. Id. at 119. The plaintiff alleged that ASORA infringed on a number of fundamental rights: the right to integrate into society, the right to privacy, the right to be let alone, and the right to pursue employment. Id. at 124. Finding that the right to privacy “is an explicitly enumerated right under the Alaska Constitution” that “should generally be considered fundamental,” the court employed strict scrutiny. Id. at 126. The court ultimately held that “ASORA's coverage is excessive to the extent it applies to sex offenders who do not present a danger of committing new sex offenses.” Id. at 132. But rather than invalidate ASORA, the Doe v. Dep't of Pub. Safety court allowed the plaintiff to file a civil action in the superior court “in which he would be permitted to attempt to prove that he no longer poses a risk to the public that justifies continued registration.” Id. at 135.
Doe v. Dep't of Pub. Safety is distinguishable because it was grounded in the Alaska constitution's guaranty of privacy. There is no case finding an analogous right in the U.S. Constitution applicable to a sex offender registry challenge. Indeed, Plaintiffs make no claim that any alleged right to privacy is at issue in this case.
Lastly, Plaintiffs cite to State v. Bani, 36 P.3d 1255 (Haw. 2001), as amended on clarification (Dec. 6, 2001). In Bani, the appellant argued that Hawaii's sex offender registration and notification statute violated his procedural due process rights by failing to provide him with notice and an opportunity to be heard before being subjected to the statute's public notification provisions. Bani, 36 P.3d at 1262-1263. As discussed above, federal precedents require rejection of any procedural due process challenge to SORA.
2. Equal Protection
Plaintiffs also argue that SORA 2021 violates equal protection by failing to provide for individualized review. Am. Compl. ¶¶ 746-756. But Plaintiffs do not develop this argument in their brief supporting their motion for summary judgment beyond mentioning equal protection in a section header and arguing that a lack of individualized review fails rational basis review. Pls. Br. Supp. Mot. for Summ. J. at 28-35. Plaintiffs do not attempt to show that they have been disparately treated as compared to similarly-situated persons, as is required of any equal protection claim. See Ctr. for Bio-Ethical Reform, Inc. v. Napolitano, 648 F.3d 365, 379 (6th Cir. 2011).
“It is not the court's responsibility to craft winning legal arguments for” the parties. United States v. Mungarro, No. 07-20076, 2020 WL 1933816, at *3 (E.D. Mich. Apr. 22, 2020). Accordingly, the Court declines to address Plaintiffs' equal protection claim as it relates to a lack of individualized review.
C. Unequal Opportunity to Petition for Removal
Plaintiffs argue that denying similarly-situated registrants the opportunity to petition for removal from the registry violates the Equal Protection Clause. Am. Compl. ¶¶ 757-772. But Plaintiffs have failed to show that they have been treated disparately than similarly-situated persons, so this claim fails.
Under SORA 2021, only two groups of registrants are allowed to petition for removal.
For those eligible to petition, SORA 2021 establishes the required procedure. Mich. Comp. L § 28.728c(4)-(11). A copy of the petition must be filed with the office of the prosecuting attorney that prosecuted the case against the registrant, and any known victim must be notified. Id. § 28.728c(7) and (8). The court must then conduct a hearing, at which the victim may speak. The court must consider the following factors in determining whether to allow the individual to discontinue registration: (i) the individual's age and level of maturity at the time of the offense, (ii) the victim's age and level of maturity at the time of the offense, (iii) the nature of the offense, (iv) the severity of the offense, (v) the individual's prior juvenile or criminal history, (vi) the individual's likelihood to commit further listed offenses, (vii) any impact statement submitted by the victim, and (viii) any other information considered relevant by the court. Id. § 28.728c(11). But the court “shall not grant the petition if the court determines that the individual is a continuing threat to the public.” Id.
First, Tier I registrants can petition for removal ten years after the later of their conviction or release from confinement if they meet certain conditions. Mich. Comp. L. § 28.728c(1), (12). Tier I registrants are required to register for 15 years, so this means that they can petition for removal from the registry up to five years early.
Tier I registrants may petition for removal ten years after the later of conviction or release from confinement if they: (i) have not since been convicted of any felony or other registrable offense, (ii) have successfully completed their assigned periods of supervised release, probation, or parole, and (iii) have successfully completed a sex offender treatment program. Mich. Compl. L. § 28.728c(1), (12).
Second, registrants subject to SORA 2021 for a juvenile adjudication can petition for removal from the registry 25 years after the later of their adjudication or release from confinement if they meet certain conditions. Mich. Comp. L. § 28.728c(2), (13). Although not emphasized by either party, the Court notes that juveniles are only required to register under SORA 2021 if their adjudication is for the commission of an offense that would classify the individual as a Tier III offender. Mich. Comp. L. § 28.722(a)(iii). So unlike Tier III adult registrants who are required to register for life and may never petition for removal from the registry, Tier III juveniles may petition after 25 years. This still allows many juvenile registrants to live free from registration requirements for many years, if not decades, of their lives.
Juvenile registrants must meet the same conditions as Tier I registrants described in the footnote above.
Tier I registrants and juvenile registrants account for 7% and 5% of all registrants, respectively. Pls. SOMF ¶ 218, 433 (citing Class Data Report ¶¶ 5, 18, 42, 93-98 (Dkt. 123-6)). This means that only 12% of SORA registrants can petition for removal, and even then, only if they meet certain conditions. The other 88% of registrants can never petition for removal. Plaintiffs argue that such a scheme “violates equal protection because (1) the barred-from-petitioning sub-class is treated differently than others . . . who are similarly situated in all material respects, and (2) the state has no rational basis for the different treatment.” Pls. Br. Supp. Mot. for Summ. J. at 35-36 (punctuation modified).
The Equal Protection Clause prevents a state from denying its citizens “equal protection of the laws.” U.S. Const. amend. XIV, § 1. The provision “does not forbid States from drawing distinctions in their laws . . . . [b]ut it does prohibit States from allocating benefits and burdens based on suspect and irrational classifications.” Gore, 107 F.4th at 555.
“To state an equal protection claim, a plaintiff must adequately plead that the government treated the plaintiff disparately as compared to similarly situated persons and that such disparate treatment either burdens a fundamental right, targets a suspect class, or has no rational basis.” Ctr. for Bio-Ethical Reform, Inc., 648 F.3d at 379 (punctuation modified). The “threshold element” of an equal protection claim is disparate treatment. Id. “[O]nce disparate treatment is shown, the equal protection analysis to be applied is determined by the classification used by government decision-makers.” Id. “Laws that discriminate based on suspect classifications, such as race or sex, receive heightened review.” Gore, 107 F.4th at 555 (citing City of Cleburne v. Cleburne Living Ctr., 473 U.S. 432, 440-441 (1985)). If a law does not discriminate based on a suspect classification, it is “presumed to be valid” and rational basis review will apply. Id.
As a threshold matter, the barred-from-petitioning subclass is not similarly situated to other SORA 2021 registrants. Plaintiffs argue that “[t]he barred-from-petitioning subclass is similarly situated to petition-eligible registrants in all respects material to that purpose: both groups have registered for ten years; have not been convicted of another registrable offense or felony; and have successfully completed supervised release, probation or parole, and any required sex offender treatment. The only difference (their tier) is not relevant to the statutory goal of providing a path off the registry for people who are rehabilitated. Individuals in any tier can be rehabilitated.” Pls. Br. Supp. Mot. for Summ. J. at 37-38. But the characterization of tier as “the only difference” between the barred-from-petitioning subclass and other registrants is inaccurate. The barred-from petitioning subclass, by definition, committed offenses that the legislature has deemed more serious than other registrants, either because of the nature of the offense itself, or the age of the offender. This is a material difference.
But even if the Court found that the barred-from-petitioning subclass was similarly situated to other SORA 2021 registrants, the statutory scheme satisfies rational basis review. Plaintiffs' various arguments to the contrary fail for the following reasons.
Plaintiffs contend that “[s]tates may punish some offenses more severely than others, but equal protection forbids line-drawing based on offense history where such line-drawing is not related to the statutory purpose.” Id. at 38. In support of this proposition, Plaintiffs rely on Baxstrom v. Herold, 383 U.S. 107 (1966). But Plaintiffs' reliance on Baxstrom is misguided.
In Baxstrom, the petitioner was certified as insane by a prison physician shortly after he was sentenced to a term of two and one-half to three years in a New York prison. Id. at 107. The petitioner was transferred from prison to a New York Department of Corrections hospital for mentally ill prisoners. Id. Shortly before the petitioner's sentence was over, the director of the hospital filed a petition requesting that the petitioner be civilly committed. Id. The petitioner was committed without a jury trial. Id. at 108-109. The Supreme Court held that denying prisoners a jury trial to determine their sanity, while affording such a right to people not in prison facing civil commitment, violated equal protection. Id. at 110-115. As the Supreme Court explained,
The capriciousness of the classification employed by the State is thrown sharply into focus by the fact that the full benefit of a judicial hearing to determine dangerous tendencies is withheld only in the case of civil commitment of one awaiting expiration of penal sentence. A person with a past criminal record is presently entitled to a hearing on the question whether he is dangerously mentally ill so long as he is not in prison at the time civil commitment proceedings are instituted. Given this distinction, all semblance of rationality of the classification, purportedly based upon criminal propensities, disappears.Id. at 115. Baxstrom, therefore, does not stand for the proposition that equal protection forbids line-drawing based on offense history; it prohibited line-drawing based on custodial status.
Plaintiffs also argue that SORA 2021's scheme is irrational because tier classifications do not correspond to risk. Pls. Br. Supp. Mot. for Summ. J. at 39. But the legislature could reasonably conclude that adults who committed more serious crimes should be monitored for longer periods of time without the opportunity for removal. Making a mistaken judgment about who is not dangerous carries potentially more severe consequences when made about a more serious offender.
Lastly, the Court notes that Plaintiffs' arguments regarding juvenile offenders are inaccurate. Plaintiffs contend that “[Requiring juveniles to wait 15 years longer than adults to petition is [] irrational” and argue that “[c]ourts have repeatedly struck down laws that subject juvenile registrants to stricter requirements than adults.” Id. at 39-40. But as already explained, SORA 2021 does not subject juvenile registrants to stricter requirements than adults. In fact, the opposite is true: SORA 2021 allows all juvenile registrants to petition for removal after 25 years, whereas Tier II and Tier III adults can never petition for removal.
For these reasons, Plaintiffs' equal protection claim based on registrants' unequal opportunity to petition for removal fails.
D. Plea Agreements
Plaintiffs argue that SORA 2021 violates due process by retroactively changing the terms of plea agreements. Am. Compl. ¶¶ 789-797. Plaintiffs bring this claim on behalf of the plea bargain subclass, which is defined as “members of the primary class who gave up their right to trial and pled guilty to a registrable offense in Michigan and who, as a result of retroactive amendments to SORA, (a) were retroactively subjected to SORA even though there was no registration requirement at the time of their plea; or (b) had their registration terms retroactively extended beyond that in effect at the time of their plea.” 5/18/22 Order at 3.
Plaintiffs argue that due process protects plea agreements in two ways: by requiring (i) defendants' waiver of constitutional rights to be knowing and voluntary and (ii) agreements with defendants to be fulfilled. Pls. Br. Supp. Mot. for Summ. J. at 48-49. According to Plaintiffs, “[b]y retroactively lengthening or imposing lifetime registration on people who reasonably expected no or shorter registration in return for pleading guilty, Michigan has changed the terms of plea deals.” Id. at 52. Plaintiffs contend that this violates both due process protections afforded to plea agreements. Id. at 50.
However, the Court has now determined that SORA 2021 violates the Ex Post Facto Clause, which ruling will prohibit retroactive enforcement of amendments to SORA. This means that no plea agreements will be retroactively modified or go unfulfilled. It does not appear that any further finding need be reached with respect to this claim. This claim is therefore moot.
