The People v. Marcus Brown

State of New York OPINION

Court of Appeals This opinion is uncorrected and subject to revision

before publication in the New York Reports.

No. 83

The People &c.,

Respondent,

v.

Marcus Brown,

Appellant.

Ava C. Page, for appellant.

William H. Branigan, for respondent.

Office of the Appellate Defender, et al., amici curiae.

RIVERA, J.:

Defendant stole money at gunpoint from his aunt in the presence of his 10-year-old

cousin for which he pleaded guilty to, inter alia, the unlawful imprisonment of the child.

For this crime, New York requires that he register as a sex offender and comply with the

-1-

-2- No. 83

Sex Offender Registration Act (SORA). It is undisputed that the crime was non-sexual and

that the SORA court found that defendant is not a sex offender and poses no sexual threat.

Nevertheless, the courts below felt constrained by People v Knox (12 NY3d 60 [2009]) to

impose SORA requirements. Defendant contends that the holdings in Knox and its

companion cases are distinguishable and do not control his as-applied challenge. We agree

and conclude that requiring defendant to register violates his due process rights and does

nothing to further the legislative purpose of SORA to protect the public from actual sex

offenders. The order of the Appellate Division should be reversed.

I.

Defendant Marcus Brown pleaded guilty to robbery in the first degree (Penal Law

§ 160.15 [4]) and other counts including, as relevant to this appeal, unlawful imprisonment

in the first degree (§ 135.10) based on allegations that he robbed his aunt at gunpoint in her

home where his 10-year-old cousin was present. Unlawful imprisonment of a minor when

the offender is not the parent of the victim is a SORA-eligible crime and brings defendant

within the statutory definition of “sex offender” which, in turn, subjects defendant to

mandatory sex offender registration, even where the crime, in fact, lacks a sexual act or

motive, and where defendant poses no risk of future harm to children (Correction Law §

168-a [2] [a] [i]).

Prior to defendant’s release from incarceration, the Board of Examiners of Sex

Offenders (Board) prepared a case summary and Risk Assessment Instrument, and assessed

90 points against defendant, for a presumptive level two (moderate) risk to reoffend

-2-

-3- No. 83

designation.1 In advance of the SORA proceedings, the Department of Corrections and

Community Supervision (DOCCS) indicated that it had determined that “defendant need

not be referred to any sex counseling and treatment program due to the nonsexual nature

of the offense committed.”

Before the SORA court, defendant claimed, in part, that requiring him to register as

a sex offender for a crime with no sexual conduct or motivation was unconstitutional as

applied to him and defied the purpose of the registry. Defendant argued that People v Knox

and its companion cases were distinguishable on the grounds that his cousin was not the

target of the offense and his cousin was not abused, abducted, or detained for days.

The prosecution conceded that defendant’s offense had no sexual component, but

nevertheless argued that registration was mandatory under Knox and that the court should

designate defendant a level two risk. Based on the record, the SORA court found that

defendant’s sole motivation was to steal money and that the offense involved “no sexual

contact or motivation” whatsoever. The court further found that defendant was not a “sex

offender” “and that he posed no risk of sexual threat at all.” Nevertheless, the court

determined that it was “constrained by state law” under Knox to mandate defendant’s

1

The Board assigns points to a defendant using certain enumerated risk factors identified

in its Sex Offender Registration Act: Risk Assessment Guidelines and Commentary.

Aggregation of these points sets a defendant’s presumptive risk of sexual reoffense

(Correction Law § 168-l [6] [a]-[c], [5]). Under this graduated point allocation, a defendant

assigned a score of 70 points or less is a presumptive level one (low) risk; a score between

70 and 110 is a presumptive level two (moderate) risk; and a score of 110 or more is a

presumptive level three (high) risk (see Sex Offender Registration Act: Risk Assessment

Guidelines and Commentary at 3 [2006]).

-3-

-4- No. 83

registration. But the court granted defendant a downward departure to level one—the

lowest risk level under SORA—because the Board “greatly overestimated” defendant’s

“risk of being a sexual offender.”

The Appellate Division affirmed, concluding that defendant’s constitutional

challenge was without merit, citing Knox and departmental precedent (203 AD3d 1183 [2d

Dept 2022], citing People v Marshall, 195 AD3d 961 [2d Dept 2021]; People v Douglas,

189 AD3d 1276 [2d Dept 2020]; People v Suarez, 147 AD3d 802 [2d Dept 2017]; People

v Edney, 143 AD3d 793 [2d Dept 2016]; People v Taylor, 42 AD3d 13 [2d Dept 2007]).

We granted leave to appeal (38 NY3d 911 [2022]), and now reverse.

II.

Defendant’s Due Process As-Applied Challenge

Defendant claims that requiring him to register as a sex offender when he never

engaged in sexual misconduct is not rationally related to the government’s interest in

protecting the public from sex offenders.2 He maintains that his underlying offense is

2

Amici, The Office of the Appellate Defender, The Legal Aid Society, and the Center for

Appellate Litigation, make a similar substantive due process argument and also contend

that SORA registration for persons like defendant who have not committed a crime

involving sexual misconduct or motivation violates procedural due process. Defendant did

not advance a procedural due process challenge and so the merits of such claim are not

before us. Amici also argue that we should revisit Knox in light of conflicting federal

decisions (see Yunus v Robinson, 2019 WL 168544 [SD NY 2019]; Pennington v Rosado,

2022 WL 3127062 [SD NY 2022]). Defendant welcomes overruling Knox but does not

rely solely on this claim. The majority agrees that SORA as applied to defendant violates

his due process rights. I and Chief Judge Wilson would overrule Knox for the reasons

discussed in Section III, infra.

-4-

-5- No. 83

factually distinguishable from Knox and its companion cases because defendant committed

a financially-motivated robbery in which the child victim was related to him, was not the

target of the crime, was not abducted, and was detained only for a short period of time.

Further, defendant asserts that the Knox Court’s concerns about dangerous sex offenders

escaping registration do not apply to him because, here, there was an uncontested judicial

finding that he presents no sexual risk. In response, the District Attorney relies on Knox as

controlling precedent and asserts that in that case the Court found a rational relationship

between the State’s goal of protecting children and SORA’s automatic registration of

persons convicted of kidnapping-related offenses—even where the offense lacks a sexual

component.3

“As the term implies, an as-applied challenge calls on the court to consider whether

a statute can be constitutionally applied to the defendant under the facts of the case”

(People v Stuart, 100 NY2d 412 [2003]). We agree with defendant that the sex offender

3

The District Attorney also claims that defendant should have raised his challenge on

appeal from his judgment of conviction, not the SORA court’s order determining his risk

level. We reject that contention. While, as a general rule, certification as a sex offender is

“appealable as part of the judgment of conviction” (People v Buyund, 37 NY3d 532, 537-

538 [2021]; see People v Hernandez, 93 NY2d 261, 267 [1999]), here, the record fails to

establish that the sentencing court certified defendant as a sex offender as required by

Correction Law § 168-d (1) (a) (cf. Hernandez, 93 NY2d at 267-268). Inasmuch as the

Board of Examiners of Sex Offenders, not the court, made the determination in the first

instance that defendant must register under SORA, the proper course was for defendant to

raise the issue before the SORA court (see People v Liden, 19 NY3d 271, 276 [2012]; see

also People v Cintron, 46 AD3d 353, 354 [1st Dept 2007], affd sub nom. People v Knox,

12 NY3d 60 [2009] [addressing an as-applied constitutional challenge to SORA

registration on appeal from the SORA court’s order, where the defendant’s conviction of

the predicate crime predated the enactment of SORA]).

-5-

-6- No. 83

designation and SORA’s mandatory registration as applied to him is not rationally related

to SORA’s purpose of protecting the public from sex offenders and thus violates his due

process rights.

A.

