Delivered: September 18, 2002.
On Appellant's Petition for Discretionary Review from the 2nd Court of Appeals Wichita County.
Anthony C. Odiorne, Wichita Falls, for appellant.
John W. Brasher, Assistant District Attorney, Wichita Falls, Matthew Paul, State's Attorney, Austin, for state.
We are called upon in this appeal to determine whether the retrospective application of Texas's sex offender registration statute to appellant violates the Ex Post Facto Clauses of the United States and Texas Constitutions. U.S. Const. art. I, § 10; Tex. Const. art. I, § 16. The Second Court of Appeals held that the retrospective application of the statute to appellant did not constitute an ex post facto violation. Rodriguez v. State, 45 S.W.3d 685 (Tex.App.-Fort Worth 2001). We granted appellant's petition for discretionary review to consider whether the Second Court of Appeals' conclusion with respect to appellant's ex post facto claim was correct. For the reasons set forth below, we will affirm.
Because we evaluate these claims under the same legal standard, we refer to these clauses in tandem as the "Ex Post Facto Clause" unless otherwise noted.
Specifically, we granted the following ground for review:
Do the 1997 Amendments to the sex offender registration law violate the "Ex Post Facto" clause of the U.S. and Texas Constitutions by requiring persons convicted prior to the amendments' effective date to register as a sex offender for life?
I. Factual and Procedural Background A. Proceedings Below
On January 26, 1987, appellant was convicted of aggravated sexual assault with a deadly weapon finding and sentenced to seventeen years' confinement. On November 6, 1992, appellant was released on mandatory supervision and given a scheduled release date of March 18, 2003. Because appellant is a Mexican national, he was deported immediately upon release. In 1997, appellant illegally re-entered the United States and moved to Wichita Falls, Texas.
During this period of time, the Texas Legislature made a series of amendments to the sex offender registration and notification statute. Among other changes, the legislature made the registration and notification requirements of the statute applicable to all defendants under state supervision who had a reportable conviction occurring on or after September 1, 1970. Act of June 1, 1997, 75th Leg., R.S., ch. 668 §§ 1, 11, 1997 Tex. Gen. Laws 2260-61, 2264. Because appellant's offense fell within the definition of "reportable conviction," he was required to register with the local law enforcement agency effective September 1, 1997. Appellant failed to do so, and on June 30, 1999, he was arrested in Wichita Falls for failure to register as a sex offender. A jury found appellant guilty, and the trial court sentenced him to 540 days' confinement in a state jail.
"Notification" is used here to refer only to the public information made available in the state's centralized database. It does not refer to newspaper publication, which was specifically exempted from retrospective application. Act of June 1, 1997, 75th Leg., R.S., ch. 668 § 1, 1997 Tex. Gen. Laws 2264 (amended 1999, 2001) (current version at Tex. Code Crim. Proc. Ann. art. 62.11) (Vernon Supp. 2002).
The record reflects that as a condition of mandatory supervision, appellant was required to comply with the sex-offender registration program. This fact relates to whether or not appellant violated the conditions of mandatory supervision that were imposed upon him, not to whether retroactive application of the statutory registration requirements violates the Ex Post Facto Clause.
Appellant then appealed his conviction to the Second Court of Appeals. On appeal, he argued, inter alia, that requiring him to register as a sex offender for life constituted an ex post facto violation. Rodriguez, 45 S.W.3d 685, 687. The court overruled the point of error, explaining that "because the registration requirement is remedial in nature, i.e., a statute enacted for the advancement of the public good, it does not impose `punishment' for constitutional purposes and is not susceptible to an ex post facto claim." Id. at 689 (citations omitted). It is this conclusion we are called upon to review.
B. Statutory Background
Texas's sex-offender registration statute was originally enacted in 1991. Act of June 15, 1991, 72nd Leg., R.S., ch. 572, Tex. Gen. Laws 2029-32 (codified at Tex. Rev. Civ. Stat. Ann. art. 6252-13c.1). Appellant challenges the amendments that were made to the law in 1997. In particular, appellant points to the expansion of the class of persons required to report as sex offenders. Prior to the amendments, the class of sex offenders with reportable convictions did not include any defendants who had been convicted prior to 1991. In 1997, however, the Legislature expanded the class to include all those who had a "reportable conviction or adjudication" since September 1, 1970, and who continued to be under some form of state supervision. Tex. Code Crim. Proc. Ann. art. 62.11; §§ 1, 11 1997 Tex. Gen. Laws 2260, 2264. Appellant also complains of the 1997 imposition of lifetime reporting requirements for several listed offenses, including aggravated sexual assault, the offense for which he was convicted. Finally, appellant notes that the legislature redesignated the statute, formerly included in the Texas Revised Civil Statutes, as Chapter 62 of the Texas Code of Criminal Procedure. § 1, 1997 Tex. Gen. Laws 2253.