E. Non-Sex Offenses
Plaintiffs make two arguments regarding non-sex offenses. Am. Compl. ¶¶ 798-808. First, Plaintiffs argue that the state cannot label someone as a sex offender, by including that person on a sex offender registry, unless that person's offense involved sex as an element or circumstance. Pls. Br. Supp. Mot. for Summ. J. at 53-55. Second, Plaintiffs argue that if the elements of an offense do not include a sexual element, then the offender has a right to a judicial determination regarding whether or not the circumstances of the offense involved sex. Id. at 55-57.
According to Plaintiffs, there are two situations where SORA 2021 requires registration for offenses without a sex element. Am. Compl. ¶¶ 799-801. The first is the “catch-all” provision, which requires registration for an offense “that by its nature constitutes a sexual offense against an individual who is a minor.” Id. ¶ 800 (citing Mich. Comp. L. § 28.722(r)(vii)). For such offenses, the statute requires a judicial determination of whether the offense “by its nature constitute[d] a sexual offense.” Id. (citing Mich. Comp. L. § 769.1(13)). Plaintiffs do not challenge the constitutionality of this “catch-all” provision. Id. ¶ 801.
Plaintiffs note that “[t]he statutory cross-references to SORA in MCL § 769.1(13) were not amended when SORA was amended, and thus refer to an older version of SORA.” Am. Compl. ¶ 800 n.26.
In addition, SORA 2021 requires registration for people convicted of certain offenses against minors, regardless of whether there is a sexual component to the crime. Id. ¶ 801. For these offenses, registration is required automatically. Plaintiffs argue that procedural due process requires a judicial determination that a person committed a sex offense before the state can brand them as a sex offender. Plaintiffs also argue that the current scheme-in which some non-sex offenders are provided a judicial determination whereas others are not-violates equal protection. Id. at 56-57.
These offenses include: (i) kidnapping a minor (Mich. Comp. L § 750.349); (ii) unlawful imprisonment of a minor (id. § 750.349b); (iii) leading away of child under 14 (id. § 750.350); or (iv) a comparable out-of-state offense. Id. § 28.722(r)(iii), (r)(x), (v)(ii), v(iii), (v)(vii). Kidnapping a minor and leading away of a child under 14 constitute Tier III offenses, which require lifetime registration. Id. § 28.722(v)(ii)-(iii). Unlawful imprisonment of a minor is a Tier offense, requiring 15-year registration. Id. 28.722(r)(iii).
Rather than responding to the merits of Plaintiffs' constitutional arguments, Defendants contend that Plaintiffs' claims are moot and that class-wide relief is not appropriate. Defs. Mot. for Summ. J. at 22-28. The Court disagrees.
1. Mootness
Defendants argue that Plaintiffs' claims are moot. Id. at 22-26. In People v. Lymon, __N.W.3d __, No. 164685, 2024 WL 3573528 (Mich. July 29, 2024), the Michigan Supreme Court affirmed a Michigan Court of Appeals decision holding that requiring non-sex offenders to register under SORA 2021 violates Michigan's constitutional protection against cruel or unusual punishment.
The Michigan Supreme Court vacated the Michigan Court of Appeals opinion “insofar as its conclusions went beyond the consideration of non-sexual offenders” but “affirm[ed] its judgment that defendant and other offenders whose crimes lacked a sexual component are entitled to removal from the sex-offender registry.” Lymon, 2024 WL 3573528, at *17.
The parties agree that after the Michigan Court of Appeals decision in Lymon, the MSP created a procedure to remove certain offenders convicted of offenses without a sex element from the registry. Defs. SOMF ¶¶ 122-123; Pls. Resp. to Defs. SOMF ¶¶ 122-123. As part of this procedure, the MSP identified people registered based solely on a Michigan offense without a sex element. Defs. SOMF ¶ 122. The MSP notified the convicting courts, law enforcement agencies, prosecutors, affected registrants, and the Prosecuting Attorneys Association of Michigan that “[a]ffected registrants would be removed from the registry unless the prosecutor provided information indicating that there was a sexual component to a registrant's crime.” Id. ¶ 123. Prosecutors were given 90 days to decide whether the affected registrant must be required to register. Pls. SOMF ¶ 458 (citing Lymon Prosecutor Letter (Dkt. 128-7)). Prosecutors were told that their assessment should be based on whether the underlying offense conduct had a sexual component. Id. (citing Lymon Prosecutor Letter). This process resulted in the removal of all people registered based solely on a Michigan offense without a sex element, except for 14 individuals for whom the prosecutor determined that the offense was sexual in nature. Defs. SOMF ¶ 124.
Defendants argue that the MSP removed “nearly all” of the non-sex offense subclass from the registry in the wake of Lymon. Defs. Mot. for Summ. J. 24. Plaintiffs disagree and estimate that about 146 members of the non-sex offense subclass remain on the registry. Pls. SOMF ¶ 460 (citing Elbakr Decl. ¶¶ 7-12 (Dkt. 123-27); Class Data Report ¶ 24.d). The exact reason for the parties' disagreement is not clear. The parties agree that the MSP's process did not apply to people subject to SORA based on out-of-state convictions for offenses without a sex element. Pls. SOMF ¶ 459; Defs. Resp. to Pls. SOMF ¶ 459. They also agree that it did not apply to people with convictions for both a sex offense and an offense without a sex element. Defs. SOMF ¶ 125; Pls. Resp. to Defs. SOMF ¶ 125. Defendants dispute Plaintiffs' claim that roughly 146 members of the non-sex offense subclass remain on the registry, but they do not explain the basis for their dispute or offer a different estimate. Defs. Resp. to Pls. SOMF ¶ 460.
Plaintiffs explain that “because the removal of the non-sex offense can affect a person's tiering (which controls both registration obligations and length of registration), some such class members may be entitled to reduced obligations or be entitled to removal if the registration term for their sex offense has run.” Pls. Lymon Br. at 3 n.2 (Dkt. 155).
“[A] case is moot when the issues presented are no longer ‘live' or the parties lack a legally cognizable interest in the outcome.” Powell v. McCormack, 395 U.S. 486, 496 (1969). Here, at the very least, the parties agree that 14 individuals remain on the registry for offenses without a sex element. As Plaintiffs note, the Michigan Supreme Court in Lymon “had no reason to consider-and did not consider-what procedures should apply to determine who is a ‘non-sex offender.'” Pls. Lymon Br. at 4. Therefore, the issue of whether these 14 individuals were afforded proper process constitutes a live issue for the Court to decide.
2. Class-Wide Relief
Defendants also argue that class wide relief is not appropriate for the offenders convicted of offenses without a sex element who were not removed as a result of Lymon. Defs. Br. Supp. Mot. for Summ. J. at 27-28. Because there are “no longer questions of fact common to the class,” Defendants contend that these registrants would have to file an independent lawsuit seeking relief under Lymon. Id. In response, Plaintiffs argue that “[t]he subclass is seeking common relief: a bar on registration absent a judicial determination that their offense was sexual in nature. Subclass members thus share a claim for procedural relief, even if the requested judicial determinations will result in varied outcomes.” Pls. Reply at 22.
Plaintiffs are correct that a common claim regarding inadequate procedures is appropriate for class treatment. See, e.g., Barry v. Lyon, 834 F.3d 706, 721-722 (6th Cir. 2016) (finding that class certification was appropriate for a group of plaintiffs challenging Michigan's procedure for terminating food assistance benefits for persons with an outstanding felony warrant).
Because Plaintiffs' non-sex claim is not moot and is appropriate for class treatment, the Court turns to the merits of Plaintiffs' claim.
3. Merits of Plaintiffs' Non-Sex Claim
First, Plaintiffs argue that the state cannot label someone as a sex offender unless that person's offense involved sex. Pls. Br. Supp. Mot. for Summ. J. at 53-55. Defendants do not dispute this point, and for good reason. As already discussed, the Michigan Supreme Court recently held that requiring non-sexual offenders to register under SORA 2021 violates Michigan's constitutional protection against cruel or unusual punishment. Lymon, 2024 WL 3573528, at *17. Multiple other courts have also held that requiring people to register as sex offenders when their offense did not involve sex violates substantive due process. See, e.g., ACLU of New Mexico v. City of Albuquerque, 137 P.3d 1215 (N.M. Ct. App. 2006); People v. Bell, 778 N.Y.S.2d 837 (N.Y. Sup. Ct. 2003); State v. Small, 833 N.E.2d 774 (Ohio Ct. App. 2005).
Second, Plaintiffs argue that the MSP's procedure violates procedural due process. Pls. Br. Supp. Mot. to Dismiss. at 55-56. As mentioned earlier, procedural due process rights “protect individuals from deficient procedures that lead to the deprivation of cognizable liberty interests.” Bambach v. Moegle, 92 F.4th 615, 624 (6th Cir. 2024) (punctuation modified). The fundamental requirement of due process “is the opportunity to be heard at a meaningful time and in a meaningful manner.” Mathews v. Eldridge, 424 U.S. 319, 333 (1976) (punctuation modified). As explained by Plaintiffs, “[t]he MSP left it to prosecutors to decide whether registration should be required, based on their assessment of the alleged facts. There was no notice, no opportunity to be heard, and no judicial determination that the offense was sexual in nature.” Pls. Lymon Br. at 2 (emphasis in original).
This due process argument is distinct from that rejected by the Supreme Court in Conn. Dep't of Pub. Safety v. Doe (DPS), 538 U.S. 1 (2003). As Plaintiffs explain, DPS “rejected a procedural due process claim because registration turned solely on whether the person had been convicted of a sex crime, a fact that a convicted offender has already had a procedurally safeguarded opportunity to contest .... [b]y contrast, for a person who was not convicted of a sex crime, there is no prior adversarial proceeding (with due process protections) where it was determined that a sex crime was committed, and so procedural protections are required.” Pls. Br. Supp. Mot. for Summ. J. at 55 (punctuation modified).
Multiple circuit courts that have considered the issue have held that persons who have never been convicted of a sex offense are entitled to procedural due process before being classified as a sex offender. See, e.g., Renchenski v. Williams, 622 F.3d 315, 320-332 (3d Cir. 2010) (holding that defendant convicted of murder that involved a sexual component was entitled to “an effective but informal hearing” before being classified as a sex offender); Coleman v. Dretke, 395 F.3d 216, 221-225 (5th Cir. 2004) (holding that state was required to provide procedural protections before imposing sex offender registration on prisoner who was convicted of burglary of a habitation); Kirby v. Siegelman, 195 F.3d 1285, 1290-1292 (11th Cir. 1999) (holding that a prisoner convicted of murder was entitled to due process before the state declared him to be a sex offender). This Court reaches the same conclusion.
The MSP's procedure of relying on unilateral input from prosecutors does not afford individuals notice or any opportunity to be heard. Sex offender registration imposes a significant burden on registrants. Registration involves regular reporting and public disclosure of personal information and can impact a registrant's career and housing opportunities. For individuals convicted of offenses without a sex element, the harm that would come from being required to register as a sex-offender despite not committing a crime that was sexual in nature would be significant. Individuals have a right to provide input regarding that determination.
Individuals also have the right to a hearing before a neutral decisionmaker. See Hicks v. Comm'r of Soc. Sec., 909 F.3d 786, 800 (6th Cir. 2018) (holding that “any time a citizen is deprived of notice of the factual basis for a governmental determination and a fair opportunity to rebut the Government's factual assertions before a neutral decisionmaker, the risk of [an erroneous deprivation] is too high.”) (punctuation modified). Unlike judges, prosecutors and police do not play a neutral role in our adversarial legal system. Allowing judges to make the final decision regarding an individual's registration requirements is the long-recognized way to reduce the risk of an erroneous deprivation of a constitutionally-protected interest.