SORA labels as “sex offender(s)” all persons convicted of a “kidnapping offense[],

provided the victim of such kidnapping related offense is less than seventeen years old and

the offender is not the parent of the victim” (Correction Law § 168-a [1]-[2] [a] [i], citing

Penal Law §§ 135.05 [second-degree unlawful imprisonment], 135.10 [first-degree

unlawful imprisonment], 135.20 [second-degree kidnapping], 135.25 [first-degree

kidnapping]). In Knox and its two companion cases, the Court rejected a constitutional

challenge to these provisions. Thus, we must first determine whether the holding and

rationale of Knox foreclose defendant’s claim.

All three appeals in Knox involved defendants subject to SORA as a result of their

convictions for attempted or completed child kidnapping and unlawful imprisonment

crimes even though there was no proof that their crimes involved any sexual act or motive.

The defendant in Knox was convicted after she approached an eight-year-old girl in a park

and tried to pull her away, apparently in an attempt to replace her own child after recently

losing custody (12 NY3d at 64). Knox pleaded guilty to attempted kidnapping. In People

v Cintron, the defendant had a long history of “violent conduct, including sexual violence”

and a conviction for sexual misconduct based on forced intercourse with a 16-year-old (id.

at 70). The defendant locked his girlfriend and her two children inside their apartment for

-6-

-7- No. 83

several days after an argument and was convicted of unlawful imprisonment of each child

(id. at 64). In People v Jackson, the defendant abducted and threatened to kill the son of a

woman if she did not continue to work for him as a prostitute (id.). Like Knox, he pleaded

guilty to attempted kidnapping.

The defendants each challenged their SORA designations on substantive due

process grounds. The Court concluded that the defendants had a “constitutionally protected

liberty interest” for purposes of substantive due process “in not being required to register

under an incorrect label” (12 NY3d at 66). The Court further concluded that because this

interest was not fundamental in the constitutional sense, the defendants’ constitutional

claims were subject to deferential rational basis review (id. at 67). Under that test, the law

must be rationally related to a legitimate governmental interest (id.).

The Court held that mandatory sex offender registration of non-parent defendants

convicted of kidnapping or unlawful imprisonment of victims under 17 years old is

rationally related to the legitimate governmental interest of protecting minors from those

“who have shown themselves capable of committing sex crimes” against children (id.).

The Court explained that the Legislature passed SORA to comply with the Jacob

Wetterling Crimes Against Children and Sexually Violent Offender Registration Act,

which, in part, conditioned federal funding on the state’s enactment of a system to register

sexual offenders and persons who commit crimes against children (id. at 67-68).4 The Court

4

The Jacob Wetterling Act was superseded by the Adam Walsh Child Protection and

Safety Act of 2006 (see 34 USC § 209).

-7-

-8- No. 83

noted that the federal law does not require the “sex offender” designation of persons

convicted of kidnapping and unlawful imprisonment of children but instead “[t]hat was the

New York Legislature’s choice” (id. at 68).

Applying the rational basis test to that legislative decision to include all persons

convicted of these crimes within SORA’s reach, even if there was no proof of a defendant’s

sexual act or motive, the Court reasoned that the Legislature could “rationally have relied

on the fact that a great many cases of kidnapping or unlawful imprisonment of children are

indeed sex offenses” (id. at 68). In support, the Court cited two studies: a 1990 finding that

“about two-thirds or so of the Non-Family Abductions [studied] involved sexual assaults”

(id., citing David Finkelhor et al., Missing, Abducted, Runaway, and Thrown-away

Children in America, at 142 [May 1990]), and a 2002 update that found a lower rate of

46% of the nonfamily abductions studied involved a perpetrator who sexually assaulted the

child (id., citing David Finkelhor et al., U.S. Department of Justice, National Incidence

Studies of Missing, Abducted, Runaway, and Thrownaway Children, Nonfamily Abducted

Children: National Estimates and Characteristics, at 10 [Oct. 2002]). The Court

acknowledged that these studies referred to actual sex crimes but opined, without any

supporting citation or resource, that “the Legislature could rationally have found that the

statistics understate the problem” and that in many cases the offender intends a sexual

assault but is thwarted by an arrest or the victim’s escape (id.). And further, that the

Legislature could have found that a child cut off from familiar surroundings “is vulnerable

to sexual abuse even if the offender’s sexual desires are not the motive of the crime” (id.).

The Court hypothesized that “[a] kidnapper may plan to prostitute a child, or may seize an

-8-

-9- No. 83

unplanned-for opportunity to do so” (id.). Thus, according to the Court, “the Legislature

had a rational basis for concluding that, in the large majority of cases where people kidnap

or unlawfully imprison other people’s children, the children either are sexually assaulted

or are in danger of sexual assault” (id. at 68-69).

The Court also concluded that “the Legislature could have considered not only that

cases where the term [sex offender] is unmerited are few, but also that the process of

separating those cases from the majority in which it is justified is difficult, cumbersome

and prone to error. It could rationally have found that the administrative burden, and the

risk that some dangerous sex offenders would escape registration, justified a hard and fast

rule, with no exceptions” (id. at 69). According to the Court, the defendants suffer no worse

injustice being called “sex offenders” instead of “child predators” (id.).

Knox is distinguishable from defendant’s case. First, and most significantly, the

SORA court here expressly found that defendant is not a “sex offender,” that “[t]he child

involved was not even the target of the crime, was not physically or sexually abused, [and]

not kidnapped or detained for any prolonged period of time,” that “[t]here was no sexual

contact or motivation on the defendant’s part at all,” and that he “posed no risk of sexual

threat at all.” No such finding was made with respect to any of the defendants in the Knox

appeals, whose crimes—although not sexually motivated—provided some basis to

conclude that each of those defendants posed a future risk of sexual harm to children.

Though not elucidated in the Court’s opinion, the Knox defendant was homeless and

suffered from substance abuse and extensive psychiatric issues. Taking the child was her

objective, and if Knox had been successful in her attempt to kidnap and remove her “from

-9-

- 10 - No. 83

the safety of everyday surroundings,” the child indeed could have been left vulnerable to

sexual abuse or exploitation (id. at 68).5 The defendants in Cintron and Jackson were both

willing to leverage children and put them in harm’s way in order to exert control over their

mothers—in Jackson’s case, to make her perform sexually (id.). Cintron also clearly posed

a risk of sexual harm to children because he had a prior conviction for the sexual assault of

a minor (id. at 70). Each of them posed a risk of sexual harm to those (or other) children.

Unlike the Knox defendants, defendant sought only to steal money from his aunt. As the

SORA court found, the child was not a target of the robbery; he just happened to be home

when defendant broke in. Defendant’s cousin was merely in the wrong place at the wrong

time. Indeed, based on the lack of a sexually motivated crime, DOCCS determined that

defendant was not eligible to participate in the Sex Offender Counseling and Treatment

Program while in prison. The SORA court acknowledged that determination, and, based

on the totality of the circumstances, concluded that defendant is not a sexual threat.

5

The dissent draws a false equivalence between Knox’s and defendant’s criminal conduct.

Knox tried to take a child to make that child her own; defendant was stealing money from

a place where his cousin happened to be present. The dissent goes so far as to disregard the

significance of each defendant’s respective motivation vis-a-vis the child (see dissenting

op at 5 n 2). If Knox, who was homeless and suffering from mental illness, had completed

the abduction, she would have put the child in a position vulnerable to abuse. Indeed, the

SORA court adjudicating Knox’s risk level noted that the legislative history “makes it clear

that not only did they want to go after sex crimes, but also specifically against child

abduction crimes” (Hearing Transcript, page 12, lines 15-21). Defendant here did no such

thing. He made no attempt to abduct his cousin or otherwise remove him from his everyday

surroundings. It simply does not follow from the fact that because defendant was willing

to rob his aunt at gunpoint that he would someday sexually assault a child or put a child at

risk of sexual abuse. Additionally, whether Knox or defendant here are sex offenders “in

the colloquial sense” (dissenting op at 5 n 2) is not the legal issue before us; we must decide

whether the State may rationally compel this defendant to register as one.

- 10 -

- 11 - No. 83

Apart from these critical factual differences between defendant’s case and the Knox

trio of appeals, none of the Knox rationales apply to defendant’s circumstances. Although

the Knox Court cited data at the time SORA was enacted and thereafter that found a

likelihood of sexual assault in nonfamily abductions, the underlying offense here was not

a nonfamily abduction; defendant was the child’s blood relative. As the Court

acknowledged, those studies reported actual sex offenses, and here the crime was a robbery

with no sexual conduct or motive. Thus, these studies do not support a sex offender

designation for defendant.