Accordingly, we limit our discussion to the 1997 amendments of which appellant complains. Furthermore, although many provisions of Texas's sex-offender registration statute have remained unchanged since 1997, all references to the statute are to the 1997 version. We will cite to the Session Laws where appropriate, but all references to Chapter 62 of the Code of Criminal Procedure are to former Chapter 62, unless otherwise noted. Finally, all future references to "Articles" are to the Code of Criminal Procedure.
C. Relevant Law
An ex post facto law: 1) punishes as a crime an act previously committed which was innocent when done; 2) changes the punishment and inflicts a greater punishment than the law attached to a criminal offense when committed; or 3) deprives a person charged with a crime of any defense available at the time the act was committed. Collins v. Youngblood, 497 U.S. 37, 42-44 (1990); Ex parte Davis, 947 S.W.2d 216, 219-20 (Tex.Crim.App. 1996); Johnson v. State, 930 S.W.2d 589, 591 (Tex.Crim.App. 1996). Appellant asserts that requiring him to comply with the Texas sex-offender registration act inflicts greater, or different, punishment upon him than that attached to the offense at the time it was committed. We must therefore determine: 1) whether the statute is being applied retroactively; and 2) whether its provisions constitute punishment for constitutional purposes. Because the State concedes the statute's retrospective application, we direct our attention solely to the question of whether the 1997 amendments to Texas's sex-offender registration statute constitute punishment for constitutional purposes. This question can be answered by application of what is known as the "intent-effects" test. See Doe v. Otte, 259 F.3d 979, 985 (9th Cir. 2001), cert. granted, 534 U.S. at 1126, 122 S.Ct. 1062 151 L.E.2d 966 (2002).
Under the "intent-effects test," a reviewing court must first ask whether the legislature intended the statute to be a criminal punishment. "Whether a particular punishment is criminal or civil is, at least initially, a matter of statutory construction. A court must first ask whether the legislature, in establishing the penalizing mechanism, indicated either expressly or impliedly a preference for one label or the other." Hudson v. United States, 522 U.S. 93, 99 (1997) (citations and internal quotation marks omitted). A reviewing court must afford a high level of deference to the legislature's stated aims in passing the statute. Id. at 101 (cautioning that courts must consider effects of statute "in relation to the statute on its face" and that "only the clearest proof will suffice to override legislative intent and transform what has been denominated a civil remedy into a criminal penalty") (citations and internal quotation marks omitted).
Hudson involved a claim under the Double Jeopardy Clause of the U.S. Constitution. Hudson, 522 U.S. at 98. Nonetheless, because the thrust of the opinion was the proper procedure for evaluating whether sanctions are civil or criminal in nature, we think it irrelevant that the claim arose under the Double Jeopardy Clause. See also Femedeer v. Haun, 227 F.3d 1244 (10th Cir. 2000) (relying on Hudson to evaluate ex post facto challenge to Utah sex-offender registration statute).
If the legislature manifests an expressly punitive intent, the inquiry is at an end and the statute is a violation of the ex post facto clause. Kennedy v. Mendoza-Martinez, 372 U.S. 144, 169 (1963). If the legislature intends to establish a civil remedy, a reviewing court then examines "whether the statutory scheme [is] so punitive either in purpose or effect as to transform what was clearly intended as a civil remedy into a criminal penalty." Hudson, 522 U.S. at 99 (citations and internal quotation marks omitted). The legislature's manifest intent will be rejected only where the party challenging the statute provides "the clearest proof" that the statute is actually criminally punitive in operation. Hendricks, 521 U.S. 346, 361 (1997). The Supreme Court has stated:
[T]he mark of an ex post facto law is the imposition of what can fairly be designated punishment for past acts. The question in each case where unpleasant consequences are brought to bear upon an individual for prior conduct, is whether the legislative aim was to punish that individual for past activity, or whether the restriction of the individual comes about as a relevant incident to a regulation of a present situation. . . .
De Veau v. Braisted, 363 U.S. 144, 160 (1960).
To evaluate whether the effects of a statute are criminally punitive, courts generally look to the factors set forth by the Supreme Court in Kennedy, 372 U.S. at 169 (the " Kennedy factors"). Courts consider: (1) whether the sanction involves an affirmative disability or restraint; (2) whether it has traditionally been regarded as a punishment; (3) whether it comes into play only on a finding of scienter; (4) whether its operation will promote the traditional aims of punishment-retribution and deterrence; (5) whether the behavior to which it applies is already a crime; (6) whether an alternative purpose to which it may rationally be connected is assignable to it; and (7) whether it appears excessive in relation to the alternative purpose assigned. 372 U.S. at 168-69.
The factors in Kennedy v. Mendoza-Martinez have been alternatively dubbed " Kennedy factors" and " Mendoza-Martinez factors" by various courts. We refer to them as the " Kennedy factors."