As to the government's interest, it is not clear how much more burdensome requiring a judicial determination of whether an offense without a sex element was sexual in nature would be as compared to MSP's current procedures. Neither party offered an estimate of the costs or burdens associated with judicial review. But given the fact that judicial review is already the procedure employed for offenses falling under the “catch-all” provision, it seems likely that requiring judicial review for all offenses where sex is not an element would not be unduly burdensome.
The Court awards summary judgment to Plaintiffs on this claim. The specific form of judicial determination that must be afforded to people convicted of offenses without a sex element will be addressed at a later phase in this case.
F. Non-Michigan Convictions
Plaintiffs argue that the MSP's process for registering people with non-Michigan convictions is unconstitutional because (i) registrants are not given notice or an opportunity to be heard before being placed on the registry, in violation of due process; and (ii) people with out-ofstate convictions are subjected to harsher treatment than people with Michigan convictions, in violation of equal protection. Pls. Mot. for Summ. J. at 67-72. The Court agrees with Plaintiffs on both points.
Plaintiffs also argue that the treatment of people with out-of-state convictions violates the Privileges and Immunities Clause, U.S. Const. art. IV, § 2, cl. 1. See Pls. Br. Supp. Mot. for Summ. J. at 70-72. Because the Court holds that such treatment violates equal protection, the Court need not address Plaintiffs' alternate argument.
Under SORA 2021, individuals with out-of-state convictions who live, work, or go to school in Michigan are required to register as sex-offenders in Michigan in two circumstances: (i) if they have been convicted of an offense that is “substantially similar” to a registrable Michigan offense, Mich. Comp. L. § 28.722(r)(x), (t)(xiii), (v)(viii); and (ii) if they are required to register or otherwise be identified as a sex or child offender or predator under a comparable statute of another state, id. § 28.723(1)(d). The parties agree that SORA does not define what it means for an offense to be substantially similar to a registrable Michigan offense. Pls. SOMF ¶ 544; Defs. Resp. to Pls. SOMF ¶ 544.
The process for deciding whether someone with a non-Michigan conviction must register under SORA-and if registration is required, at what tier-is currently handled by the Sex Offender Registry (SOR) Unit of the MSP. Pls. SOMF ¶ 536; Defs. Resp. to Pls. SOMF ¶ 536. In making this determination, SOR Unit staff may utilize flow charts approved by MSP's legal division. Pls. SOMF ¶ 537 (citing Jegla Dep. Tr. at 92-98, 146 (Dkt. 126-4), Beatty Dep. Tr. at 138-139 (Dkt. 126-1), MSP Flowcharts (Dkt. 127-10)); Defs. Resp. to Pls. SOMF ¶ 537 (citing Beatty Dep. Tr. at 137-140). The flow chart for out-of-state adult convictions asks whether there is a comparable Michigan conviction, whether the convicting state requires registration, whether the conviction was discharged on or after October 1, 1995, and whether the comparable Michigan offense, if there is one, would be a Tier III offense. MSP Flowcharts at PageID.6546.
To determine whether an out-of-state conviction has a comparable Michigan offense, SOR Unit staff utilize a spreadsheet listing more than 2,200 out-of-state offenses, 1,471 of which are assigned a SORA tier classification. Pls. SOMF ¶ 540 (citing PACC Code Chart (Dkt. 127-1), Jegla Dep. Tr. at 101, Elbakr Decl. ¶ 21); Defs. Resp. to Pls. SOMF ¶ 540. But only 43 offenses in the spreadsheet are assigned a “comparable Michigan code,” which Defendants contend is because some out-of-state offenses have language that requires review on a case-by-case-basis. Pls. SOMF ¶ 541 (citing PACC Code Chart); Defs. Resp. to Pls. SOMF ¶ 541 (citing Morris Dep. Tr. at 149-152 (Dkt. 126-8)). Of those 43, only 18 are shown as “approved by MSP legal.” Pls. SOMF ¶ 541 (citing PACC Code Chart).
MSP's legal advisor has admitted that different people can reach different conclusions about whether a given non-Michigan offense is “substantially similar” to a given Michigan offense. Pls. SOMF ¶ 545 (citing Beatty Dep. Tr. at 158, 186, 214-215); Defs. Resp. to Pls. SOMF ¶ 545 (citing Beatty Dep. Tr. at 157-160). It is also possible that a non-Michigan offense may be similar to multiple Michigan offenses. Pls. SOMF ¶ 545; Defs. Resp. to Pls. SOMF ¶ 545. Defendants explain that this is “because a non-Michigan offense may exist under a statute that has alternative elements.” Defs. Resp. to Pls. SOMF ¶ 545 (citing Beatty Dep. Tr. at 157-160). Therefore, MSP must examine the facts of the non-Michigan offense to determine the basis for the conviction. Id. (citing Beatty Dep. Tr. at 157-160).
The Court agrees with Plaintiffs that the current procedure for determining whether people with out-of-state convictions should register under SORA is constitutionally deficient. As already explained, the fundamental requirement of procedural due process is the opportunity to be heard “at a meaningful time and in a meaningful manner.” Mathews, 424 U.S. at 333 (1976) (punctuation modified). Under the MSP's current procedure, people with out-of-state convictions have no notice or opportunity to be heard before the MSP decides whether an individual must register.And as Defendants admit, the decision is not a simple matter of mapping an offense to an identical Michigan offense. It can be fact-intensive, and different people can arrive at differing conclusions.
Defendants admit that there is no hearing offered before an individual is required to register, but they argue that individuals can challenge registration determinations after they are made, and that individuals have made such challenges in the past. Defs. Resp. to Pls. SOMF ¶ 560 (citing Beatty Dep. Tr. at 225-236). But the process for challenging registration determinations, as they are described by MSP Legal Advisor Steven Beatty, are informal at best. See Beatty Dep. Tr. at 233 (“[I]f someone brings to our attention . . . a concern over a particular registration or obligation, the SOR Unit will typically look into that and then take whatever appropriate action is necessary. That's not to say that we have a published, ‘if you believe,' blah, blah, blah. I think it's just implied.”). This does not constitute a meaningful opportunity to be heard.
A meaningful opportunity to be heard includes the right to a hearing before a neutral decisionmaker. See Hicks, 909 F.3d at 800. The MSP does not fill that job description. It is not an independent and disinterested arbiter of the law. It is a law enforcement agency tasked with maintaining the registry. As with the determination of what constitutes a non-sex offense, allowing judges to make the final decision regarding an individual's registration requirements is the best way to protect the constitutional interests of that individual.
Neither party discussed how burdensome it would be to require a judicial determination of whether a person with a non-Michigan offense must register under SORA. But given that only 7% of current registrants, or 3,100 people, have convictions from another jurisdiction, the Court believes that the costs of providing judicial determinations will be outweighed by the benefits. Class Data Report ¶¶ 16, 36, 86. Individuals required to register as a result of Michigan offenses are afforded procedural due process when they are convicted. People with out-of-state convictions are afforded similar due process when they are convicted, but not when it is determined that they will be branded as a sex offender in Michigan, potentially for life. See Kvech v. N.M. Dep't of Pub. Safety, 987 F.Supp.2d 1162, 1214 (D.N.M. 2013) (holding that a plaintiff convicted of a sex offense in Colorado was entitled to “notice of the charges, an opportunity to present witnesses and evidence in defense of those charges, and a written statement by the factfinder of the evidence relied on and the reasons for the disciplinary action” before being required to register in New Mexico).
The Court also finds that SORA's treatment of people with non-Michigan convictions violates equal protection with regard to certain disparities between Michigan and other states. Plaintiffs contend that SORA treats people with non-Michigan convictions more harshly than people with Michigan convictions in two ways. Pls. Br. Supp. Mot. for Summ. J. at 71; Pls. SOMF ¶¶ 552-553. First, people convicted of “substantially similar” out-of-state offenses must register for the duration required by Michigan or the adjudicating state, whichever is longer. Pls. SOMF ¶ 552 (citing SORA Op. Proceds., 307, 315 (Dkts. 127-5, 127-9), Jegla Dep. Tr. ¶ 95, 135, 145, Morris Dep. Tr. at 143-144, Beatty Dep. Tr. at 145); Defs. Resp. to Pls. SOMF ¶ 552. Second, if an offense would not require registration in Michigan, but the offense requires registration in the convicting state, the person must register in Michigan. Pls. SOMF ¶ 553 (citing MSP Flowcharts, Jegla Dep. Tr. at 94-99, Beatty Dep. Tr. at 140-141, 148); Defs. Resp. to Pls. SOMF ¶ 553. In such a case, the MSP assigns the tier and duration requirements applicable to the substantially similar Michigan offense. Id.
In addition to these two theories, Plaintiffs also contend that people with non-Michigan convictions are treated more harshly than people with Michigan convictions because MSP staff use unproven allegations about offense conduct to make registration decisions for people with out-of-state convictions. Pls. Br. Supp. Mot. for Summ. J. at 71; Pls. SOMF ¶ 556. Because the Court holds that a judicial determination is required before an individual with an out-of-state conviction can be required to register, the Court need not address this argument.
Plaintiffs offer the following example, which Defendants do not contest: “if the offense is ‘substantially similar' to a Michigan 25-year offense, but the convicting state requires a longer period, Michigan will impose the convicting state's longer period.” Pls. SOMF ¶ 552; Defs. Resp. to Pls. SOMF ¶ 552.
Plaintiffs reference age-of-consent laws, which vary by state, as an example. Pls. SOMF ¶ 554. “A person required to register in another state for consensual sex with a 17-year-old would have to register in Michigan though the same act is legal in Michigan.” Id. (citing Beatty Dep. Tr. at 150-151). Defendants do not dispute this example. Defs. Resp. to Pls. SOMF ¶ 554.
Defendants do not dispute that SORA treats people with non-Michigan convictions more harshly in these ways. Defs. Resp. to Pls. SOMF ¶¶ 552-553. Nor do Defendants offer any explanation or justification for the disparate treatment. Instead, Defendants argue that because rational-basis review applies, Plaintiffs bear “the burden of negating all possible rational justifications for the classification.” Defs. Mot. for Summ. J. at 70 (citing League of United Latin Am. Citizens v. Bredesen, 500 F.3d 523, 536 (6th Cir. 2007)).
In their response to Plaintiffs' statement of material facts, Defendants reference giving “full faith and credit” to the convicting state's laws. Defs. Resp. to Pls. SOMF ¶ 553. But this argument is not developed in their motion for summary judgment briefing, and Defendants do not explain how the Full Faith and Credit Clause is relevant to the treatment of people with out-of-state convictions. This passing reference is precisely the sort of argument that Courts deem waived. See McPherson v. Kelsey, 125 F.3d 989, 996-997 (6th Cir. 1997) (“Issues adverted to in a perfunctory manner, unaccompanied by some effort at developed argumentation, are deemed waived. It is not sufficient for a party to mention a possible argument in the most skeletal way, leaving the court to put flesh on its bones.”) (punctuation modified).
While it is correct that “[a] statute will be considered constitutional under rational basis review if there is ‘any reasonably conceivable set of facts that could provide a rational basis for' it ....the Supreme Court has nonetheless instructed that ‘even in the ordinary equal protection case calling for the most deferential standards, we insist on knowing the relation between the classification adopted and the object to be obtained.'” Doe v. Penn. Bd. of Prob. & Parole, 513 F.3d 95, 107 (3d Cir. 2008) (citing FCC v. Beach Commc'ns, 508 U.S. 307, 313 (1992); Romer v. Evans, 517 U.S. 620, 632 (1996)). Here, Defendants have offered no supposed relation between the classification and any particular objective. For this reason, Michigan's imposition of harsher registration terms on people with out-of-state convictions does not survive rational-basis review.
The Court awards summary judgment to Plaintiffs with respect to their non-Michigan convictions claim. The specific form of judicial determination that must be afforded to people with out-of-state convictions will be addressed at a later phase in this case.