Nor is there support to be found in the Knox Court’s observation that “a child cut

off from the safety of [their] everyday surroundings is vulnerable to sexual abuse even if

the offender’s desires are not the motive of the crime” (id. at 68). It is undisputed that

defendant never removed the child from his home. Unlike the child grabbed at a playground

in Knox, who the defendant intended to make her own, the children in Cintron who were

kept captive in their apartment for days, and the child in Jackson who was abducted upon

threat of being killed if his mother did not agree to the demands of that defendant, here

defendant’s cousin, as the SORA court found, was “not kidnapped or detained for any

prolonged period of time” and the cousin’s detention did not put him at risk of a sexual

assault. The child was not “cut off from his everyday surroundings” nor at risk of sexual

violence as was the case in Cintron where the defendant had previously been convicted of

forced sexual intercourse with a minor, or Jackson where the defendant profited from

facilitating prostitution. In both of those cases the record supported a finding of the

respective defendant’s risk of future sexual harm to children. Here, there was no past sexual

- 11 -

- 12 - No. 83

offense and an affirmative judicial finding that defendant posed no future risk.6 There were

no such findings in Knox.

It is one thing to say that the Legislature may cast a wide net by “employ[ing]

overinclusive terms” to include within SORA’s reach those who commit a non-sexual

crime but nonetheless present a future risk of sexual harm to children (dissenting op at 3).

It is another to say that there can be no remedy for those whose crime did not involve a

sexual act or motive, in no way targeted a child, and who pose no future risk of sexual

harm. Contrary to the dissent’s presumption, Knox held that the former was constitutionally

permissible, but it did not close the door on a judicial remedy for those who are

undisputedly not a sexual danger to children. The Court in Knox focused on each

defendant’s underlying SORA-eligible crime as a basis for a legislative determination that

they could be labeled as sex offenders without inflicting constitutional harm. We cannot

say that is true of defendant here; defendant is one of the “few” individuals, recognized by

the Knox Court, for whom the sex offender designation “is unmerited” (Knox, 12 NY3d at

69). Knox never held that a person who is found to pose no risk of sexual harm to children

6

Having no legitimate basis to escape the SORA court’s explicit finding on this point, the

dissent attempts to recast that court’s use of the past tense as “merely another way of saying

that ‘there was no actual or potential sexual aspect’ to defendant’s crime” (dissenting op at

6). This is an inventive way to find hidden meaning in the court’s unambiguous words: that

“the [c]ourt does not find him to be a sex offender and that he posed no sexual risk of sexual

threat at all [emphasis added].” Along with the court’s other findings regarding the total

absence of any sexual motivation on defendant’s part and the role of SORA courts in

assessing future risks (see Corrections Law §§ 168-d, 168-n; People v Johnson, 11 NY3d

416, 421 [2008]), there is no way to interpret the court’s words as anything other than its

recognition that defendant did not pose a risk of future harm and was not a sexual threat to

children.

- 12 -

- 13 - No. 83

can have no judicial remedy for the constitutional harm inflicted simply because it is

administratively convenient. It is precisely because there is a factual finding of defendant’s

lack of risk—a finding which we have no power to disturb—that defendant’s as-applied

challenge must succeed.

The SORA court is tasked with determining each defendant’s future risk of sexual

offense based on the court’s assessment (Johnson, 11 NY3d at 421 [explaining that the

SORA court “is to make its determination after considering (the Board’s) recommendation,

and any other materials properly before it”]). Were this not so, there would be no need for

a hearing where the defendant has an opportunity to present evidence and opposing

arguments regarding their likelihood of recidivism, and for the court to make a legal

determination based on the parties’ submissions. By suggesting that the SORA court in this

case went “well beyond” the statutory framework when acknowledging defendant’s lack

of risk, the dissent appears to discourage SORA courts from making such findings

(dissenting op at 9 n 6). Members of the judiciary should decline the dissenters’ invitation

to ignore their constitutional duties and statutory factfinding responsibilities.7

7

Worryingly, the dissent seems to suggest that the Legislature has the power to foreclose

as-applied constitutional challenges simply by using “overinclusive terms” (see dissenting

op at 7-8). The dissent misunderstands basic constitutional principles and rudimentary

aspects of judicial review. When presented with an as-applied challenge like this one, it is

our job to decide whether the legislation is unconstitutional based on the record before us

(see People v Stuart, 100 NY2d 412, 421 [2003]). The dissent’s view of the SORA court

as a rubber stamp is also contrary to our constitutional principles and contrary to the

Legislature’s mandate (dissenting op at 8-9).

- 13 -

- 14 - No. 83

As the Court in Knox recognized, a sex offender designation imposes a stigma that

broadly impacts a defendant’s life and ability to participate in society, but, the Court

reasoned, that stigma is no worse than that associated with being labeled a “child predator”

(12 NY3d at 69). That reasoning cannot support designating defendant a sex offender

where his crime is not sexual in nature and his conduct provides no basis to predict risk of

future sexual harm. The stigma of this misdesignation unconstitutionally impacts

defendant’s liberty interest in a criminal designation that rationally fits his conduct and

public safety risk.

Lastly, imposing SORA requirements on this defendant does not further “the

protection of the community against people who have shown themselves capable of

committing sex crimes”—the purpose of SORA (Knox, 12 NY3d at 67). Put another way,

the concerns we identified in Knox are not present here in light of the SORA court’s

findings. Defendant committed a terrible crime, but he has not shown himself to be capable

of committing sex crimes. The law should not mislead the public into thinking otherwise

by listing him in the registry alongside rapists and child molesters.

In sum, the core holding and reasoning in Knox does not control here where

defendant’s criminal behavior and his lack of future risk of sexual harm to children

distinguish him from the Knox defendants. The Knox Court’s conclusion that designating

those three defendants as sex offenders and mandating their SORA registration was

rationally related to the government’s interest in protecting children from sexual assault

lacks force here, where defendant’s offense belies any such threat. Where a SORA court

expressly finds that there was no sexual conduct or motive and no risk of future sexual

- 14 -

- 15 - No. 83

offense, application of the sex offender label cannot be justified on the ground that a

defendant may pose a risk of future sexual misconduct, and there is no further

administrative burden in reaching that conclusion. Applying SORA to defendant violates

his due process rights by impinging on his liberty interest to be free of the improper

designation and registration as a “sex offender.”

III.

Revisiting Knox

Separately, I write for myself and Chief Judge Wilson that automatic registration

for nonparents convicted of child abduction and unlawful imprisonment of a minor where

the crime involves no sexual act or motivation and who present no risk of sexual threat to

children is facially unconstitutional. To the extent Knox holds otherwise we would overrule

it.

Knox correctly concluded that where a nonparent defendant who kidnaps or

unlawfully imprisons a minor poses a risk of future sexual harm, the Legislature had a

legitimate basis for mandating the defendant’s registration as a sex offender. But Knox had

no analytically sound basis for its conclusion that the Legislature had a legitimate basis for

requiring defendants whose crimes lack any sexual component and pose no risk of future

sexual harm to children to be designated and registered as a sex offender. Recent federal

district court decisions have so held, putting the continued application of Knox to this class

of defendants in question (see Yunus v Robinson, 2019 WL 168544 [SD NY 2019];

Pennington v Rosado, 2022 WL 3127062 [SD NY 2022]).

- 15 -

- 16 - No. 83

Moreover, the stigma attached to the sex offender designation recognized by the

Court in Knox has increased in severity and impact with the expanded use of the SORA

registry by the public. And the concerns raised by the Court regarding administrative

burdens have not materialized, as courts readily distinguish sex offenders from other

criminals.

A.

