I am an attorney in the state of New York and currently pursuing a Ph.D. in politics at
Princeton University. I primarily research constitutional law and theory. I am also a
member of the subcommittee on sentencing for Connecticut’s Special Committee on Sex
Offenders, and a member of Connecticut for One Standard of Justice (CTOSJ), an
organization that works to advance offenders’ rights.
I have no personal stake in the policy proposals before this Commission. I have no family or
close friends who are sex offenders. I have, however, met dozens of registered citizens
convicted of non-violent sex crimes in Connecticut, and their family members, through my
work with CTOSJ. From these citizens I have repeatedly seen decency, politeness, hard
work, and determination to rebuild their lives. No doubt these men are but a fraction of
their ilk among those registered in Connecticut. I firmly believe that none of these offenders
deserves the lifelong branding that registration entails. The only stake I have, therefore, is
one in justice and fairness.
With that said, I turn to the main argument of this testimony. The Commission has
proposed dividing offenders into risk-based tiers, placing only those with high risk on the
publicly accessible registry. This makes good policy sense. However, some have suggested
making this change prospective only, leaving the approximately 5,700 offenders now on the
publicly accessible registry there—in some cases, for their entire lives. This makes no
policy sense, for at least five reasons.
First, all of the principled reasons for reducing the size of the public registry in the first
place apply with equal force to future and past offenders. We have all heard these reasons,
but they’re worth reiterating: existing studies demonstrate that publicly accessible
registries do not reduce (already very low1) sexual recidivism2; registries may, if anything,
increase recidivism slightly3; registration imposes burdens on low-level, first-time
offenders that are disproportionate to their crimes4; and offenders’ families, including their
1 “Recidivism among sex offenders in Connecticut,” Office of Policy and Management (2012) (reporting a
re-incarceration rate of sex offenders in Connecticut (for sex crimes) of 1.7%, and a re-conviction rate of
2 See, e.g., Richard Tewksbury & Wesley G. Jennings, Assessing the Impact of Sex Offender Registration and
Community Notification on Sex-Offending Trajectories, 37 CRIM. JUSTICE & BEHAVIOR 570, 579 (2010)
(concluding that [the set of state sex offender registries stemming from the federal Sex Offender
Registration and Notification Act] “has not reduced the rate of sex offender recidivism, nor has it led to a
decrease in the number of offenses committed by sex offenders”); Washington State Institute for Public
Policy, “Does Sex Offender Registration and Notification Reduce Crime? A Systematic Review of the
Research” (2009) (reviewing seven studies on the subject and concluding, because of varying results, that
notification laws “have no statistically significant effect on recidivism,” though they might deter crimes by
3 See J.J. Prescott & Jonah E. Rockoff, Do Sex Offender registration and Notification Laws Affect Criminal
Behavior? 54 J. L. & ECON. 192 (2011) (finding that average-to-large-size public registries increase the
number of sex offenses by more than 1.5%).
4 See, e.g., Richard Tewksbury, Collateral Consequences of Sex Offender Registration, 21 J. CONTEMP. CRIM.
JUSTICE 67, 75 (2005) (recording that roughly 42.7% of offenders surveyed reported losing a job due to
their registration; 23.1% a denial of a promotion; and 45.3% a loss or denial of housing—and that these
numbers are not much better where the offender did not have a child victim); Jill S. Levenson & Leo P.
Cotter, The Effect of Megan’s Law on Sex Offender Reintegration, 21 J. CONTEMP. CRIM. JUSTICE 49, 58 (2005)
(recording, from a survey of Florida offenders, that 27% reported losing a job because coworkers found
children, suffer as a byproduct. To heed these reasons in the case of future but not past
offenders is to draw arbitrary distinctions.
Second, without retroactivity, the proposed changes will increase burdens on lower-risk
offenders who remain on the registry. Once the change is announced, members of the
public will likely come to believe, in greater numbers than they already do, that everyone
on the registry is highly dangerous. (Subtleties like non-retroactivity are often lost through
the imprecise medium of political communication.) Facing even stronger assumptions that
they are predators, lower-level offenders will likely struggle more to find jobs and housing;
be threatened or harassed more often; and become more stigmatized and isolated than they
already are. This could stymie (already difficult) efforts toward rehabilitation; it is
extremely hard to improve personally when the public believes that you are worse than
your worst deed.
Third, if the registry at least makes some people feel safer, this effect would be diluted by
the inclusion of lower-risk offenders. With both low- and high-risk individuals on the
registry and minimal information included about them, it can be hard to assess whether the
offender who lives in your neighborhood poses a genuine threat: a person convicted of
“sexual assault” may be a violent rapist, or he may have had sex with his teenage girlfriend
when he was nineteen; a person convicted of “kidnapping,” with a child victim, may have
kidnapped and molested a young stranger, or he may have kidnapped his own child during
a custody dispute.5 If the state excludes all but high-risk offenders, it will permit the public
to focus their attention on the few true predators on the registry. If, by contrast, the
Commission recommends retaining all 5,700 offenders on the registry, Connecticut citizens
may live in ill-informed, unnecessary fear for decades to come.