G. Compelled Speech
Plaintiffs argue that SORA 2021's reporting requirements compel their speech in violation of the First Amendment by (i) forcing Plaintiffs to provide certain personal information that they otherwise would not and (ii) “us[ing] the compelled information to widely disseminate a message that they are dangerous sex offenders”-a message with which they disagree. Pls. Br. Supp. Mot. for Summ. J. at 41-42. According to Plaintiffs, SORA 2021's reporting requirements “force [Plaintiffs] to alter their own speech to assist the government in disseminating a message they believe to be false,” and therefore, SORA 2021 is a “content-based regulation subject to strict scrutiny.” Id. at 43.
SORA 2021 requires registrants to report their legal names, aliases, social security numbers, birth dates, home addresses, employer names and addresses, telephone numbers, email addresses, internet identifiers, driver's license numbers, occupational licensing information, and physical descriptors. See Am. Compl. ¶ 365 (citing Mich. Comp. L. § 28.727). The online public registry website includes registrants': name, date of birth, registration number, photo, registration details, physical characteristics, known alias(es), scars, marks, tattoos, offense information, address information, campus information, employment information, and vehicle information. See MSP, Sex Offender Registry FAQ, https://www.michigan.gov/msp/services/sex-offender-reg.
Defendants contest this view. They argue that the reporting requirements are contentneutral because the SORA 2021's reporting regulations “make no reference to the content of speech,” and therefore, are subject to intermediate scrutiny. Defs. Br. Supp. Mot. for Summ. J. at 45-46. In Defendants' view, SORA 2021's reporting requirements survive such scrutiny because the statute is narrowly tailored, serves a significant government interest, and leaves open ample alternatives for the communication of information. Id. at 47-48 (citing Doe v. Lee, 518 F.Supp.3d 1157, 1211 (M.D. Tenn. 2021)).
As discussed below, the Court declines to adopt wholesale either of the positions advocated by the parties. The Court concludes that SORA 2021's disclosure requirements do not implicate the First Amendment's prohibition of compelled speech such that a heightened-scrutiny-i.e., either intermediate or strict scrutiny-analysis is required. However, even if the statute's disclosure requirements were to trigger heightened scrutiny, as Plaintiffs contend, such disclosure requirements would survive strict scrutiny.
1. Compelled Speech Under the First Amendment
The First Amendment protects “both the right to speak freely and the right to refrain from speaking at all.” Wooley v. Maynard, 430 U.S. 705, 714 (1977). “[O]utside [of the commercial] context [the government] may not compel affirmance of a belief with which the speaker disagrees .... this general rule, that the speaker has the right to tailor the speech, applies not only to expressions of value, opinion, or endorsement, but equally to statements of fact the speaker would rather avoid.” Hurley v. Irish-Am. Gay, Lesbian & Bisexual Grp. of Bos., 515 U.S. 557, 573 (1995) (punctuation modified). Regulations “[m]andating speech that a speaker would not otherwise make necessarily alter[] the content of the speech” and are therefore content-based. Riley v. Nat'l Fed'n of the Blind of N.C., Inc., 487 U.S. 781, 795 (1988). “Content-based laws- those that target speech based on its communicative content-are presumptively unconstitutional and may be justified only if the government proves that they are narrowly tailored to serve compelling state interests.” Reed v. Town of Gilbert, 576 U.S. 155, 163 (2015). Laws that are content-neutral, however, are subject to intermediate scrutiny. Phelps-Roper v. Strickland, 539 F.3d 356, 361 (6th Cir. 2008).
To argue that SORA 2021 is a content-based regulation such that strict scrutiny applies, Plaintiffs rely on cases that fall roughly in two camps: (i) cases addressing regulations requiring dissemination of a particular message and (ii) cases addressing the regulations that impinge on associational rights. As the Court discusses below, however, neither context applies readily to the sort of compelled-speech claim that Plaintiffs assert here.
See Pls. Br. Supp. Mot. for Summ. J. at 40-45. As to the cases involving regulations requiring the dissemination of messages, Plaintiffs cite Wooley v. Maynard, 430 U.S. 705 (1977); Pacific Gas & Electric Co. (PG&E) v. Public Utilities Comm'n of Cal., 475 U.S. 1 (1986); Riley v. Nat'l Fed'n of the Blind of N.C., Inc., 487 U.S. 781 (1988); Nat'l Inst. of Family and Life Advocs. v. Becerra, 585 U.S. 755 (2018). Plaintiffs rely on Ams. for Prosperity Found. v. Bonta, 594 U.S. 595 (2021), which involved association rights under the First Amendment.
In Wooley, the Supreme Court held that a New Hampshire law requiring license plates to show unobstructed the state's motto “Live Free or Die” violated the First Amendment. Wooley, 430 U.S. 705. In that case, the court explained that the First Amendment protects the right of individuals to “hold a point of view different from the majority” and found that the law effectively rendered private property a “mobile billboard for the State's ideological message . . . .” Id. at 715. Given the New Hampshire law's restriction on the appellants' speech rights, the court applied strict scrutiny, examining whether the state's interest was sufficiently compelling to justify the regulation. Id. at 715-716. The court found that New Hampshire's stated interests of facilitating identification of passenger vehicles and promotion of state history were not sufficiently compelling. Id. at 716-717. The court further reasoned that the state could achieve these goals through less restrictive means. Because the regulation did not meet the requirements of the heightened-scrutiny analysis, the Court held that the regulation violated appellees' First Amendment rights against compelled speech.
Although Wooley did not employ the term “strict scrutiny” it nevertheless applied that analysis by determining whether New Hampshire's interests were “sufficiently compelling” and whether the means used to achieve that goal could be more “narrowly achieved.” Wooley, 430 U.S. at 715.
Pacific Gas & Electric Co. (PG&E) v. Public Utils. Comm'n of Cal., 475 U.S. 1 (1986) reached a similar result. There, the Supreme Court invalidated a California utility commission order requiring the state's privately-owned utility provider, PG&E, to include in its billing envelopes political editorials of a third-party interest group, with which PG&E disagreed. The Court reasoned that, because the commissioner's order “required [PG&E] to assist in disseminating [the third party's] views” the commissioner's order violated PG&E's “right to be free from government restrictions that abridge its own rights in order to enhance the relative voice of its opponents.” Id. at 14 (punctuation modified). Finding that the commission order was not narrowly tailored to support a compelling state interest, the court held the order invalid under the First Amendment. Id. at 20.
Plaintiffs also rely on Riley v. Nat'l Fed'n of the Blind of N.C., Inc., 487 U.S. 781, 798 (1988) which held unconstitutional North Carolina's regulation requiring professional charity fundraisers to disclose to potential donors the percentage of contributions collected during the previous year that were actually given to charity. Because the regulation “mandat[ed] speech that a speaker would not otherwise make,” the regulation was content-based and subject to “exacting First Amendment scrutiny.” Riley, 487 U.S. at 795. Notably, the court rejected the state's argument that the regulation did not burden a speaker's right under the First Amendment because the regulation required statements of opinion, rather than fact. Id. at 797-798. In applying “exacting scrutiny,” the court found that (i) the state's asserted interest of informing charitable donors how the charities spend donated funds was not “as weighty” as the state asserted and (ii) the disclosure rule was not sufficiently tailored because the state could have achieved the same goal via narrower means, such as by publishing charities' financial disclosure forms. Id. at 798800. The court accordingly held that the regulation compelled speech in violation of the First Amendment. Id.
The “exacting scrutiny” standard under Riley required the regulation to be narrowly tailored to the state's asserted interest. See Riley, 487 U.S. at 787-784.
Plaintiffs also cite Nat'l Inst. of Family and Life Advocs. v. Becerra, 585 U.S. 755 (2018), which held that a law requiring clinics that serve pregnant women to notify clients that California provides low-cost services-including abortions-violated the First Amendment. As the court explained, by requiring the clinics to “provide a government-drafted script,” the regulation “alter[ed] the content” of the clinics' speech, and therefore, was subject to strict scrutiny. Id. at 766-767. The court held that the notice did not survive strict scrutiny because the notice was “not sufficiently drawn” to achieve the state's purpose of providing low-income women with information about state-sponsored services. Id. at 773. Viewing the disclosure as furthering the goal of educating women about health care services the state provides, the notice requirement was “wildly underinclusive” in that it applied only to family-planning clinics. Id. at 774. The court further noted that the same goal could be achieved without burdening a speaker with unwarranted speech via a public-information campaign or other public postings. Id. at 775. Because the regulation was not narrowly tailored to the government's asserted goal, it failed to survive strict scrutiny.
Here, Plaintiffs' fail to establish that SORA 2021's registration requirements constitute compelled speech such that the statute must survive strict scrutiny. As an initial matter, unlike the challenged provisions in Wooley, Pacific Gas, Riley, and Becerra, SORA 2021 does not require registrants to disseminate a message. The message-whatever it might be-is being disseminated by the state. This alone distinguishes the present case from the cases cited by Plaintiffs.
Further, case law does not establish that Plaintiffs' supplying of information to MSP constitutes Plaintiffs' “support” of the message. None of Plaintiffs' cases involved a compelled-speech claim based on the mere disclosure of factual information that is later publicized by the state. As discussed below, such disclosures that assist essential government operations do not violate the First Amendment. SORA 2021 bears little resemblance to regulations compelling individuals to disseminate government messaging.
Plaintiffs are no better served by their reliance on compelled disclosure cases in the context of the right to associate. See Pls. Br. Supp. Mot. for Summ. J. at 43. Plaintiffs chiefly rely on Ams. for Prosperity Found. v. Bonta, 594 U.S. 595 (2021). In that case, the court invalidated California's regulatory requirement that charities disclose to the state their top donors, donor addresses, and contributions. The court evaluated the regulation as a burden on charities' right to associate with others. Id. at 605-606. As the court explained, “‘compelled disclosure of affiliation with groups engaged in advocacy may constitute as effective a restraint on freedom of association as [other] forms of governmental action.'” Id. at 606 (quoting NAACP v. Patterson, 357 U.S. 449, 462 (1958)).
Contrary to Plaintiffs' briefing, Bonta declined to apply strict scrutiny in this context. It explained that the compelled disclosures are subject to a lesser, but still “exacting scrutiny,” which Bonta defined as a “substantial relation between the disclosure requirement and a sufficiently important governmental interest.” Bonta, 594 U.S. at 607 (punctuation modified). Notably, the court rejected the argument that the lesser “exacting scrutiny” standard should be applied only in the context of election-related disclosures, explaining: “[r]egardless of the type of association, compelled disclosure requirements are reviewed under exacting scrutiny.” Id. at 608. Applying exacting scrutiny, the court reasoned that although California had an important interest in preventing fraud by charities, the disclosure regime was overbroad. It noted that California's disclosure regulation failed to produce a “single, concrete instance” of advancing the state attorney general's fraud enforcement efforts. Id. at 613. Given the mismatch between the amount of information required by the disclosure regime and its limited utility, the court concluded that the regime failed to survive exacting scrutiny. Id. at 614-615.
Here, Plaintiffs' compelled-speech claim does not involve the sort of associational rights at issue in Bonta. Although Plaintiffs vaguely assert that that registration requirements “implicate” their ability to associate, Pls. Br. Supp. Mot. for Summ. J. at 47 n.22, they fail to explain how registration requirements burden their associational rights for purposes of speech. See Hartwell v. Houghton Lake Cmty. Sch., 755 Fed.Appx. 474, 477 (6th Cir. 2018) (punctuation modified) (explaining that the First Amendment implicitly protects “the right to associate for the purpose of speaking”). To the extent Plaintiffs contend that their inclusion on the registry burdens their ability to form relationships, such allegations do not give rise to a First Amendment claim. Doe v. Lee, No. 3:21-cv-00028, 2021 WL 1907813, at *18 (M.D. Tenn. May 12, 2021) (“There is simply no ground for concluding, under current caselaw, that an accurate government record that reflects poorly on a person, in and of itself, gives rise to a plausible association-based claim.”).