“Under stare decisis principles, a case ‘may be overruled only when there is a

compelling justification for doing so’ ” (People v Peque, 22 NY3d 168, 194 [2013], quoting

People v Lopez, 16 NY3d 375, 384 n 5 [2011]). A “compelling justification may arise when

the Court’s prior holding ‘leads to an unworkable rule, or . . . creates more questions than

it resolves’ ” (id., citing People v Taylor, 9 NY3d 129, 149 [2007]) or when “a preexisting

rule, once thought defensible, no longer serves the ends of justice or withstands the cold

light of logic and experience” (id., citing Policano v Herbert, 7 NY3d 588, 604 [2006]).

Generally, “a court should ‘not . . . apply stare decisis as rigidly in constitutional as in

nonconstitutional cases’ ” (Taylor at 149, quoting Glidden Co. v Zdanok, 370 US 530, 543

[1962]). For the reasons I discuss, the Knox Court’s conclusion that the Legislature may

designate as “sex offenders” all persons convicted of an attempted or completed act of child

kidnapping and unlawful imprisonment, even when a defendant poses no sexual risk of

harm to children, does not serve the ends of justice (12 NY3d at 69). Such designation

imposes a harm of constitutional proportions on a defendant that poses no risk without

furthering SORA’s public safety purpose.

B.

- 16 -

- 17 - No. 83

The Knox analysis is doctrinally flawed and lacks a relevant factual predicate. First,

the Knox Court misunderstood the liberty interest at stake in cases where a person is treated

as guilty of having committed an offense involving sexual misconduct or motivation when,

in fact, they are innocent of such crime, and pose no risk to the general public. That

defendant’s interest is not simply the right to avoid being mislabeled, it is the right not to

be treated guilty of a crime the individual never committed. Put another way, a defendant

has an interest in retaining the status of a non-sex offender. The difference matters because

our criminal law system depends on the proper finding of guilt as related to the actual

criminal behavior for which the individual is accused. Indeed, our system treats crimes

differently based on the seriousness of the crime and the scope of public harm that flows

from a particular criminal action (see Penal Law §§ 70.00, 70.02 [classifying felonies and

violent felonies and setting the maximum sentences by class], 70.15 [maximum sentences

for misdemeanors and violations]).

The misclassification as a sex offender has severe consequences. Although some

amount of stigma results from the simple fact of a criminal conviction, the extent of that

stigma flows from the nature of the particular crime committed. For example, the stigma

attached to a conviction for burglary is not the same as the stigma attached to a conviction

for child sexual assault. Sex offenders are societal pariahs. “More than ‘name calling by

public officials,’ ” a sex offender label ‘is a determination of status’ that can have a

considerable adverse impact on an individual’s ability to live in a community” (People v

David W., 95 NY2d 130, 138 [2000], citing Paul v Davis, 424 US 693, 703 [1976]). “As

the experiences of convicted sex offenders . . . has shown,” widespread public

- 17 -

- 18 - No. 83

dissemination of an individual’s sex offender status and other personal information “is

likely to carry with it shame, humiliation, ostracism, loss of employment and decreased

opportunities for employment, perhaps even physical violence, and a multitude of other

adverse consequences” (Doe v Pataki, 3 F Supp 2d 456, 467-468 [SD NY 1998]). The

“consequences of community notification are unlimited” and “the stigma created by

[SORA registration] pervades into every aspect of an offender’s life” (Doe v Pataki, 120

F3d 1263, 1279 [2d Cir 1997] [internal quotation marks and brackets omitted]).

In the years since this Court decided Knox, that stigma has only expanded in scope

and severity, affecting a sex offender’s ability to find and retain employment and housing,

and limiting their efforts to reenter society and to engage with family and friends; it

“consigns them to years, if not a lifetime, of existence on the margins” (Does #1-5 v Snyder,

834 F3d 696, 705 [6th Cir 2016], cert denied 583 US 814 [2017]). For example, at the time

Knox was decided, the sex offender registry was searchable by name. In 2015, a renter

screening site debuted an online tool compiling data from the public registry into a map

searchable by neighborhood (Viewing NYC, This Map Shows You Every Registered Sex

Offender in the City, available at https://viewing.nyc/this-map-shows-you-every-

registered-sex-offender-in-the-city/ [last accessed Nov. 6, 2023]). The site explained that,

for example, “if you live on the Lower East Side, you [couldn’t] see all the offenders near

you. Until now.” (id.). Similar map tools searchable by address or zip code are available

now on other websites such as “Offender Radar” (Offender Radar, National Offender

Registry and Offender Resource Center, available at https://www.offenderradar.com/ [last

accessed Nov. 9, 2023]). Such tools make identifying offenders in a particular area easier

- 18 -

- 19 - No. 83

and increase the risk of vigilante violence against sex offenders (see Matt Clarke,

Vigilantes Assault, Rob and Murder Registered Sex Offenders, Prison Legal News,

available at https://www.prisonlegalnews.org/news/2017/may/5/vigilantes-assault-rob-

and-murder-registered-sex-offenders/ [last accessed Nov. 9, 2023] [discussing more than

ten examples of vigilante attacks against actual and suspected sex offenders between 2005

and 2017]).

Given the impact of a sex offender designation, there should be no room for error

in classification. The label must be limited to those who are actual sex offenders, as society

understands that term: those persons who pose a risk to public safety of sexual recidivism

against vulnerable members of our population. Indeed, proper identification of these

criminal actors is the constitutional justification for sex offender registries (see Smith v

Doe, 538 US 84, 85 [2003] [“an imposition of restrictive measures on sex offenders

adjudged to be dangerous is a legitimate nonpunitive governmental objective . . . to protect

the public from harm”]; citing Kansas v Hendricks, 521 US 346, 363 [1997] [“The State

may take measures to restrict the freedom of . . . individuals who present a danger to the

public . . . This is a legitimate nonpunitive governmental objective and has been historically

so regarded”]).

The Knox Court concluded that a defendant’s right not to be mislabeled as a sex

offender was outweighed by the greater concern of ensuring that no person who places a

child at risk for sexual assault ‘slips through the cracks’ and escapes SORA registration.

This conclusion ignores that our Constitution does not tolerate wrongful criminal

designations. A “sex offender” may indeed be a “child predator” and therefore suffer no

- 19 -

- 20 - No. 83

injustice by bearing one label instead of the other—as Knox opined—but a person

convicted of a crime that lacks any sexual act or motivation and who poses absolutely no

sexual threat against children is not a “child predator” for purposes of SORA or any other

purpose. That person suffers a true injustice by being designated a “sex offender.”

C.

The Knox Court was also incorrect that SORA’s designation as “sex offenders” of

defendants convicted of attempted or completed child kidnapping and unlawful

imprisonment crimes survives rational basis review.8 The Court’s reasons for concluding

that the Legislature had a rational basis for labeling persons who do not merit a sex offender

designation—studies showing a potential risk of sexual abuse during nonfamily abductions

8

The Knox Court concluded that this liberty interest was not a “fundamental right” in due

process terms because fundamental rights are those “deeply rooted in this Nation’s history

and tradition” (id. at 67, citing Washington v Glucksberg, 521 US 702 [1997], quoting

Moore v East Cleveland, 431 US 494 [1977] [plurality op]). By way of comparison, the

Court explained that “[t]he right not to have a misleading label attached to one’s serious

crime” is unlike the right to marry, to have children, to decide how to educate your child,

and to engage in private consensual sexual activity—rights previously identified as

fundamental by the United States Supreme Court (id. at 67 [internal citations omitted]).

This analogy is open to criticism because the cases referenced by the Knox Court do not

involve criminal behavior or public risk of sexual offense recidivism. Arguably, the correct

analysis should be grounded on criminal justice principles and the core due process rights

of criminal defendants (see e.g. In re Winship, 397 US 358, 368 [1970] [due process

requires proof beyond all reasonable doubt for criminal convictions and non-punitive

juvenile delinquency adjudications alike]). In accordance with those principles, and given

the weighty rights guaranteed to criminal defendants, SORA’s sex offender designation

and mandatory registration requirement would be subject to strict scrutiny. Under that

standard, the law passes constitutional muster only if it is “narrowly tailored to serve a

compelling state interest” (Reno v Flores, 507 US 292, 302 [1993]). SORA fails under that

heightened standard. Of course, as discussed, supra, SORA’s application to defendants

who pose no risk of sexual offense fails even under the more deferential rational basis

standard.