Courts have taken various approaches with respect to these factors. The Supreme Court has never explicitly adopted the Kennedy factors for application in the ex post facto context. Nonetheless, the Court applied many of the factors in the context of the ex post facto challenge raised in Hendricks. In reviewing the double jeopardy claim raised in Hudson, the Court explained that the factors are "useful guideposts" but that none are dispositive. 522 U.S. at 99, 101. We will consider each of the useful guideposts the Supreme Court provided to us in Kennedy. Furthermore, because "our task is not simply to count the factors, but to weigh them," we will also discuss the varying weight to be afforded to each factor. See Arizona v. Noble, 829 P.2d 1217, 1224 (Ariz. 1992).
For instance, the Kansas Supreme Court concluded that the Kennedy factors apply in the punitive/nonpunitive effect analysis, but the court chose to apply only those factors that provided "significant guidance" and disregard those factors it determined added little or nothing to the discussion. State v. Myers, 923 P.2d 1024, 1040 (Kan. 1996). The Third Circuit has used a three-part test that it harmonized with relevant Supreme Court precedent in E.B. v. Verniero, 119 F.3d 1077, 1094-96 (3d Cir. 1997), cert. denied, 522 U.S. 1109 (1998). The Sixth Circuit has noted that the Supreme Court "although not expressly adopting the Kennedy factors, recently applied many of the same factors in deciding [the] ex post facto challenge in Hendricks." Cutshall v. Sundquist, 193 F.3d 466, 477 (6th Cir. 1999), cert. denied, 529 U.S. 1053 (2000).
II. Application A. Intent of the 1997 amendments
The State provides considerable support for the proposition that the sex offender registration statute was enacted to promote public safety. It quotes relevant portions of the legislative debates on the statute's initial enactment and subsequent amendment in 1995. It also points to the Fort Worth Court of Appeals' opinion in In re M.A.H., in which the court noted:
The State provides proof that in enacting the current registration and notification plan, the legislature considered the unique threat sex offenders present to public safety, the high rate of recidivism among sex offenders, the low incidence of rehabilitation among sex offenders, and that sexual misconduct often begins as a juvenile. The State provides further proof that the legislature's goal in passing the registration and notification provisions was to advance public safety objectives by facilitating law enforcement's monitoring of sex offenders and by alerting members of the public who may be in an especially vulnerable situation to take appropriate precautions which could deter or prevent further crimes.
20 S.W.3d 860, 863 (Tex.App.-Fort Worth 2000, no pet.).
With respect to the 1997 amendments, however, the State points only to the comments of the sponsoring senator, who indicated that the provisions were being moved into the Code of Criminal Procedure to facilitate compliance by trial judges and school administrators with respect to special notification and registration mandates. The State does not address the expansion of the class of defendants required to register or the extension of the period during which the offenders had a duty to register.
Neither the text of the bill nor the bill analysis provides an explicit statement of purpose. At the same time, however, the State is aided by this same absence of an express punitive intent: Whenever we are confronted with an attack upon the constitutionality of a statute, we presume that the statute is valid and that the Legislature has not acted unreasonably or arbitrarily. Ex parte Granviel, 561 S.W.2d 503, 511 (Tex.Crim.App. 1978). The burden rests upon the individual who challenges the statute to establish its unconstitutionality. Id. In the absence of contrary evidence, we will presume that the legislature acted in a constitutionally sound fashion. In this case, appellant cites no evidence of a punitive intent on the part of the Texas legislature with respect to the 1997 amendments. Thus, we presume that the legislature's intent was civil and remedial, and not criminal or punitive in relation to the claim as stated here as being an ex post facto violation. See also State v. Pickens, 558 N.W.2d 396, 398 (Iowa 1997) (reasoning that although court was not aided by an express statement of legislative intent, the state was "aided by the principle that statutes are presumed to be constitutional and a challenger has the burden to negate every reasonable basis upon which the statute may be sustained").
B. Effect of the 1997 amendments
Having determined that the intent of the 1997 amendments was civil and remedial, and not criminal or punitive, we must now consider whether the amendments are so punitive in purpose or effect as to transform what was clearly intended as a civil regulation into a criminal penalty. Hudson, 522 U.S. at 99 (citations omitted). As discussed above, in order to make this determination, we look to the "useful guideposts" set forth by the Supreme Court in Kennedy.
1. Affirmative disability or restraint
One of the more recent pronouncements on the meaning of this factor came in Hudson. There, the Supreme Court held that occupational disbarment and the imposition of monetary fines "do not involve an `affirmative disability or restraint,' as that term is normally understood. While petitioners have been prohibited from further participating in the banking industry, this is certainly nothing approaching the infamous punishment of imprisonment." 522 U.S. at 104 (citing Flemming v. Nestor, 363 U.S. 603 (1960)) (internal quotation marks partially omitted). In addition, as the Third Circuit has noted, the burden imposed by registration and notification "pales in comparison to the civil commitment of sex offenders sanctioned in Hendricks." Verniero, 119 F.3d at 1104.