In short, the claims at issue in Plaintiffs' cited case law are distinguishable from the claim asserted by Plaintiffs here. Plaintiffs have not established that SORA 2021's reporting requirements implicate their First Amendment rights against compelled speech; therefore, heightened scrutiny is not applicable.
In addition to Bonta, Plaintiffs also cite Brown v. Socialist Workers '74 Campaign Comm., 459 U.S. 87 (1982); NAACP v. Patterson, 357 U.S. 449 (1958); Buckley v. Valeo, 424 U.S. 1, 74 (1976); Shelton v. Tucker, 364 U.S. 479 (1960). Pls. Br. Supp. Mot. for Summ. J. at 43 n.19. These cases are not applicable to Plaintiffs' claims here because they address the validity of disclosure requirements under First-Amendment association rights, which, as discussed, are not implicated by SORA 2021's disclosure requirements.
As the Court discusses below, courts have rejected compelled-speech claims related to sexoffender registry requirements.
In United States v. Arnold, 740 F.3d 1032 (5th Cir. 2014), the Fifth Circuit held that federal SORNA's registration requirements did not compel speech in violation of the First Amendment. The court did not characterize the plaintiff's claim as being subject to any form of heightened scrutiny. Instead, it reasoned that the plaintiff's speech had not been compelled in a manner that violates the First Amendment. While it acknowledged that the First Amendment “protects the right to remain silent,” the court further noted that “[t]here is no right to refrain from speaking when essential operations of government require it for the preservation of an orderly society ....” Id. at 1034-1035 (quoting United States v. Sindel, 53 F.3d 874, 878 (8th Cir.1995) (rejecting claim that compelled disclosure of information on IRS form was compelled speech)). Because the court viewed SORNA's registration requirements as a function of the “essential operation of the government,” similar to requiring disclosures of information for tax collection, SORNA's registration requirements did not unlawfully compel the plaintiff's speech. Id. at 1035.
Other courts have similarly recognized that disclosures made in furtherance of “essential operation[s]” of government do not implicate the First Amendment's prohibition of compelled speech. See, e.g., Full Value Advisors, LLC v. S.E.C., 633 F.3d 1104, 1108-1109 (D.C. Cir. 2011) (holding that SEC rules requiring institutional investment managers to disclose “among other things, the names, shares, and fair market value of the securities over which the institutional managers exercise control” was “indistinguishable from oft unnoticed forms of disclosure the Government requires for its ‘essential operations'”); Sindel, 53 F.3d at 878 (explaining that “[t]here is no right to refrain from speaking when essential operations of government may require it for the preservation of an orderly society[]-as in the case of compulsion to give evidence in court”) (punctuation modified); Morales v. Daley, 116 F.Supp.2d 801, 816 (S.D. Tex. 2000) (holding that government requirement that plaintiffs classify themselves by racial categories for purposes of the census did not implicate First Amendment right against compelled speech).
The Court adopts the Fifth Circuit's approach set forth in Arnold. SORA 2021's reporting requirements are in furtherance of the State of Michigan's essential operation of operating the sex offender registry website. The law does not require that Plaintiffs speak a particular message, nor does it burden Plaintiffs' associational rights. SORA 2021, therefore, does not implicate Plaintiffs' First Amendment rights against compelled speech.
2. SORA 2021 Reporting Requirements and Strict Scrutiny
While the Court concludes that SORA 2021 does not implicate Plaintiffs' First Amendment rights against compelled speech such that the statute is subject to heightened scrutiny, the Court recognizes that other courts have nonetheless proceeded to scrutinize sex offender registration requirements under strict scrutiny. See United States v. Doby, No. 18-cr-40057, 2019 WL 5825064, at *3 (D. Kan. Nov. 7, 2019) (concluding that, “to the extent [federal SORNA] compels speech, it passes strict scrutiny”) (punctuation modified); United States v. Fox, 286 F.Supp.3d 1219, 1222-1224 (D. Kan. 2018) (following Arnold and proceeding to find that the federal SORNA statute's reporting requirements survived strict scrutiny). Even if SORA's 2021 disclosure requirements did compel speech such that strict scrutiny applied, the Court finds that SORA 2021 withstands scrutiny.
To survive strict scrutiny, the government must prove that the restriction (i) furthers a compelling interest and (ii) is narrowly tailored to achieve that interest. Reed v. Town of Gilbert, 576 U.S. 155, 171 (2015). SORA 2021 satisfies both prongs of the analysis.
As to the first prong-the asserted interest-Defendants assert that SORA 2021's intent is to “assist law enforcement officers and the people of this state in preventing and protecting against the commission of future criminal sexual acts by convicted sex offenders.” Defs. Br. Supp. Mot. for Summ. J. at 48 (citing Mich. Comp. L. § 28.721a). Plaintiffs agree that such an interest is sufficiently compelling, acknowledging that the state has a “strong interest in preventing sexual recidivism.” Pls. Br. Supp. Mot. for Summ. J. at 46. SORA 2021's registration requirements serve a compelling government interest.
Turning to the second prong-tailoring-Defendants argue that SORA 2021's registration requirements are narrowly tailored because they “efficiently and effectively provide[] all the relevant information that the people of Michigan need to know to prevent and protect against the potential commission of future criminal sexual acts.” Defs. Br. Supp. Mot. for Summ. J. at 4950.
The Court agrees with Defendants that the reporting requirements are narrowly tailored. The viability of a registry system depends on the ability-of the public or law enforcement-to use the information provided on the registry to accurately locate and identify registrants to protect against future sex crimes. For this reason, courts analyzing the validity of sex offender registry disclosure requirements under the First Amendment have repeatedly found that such disclosure requirements survive strict scrutiny. Fox, 286 F.Supp.3d at 1224 (explaining that federal SORNA's disclosure requirements-which include name, social security number, certain addresses registrants frequent, and international travel plans-satisfied strict scrutiny because such information “provide[d] the public and state law enforcement officials with the necessary information to track [registrants] effectively”); Doby, 2019 WL 5825064, at *4 (relying on the reasoning provided in Fox to find that federal SORNA is narrowly tailored).
Like the required disclosures in Fox and Doby, the disclosures required under SORA 2021 serve the purpose of tracking registrants for protective purposes. Specifically, registrants are required to disclose information regarding their (i) identity (names/aliases, conviction summaries), (ii) appearance (photograph, weight, hair and eye color, tattoos) and (iii) location (address, associated vehicle information, employer and/or school address). Each of these categories serves the purpose of empowering members of the public to identify registrants and take appropriate precautions to protect themselves.
Notably, SORA 2021's disclosure requirement mirrors the sort of regime endorsed by the Supreme Court in Riley. As the court explained in that case, “as a general rule, the State may itself publish the detailed financial disclosure forms it requires professional fundraisers to file [with the State]. This procedure would communicate the desired information to the public without burdening a speaker with unwanted speech during the course of a solicitation.” Riley, 487 U.S. at 800. Here, SORA 2021 adopts this same approach in that the state publicizes the disclosure information it requires from registrants through its public sex offender registry website.
Plaintiffs counter that SORA 2021's registration requirements are not narrowly tailored because (i) “registries nor reporting requirements reduce recidivism,” (ii) registrants are not subject to individualized risk reviews, so SORA's speech burdens are not limited to those who may pose a risk, and (iii) in-person reporting requirements “have no relationship to public safety at all.” Pls. Br. Supp. Mot. for Summ. J. at 47. None of these arguments has merit.
Plaintiffs' first and second arguments fail for the same reason: even if SORA does not reduce recidivism and includes some convicted offenders who no longer post a risk, the statute's compelling interest is not limited to that goal. As noted above, an asserted goal of SORA is to assist the public in protecting themselves against future sexual crimes. The registry's disclosure requirements provide a narrow means of accomplishing that goal. And while the requirement that Plaintiffs report some information in person does impose a burden on Plaintiffs, they fail to explain how that requirement impinges on their speech rights. Id.
Although Plaintiffs also aver that much of the information on the registries is available to law enforcement through other sources, Pls. Br. Supp. Mot. for Summ. J. at 46; Pl. SOMF ¶ 447, the same cannot be said for the general public, who would not have the access or ability to obtain the same information made available on the public registry.
Plaintiffs also rely on State v. Hill, 341 So.3d 539 (La. 2020) and Doe 1 v. Marshall, 367 F.Supp.3d 1310, 1318 (M.D. Ala. 2019) to argue that the registration requirements do not survive strict scrutiny. Pls. Br. Supp. of Mot. Summ. J. at 45. Both cases are distinguishable, however, because they did not involve challenges to the reporting aspect of the applicable sex offender laws.
In Hill, the Louisiana Supreme Court held that a statute requiring convicted sex offenders to carry an identification card branded “sex offender” was not narrowly tailored because the state could have adopted another “symbol, code, or . . . letter designation” to inform law enforcement of an individual's sex-offender status. Hill, 341 So.3d at 553. Importantly, the court found that the regulation was overbroad because the state already had in place a “sex offender registry” and “notification [was] available to those who have a need to seek out that information, while also not unnecessarily requiring disclos[ure] [of] that information to others via a branded identification.” Id.
Marshall reached a similar result. In that case, the district court assessed whether Alabama's statute requiring convicted sex offenders to carry an ID reading “CRIMINAL SEX OFFENDER” survived strict scrutiny. Marshall, 367 F.Supp.3d at 1324-1327. The court found that the statute lacked tailoring because was not the least restrictive means-i.e., the ID could have “use[d] a single letter to designate sex offenders.” Id. at 1326-1327.
SORA 2021's registration requirements bear no resemblance to the branded-ID cards at issue in Hill and Marshall. SORA 2021 does not require registrants to carry IDs or otherwise broadcast such a state-sponsored message. Neither case supports the proposition that sex offender registry reporting requirements-which are essential for the functioning of a public registry-are not narrowly tailored to furthering the goal of empowering the public to protect themselves against future sexual crimes by convicted offenders.
Federal SORNA does require that sex offenders' passports contain a unique identifier. See 22 U.S.C § 212b. (“Except as provided under paragraph (2), the Secretary of State shall not issue a passport to a covered sex offender unless the passport contains a unique identifier, and may revoke a passport previously issued without such an identifier of a covered sex offender.”).
SORA 2021's reporting requirements do not compel speech in a manner that implicates Plaintiffs' First Amendment rights because they are in furtherance of an essential government operation-i.e., maintenance of the registry. To the extent the registration requirements do compel speech, these requirements do not violate the First Amendment because they are narrowly tailored to serve a compelling interest of allowing the public to protect themselves against future sexual crimes committed by committed sex offenders. Defendants are entitled to summary judgment on Plaintiffs' compelled-speech claim.
H. Forced Admission of Understanding
Plaintiffs allege that SORA 2021 violates the First Amendment by requiring them to sign a form attesting: “I [the registrant] understand my registration duties.” Pls. Br. Supp. Mot. for Summ. J. at 61 (citing Explanation of Duties Form (Dkt. 126-17), Mail-in Update Form (Dkt. 12618)). Plaintiffs assert that they are compelled, under threat of prosecution, to state that they understand SORA 2021's registration requirements even if that is not true. Id.
Turning to the relevant statutory provisions, SORA 2021 provides that a registrant “shall sign a registration and notice ....” Mich. Comp. L. § 28.727(4). An “individual who willfully fails to sign a registration and notice as provided in [§ 28.727(4)] is guilty of a misdemeanor punishable by imprisonment . . . or a fine . . . .” Id. § 28.729(3). Although SORA 2021's statutory language does not require that Plaintiffs' attest to understanding their registration requirements, the registration and notice form that Plaintiffs are required to sign provides: “I [the registrant] understand my registration duties.” Explanation of Duties Form at PageID.6208. The form reiterates the registrant's obligation to sign the form, providing: “I am required to sign the required registration form(s). Failure to sign the required registration form(s) is a misdemeanor and may result in criminal prosecution.” Id. at PageID.6206.