- 20 -

- 21 - No. 83

and the administrative burden of identifying defendants convicted of kidnapping and

imprisonment of children who pose no risk of sexual harm to children—do not support

such broad application to defendants who are not sexual threats to the public.

The two studies cited in Knox refer to the percentages of nonfamily abductions that

involve actual sexual assaults and therefore do not support a sex offender designation for

those persons who, in fact, have not committed or attempted a sexual crime against a child

and who the court finds do not pose a risk of committing one in the future. In addition,

recent studies show that the vast majority of abductions are committed by parents and not

the class of defendants covered by the SORA provisions at issue in this appeal (Wolak,

Finkelhor, & Sedlak, Child Victims of Stereotypical Kidnappings Known to Law

Enforcement in 2011, U.S. Department of Justice – Juvenile Justice Bulletin [June 2016];

Finkelhor et al., Family abduction in a national sample of US children, 67 Child Abuse &

Neglect 403 [May 2017], https://www.sciencedirect.com/science/article/

abs/pii/S0145213416302599 [last accessed Oct. 29, 2023]; Correction Law § 168-a [2] [a]

[i]).

The administrative burden justification also fails the rational basis test. This concern

is wholly a creation of the Knox Court. The prosecution did not raise this as a ground in

any of the three appeals, and there is no legislative history to suggest the Legislature saw

this alleged problem. Indeed, it is hard to square the Knox Court’s conclusion that it is

simply too difficult and burdensome to accurately distinguish non-sex offenders from sex

offenders, and defendants who pose no risk to the community of future sex crimes from

defendants who present a risk of sexual recidivism, when distinguishing criminals from

- 21 -

- 22 - No. 83

non-criminals is exactly the work of the criminal legal system. Indeed, the Board and the

SORA court are tasked with of the individual assessment of a defendant’s risk, based on

the available information and upon findings of fact (Correction Law § 168-l).

Nor does the sex offender registration of a person who poses no risk of sexual harm

to children further the goals of SORA and the interests of the community. The purpose of

SORA is simple: “to protect the public from sex offenders” (People v Mingo, 12 NY3d 563

[2009]). The legislative history makes clear that “protection of the public” from “the danger

of recidivism posed by sex offenders, especially [sexually violent offenders] . . . is of

paramount concern or interest to the government” and that the sex offender registry is

meant to aid law enforcement agencies and the justice system in “identify[ing],

investigat[ing], apprehend[ing] and prosecut[ing]” those offenders (Doe, 120 F3d at 1276).

Thus, the purpose of SORA is clearly to track those who are actual sex offenders—it is not

intended as a registry of anyone who commits a crime. Populating the registry with the

names and information of individuals who do not pose a danger to children of sexual

recidivism “undermin[es] the usefulness of the registry” and wastes government resources

on tracking people who are not the intended targets of SORA nor implicate the public risk

and law enforcement needs that first necessitated SORA registries (Yunus, 2019 WL

168544 at *11, citing People v Diaz, 150 AD3d 60, 66 [1st Dept 2017]; Pennington, 2022

WL 3127062 at *8; see also Senate Introducer’s Mem in Support, Bill Jacket, L 1995 ch

192 at 19-20 [“Sexual offenses and sexual assault are among the most heinous of crimes

and often the victims are those least able to protect and defend themselves – children. . . .

- 22 -

- 23 - No. 83

By establishing a tracking/notification system local law enforcement agencies will be able

to identify and monitor sex offenders”]).

At the time Knox was decided, three of the four other state courts that had considered

the issue had reached a contrary conclusion (State v Robinson, 873 So2d 1205 [Fla 2004];

American Civ. Liberties Union of N.M. v City of Albuquerque, 139 NM 761 [NM 2006];

State v Reine, 2003 WL 77174 [Ct App OH, 2d Dist 2003]).9 Knox declined to follow those

9

The dissent’s assertion that post-Knox, the trend of courts in other jurisdictions has been

“in Knox’s direction” is overstated (dissenting op at 12-13 & n 7). There is no “trend”

supporting the dissent’s position that because the great majority of child abductors are

found to pose a risk of sexual harm to children based on past conduct or a future risk

assessment, an as-applied challenge cannot succeed. Quite the opposite. Other courts have

held “that sex offender registration requirements violated due process as applied to persons

convicted of kidnapping a child” and that “typically, these cases have involved situations

where the State conceded, or the facts conclusively demonstrated, that there was no sexual

component or motivation in the kidnapping” (Doe v Sex Offender Registry Bd. (488 Mass

15, 25, 170 NE3d 1143 [2021] citing State v Robinson, 873 So 2d 1205, 1217 [Fla 2004]).

Here, defendant’s as-applied challenge is even stronger because not only did the State

concede and the SORA court find that defendant’s restraint of his cousin was not sexually

motivated, but the SORA court also found that defendant did not pose a future risk of sexual

misconduct (see id.) (SORA hearing examiner concluded that the defendant, who

physically assaulted a ten-year-old girl, punching her repeatedly and dragging her down a

stairwell and who had been previously charged with aggravated rape of a child, “posed a

high risk of reoffense” (id. at 18-19).

In any event, the cases cited by the dissent are factually distinguishable (see Moffitt v Com.,

360 SW3d 247 [Ky Ct App 2012] [the “guilty verdict in this case necessarily required the

jury to find beyond reasonable doubt that (the offender) intended to commit a sex crime”

so the kidnapping in fact “did include a sexual component”]; State v Coleman, 241 Ariz

190, 385 P3d 420 [Ariz Ct App 2016] [the offender grabbed a baby’s arm to pull her away

from her mother and physically assaulted the mother]; Rainer v State, 286 Ga 675, 690

SE2d 827 [2010] [the offender planned to rob the 17-year-old victim in a drug deal and a

physical “struggle” ensued]; State v Smith, 323 Wis 2d 377, 780 NW2d 90 [2010] [after

forcing the minor victim into a vehicle, the offender “physically threatened the minor”]).

- 23 -

- 24 - No. 83

cases and instead followed People v Johnson (225 Ill 2d 573 [Ill 2007]), where the Illinois

Supreme Court upheld their version of SORA under rational basis review. But as the

dissent in Johnson points out, the Illinois Legislature amended its SORA while that appeal

was pending, making kidnapping a “sex offense” only where “the offense was sexually

motivated” (id. at 593-594 [Burke, J., dissenting], citing Ill Pub Act 94-945 § 2 [B] [1.5]).

The amendment also “create[d] a new registry for offenders who commit certain violent

crimes, including aggravated kidnapping of a child by a nonparent, when those crimes are

committed without sexual motivation. . . . [Such offenders] will not, however, be a

registered sex offender” (id. at 594). Given that new legislation, which did not make a non-

sexually motivated kidnapping a sex offense, Knox’s reliance on Johnson was inapt.

IV.

If all this were not reason enough to revisit Knox, there are recent federal decisions

rejecting the Knox analysis for this class of defendants. So long as Knox remains the law,

defendant and others similarly situated will have meritorious grounds for federal relief, as

did the petitioners in Yunus (2019 WL 168544) and Pennington (2022 WL 3127062), who

likewise committed kidnapping offenses with no sexual motivation.10 The magistrate judge

in Yunus agreed with the petitioner that the Knox court had “failed to appreciate the costs

of sex offender registration” and acknowledged that a sex offender label comes with

10

Amici note there is no right to court-appointed counsel in federal habeas corpus

proceedings. This leads to the disparate treatment of similarly situated defendants, with

one class having greater access to federal relief solely because they can afford it.