Appellant asserts that the 1997 amendments imposed "a significant affirmative disability by subjecting offenders to onerous conditions that in some respects are similar to probation or supervised release." In support of this argument, he relies on Doe v. Otte, 259 F.3d 979, 987 (9th Cir. 2001), cert. granted, 534 U.S. at 1126, 122 S.Ct. 1062 151 L.E.2d 966 (2002). There, the Ninth Circuit struck down Alaska's sex offender registration statute as a violation of the Ex Post Facto Clause. With respect to this first Kennedy factor, the Ninth Circuit found an affirmative disability in the onerousness of the registration provisions as well as "community obloquy and scorn" that resulted from public notification. Id.
Appellant also relies on State v. Myers, 923 P.2d at 1041, in support of his contention that the public disclosure provisions of Texas's statute impose a disability on registrants. In Myers, the Kansas Supreme Court held that although there was no affirmative disclosure requirement in the Kansas sex-offender statute, the unlimited disclosure of registrant information upon request constituted an affirmative disability or restraint. Id. at 1041. "The information could be routinely published in the newspaper or otherwise voluntarily disseminated by anyone. The practical effect of such unrestricted dissemination could make it impossible for the offender to find housing or employment." Id. The State argues that the provisions of the Texas statute do not work an affirmative disability on registrants because registrants' movements within and without the state remain unrestricted.
The court held specifically that the registration requirement did not impose an affirmative disability or restraint because it did not restrict the offender's movements within or without the community. Id.
Although the question is closer than the State suggests, on balance, we agree that the 1997 amendments to the Texas statute do not constitute an affirmative disability or restraint. First of all, the facts of Otte were different because the statute at issue was different. In Otte, the Ninth Circuit placed particular weight on the onerousness of the registration requirements: under the Alaska statute, offenders such as the plaintiffs, who had been convicted of aggravated sex offenses, were required to re-register four times per year in person at the local law enforcement authority for the remainder of their lives. This re-registration required them to "appear in person at a police station on each occasion, and provide, under oath, a wide variety of personal information, including address, anticipated change of address, employer address, vehicle description, and information concerning mental health treatment for any `mental abnormality or personality disorder.'" Otte, 259 F.3d at 987. By contrast, unless an offender in Texas has been convicted two or more times of a sexually violent offense, he must only register once for each move or anticipated move; after this, he must verify his registration only once per year.
Offenders with two or more convictions for a sexually violent offense are required to verify registration every ninety days. Art. 62.06. That requirement is not before us because it does not apply to appellant.
Moreover, much of the Otte court's determination that the Alaska statute worked an affirmative disability was based on the "community obloquy and scorn" to which an offender would be subjected as a result of public disclosure. Id. at 987. Like the Texas statute, the Alaska statute at issue in Otte permitted sex-offender information to be made available to the public via the Internet. The information made available to the public under the Alaska statute included the offender's name, address, color photograph, physical description, conviction information and employer addresses. Id. at 984, 987. Thus, the court explained, "the procedures employed under the Alaska statute are likely to make the plaintiffs completely unemployable. Alaska publishes the names and addresses of the registrants' places of employment on its sex-offender internet site, and makes it simple for users of the site to search for the presence of any sex offenders working at a particular place of employment." Id. at 987. In Texas by contrast, several items are exempted from public disclosure. Art. 62.08. The address of the registrant's employer is not included within the required registration information. Art. 62.02. In this sense, the Texas statute is considerably narrower in scope than the Alaska statute in Otte.
Second, although we sympathize with the conclusion reached by the Kansas Supreme Court in Myers, that public availability of sex-offender registrant information might make it "practically impossible" for registrants to find work or housing, we think that those effects are not the relevant considerations under this factor. The question in this regard is whether the provisions of the statute, and not the speculative response of the community, work an affirmative disability or restraint. See, e.g., Femedeer, 227 F.3d at 1250 ("[N]otification does not by itself prohibit sex offenders from pursuing any vocation or avocation available to other members of the public, and we therefore conclude that this factor weighs against finding that the statute is punitive in purpose or effect"); see also Verniero, 119 F.3d at 1102 (noting that the concern of registrants is often the "indirect effects" of registration statutes).
In light of the foregoing, we conclude that although registration and notification impose a burden upon those required to register, it does not impose an affirmative disability or restraint as the term is commonly understood. The registration and verification requirements are not onerous in the manner of the Alaska statute at issue in Otte. Additionally, although registrants may face public embarrassment once their convictions are disclosed, much of this information is attributable to the convictions themselves, which in our criminal justice system, are public. "Dissemination of information about criminal activity has always held the potential for substantial negative consequences for those involved in that activity. Dissemination of such information in and of itself, however, has never been regarded as punishment when done in furtherance of a legitimate governmental interest." Verniero, 119 F.3d at 1099-1100.