1. Applicable Scrutiny
Plaintiffs argue that strict scrutiny applies. Pls. Br. Supp. Mot. for Summ. J. at 61-62. Defendants, on the other hand, presume that intermediate scrutiny applies. Defs. Br. Supp. Mot. for Summ. J. at 50-51. Neither party cites case law squarely addressing the sort of claim that Plaintiffs assert. However, the Court finds that application of the First Amendment's prohibition against compelled speech requires that the Court apply strict scrutiny.
Plaintiffs cite Johanns v. Livestock Mktg. Ass'n, 544 U.S. 550, 557 (2005) and Jackler v. Byrne, 658 F.3d 225, 241 (2d Cir. 2011) for the proposition that the First Amendment prohibits the government from forcing an individual to express a message with which he or she disagrees. Pls. Br. Supp. Mot. for Summ. J. at 61. But neither case involves the sort of facts at issue here. Johanns held that a federal program financing advertising to promote beef products did not violate the First Amendment because the advertising at issue was the Government's speech. Johanns, 544 U.S. at 566-567. In Jackler, the Second Circuit held that a probation officer stated a First Amendment claim based on retaliation he faced after he retracted a false statement regarding a fellow officer's use of force. Jackler, 658 F.3d at 241-242. These cases provide the Court with little guidance in assessing Plaintiffs' claim centered on the attestation of understanding. In any event, the Court agrees with Plaintiffs that strict scrutiny applies.
As discussed above, the First Amendment protects “both the right to speak freely and the right to refrain from speaking at all.” Wooley, 430 U.S. at 714 . “[O]utside [of the commercial] context [the government] may not compel affirmance of a belief with which the speaker disagrees .... this general rule, that the speaker has the right to tailor the speech, applies not only to expressions of value, opinion, or endorsement, but equally to statements of fact the speaker would rather avoid.” Hurley, 515 U.S. at 573 (punctuation modified). “Content-based laws-those that target speech based on its communicative content-are presumptively unconstitutional and may be justified only if the government proves that they are narrowly tailored to serve compelling state interests.” Reed, 576 U.S. at 163.
Here, SORA 2021 is a content-based regulation of speech because it requires registrants to state that they understand their registration obligations regardless of whether that is true. See Riley, 487 U.S. at 795 (explaining that “[m]andating speech that a speaker would not otherwise make necessarily alters the content of the speech” and therefore, such regulations are “contentbased regulation of speech”). Because the required statement of understanding alters the content of speech for Plaintiffs who do not understand their registration duties, the statement of understanding is a content-based regulation subject to strict scrutiny.
2. Analysis of the Statement of Understanding Under Strict Scrutiny
Defendants argue that the statement of understanding is narrowly tailored to serve the state's goal of “ensur[ing] that registrants make efforts to understand and take seriously their ongoing requirements under the law” as part of its efforts to “monitor[] offenders by ensuring compliance.” Defs. Br. Supp. Mot. for Summ. J. at 51. Plaintiffs protest that the purpose of the regulation is to “make it easier for prosecutors to establish willfulness” when bringing charges against registrants for failure to register, and, therefore, the government lacks a compelling interest. Pls. Br. Supp. Mot. for Summ. J. at 62.
Assuming the state has a compelling interest in ensuring compliance with SORA 2021's registration requirements, Defendants fail to establish that the same goal could not be achieved without burdening Plaintiffs' speech. Defendants do not, for example, argue that they were unable to effectively monitor registrants' compliance before the form included the statement of understanding. As Plaintiffs note, registrants were not required to sign an Explanation of Duties form that attested to understanding the registration requirements until after SORA 2021's passage. Pls. SOMF ¶ 508 (citing Pre-2021 Explanation of Duties Form (Dkt. 126-19)); Defs. Resp. to Pls. SOMF ¶ 508. Defendants offer no explanation as to why a less burdensome regime, such as providing registrants with FAQs, training sessions, or similar informational aides, would not accomplish the goal of ensuring that registrants understand their reporting requirements.
The parties do not point to case law specifically addressing whether the government has a compelling interest ensuring that registrants understand their registration obligations for purposes of monitoring their compliance.
In attempting to defend the attestation requirement, Defendants point out that the federal SORNA statute similarly requires that a state's registering agency employee must require the sex offender to read and sign a form stating that the duty to register has been explained and that the sex offender understands the registration requirement. See Defs. Br. Supp. Mot for Summ. J. at 50-51 (citing 34 U.S.C.A. 20919(a)(2)). Unlike SORA 2021, however, federal SORNA does not penalize a registrant's failure to attest to understanding their registration requirements. Thus, where SORA 2021 compels registrants by penalizing them for failing to sign the statement of understanding, federal SORNA does not. See Parents Defending Educ. v. Olentangy Loc. Sch. Dist. Bd. of Educ., No. 23-3630, 2024 WL 3565635, at *7 (6th Cir. July 29, 2024) (“A general principle of compelled speech jurisprudence is that a violation of the First Amendment right against compelled speech occurs only in the context of actual compulsion.”) (punctuation modified). SORA 2021 compels registrants to state that they understand their registration requirements and penalizes them if they fail to do so. Federal SORNA carries no such penalties and, therefore, lacks compulsion.
And as Plaintiffs point out, SORNA does not implement its own registration scheme: “There is no way to register directly with the federal government. Nor does the federal government provide any notice about any federal obligations. State registration schemes are the only way for registrants to report and to be notified about registration requirements. There is also no way for registrants to report information not required by the state.” Pls. Resp. to Defs. SOMF ¶ 35.
Requiring registrants to attest to understanding their registration requirements-even if they do not-constitutes compelled speech under the First Amendment and is subject to strict scrutiny. Because Defendants fail to establish that the regulation is narrowly tailored to further the state's compelling interest of monitoring registrants by ensuring that registrants understand their registration obligations, the requirement does not survive scrutiny. Plaintiffs are entitled to summary judgment on their forced understanding claim.
I. Reporting of Internet Identifiers
SORA 2021 requires individuals required to be registered after July 1, 2011 to report “all electronic mail addresses and internet identifiers registered to or used by the individual.” Mich. Comp. L. § 28.727(1)(i). Registrants subject to this regulation must continue to report any change to their registered email addresses or internet identifiers within three business days of that change. Id. § 28.725(2)(a). SORA 2021 defines the term “internet identifier” to mean “all designations used for self-identification or routing in internet communications or posting.” Id. § 28.722(g).
“An author's decision to remain anonymous is an aspect of the freedom of speech protected by the First Amendment.” Signature Mgmt. Team, LLC v. Doe, 876 F.3d 831, 835 (6th Cir. 2017) (punctuation modified). “Internet speech receives the same First Amendment protection as other speech.” Id. Courts analyzing the validity of similar internet-identifier reporting requirements of other sex offender registration statutes have explained that such regulations are content-neutral and subject to intermediate scrutiny. Doe v. Shurtleff, 628 F.3d 1217, 1223 (10th Cir. 2010); Doe v. Harris, 772 F.3d 563, 574-575 (9th Cir. 2014). “Under this test, the government may impose reasonable content-neutral restrictions on the time, place, or manner of protected speech, provided the restrictions: (1) serve a significant governmental interest; (2) are narrowly tailored; and (3) leave open ample alternative channels for communication of the information.” Phelps-Roper v. Strickland, 539 F.3d 356, 362 (6th Cir. 2008) (punctuation modified).
Although Plaintiffs assert that Defendants' motion to dismiss “conceded that strict scrutiny applies” to the internet-identifier requirements, Plaintiffs do not cite any cases applying strict scrutiny to internet identifiers. Pls. Br. Supp. Mot. for Summ. J. at 66. The case law appears to be uniform in applying intermediate scrutiny to such provisions.
Plaintiffs contend that having to report their “internet identifiers” chills internet use and anonymous speech. Pls. Br. Supp. Mot for Summ. J. at 63. Plaintiffs further submit that, because SORA 2021 contains no provision restricting law enforcement from publicizing such identifiers, registrants' speech is chilled for fear that the reported identifiers will be made public. Id. at 64. As Plaintiffs point out, prior iterations of the statute prohibited the government from publicizing registrants' internet identifiers on the sex offender registry website. Pls. SOMF ¶ 516 (citing Mich. Comp. L. § 28.728(3)(e) (2020). SORA 2021 no longer contains an express restriction against sharing publicly the internet identifiers reported by registrants. Id. Defendants counter that the internet-reporting requirement does not limit Plaintiffs' online speech or registrants' anonymity. Defs. Br. Supp. Mot. for Summ. J. at 52. While Defendants acknowledge that SORA 2021 does not prohibit publication of internet identifiers, they submit that federal regulations prohibit such public posting. Id. at 52-53.
Although the Sixth Circuit has not squarely addressed whether the sort of internet-reporting requirements like those contained in SORA 2021 impermissibly chill speech, the weight of authorities that have addressed this issue supports Plaintiffs' view.
In Harris, the Ninth Circuit Court examined a provision requiring sex offenders to report “a list of any and all internet identifiers established or used by the person” and “a list of any and all internet service providers used by the person,” as well as requiring registrants to send written notice to law enforcement within 24 hours of any additions or changes to this information. Harris, 772 F.3d at 567-568 (punctuation modified). Applying intermediate scrutiny, the Ninth Circuit concluded that the Act “unnecessarily chills protected speech in at least three ways: [i] the Act does not make clear what sex offenders are required to report, [ii] there are insufficient safeguards preventing the public release of the information sex offenders do report, and [iii] the 24-hour reporting requirement is onerous and overbroad.” Id. at 578. With respect to the law's treatment of publicly reporting internet identifiers, the court noted that this provision “contains no standards for judging what is necessary to ensure the public safety,” and that, “[w]ithout such standards, the Act impermissibly places unbridled discretion in the hands of a government official or agency.” Id. at 580 (punctuation modified). Thus, “registered sex offenders are unnecessarily deterred from engaging in anonymous online speech.” Id. at 581.
Harris was in the context of a motion for preliminary injunction.
Similarly, in Cornelio v. Connecticut, 691 F.Supp.3d 529 (D. Conn. 2023), the court held that a plaintiff challenging a statute requiring offenders to disclose email addresses and other internet identifiers was entitled to summary judgment on a First Amendment claim alleging chilled speech. In that case, the applicable statute required sex offenders to disclose their “electronic mail address, instant message address or other similar Internet communication identifier” on “such forms and in such locations as the commissioner shall direct.” Id. at 535. The court held that the statute failed to survive intermediate scrutiny because (i) the state failed to produce evidence showing that the disclosure requirement “deterred a sex offender or helped solve a crime” and (ii) the statute was overbroad because it required registrants to disclose any internet identifier regardless of whether the identifier was associated with a platform that could be used to perpetrate sex crimes. Id. at 543-544.
Plaintiffs' brief cites the Second Circuit's opinion in Cornelio v. Connecticut, 32 F.4th 160, 171 (2d Cir. 2022), which reversed the district court's prior grant of the state's motion to dismiss. See Pls. Br. Supp. Mot. for Summ. J. at 76. Because the district court's summary judgment opinion following remand from the Second Circuit is helpful in evaluating the parties' summary judgment briefing, the Court discusses the district court opinion.
The court further observed that the registrants would be required to disclose internet identifiers “on a myriad of [i]nternet platforms,” such as Amazon.com, the Washington Post, and WebMD- none of which is likely to be used as a platform to perpetuate sex crimes. Cornelio, 691 F.Supp.3d at 544.