- 24 -

- 25 - No. 83

“profound stigma,” “a long list of filing requirements, employment prohibitions, both de

jure and de facto, residency restrictions, and other civil disabilities,” as well as “a host of

expanded criminal risks, including state and federal penalties for failure to register” (2018

WL 3455408, *24 [SD NY 2018]). Given those “devastating consequences,” the court

could not “conclude that due process is satisfied by a statistical showing that 46% of the

kidnappers placed on the sex offender registry—a group which [did] not include the

plaintiff at bar—deserve to be there” (id. at *25). The court also rejected that “the

‘administrative burden’ of determining which kidnappers are sex offenders makes it

rational for the State of New York to subject [the petitioner] to 20 years of public

opprobrium . . . after conceding that—for him—the term is ‘unmerited’” (id.). District court

judge Alison Nathan—now sitting on the Second Circuit—agreed and adopted the report

and recommendation, finding the “slipping-through-the-cracks argument” “insufficient to

provide a rational basis for imposing extensive civil and stigmatizing burdens” (2019 WL

168544 at *10; see also Pennington, 2022 WL 3127062 at *8 [holding that the State may

“impose a designation” that is “rationally related to any non-sexual risk (the petitioner)

might pose to children,” but may not subject him to “a specifically stigmatizing designation

and restrictions designed to prevent sexual abuse”]). According to the court, even if such a

rational basis existed, “no further administrative effort is required” where “the absence of

a sexual element is undisputed” (id. at *11).11

11

The presiding judge at the SORA hearing in Yunus “found that there was virtually no

likelihood that Plaintiff would ever commit a sex crime” (2019 WL 168544 at *1).

- 25 -

- 26 - No. 83

Under stare decisis principles, these federal court decisions provide a reason to

reconsider and overturn Knox for this class of defendants that pose no risk of sexual harm

to children. The analysis by the federal courts is persuasive and thus Knox’s “preexisting

rule, once thought defensible, no longer serves the ends of justice or withstands the cold

light of logic and experience” (Policano, 7 NY3d at 604). Moreover, given the federal

rulings, the Knox rule is unworkable as it leads to increased litigation in federal court. There

is simply no legitimate interest served by a rule that requires individuals like defendant to

exhaust a constitutional challenge in state court or seek federal relief, all while bearing the

“sex offender” label, which, upon success in federal court, will be belatedly removed—

although the stigma it brings can never be fully erased.

Defendants have a liberty interest in not being misclassified as a sex offender and

required to register under SORA. And while defendants are clearly harmed by this

misclassification, there is no countervailing public benefit from a rule that does not achieve

the SORA provisions’ child safety purpose—and even undermines that purpose by making

the registry less reliable. Judicial inertia in the face of logic, reason and evidence of the

mismatch between the legislative goal and the means adopted is indefensible. We would

overrule Knox to the extent it upholds that mismatch and hope that it will not be long before

a majority of the Court agrees.

V.

Defendant stands convicted of a crime that lacks a sexual act or motive. The SORA

court found that he is not a sex offender and is not a sexual threat. Under these

- 26 -

- 27 - No. 83

circumstances, and unlike the defendants in Knox, application of SORA’s sex offender

designation and registration requirement to defendant is a clear violation of his due process

rights.

Accordingly, the order of the Appellate Division should be reversed, with costs, and

defendant’s Sex Offender Registration Act designation vacated.

- 27 -

CANNATARO, J. (dissenting):

A little more than a decade ago, in People v Knox, this Court unanimously decided

that defendants who commit, or attempt to commit, kidnapping and unlawful imprisonment

-1-

-2- No. 83

of other people’s children can be compelled to register as “sex offenders” under the Sex

Offender Registration Act (SORA) without offending the due process guarantees of our

state and federal constitutions, even if “there was no actual or potential sexual aspect in

any of the[ir] crimes” (12 NY3d 60, 66 [2009]). Today the majority all but explicitly

overturns that recent precedent. Because in doing so the majority evades our well-

established principles of stare decisis, I respectfully dissent.

I.

There are two ways to go about overturning precedent. The first, straightforward,

approach at least attempts to apply stare decisis principles (see op of Rivera, J., parts III-

IV [writing solely for herself and Chief Judge Wilson]). The more indirect method, chosen

by the majority, “distinguishes” precedent to the point that it unravels. Although I disagree

with Judge Rivera that defendant has made the compelling showing necessary to justify

departure from stare decisis in this case (see part II, infra), my greater concern is with the

circumvention of those core jurisprudential principles by the majority.

In Knox, we held that it was “plainly rational”—and therefore constitutional—for

the legislature to require all persons who kidnap or unlawfully imprison other people’s

children to register as sex offenders under SORA, even if that rule impacts some

individuals for whom the term is technically “unmerited” (12 NY3d at 69). We concluded

that application of “a hard and fast rule, with no exceptions,” was justified by

administrative convenience and the rational inference that “in the large majority of cases

-2-

-3- No. 83

where people kidnap or unlawfully imprison other people’s children, the children are either

sexually assaulted or in danger of sexual assault” (id. at 68-69 [emphasis added]).

Our holding in Knox respected the prerogative of the legislature to employ

overinclusive terms when doing so does not impinge on a fundamental right and rationally

furthers a legitimate State objective—here, the government’s compelling interest in

protecting the public, and especially children, from recidivism by sex offenders (see id. at

67). Application of the rational basis test—which is “not a demanding one”—begins with

the “strong presumption that legislative enactments are constitutional” (id. at 69). The test

requires the party contending otherwise to bear “the heavy burden of showing that a statute

is ‘so unrelated to the achievement of any combination of legitimate purposes’ as to be

irrational” (id. at 69, quoting Affronti v Crosson, 95 NY2d 713, 719 [2001]). Under that

standard, “[a] classification that has a rational basis or relationship to a legitimate State

interest will be upheld even though it ‘is not made with mathematical nicety or because in

practice it results in some inequality’” (Lovelace v Gross, 80 NY2d 419, 427 [1992],

quoting Dandridge v Williams, 397 US 471, 485 [1970]; accord Vance v Bradley, 440 US

93, 108 [1979] [“Even if the classification involved here is to some extent both

underinclusive and overinclusive, and hence the line drawn by Congress imperfect, it is

nevertheless the rule that perfection is by no means required” (internal quotation marks

omitted)]; Rojas-Reyes v Immigration & Naturalization Serv., 235 F3d 115, 123-124 [2d

Cir 2000] [“a law will be upheld under rational basis review if any reasonably conceivable

state of facts could demonstrate that the statute is rationally related to a legitimate

-3-

-4- No. 83

government purpose . . . because the judicial system has long recognized that the problems

of government are practical ones and may justify, if they do not require, rough

accommodations—illogical, it may be, and unscientific” (internal quotation marks and

citations omitted)]; People v Turnage, 55 Cal 4th 62, 77 [2012] [“When conducting rational

basis review, we must accept any gross generalizations and rough accommodations that the

Legislature seems to have made”]). The deference afforded by the rational basis test makes

it “a paradigm of judicial restraint” (see Knox, 12 NY3d at 69 [internal quotation marks

omitted]; Affronti, 95 NY2d at 719).

Here, the record establishes that defendant is one of the “few” persons who must

register as a sex offender even though his crime involved no sexual component or motive1

(see Knox, 12 NY3d at 69). However, this puts him in no better position than any of the

three defendants in Knox, who we assumed fell within the “small minority” of cases “in

which there was neither a sexual assault nor any discernible risk of one” (id.). Indeed, in

the opening paragraph of Knox, we observed that “there was no proof that [the defendants’]

crimes involved any sexual act or sexual motive” (12 NY3d at 64 [emphasis added]). We

also noted the People’s “acknowledge[ment]” that there was “no evidence of sexual

misconduct in any of these cases” (id. at 66). Our ultimate conclusion was that even “on

1

Although unclear from the record, defendant does not contend that his trial counsel failed

to inform him of this consequence of his plea. If defendant voluntarily agreed to plead

guilty to a SORA-qualifying offense with knowledge that doing so would require him to

register as a sex offender, that would significantly undermine his claim here.

-4-

-5- No. 83

the assumption that there was no actual or potential sexual aspect in any of these crimes

. . . defendants’ constitutional rights have not been violated” (id.).