2. Historical Treatment
Appellant argues that because sex-offender registration statutes are of fairly recent origin, they should bear little on our decision. The State argues that this factor militates in favor of a finding that the Texas statute is nonpunitive. For the most part, we think that appellant is correct. Nevertheless, in order to fully explore the statute's character, and to address the arguments made by the State, we think it best to discuss the factor but afford it relatively little weight.
The State cites to Dean v. State, 60 S.W.3d 217, 223-24 (Tex.App.-Houston [14th Dist.] 2001, no pet.), a case which in turn relied on the analysis of the Third Circuit in Verniero. In Verniero, the plaintiffs alleged that the disclosure provisions of Tennessee's sex offender registration statute branded them sex offenders, and that the resulting ignominy rendered the statutes "closely analogous to the well-recognized historical punishments of public shaming, humiliation and banishment as those practices were employed in colonial times." Verniero, 119 F.3d at 1099. The court rejected the analogy, reasoning that such shaming punishments involved more than the mere dissemination of information: the colonial practices either physically restrained or removed the persons being shamed. Id. Rather, the court stated that "far more compelling analogies" were to be found in "the required dissemination of information generated by our criminal justice system and the subsequent dissemination of `rap sheet' information to regulatory agencies, bar associations, prospective employers and interested members of the public." Id. at 1100. The court also likened the public disclosure provisions to quarantine notices and posters warning the community of armed and dangerous criminals.
We think that there is in truth no precise historical analogue to sex offender registration and notification statutes. On the one hand, many of the provisions bear similarities to those quasi-administrative adjuncts of a criminal sentence that have not traditionally been considered criminal sanctions: for example, the loss of voting privileges, license suspensions or revocations, as well as the aforementioned dissemination of rap sheet information. On the other hand, for no other type of offense except for death penalty cases does the general public have access to a centralized collection of information about the convictions and offenders. Even if such information could accurately be termed "public" in the sense of a public trial, the information is not assembled in a central repository for any offenses other than sex offenses and death penalty capital murders. See also Doe v. Poritz, 662 A.2d 367, 411 (N.J. 1995) (relying upon United States Department of Justice v. Reporters Comm. for Freedom of the Press, 489 U.S. 749, 763 n. 15 for the proposition that "a privacy interest is implicated when the government assembles those diverse pieces of information into a single package and disseminates that package to the public, thereby ensuring that a person cannot assume anonymity — in this case, preventing a person's criminal history from fading into obscurity and being wholly forgotten").
Similarly, although quarantine notices and law enforcement warning posters might cause those depicted therein some public embarrassment, there is not a one-to-one correspondence to the public disclosure provisions embodied in the Texas sex offender statute. If a poster warns of a quarantine or an armed and dangerous criminal, it is generally because there has been a determination that there is, in fact, a quarantine or an armed and dangerous criminal. In Texas, the registration and notification provisions become operational automatically upon conviction for certain offenses. Art. 62.11.
Owing to these statutes' relatively recent arrival and to the lack of a precise historical analogue, then, we weigh this factor lightly. Nevertheless, to the extent this factor is weighed, we weigh it in favor of a finding that the statute is nonpunitive. Any consideration of the Kennedy factors must always be carried out in light of the statute on its face. Kennedy, 372 U.S. at 169. Because the legislature intended to create a remedial, regulatory scheme to monitor the whereabouts of sex offenders, and to warn community members of their presence, we characterize this statutory scheme as we characterize other "regulatory disabilities" that accompany certain convictions: as nonpunitive in effect. Doe v. Department of Public Safety ex rel. Lee, 271 F.3d 38, 61 (citation omitted), cert. granted sub nom. Connecticut Dep't of Pub. Safety v. Doe, ___ U.S. ___, 122 S.Ct. 1959, 152 L.Ed.2d 1020 (2002).
3. Finding of scienter
Appellant argues that the statute does not become operative unless a defendant is convicted of one of the listed offenses, and because the offenses require a culpable mental state, the statute is criminal in nature. The State relies on Dean for the proposition that because the provisions become operative upon release of a sex offender into the community, no culpable mental state is required. We agree with the State. Although a culpable mental state may be required with respect to some of the underlying offenses, this does not answer the question of whether the registration statute requires a culpable mental state. As noted above, the statute becomes operational automatically upon an offender's release. Therefore, the statute does not require a culpable mental state with respect to the registration and notification provisions, and we weigh this factor in favor of a finding that the statute is nonpunitive. Cf. Hudson, 522 U.S. at 104 (no finding of scienter required where provisions allowed for assessment of a penalty against any person "who violates" any of the underlying banking statutes).