In Doe v. Marshall, the court held that an Alabama law requiring registrants to report email addresses and other “designations or monikers used for self-identification in Internet communications or postings, and any and all [i]nternet service providers used by the sex offender” did not survive intermediate scrutiny. Marshall, 367 F.Supp.3d at 1320, 1329. Like the court in Cornelio, the Marshall court found that the regulation lacked tailoring because it included reporting related to registrants' use of internet platforms that do not raise safety concerns. As posited by the court: “if a sex offender establishes a username on a news outlet's website for purposes of posting comments to news articles, it is hard to imagine how speedily reporting that identifier will serve the government's interests.” Id. at 1329 (punctuation modified). Because the reporting requirements “chill[ed] a wide swath of protected speech,” the Court concluded that they were overbroad in violation of the First Amendment. Id. at 1331.
Here, SORA 2021 shares the same lack of tailoring that doomed similar provisions in Harris, Cornelio, and Marshall. It applies too broadly, requiring registrants to report their email addresses, usernames, or other identifiers, regardless of whether a particular identifier is associated with a platform that could be used to commit a sex crime. In addition, the statute lacks tailoring in that it does not limit the disclosure of registrants' internet identifiers to law enforcement. Nor do Defendants offer any explanation as to how the reporting of internet identifiers would promote public safety.
SORA 2021 does not limit law enforcement's ability to publicly disclose the registrants' internet identifiers. Although Defendants assert that the federal SORNA statute prohibits such disclosure, see Defs. SOMF ¶ 148 (citing 28 U.S.C. § 20916(c); 34 U.S.C. § 20920(b)(4); 76 Fed.Reg. 1630, 1637; 86 Fed.Reg. 69856, 69858), that statute does not prevent the State of Michigan from publicly disclosing registrants' internet identifiers. The absence of a limitation on law enforcement's ability to publicly disclose internet identifiers counsels further in favor of finding the statute overbroad. See Harris, 772 F.3d at 580-581 (finding that a similar internet-reporting statute chilled speech, despite including a provision that allowed law enforcement to disclose internet identifiers “when necessary to ensure public safety” because the statute contained “no standards” for judging what is necessary to ensure public safety).
28 U.S.C. § 20916(c) provides that “[t]he Attorney General . . . shall exempt from disclosure” the internet identifiers that offenders are required to report under § 20916(a). Section 20920(b)(4) mandates that “[a] jurisdiction shall exempt from disclosure . . . (4) any other information exempted from disclosure by the Attorney General.” As Plaintiffs point out, states are not required to comply with every provision of SORNA. See Pls. Resp. to Defs. SOMF ¶ 35 (citing United States v. Felts, 674 F.3d 599, 604 (6th Cir. 2012) (explaining that a state has a “sovereign prerogative” to choose to comply with SORNA, but rejecting defendant's argument that SORNA requires persons to register only when states “fully implement[]” SORNA)). SORA 2021 does not contain any such exemption related to its internet-identifier disclosure requirements. That does not mean that the state is in violation of SORNA. Rather, to receive federal funding, states must achieve substantial compliance, which Michigan has done. See 3/16/22 Letter to Gov. Gretchen Whitmer at PageID.8281 (“Based on the response provided on July 21, 2021, the additional information provided on March 10, 2022, and information and analysis currently available, the SMART Office has determined that Michigan continues to substantially implement SORNA as of the [March 16, 2022] date of this letter.”) (Dkt. 131-9); Morris Dep. Tr. at 129-130 (Dkt. 126-8) (explaining that SMART has not declined funding to the state based on changes made to the statute following the Does I and Does II decisions). Thus, while one requirement of SORNA is that states exempt internet identifiers from disclosure, Michigan has apparently chosen not to comply with this requirement and has nonetheless retained its status as a state that substantially implements SORNA. Defendants' citation to 76 Fed.Reg. 1630-1631 actually undercuts its contention that federal regulations prohibit the state from posting internet identifiers. As the guidelines explain, “jurisdictions cannot, consistent with SORNA, include sex offenders' Internet identifiers (such as e-mail addresses) in the sex offenders' public Web site postings or otherwise list or post sex offenders' Internet identifiers on the public sex offender Web sites.” 76 Fed.Reg. 1630, 1637. Importantly, however, the guidance further provides that SORNA's mandatory exemption of internet identifiers “does not limit the discretion of jurisdictions to include on their public Web sites functions by which members of the public can ascertain whether a specified e-mail address or other Internet identifier is reported as that of a registered sex offender . . . or to disclose Internet identifier information to any one by means other than public Web site posting.” Id. (punctuation modified). Thus, even assuming SORNA did require that Michigan refrain from publicly disclosing internet identifiers on the registry website, it does not prohibit other means by which the public may obtain registrants' internet identifiers. Finally, Plaintiffs' citation to 86 Fed.Reg. 69856-69858 is not helpful, as that guidance provides: “Disclosure of sex offender information is separately addressed in statutory provisions that are not implicated by this rulemaking and in the SORNA Guidelines and SORNA Supplemental Guidelines.” 86 Fed.Reg. 69856.
Tellingly, Defendants do not dispute that SORA 2021 has a chilling effect on Plaintiffs' speech. Rather, Defendants argue that Plaintiffs' claim fails because any burden on Plaintiffs' internet usage is the result of their “own subjective chill.” Defs. Br. Supp. Mot. for Summ. J. at 53-54 (citing ACLU v. Nat'l Sec. Agency, 493 F.3d 644, 654, 661 (6th Cir. 2007) (holding that plaintiffs who alleged a “well founded belief” that the National Security Agency was surveilling their communications did not state a First Amendment claim because they failed to allege an injury other than a chilling effect on speech caused by “their subjective belief that the NSA might be intercepting their communications”)).
This argument misses the mark. In ACLU, the plaintiffs lacked standing because they alleged only a speculative, subjective fear of surveillance that could lead to future government action. But the court pointed out that if there was something “more” to a plaintiff's claim of injury beyond subjective “chill,” then standing could be established. Id. at 663. That something “more” included “the exercise of governmental power that is regulatory, proscriptive, or compulsory in nature, and that directly regulates, proscribes, or compels the plaintiffs.” Id. That exactly describes the circumstances of Plaintiffs here, given the myriad ways in which the sex registration system “regulates, proscribes, or compels” them. Indeed, SORA 2021 compels Plaintiffs by requiring them to report their internet identifiers to the state. Further, unlike in ACLU, there is no speculation regarding the state's receipt of information regarding Plaintiffs. The issue is whether that receipt-coupled with no limitation on the state's ability to disclose that Plaintiffs' information publicly-would likely engender a “chilling effect.” Defendants do not dispute that it is reasonable to infer a chilling effect.
Defendants also attempt to rely on Doe v. Shurtleff 628 F.3d 1217, 1222 (10th Cir. 2010) and Does I, 101 F.Supp.3d 672, 701 (E.D. Mich. 2015) to argue that SORA 2021's internetreporting requirements withstand scrutiny. Defs. Br. Supp. Mot. for Summ. J. at 52. Neither case supports Defendants' position.
In Shurtleff, the Tenth Circuit considered whether a Utah law requiring registrants to report their internet identifiers violated the First Amendment. Shurtleff, 628 F.3d 1217. The plaintiff in that case principally argued that the law chilled their speech by permitting state officials to make required disclosures public. Id. at 1224. Rejecting this argument, the Tenth Circuit read the applicable law narrowly, interpreting it to allow “sharing only among law-enforcement agencies, not the public at large, and only for recited law-enforcement purposes.” Id. at 1225. Given this narrow interpretation of the statute, the court concluded that it survived intermediate scrutiny.
Does I analyzed a First Amendment challenge to a prior version of SORA, which required registrants (i) to report “all electronic mail addresses and instant message addresses assigned to the individual or routinely used by the individual and all login names or other identifiers used by the individual when using any electronic mail address or instant messaging system” and (ii) to report in person and notify the registering authority within three business days after the individual establishes any electronic mail or instant message address, or any other designations used in internet communications or postings.” Does I, 101 F.Supp.3d at 699 (punctuation modified). Importantly, the 2011 version of SORA at issue in Does I expressly provided that the internet identifiers “shall not be made available on the public website.” Id. at 703. Noting the differing outcomes in Harris and Shurtleff, the court concluded that SORA 2011 was “more akin to the Utah legislation [in Shurtleff] than the enjoined California statutes [in Harris]” because neither SORA 2011 nor the Utah statute in Shurtleff permitted registrants' identifiers to be made available to the public. Id. at 703. The court further reasoned that while SORA “unveil[ed] registrants' anonymity to law enforcement” that “does not, by itself, infringe upon [the plaintiffs'] First Amendment rights.” Id. Ultimately, however, the court concluded that the internet-reporting requirements were not narrowly tailored because the statute (i) was ambiguous with respect to its use of the term “routinely used” and (ii) required registrants to report in person.
Neither Shurtleff nor Does I provides strong support for Defendants' position. The crux of Shurtleff was that the Utah statute did not impermissibly chill the plaintiffs' speech because it did not permit public disclosure of internet identifiers. Shurtleff, 628 F.3d at 1225. Snyder is similarly unhelpful to Defendants. The version of SORA in that case prohibited the public disclosure of registrants' internet identifiers. No similar limitation on the disclosure of Plaintiffs' internet identifiers exists here.
The Court recognizes that, like the plaintiffs in Does I, here, Plaintiffs also argue that SORA 2021's internet-reporting requirement is vague and ambiguous. Pls. Br. Supp. Mot. for Summ. J. at 65. The Court addresses Plaintiffs' void-for-vagueness claims below.
All told, SORA 2021's internet-identifier reporting requirements do not withstand intermediate scrutiny. The statute chills a wide swath of speech activity-regardless of whether such activity could further the commission of a sex crime. Moreover, Defendants have made no showing of whether or how law enforcement uses internet identifiers to protect the public against the commission of sex crimes. Nor have they made any showing that the public could effectively use such information to protect themselves. On this record, Defendants cannot show that the internet reporting requirements serve any government interest, much less a significant interest. The Court concludes that Plaintiffs are entitled to summary judgment on this claim. The precise contours of the relief that follows will be addressed at a later phase in this case.
J. Vagueness
Plaintiffs argue that SORA 2021 is unconstitutionally vague. Pls. Br. Supp. Mot. for Summ. J. at 57. Under the void-for-vagueness doctrine, courts assess whether (i) “ordinary people, exercising ordinary common sense, can understand [the statute] and avoid conduct which is prohibited” and (ii) the statute “establishes minimal guidelines to govern enforcement.” United States v. Salisbury, 983 F.2d 1369, 1378 (6th Cir.1993). The doctrine is intended to “ensure fair notice to the citizenry” and “provide standards for enforcement by the police, judges, and juries.” Columbia Nat. Res., Inc. v. Tatum, 58 F.3d 1101, 1104 (6th Cir.1995). “The second prong- providing minimal guidelines to govern the conduct of law enforcement-constitutes the more important aspect of the vagueness doctrine.” Belle Maer Harbor v. Charter Twp. of Harrison, 170 F.3d 553, 556-557 (6th Cir. 1999). “[A] failure to define a term within a statute or ordinance does not render the statute unconstitutionally vague, where the common meaning of the word provides both adequate notice of the conduct prohibited and the standards for enforcement.” Id. at 558.
Plaintiffs' brief provides a list containing eight categories of provisions under SORA 2021 that they assert are vague: “Employment, changes to employment, employment locations, and volunteer work; phone numbers; vehicles; residential addresses; travel; email and internet usage; nicknames and physical descriptions; higher education and other classes.” Pls. Br. Supp. Mot. for Summ. J. at 60 (capitalization and punctuation modified) (citing Mich. Comp. L. §§ 28.722, 28.724a, 28.725, 28.727). Plaintiffs submit that they “are unsure both about what they must do and what they cannot do” under SORA 2021. Id. at 59. And according to Plaintiffs, “[l]aw enforcement officials fare no better, offering widely divergent interpretations of [the above cited] provisions.” Id. at 60 (citing MSP Response Chart (Dkt. 123-5); Law Enforcement Survey (Dkt. 123-23)).