Because we decided Knox on the premise that the three defendants’ crimes involved

no actual sexual component—and further, that even “the possibility of . . . threatened sexual

misconduct” had been “ruled out” (see id.)—the existence of findings to that effect in the

instant case is not a “significant[]” distinguishing feature (majority op at 9; see Mackey v

Warden, Lebanon Corr. Inst., 525 Fed Appx 357, 361 [6th Cir 2013] [rejecting distinction

based on proposition that was assumed in prior case]). Although the legislature “might

well have made such cases an exception to the requirement that the convicted people

register as sex offenders . . . it acted rationally in not doing so” (Knox, 12 NY3d at 69).

Unable to distinguish Knox on this basis, the majority instead invents a rationale

that appears nowhere in Knox: that the underlying facts of the Knox defendants’ crimes

“provided some basis to conclude that each of those defendants posed a future risk of sexual

harm to children” (majority op at 9 [emphasis added]). Preliminarily, that is not true with

respect to the lead defendant, Knox.2 But even if it were, we made no such finding or

2

Knox’s SORA-qualifying offense consisted of attempting to abduct an eight-year-old girl

in a park for the apparent purpose of replacing one of her own children, of whom she no

longer had custody (12 NY3d at 64; majority op at 6). The majority strains to explain why

Knox “posed a future risk of sexual harm,” mustering only that Knox was “homeless and

suffering from mental illness,” and if successful purportedly “would have put the child in

a position vulnerable to abuse” of some kind (majority op at 9-10 & n5 [acknowledging

that these facts were “not elucidated in the Court’s opinion” in Knox]). That rationalization

is not one I can imagine the majority accepting if Knox’s appeal were actually before the

Court today, and it certainly was not the basis for our decision in 2009. It is utterly

speculative and appears to punish Knox for the actions of some unidentified and completely

theoretical third party.

-5-

-6- No. 83

observation in Knox, nor did the facts identified by the majority play any role in our

constitutional analysis—they function merely as a post hoc justification for setting aside

our actual holding that the legislature was constitutionally permitted to define “sex

offender” to encompass persons like defendant here.

The majority also mischaracterizes the SORA court’s findings in this case. Contrary

to the majority’s repeated assertion, there was no judicial finding below that defendant

poses “no future risk” of sexual harm (majority op at 12, 5). What the SORA court actually

stated was that defendant “posed [past tense] no risk of sexual threat”—an unambiguous

reference to his conduct during the robbery. That is merely another way of saying that

“there was no actual or potential sexual aspect” to defendant’s crime, which is precisely

what we assumed in Knox (12 NY3d at 66). With respect to defendant’s risk of future

sexual harm, the SORA court reached the same conclusion as the courts in Knox with

respect to defendants Knox and Jackson—it classified him as a level one sex offender, the

applicable designation for all offenders who pose a “low risk” of future sexual harm (see

To be sure, the underlying facts of the crimes committed by Knox and defendant

were different. Knox was unstable and clearly intended no harm to the child; defendant

willfully held his ten-year-old close relative at gunpoint while restraining and robbing the

child’s mother, whom he had lived with for years. But at the end of the day, it cannot be

disputed that neither Knox nor defendant is a “sex offender” in the colloquial sense (the

majority claims this “is not the legal issue” but inconsistently attempts to distinguish this

case based in part on the SORA court’s finding that defendant was “not a sex offender”)

(id. at 3, 9). Nor did either of their underlying crimes or backgrounds suggest that they

posed any significant future risk of sexual harm. Defendant’s and Knox’s facts are not

distinguishable in any material aspect, and the majority’s assertions to the contrary are

baseless.

-6-

-7- No. 83

id. at 70 [noting that defendants Knox and Jackson were classified as “level one (low risk)”

sex offenders, and affirming the classification of defendant Cintron as a “level three (high

risk)” offender based on the facts identified in today’s majority opinion]).3 The majority

is forced to rewrite the SORA court’s findings in this manner because its ends-oriented

analysis demands it in order to distinguish Knox.

The majority’s further assertion that “none of the Knox rationales apply to

defendant’s circumstances” (majority op at 11) ignores the ultimate holding of Knox, which

is that the applicability of those rationales in the “large majority” of cases involving these

SORA-qualifying offenses justifies the marginally overinclusive application of the rule

against persons like defendant. It is that principle which drove the decision to reject the

claims of the defendants in Knox, despite our assumption that they, too, fell into the

“minority” class (12 NY3d at 68-69). This appeal merely confirms the existence of outlier

cases we assumed could exist in Knox; but it in no way distinguishes that decision or

exposes any irrationality in its reasoning.4

If anything, this case demonstrates that the legislature’s inference that outlier cases

are extremely rare was, in fact, rational. At oral argument, defense counsel was asked if

she had any “sense of the number of individuals that fit within this narrow category we’re

3

This will generally be the case where an individual’s SORA-qualifying offense involved

no actual or potential sexual aspect.

4

The attempt to distinguish this case based on the “blood” relationship (see majority op at

11) between defendant and his cousin is equally unavailing. Because the record is devoid

of any data on kidnappings and false imprisonments by cousins, I cannot conclude that the

legislature acted irrationally in declining to exclude them from the sex offender registration

requirement as they did parents.

-7-

-8- No. 83

talking about”—specifically, persons whose offenses involved “no sexual motivation[ or]

sexual act” (oral argument tr at 14, lines 1-16 [Oct 19, 2023]). Counsel forthrightly

admitted that the class of individuals who fit within this narrow category amounts to “0.001

percent” of all SORA cases (id. at 14, line 8). In crafting its purported distinction among

such an infinitesimal class of persons, the majority has usurped the legislative function and

disrespected principles of stare decisis. Indeed, in holding that there must be a remedy for

any defendant as to whom the overarching rationales of Knox do not individually apply

(majority op at 12), the majority is effectively holding that the legislature has no

prerogative to use overinclusive terms in this arena. That can only be true if one ignores

the holding of Knox that the rights at stake are not “fundamental” and subjects the entire

statutory process to a higher level of judicial review in order to reach the desired outcome.5

Finally, in pointing to the SORA court’s “affirmative judicial finding[s]” as the

“most significant[]” distinguishing factor here (see majority op at 12, 9), the majority

arguably creates a standard under which the success of a constitutional challenge turns not

on the underlying conviction, but on the election of a SORA court to analyze the criminal

record and make findings that are neither statutorily nor constitutionally required. As we

recognized in Knox, one purpose of the legislature’s “hard and fast rule” was to relieve

5

Unable to dispute these points, the majority employs straw man tactics, accusing me of

arguing that “the Legislature has the power to foreclose as-applied constitutional

challenges simply by using ‘overinclusive terms’” (majority op at 13 n 7). Not even

remotely. My point is that this Court has already decided an as-applied challenge to

SORA on materially indistinguishable facts and determined that the legislature’s use of

overinclusive terms is constitutional, resulting in application of stare decisis.

-8-

-9- No. 83

SORA courts of the administrative burden of analyzing each kidnapping and false

imprisonment offense in order to identify the few cases which lack any actual or potential

sexual aspect (see 12 NY3d at 69). This Court unanimously and unambiguously held that

the legislature’s decision to relieve courts of that administrative burden was rational and

constitutionally permissible. Consequently, there can be no requirement—not under the

constitution, and certainly not under the statute, which speaks for itself—to make the kinds

of findings on which the majority now attempts to distinguish Knox.6 This gives rise to a

risk of arbitrary decision-making.

Today’s majority responds by asserting that a court that fails to make such findings

“ignore[s] their constitutional duties and statutory factfinding responsibilities” (majority

op at 13). But if that is true, then the majority is being disingenuous in stating that it is

merely distinguishing and not overturning Knox. It is not possible to have it both ways.

Either all SORA courts are required to analyze whether a defendant’s crime involved a

sexual component or motive before requiring sex offender registration as a matter of

constitutional due process (in which case Knox is truly dead law), or the presence or

absence of such findings is irrelevant on the theory of Knox (i.e., that the legislature had a

rational basis on which to define “sex offender” to encompass these crimes, even if in doing

so it captured a small number of persons who do not fit within the colloquial definition of

6

As previously noted, SORA courts do have the statutory obligation to determine a

defendant’s risk-classification level. However, that determination depends on a weighing

of specific, carefully developed statutory factors (see Correction Law § 168-l [5]). The

findings here go well beyond application of those factors.