Fourth, the state gains no fiscal advantage from making the registry reforms prospective
only. In fact, because of the expense of registry administration, Connecticut would
potentially save millions of dollars if it removed past lower-risk offenders from its
registers.6 If the state pared down the total number of registrants, it could use the saved
funds to finance other changes proposed by this Commission, such as the much-needed
registry removal mechanism, which will require court hearings and possible state-funded
legal defense. Some savings might even be redirected toward actually effective means of sex
Fifth, offenders who were convicted of sex crimes prior to the establishment of the
Connecticut sex offender registry in 1998, and required to register retroactively, have even
out about registration, 35% had had to move from a home or apartment because neighbors found out;
and that 33% of offenders reported being threatened or harassed by neighbors because of their offender
status; 5% reported being physically assaulted or injured for the same reason; and 19% reported that a
person who lived with them had been threatened, harassed, or injured because of the offender’s status).
5 Conn. Gen. Stat. §§ 53a-71 (second-degree assault), 53a-94 (second-degree kidnapping).
6 The exact budget for the sex offender registry is not published anywhere that I have found. However, an
estimate of several million is reasonable. The Justice Policy Institute projected in 2009 that it would cost
Connecticut $5.6 million to comply with federal sex offender registration requirements. While
Connecticut is not fully compliant today, its registry includes more people than it did in 2009. “What will
it cost states to comply with the Sex Offender Registration and Notification Act?” Justice Policy Institute
(2008), available at http://www.justicepolicy.org/images/upload/08-08_FAC_SORNACosts_JJ.pdf.
stronger claims of justice to be exempted from the registry than future offenders. These
individuals were not told at sentencing that they would later be subject to registration.
Indeed, there is a real concern that some of them—including potential innocents—might
not have taken a plea deal if the sentence offered them had explicitly included a decade or
lifespan of public shaming. Moreover, the retroactivity of registries has recently come
under constitutional doubt in one federal jurisdiction, and such challenges are common7; in
case the Second Circuit follows suit, removal of these retroactively included offenders from
the registry now, at the state’s pace, would spare the state future hassles.
Let me close with a constitutional observation. Under our Constitution, any penalty
imposed on individuals is classified in one of two ways: it is either a criminal punishment,
or a civil regulation.8 Each of these classifications has its own limitations. Punishment is
subject to the constitutional dictates that it be neither “cruel and unusual” (Eighth
Amendment) nor retroactive (Ex Post Facto Clause)9; civil regulations, while exempt from
these constitutional provisions, must be justified by non-punitive considerations, such as
public healthy and safety.10 According to the U.S. Supreme Court, sex offender registration
is a civil regulation, and it therefore technically can be disproportionate and retroactive.11
But its justification must still be non-punitive.
If this Commission recommends a high-risk-only registry, but applies the change only
prospectively, it will lose any claim to be motivated by concerns of public health and safety.
There is no reason based in public safety for distinguishing among lower-risk offenders
based on when they committed their crimes.12 The only apparent alternative motives are
retributive—that is, quintessentially punitive. If these are the motivations underlying this
policy, I urge the Commission (and, ultimately, the General Assembly) to frankly
acknowledge them so that registration may be constitutionally reviewed as punishment.
The state of Connecticut cannot in good conscience both enact retributive sanctions against
offenders and expect to remain immune from constitutional rights review. To do so would
be a grave transgression of our constitutional order.
This Commission is charged with the awe-some task of reviewing all of the sex offense laws
in this state. For its recommendations to be legitimate, they must be founded on reason, on
evidence, and on compassion for all persons. I can confidently say that a recommendation
to make a risk-based registry prospective only would be unprincipled, ungrounded in
evidence, and frankly arbitrary; given the registry’s expense, it would indeed amount to
paying to be arbitrary. When the toll on human life is as severe as it is in this case,
arbitrariness is inexcusable.
7 Does #1-5 v. Snyder, 834 F.3d 696 (6th Cir. 2016).
8 U.S. v. Ward, 448 U.S. 242, 248 (1980).
9 U.S. CONST. amend. VIII; U.S. CONST. art. I, §9.
10 Kansas v. Hendricks, 521 U.S. 346, 361 (1996).
11 Smith v. Doe, 538 U.S. 84, 96 (2003).
12 There may also be no reason based in public safety for distinguishing between high- and lower-risk
individuals; I have not found a study that reports different impacts on recidivism for these different
groups. However, any argument that there is a public safety reason for distinguishing among lower-risk
offenders based on the dates of their crimes would be uniquely implausible.
--Erin L. Miller
January 30, 2017