Defendants disagree. They argue that SORA 2021 provides fair notice of the conduct it punishes and does not invite arbitrary enforcement of its terms. Defs. Br. Supp. Mot. for Summ. J. at 37, 41. Defendants briefly discuss each category of SORA 2021 challenged by Plaintiffs, arguing that the provisions are not lacking in clarity. Id. at 37-40.
Neither party provides a record from which the Court can render a determination regarding Plaintiffs' vagueness claims. Aside from the allegations summarized above, Plaintiffs' brief does not explain with any specificity how each of the challenged provisions is vague. Instead, Plaintiffs direct the Court to their Statement of Material Facts, which provides examples of provisions that Plaintiffs contend are vague. This is not a proper way of presenting its argument to the Court. See McPherson v. Kelsey, 125 F.3d 989, 996-997 (6th Cir. 1997) (“Issues adverted to in a perfunctory manner, unaccompanied by some effort at developed argumentation, are deemed waived. It is not sufficient for a party to mention a possible argument in the most skeletal way, leaving the court to put flesh on its bones.”) (punctuation modified).
Defendants' presentation fares no better. In addressing each category challenged by Plaintiffs, Defendants direct the Court to review Defendants' initial motion to dismiss, arguments which it claims are incorporated by reference. See Defs. Br. Supp. Mot. for Summ. J. at 37-40 (citing Fed.R.Civ.P. 10(c)). But Defendants' reliance on Rule 10(c) is misguided. Rule 10(c) allows parties to adopt by reference statements made in pleadings, a term that does not encompass a motion to dismiss. See Fed. R. Civ. Pro. 7(a) (listing seven specific documents that are the “only” pleadings “allowed,” none of which is a motion). Defendants, therefore, cannot rely on the arguments they made in their earlier motion. See, e.g., Ortiz v. New Mexico, 550 F.Supp.3d 1020, 1078-1079 (D.N.M. 2021); Marco Int'l, LLC v. Como-Coffee, LLC, No. 17-cv-10502, 2018 WL 1790171, at *1 n.1 (E.D. Mich. Apr. 16, 2018).
Given the inadequacy of the record before it, the Court will require the parties to confer to reach agreement on the meaning of the terms challenged by Plaintiffs. If agreement can be reached on the meaning of terms, those definitions could then be the subject of a court order. If no agreement is reached, the Court will require further briefing from the parties.
K. Statute of Limitations and Mootness Due to Federal SORNA
Defendants make two final arguments in an attempt to dispose of the case: (i) that the statute of limitations has lapsed on many of Plaintiffs' claims, and (ii) that Plaintiffs' claims are moot because even if SORA is struck down, federal SORNA is still in effect and would require registration. Defs. Mot. for Summ. J. at 73-74. Both arguments fail.
As an initial matter, Defendants do not develop either argument. Their brief in support of their motion for summary judgment simply “adopt[s] by reference the arguments previously made” in their earlier motion to dismiss, which they again contend is “consistent with Fed. R. Civ. Pro. 10(c).” Id. at 73. But as already discussed, Defendants reliance on Rule 10(c) is misguided, and they cannot rely on the arguments they made in their earlier motion.
Without their motion to dismiss, Defendants' arguments regarding statute of limitations and mootness are, as Plaintiffs describe them, “perfunctory.” Pls. Reply at 27. As discussed above, arguments adverted to in this matter are deemed waived. McPherson, 125 F.3d at 996-997.
But even if the Court were to consider Defendants' earlier arguments, they are without merit. With respect to statute of limitations, Defendants contend that “many of the provisions being challenged in this lawsuit have remained unchanged for over a decade” because they were initially introduced in the 2011 amendments to SORA. Defs. Br. Supp. Mot. to Dismiss at 49. Defendants correctly note that Michigan's three-year statute of limitations period for personal injury claims is applied to Plaintiffs' claims brought under 42 U.S.C. § 1983. Id. at 47-48 (citing Wolfe v. Perry, 412 F.3d 707, 713-714 (6th Cir. 2005); Garza v. Lansing Sch. Dist., 972 F.3d 853, 867 n.8 (6th Cir. 2020)). According to Defendants, the statute of limitations for many of Plaintiffs' claims began to run in 2011, when the Plaintiffs “[knew] or [had] reason to know that the act providing the basis of his or her injury has occurred.” Id. (citing Garza, 972 F.3d at 867 n.8).
The Court disagrees. As Plaintiffs note in their response to Defendants' motion to dismiss, the Sixth Circuit views a constitutional challenge to a statute that is alleged to continue to violate a plaintiff's rights on an on-going basis as timely so long as the violation continues. Pls. Resp. to Defs. Mot. to Dismiss (Dkt. 44) at 65-66 (citing Kuhnle Bros., Inc. v. Cnty. of Geauga, 103 F.3d 516 (6th Cir. 1997)). In Kuhnle, the court considered a challenge under § 1983 based on a county resolution that interfered with the plaintiff's trucking operations. Id. at 518. The court distinguished between the plaintiff's (i) “takings” claim and substantive due process claim for deprivation of property, and (ii) substantive due process claim for interference with interstate travel. Id. at 520-523. The former claims were found to be time-barred. Id. at 521. The takings claim was time-barred because the “taking”-or the diminution in value of the plaintiff's property interest-occurred when the resolution was enacted. Id. The substantive due process claim for deprivation of property was time-barred for the same reason. Id.
The court found that the interference with travel claim, however, was a continuing violation so that plaintiff's claim was timely. Id. at 521-523. As the court explained, “each day that the invalid resolution remained in effect, it inflicted continuing and accumulating harm on [the plaintiff] ” Id. at 522 (punctuation modified). “A law that works an ongoing violation of constitutional rights does not become immunized from legal challenge for all time merely because no one challenges it within [the first few] years of its enactment.” Id.
The same principle has been recognized outside the Sixth Circuit. See Flynt v. Shimazu, 940 F.3d 457, 462 (9th Cir. 2019) (deeming timely a § 1983 challenge by gambling licensees to state statute limiting their ownership of out-of-state casinos because “[w]hen the continued enforcement of a statute inflicts a continuing or repeated harm, a new claim arises (and a new limitations period commences) with each new injury”); Leal v. Azar, No. 2:20-cv-185-Z, 2020 WL 7672177, at *7 (N.D. Tex. Dec. 23, 2020) (deeming timely a challenge to a federal agency rule because “these types of suits are seeking prospective relief for ongoing injuries” and “[s]tatutes of limitations are simply inapplicable to such injuries”) (emphasis in original), vacated and remanded on other grounds, Leal v. Becerra, No. 21-10302, 2022 WL 2981427 (5th Cir. July 27, 2022).
Here, Plaintiffs claim ongoing injuries, as they allege that ongoing enforcement of SORA 2021 is unconstitutional. The remedy they seek-declaratory and injunctive relief-is prospective only. Their claims are not barred by any statute of limitations.
In their motion to dismiss, Defendants cited Doe v. Rausch, No. 3:20-cv-00728, 2022 WL 481240, at *1 (M.D. Tenn. Feb. 16, 2022), which the Court finds persuasive only in part. In Rausch, the plaintiff pled guilty to two counts of aggravated sexual battery in 1999. Id. at *1. At that time, the plaintiff was required to register as a “sex offender” in Tennessee but was allowed under the statute to seek removal from the registry ten years after completing his sentence. Id. Plaintiff completed his sentence in 2006. Id. In 2004, the Tennessee General Assembly passed a new version of its SORA, under which the plaintiff was reclassified as a “violent sexual offender” and subject to lifetime registration. Id. The plaintiff waited until 2020 to file a lawsuit challenging the amendment as applied to him as unconstitutional, bringing ex post facto, due process, vagueness, and First Amendment claims. Id.
The defendant in Rausch argued that the plaintiff's claims were untimely because he filed his action more than one year after he knew or should have known that he was subject to lifetime registration, which was the relevant statute of limitations under Tennessee law. Id. at *2. In response, the plaintiff argued that the continuing violation doctrine applied to his causes of action. Id. The court held that the plaintiff's ex post facto, vagueness, and First Amendment claims were timely, but that plaintiff's due process claim was not because it was “directed at the original imposition of SORA's requirements on [p]laintiff rather than at the ongoing effects of the statutory scheme.” Id. at *3.
This Court agrees with Rausch, except as to it ruling regarding the due process claim. It would seem that any deprivation of due process was not confined to the date of the statute's enactment, but continued as long as it remained in effect. Unless that court's opinion obscured the peculiarities of the statute's enactment as it might bear on a due process claim, it would seem that the claim should have been treated as timely, like the other claims whose timeliness was sustained. Except for its inexplicable treatment of the due process claim, Rausch confirms this Court's conclusion that challenges seeking prospective relief for ongoing harms from unconstitutional statutes are timely so long as the allegedly unconstitutional conduct continues.
Defendants' mootness claim is also without merit. Because Defendants characterize this issue as touching on the question of jurisdiction, Defs. Br. Supp. Mot. to Dismiss at 51, the Court proceeds to address it.
According to Defendants, Plaintiffs' claims are moot because even if the Court grants relief from SORA, Plaintiffs will have a “continuing duty to comply with federal SORNA registration requirements, which are practically identical to the requirements of [SORA 2021].” Id. at 53. But Defendants provide no case that says the existence of SORNA makes a challenge to a state sex offender registry moot. As already discussed, SORNA does not establish its own sex offender registry; individuals are only required to register with states. Further, Plaintiffs' obligations under a state registration system are distinct from any obligations under SORNA. Willman v. Att'y Gen. of U.S., 972 F.3d 819 (6th Cir. 2020). Regardless of Plaintiffs' federal obligations under SORNA, they have distinct state-law obligations, which they are entitled to challenge. Plaintiffs' claims regarding SORA are not mooted by the existence of federal SORNA.
L. Whether the Governor is a Proper Defendant
In their supplemental brief regarding Doe v. Lee, 102 F.4th 330 (6th Cir. 2024), Defendants argue for the first time that the Court should dismiss the Governor as a party because Plaintiffs do not have standing to bring this challenge against her. Defs. Lee Br. at 5. In their supplemental brief, Plaintiffs submit that the Court “need not address Defendants' late request before deciding the merits.” Pls. Lee Br. at 4. The Court agrees with Plaintiffs. Defendants raised this argument too late for it to be ruled on presently, and whether the Governor remains a Defendant in this matter does not affect the Court's other rulings. The Court will direct counsel for the parties to confer on whether additional briefing on this issue is required.
IV. CONCLUSION
For the reasons explained above, the Court grants summary judgment to Plaintiffs as to their (i) ex post facto claims, (ii) non-sex offense claim, (iii) non-Michigan offense claim, and (iv) First Amendment claims alleging forced admission of understanding and chilled speech rights stemming from SORA 2021's internet-identifier reporting requirements. The Court grants summary judgment to Defendants as to Plaintiffs' claims regarding (i) lack of individualized review, (ii) unequal opportunity to petition, and (iii) compelled speech. Plaintiffs' claim regarding plea agreements is moot.
The Court orders counsel for the parties to meet and confer regarding the issues set forth below and then file a joint statement on or before October 18, 2024 setting forth their points of agreement and disagreement regarding each one.
1. What further proceedings are required.
2. The statutory terms that the parties agree can be narrowed to avoid vagueness and the terms upon which they cannot agree.
3. Whether additional briefing is required to address the issue of the propriety of the Governor of Michigan being a defendant in this action.
4. The terms of a proposed judgment upon which the parties can agree and the terms upon which they cannot agree.
SO ORDERED