-9-

- 10 - No. 83

that term). Today’s decision creates rather than alleviates irrationality in the sex offender

registration scheme.

II.

Any analysis of whether to overturn precedent must be “guided, first and foremost,

by the principle of stare decisis” (People v Taylor, 9 NY3d 129, 148 [2007]). The doctrine

of stare decisis provides that once a court has decided a legal issue, subsequent appeals

presenting similar facts should be decided in conformity with that decision (see People v

Peque, 22 NY3d 168, 194 [2013]). The doctrine “promotes the evenhanded, predictable,

and consistent development of legal principles, fosters reliance on judicial decisions, and

contributes to the actual and perceived integrity of the judicial process” (Taylor, 9 NY3d

at 148 [internal quotation marks omitted]), through its assurance to the public that “a court

is an institution, not merely a collection of individuals, and that governing rules of law do

not change merely because the personnel of the court changes” (People v Garvin, 30 NY3d

174, 185 [2017] [internal quotation marks omitted]; see 1 William Blackstone,

Commentaries on the Laws of England 69 [1765] [stare decisis “keep(s) the scale of justice

even and steady, and not liable to waver with every new judge’s opinion”]).

“Even under the most flexible version of the doctrine of stare decisis, prior decisions

should not be overruled unless a compelling justification exists for such a drastic step”

(Grady v Chenango Val. Cent. Sch. Dist., 40 NY3d 89 [2023] [internal quotation marks

and alterations omitted]). This means that “the Court must have a good reason to do so

over and above the belief that the precedent was wrongly decided. It is not alone sufficient

- 10 -

- 11 - No. 83

that we would decide a case differently now than we did then” (Dobbs v Jackson Women’s

Health Organization, ___US___ , ___, 142 S Ct 2228, 2334 [2022] [Breyer, J., dissenting]

[internal quotation marks, citation, and brackets omitted]; accord Grady, 40 NY3d at 96

[“Our stare decisis doctrine does not permit overturning precedent merely because ‘it’s

time’”]). The doctrine generally requires “evidence that [the existing precedent] has

become unworkable, is unjust or has created an irreconcilable conflict in our case law”

(Matter of State Farm Mut. Auto. Ins. Co. v Fitzgerald, 25 NY3d 799, 820 [2015]

[emphasis added]).

Here, defendant did not preserve a challenge to Knox and only “invite[d]” us to

overturn it when directly asked at oral argument (see oral argument tr at 4, line 9).

Although “[i]nvitations to judicial reconsideration carry more weight when addressed to

constitutional issues” (Matter of Higby v Mahoney, 48 NY2d 15, 18 [1979]), it is improper

for judges to effectively invite themselves to revisit our cases and declare duly enacted

laws null and void. Moreover, even when properly invited to overturn prior constitutional

precedent, the “‘vitality of . . . constitutional principles cannot be allowed to yield simply

because of disagreement with them’” (Thornburgh v American College of Obstetricians

and Gynecologists, 476 US 747, 759 [1986], quoting Brown v Board of Education, 349 US

294, 300 [1955]).

Because defendant did not preserve any argument that Knox should be overturned,

the record unsurprisingly fails to show that our decision “has become unworkable, is unjust

or has created an irreconcilable conflict in our case law” (see Fitzgerald, 25 NY3d at 819-

- 11 -

- 12 - No. 83

820). Nor has Judge Rivera succeeded in making that showing on defendant’s behalf (see

op of Rivera, J., parts III-IV). Although her criticisms are worthy of debate, virtually all

of them were considered and rejected in Knox, and I question whether her analysis

faithfully applies the presumption of constitutionality and correctly allocates the burden of

proof (see op of Rivera, J., at 20-21 [criticizing certain gaps in the studies cited in Knox,

ignoring that it is a defendant’s burden to prove irrationality]).

Knox is plainly “workable”: it has been applied for 13 years in this State without

complication or issue. Judge Rivera asserts that “the stigma attached to the sex offender

designation recognized by the Court in Knox has increased in severity and impact” (op of

Rivera, J., at 16), but offers no significant factual basis or support for that assertion. She

correctly observes that Knox was an outlier among state courts when it was decided, and

that the U.S. District Court for the Southern District of New York has expressed

disagreement with Knox in two recent opinions (id. at 24-25; see Yunus v Robinson, 2019

WL 168544, 2019 US Dist LEXIS 5654 [SD NY, Jan. 11, 2019, No. 17-cv-5839 (AJN)];

Pennington v Rosado, 2022 WL 3127062, 2022 US Dist LEXIS 143233 [SD NY, May 31,

2022, No. 21-cv-7322 (GHW)]). However, the modern trend seems to be in Knox’s

direction.7 Further, most states—not just New York—require persons convicted of

7

Compare State v Robinson, 873 So2d 1205 (Fla 2004), American Civ. Liberties Union of

N.M. v City of Albuquerque, 2006-NMCA-078, 137 P3d 1215 (2006), and State v Reine,

2007 WL 77174, 2003 Ohio App. LEXIS 52 (2003), cause dismissed 99 Ohio St 3d 1549,

795 NE2d 686 (2003) (all finding constitutional violations prior to Knox, and all considered

in Knox), with Doe v Sex Offender Registry Bd., 488 Mass 15, 690 SE2d 827 (2021); State

v Coleman, 241 Ariz 190, 385 P3d 420 (Ariz Ct App 2016); Moffitt v Commonwealth, 360

SW3d 247 (Ky Ct App 2012); Rainer v State, 286 Ga 675, 690 SE2d 827 (2010); State v

- 12 -

- 13 - No. 83

kidnapping or imprisoning other people’s children to register as sex offenders, even if there

is no sexual purpose motivating the crime (see Smith, 323 Wis 2d 377, 385, 780 NW2d 90,

94 [2010] [observing that at least 41 other states and the District of Columbia “require

individuals convicted of false imprisonment or kidnapping of a minor to register as sex

offenders—regardless of whether the crime was of a sexual nature”]). The federal Sex

Offender Registration and Notification Act, which was part of the legislation that replaced

the Jacob Wetterling Act (see majority op at 7 & n 4), similarly characterizes persons in

defendant’s position as sex offenders (see 34 USC § 20911 [1], [5] [A] [ii], [7] [B]).

Although I share Judge Rivera’s concern that the two federal cases referenced in her

decision are creating inequities, they do not “provide a reason to reconsider and overturn

Knox” (see op of Rivera, J., at 26)—the two cases both involved the same kidnapping, and

neither binds this Court.

In the end, Judge Rivera has made a good case for the Second Circuit, the U.S.

Supreme Court,8 or even this Court to address this issue on a fully developed record in an

appropriate future case, but not for this Court to abandon its precedent on this record.

Regrettably, a true test of Knox’s reasoning will now never occur because that decision is

effectively no longer good law: by virtue of today’s decision, the “hard and fast rule, with

no exceptions” we approved in Knox no longer exists (see 12 NY3d at 69). The current

Smith, 323 Wis 2d 377, 780 NW2d 90 (2010) (all rejecting constitutional challenges post-

Knox). Judge Rivera’s attempts to distinguish the post-Knox cases have no more merit than

her attempt to distinguish Knox.

8

Notably, the U.S. Supreme Court denied certiorari in Knox (558 US 1011 [2009]).

- 13 -

- 14 - No. 83

majority has at once knee-capped and provided future litigants with a sword to slay our 14-

year-old, unanimous prior decision.

III.

The majority’s desire to protect the due process rights of criminal defendants is

commendable. I only wish it showed the same respect for the process by which we overturn

precedent. For all of the foregoing reasons, I would affirm the Appellate Division order.

Order reversed, with costs, and defendant's Sex Offender Registration Act designation

vacated. Opinion by Judge Rivera. Chief Judge Wilson concurs. Judges Troutman and

Halligan concur in result for reasons stated in sections I, II and V of the opinion. Judge

Cannataro dissents and votes to affirm in an opinion, in which Judges Garcia and Singas

concur.

Decided November 21, 2023

- 14 -