4. Traditional aims of punishment
When a statute promotes the traditional aims of punishment, retribution and deterrence, its effect is more likely to be considered punitive. Hudson, 522 U.S. at 104. The existence of a deterrent effect alone, however, will generally be insufficient to transform a civil sanction into a criminal one. Id. at 105. We must always assess this factor with an eye to the statute's stated aims. Where the legislature has not manifested an intent to promote the traditional aims of punishment, it behooves us to refrain from searching for one in the statute's indirect effects. See Kennedy, 372 U.S. at 169.
Appellant relies on Otte and Myers and argues that the statute promotes deterrence. According to him, the threat of registration and disclosure could prevent people who might otherwise have become offenders from offending. He relies on the same cases in support of his assertion that the statute is also retributive in nature. First, he analogizes the registration requirements to the requirements of supervised release, which are "part of the punishment meted out through a defendant's criminal sentence." Next, he complains that because public disclosure is not limited to that necessary to protect the public, the statute exceeds its stated nonpunitive aims and becomes retributive in nature. The State responds by asserting that a deterrent effect, alone, does not render the entire act punitive in nature. In this regard, it again relies upon Dean, 60 S.W.3d at 224.
We disagree with appellant that the deterrent effect of the statute is that persons who might otherwise have become offenders will refrain from offending. While we think that there is a marginal possibility this would occur, this possibility is far removed from the true operation of the sex offender registration statutes. The statutes were enacted amid public outcry that sex offenders were particularly likely to re-offend and that the communities in which sex offenders lived were entitled to be warned of the offenders' presence in order to protect themselves from future harm. The deterrent force of these statutes is claimed to lie in the fact that law enforcement authorities will remain informed of offenders' whereabouts and the fact that members of the general public will be able to take precautions to prevent re-offense by offenders living in their communities. See Cutshall, 199 F.3d at 475.
See Verniero, 119 F.3d at 1081-82 (detailing events leading up to enactment of original "Megan's Law" for disclosure of ex-offender whereabouts). For an exposition of the history of sex offender laws nationwide and in Texas in particular, see Children in the War on Crime, supra n. 9.
We agree with appellant that there is some measure of retribution incident to the statute. As with the Alaska statute considered in Otte, the Texas statute links the duration of the registration requirements to the extent of an offender's wrongdoing, rather than an assessment of the future dangers the offender poses to the community. Art. 62.12. We disagree with appellant, however, that the registration requirements are so similar to supervised release that they become part of the punishment meted out through a defendant's criminal sentence. A criminal sentence is bounded by certain guidelines, but is generally an individually tailored response to the type and extent of wrongdoing in a given case. Sex-offender registration is in operation far more regulatory in nature: its provisions become and remain effective automatically upon conviction for certain offenses. Although law enforcement agencies collect and monitor the data, the individual circumstances of such collecting and monitoring do not vary among offenders. Only the length of the duty to register varies, and this variation is, in turn, prescribed by statute.
Overall, the retributive and deterrent effects of Texas's sex-offender registration statute are incidental, and not primary, to the statute's operation. Therefore, we also weigh this factor in favor of a nonpunitive intent.
5. Application to behavior that is already a crime
A statute that applies to behavior that is already a crime is more likely to be characterized as a criminal sanction. See Kennedy, 372 U.S. at 168. The State concedes that this factor is indicative of a punitive effect. As noted by appellant, the statute applies only to defendants who have "reportable convictions." Other states that have found their sex-offender registration statutes nonpunitive in this regard have noted that the statutes apply not only to defendants, but also those persons who are civilly committed or adjudicated not guilty on the ground of mental incapacity. See, e.g., Otte, 259 F.3d at 991. Presumably, if Texas's statute were entirely nonpunitive in character, it would apply to all those from whom the community would desire protection, including, and perhaps especially, those who are determined to be mentally unfit even to stand trial.
At the same time, however, the Supreme Court has noted that this factor alone is insufficient to render sanctions criminally punitive. Hudson, 522 U.S. at 105. That registration is triggered by a conviction is a characteristic "common to all regulatory disabilities that follow from a prior conviction, such as the loss of the right to vote." Lee, 271 F.3d at 61 (citation omitted). Therefore, although this factor is indicative of a punitive intent, it is not especially crucial in ferreting out the true character of the sanction in question.
6. Alternative Purpose
Under this factor, we inquire whether there is an alternative, nonpunitive purpose that may rationally be connected to the statute. Kennedy, 372 U.S. at 168-69. Appellant concedes that the Texas statute serves the nonpunitive purpose of promoting public safety by alerting the public to the presence of sex offenders. He suggests, however, that this is not the primary goal of the statute because if public safety were the primary goal, it would also be appropriate to enact registration requirements for those convicted of offenses such as theft or murder. The State does not directly respond to the suggestion that public safety may not be the primary goal of the statute. Rather, it again relies on Dean and asserts that the act serves the dual purposes of protecting public safety and providing information to law enforcement authorities.
The burden of demonstrating that there is no rational connection between the statute and the alternative purpose assigned is a high one. The Third Circuit's analysis in this regard is instructive:
Nothing in [the relevant case law] requires a perfect fit between end and means. An absence of remedial, objective purpose is not demonstrated by pointing out that the legislature did not address what might be perceived as another aspect of the same problem or that there may be a means of serving the legislative end that would be more effective than the means chosen. If a reasonable legislator motivated solely by the declared remedial goals could have believed the means chosen were justified by those goals, then an objective observer would have no basis for perceiving a punitive purpose in the adoption of those means.
Verniero, 119 F.3d at 1093. We owe the same deference to the Texas statute. A reasonable legislator could have believed that the assembly and maintenance of a database to track the whereabouts of sex offenders convicted since 1970 bore a rational connection to the promotion of public safety. This is so even if the legislature did not impose similar requirements on other types of offenders. See id. Therefore, we conclude that this factor is, on balance, indicative of a nonpunitive effect.
We must now determine whether the 1997 amendments to Texas's sex offender registration statute are excessive in relation to the alternative purpose assigned, the promotion of public safety. In the context of a double jeopardy claim, the Supreme Court made it clear that this factor should not be elevated to dispositive status. Hudson, 522 U.S. at 101. Nevertheless, of all the Kennedy factors, this factor cuts most directly to the question of which statutes cross the boundaries of civil sanctions, and which do not. See, e.g. Otte, 259 F.3d at 991 ("[t]he final and, in this case, a highly significant factor in the Mendoza-Martinez analysis is whether the [act] appears excessive in relation to the alternative purpose assigned") (citation omitted); Myers, 923 P.2d at 1041 (excessiveness "is the key factor in our analysis"). Accordingly, we afford this factor considerable weight in deciding whether the amendments are punitive-in-fact.
Appellant's argument in this regard is that the statute is excessive because neither registration nor disclosure is tied to a finding that the safety of the public is threatened. In support of this, he points to the decisions of other jurisdictions that have rejected ex post facto challenges to sex-offender registration statutes. All of those jurisdictions, he alleges, specifically noted that disclosure was limited to that necessary to public safety, and/or that an individualized finding of future dangerousness was made. He notes that the Texas legislature provided a limited exemption from registration for juveniles, and he argues that such a provision should be extended to adult sex offenders.
Until 1999, similar exemptions were available for adult sex offenders convicted of sexually violent offenses. Art. 62.11 (c), repealed by Act of June 19, 1999, 76th Leg., R.S., ch. 1415 § 25, 1999 Tex. Gen. Laws 4831, 4841 (effective September 1, 1999).
According to the State, the act is not excessive because disclosure under the statute is limited. The State bases this argument on Dean, in which the Fourteenth Court of Appeals determined that the 1999 amendments to the sex-offender registration statute did not offend the Ex Post Facto Clause. Because Dean approved of certain newspaper publication provisions that the State characterizes as less restrictive than the public disclosure provisions of the 1997 version of the statute, the State reasons that the Dean court holding should be extended to the "less restrictive" 1997 version. The State also alleges that the act is not excessive because it limits the lifetime reporting requirements to those defendants convicted of certain offenses.
We agree with appellant that the cases in which ex post facto challenges to sex-offender registration statutes have been rejected do not precisely correspond to the issues raised by this appeal. With one exception that is discussed below, all of the federal circuit courts that have specifically addressed this Kennedy factor have considered statutes that do not provide for unlimited public disclosure, and/or that provide for an individualized determination of future risk. See, e.g., Doe v. Pataki, 120 F.3d 1263, 1281-83 (2nd Cir. 1997) (upholding New York sex-offender statute based on its tiered structure, which tied the harshness of the registration requirements to an individualized assessment of the risk that each offender posed to the community. Significantly, despite the outcome the court reached, it noted that the question of the statute's punitive-in-fact aspect was "not free from doubt." Id. at 1265), cert. denied, 522 U.S. 1122 (1998); Verniero, 119 F.3d at 1082-86) (discussed below); Moore v. Avoyelles Correction Center, 253 F.3d 870, 872-73 (5th Cir. 2001) (upholding disclosure provisions of Louisiana sex-offender statute, which provides for neighborhood notification, and not unlimited public disclosure), cert. denied sub nom. Moore v. Kaylo, 534 U.S. 1008, 122 S.Ct. 492, 151 L.Ed.2d 404 (2001); Cutshall v. Sundquist, 193 F.3d at 476-77 (rejecting ex post facto challenge to Tennessee sex-offender registration statute, which provided that the Tennessee Bureau of Investigation "or the local law enforcement agency may release relevant information deemed necessary to protect the public concerning a specific sexual offender who is required to register." Tenn. Code § 40-39-106 (1994)). See also Otte, 259 F.3d at 988-89, 992 (explaining that although prior case law of the circuit had approved of retroactive application of sex-offender statutes, the decisions were based both on the limited nature of disclosure and on less onerous registration requirements than those imposed by the Alaska statute at issue in Otte. The court in Otte also specifically noted that similar statutes approved of by federal courts of appeals have "tailored the provisions of the statute to the risk posed by the offender." Id. at 992).
A comparison of statutory provisions is also instructive in understanding why the case law cited in Dean is distinguishable from the 1997 amendments that are before us today. The operation of the Texas sex-offender registration statute that became effective in September 1997 was, in relevant part, as follows: the statute's requirements applied only to those under state supervision who had a reportable conviction or adjudication occurring on or after September 1, 1970. Art. 62.11. A person required to register was required to do so for either a ten-year period or for the duration of the person's lifetime. Lifetime reporting requirements were imposed on defendants convicted of one of the following offenses: indecency with a child by sexual contact; sexual assault; aggravated sexual assault; sexual performance by a child; aggravated kidnapping with intent to violate or abuse the victim sexually; first-degree-felony burglary with intent to commit indecency with a child, sexual assault, aggravated sexual assault or prohibited sexual conduct; first-degree-felony burglary with intent to commit aggravated kidnapping with intent to violate or abuse; a similar offense in another state; prohibited sexual conduct; compelling prostitution by a minor; or possession or promotion of child pornography. Art. 62.12 (a), (b). However, there was also a provision under which offenders convicted of "sexually violent offenses" could seek an exemption from the lifetime reporting requirements. Art. 62.12 (c), repealed by § 25, 1999 Tex. Gen. Laws 3841. Finally, although the statute provided for a risk assessment review committee, all of the public information about the offenders is published in the centralized database maintained on the Internet, without regard to the particular risk level assessed. Art. 62.08.
Compare these provisions to the New Jersey statute approved of by the Third Circuit in Verniero. Under that law, the original "Megan's Law," there are three separate tiers that correspond to an offender's risk level. Tiers II and III were challenged on ex post facto grounds in Verniero. Only one of the tiers, Tier III, provided for general community notification. Verniero, 119 F.3d at 1083. For those offenders placed in Tier II, notification was to "law enforcement, school and community organizations." Id. The statute itself specified certain considerations that must factor into the attorney general's required risk assessment guidelines. Id. Those guidelines, in turn, play into the "Registrant Risk Assessment Scale" that is used by prosecutors in placing offenders into the various tiers. Id. Prosecutors must also consider the applicability of certain exceptions. Into these provisions the New Jersey courts also read several additional safeguards: tailoring of the notification to those parties who might be at risk, written notice to the offender specifying his right to counsel, and a prenotification judicial review process for those who wish to contest their status. Id. 1086.
To cite such a case for the proposition that Texas's statute provides for limited disclosure is, to say the least, problematic. Indeed, the only case that addressed a statute similar to Texas's in unlimited disclosure and lack of individualized assessment is Femedeer, 227 F.3d 1244 (10th Circ. 2000). In Femedeer, the district court granted partial summary judgment in favor of plaintiffs who had completed their criminal sentences and were subsequently required to comply with New Jersey's registration requirements. Id. at 1246. The court granted partial summary judgment on the plaintiffs' ex post facto and double jeopardy claims, but only the former is relevant to our discussion. In considering whether the Utah scheme at issue was excessive in light of its nonpunitive aims, the Tenth Circuit criticized the district court's reliance on the fact that unlimited disclosure would disseminate sex-offender information to many people who would not be at risk of encountering an offender:
The district court failed to consider, however, that there will be a concomitant decline in the global audience's desire to receive Utah's registry information. That is, the farther removed one is from a sex offender's community and from Utah generally, the less likely one will be to have an interest in accessing this particular registry. To a large extent, therefore, the dangers resulting from widespread availability are negated by the decreasing likelihood that the information will actually be obtained.
227 F.3d at 1253. However, this analysis begs the very question of vigilante justice and statutory excesses. If sex offender registration statutes are enacted and upheld in the name of public safety, then there must be at least some correspondence between the burden placed upon registrants and the nonpunitive aims of the statute. If dissemination of the information routinely exceeds that which is necessary, or even helpful, in protecting the public, the link weakens considerably. "If public availability of information about the plaintiff serves no remedial purpose, that availability is unnecessary to provide protection for those whom the act was designed to protect." Doe v. Attorney General, 686 N.E.2d 1007, 1012 (Mass. 1997). To argue that parties from other states might have no interest in a state's registry is to ignore the true dangers that registrants face: bearing the brunt of the statute's negative effects even among those whose safety is not endangered.
That such excess occurs routinely seems relatively likely. In Connecticut, for instance, the sex offender website received over three million hits in its first five months of operation. Lee, 271 F.3d at 44, n. 14. According to the 2000 Census, the population of the entire state was 3,405,565. See www.state.ct.us/ecd/research/census2000/2000%20Census%20Profile_ DP-1.xls.
Nevertheless, any excessiveness in Texas's statute does not rise to the level of that in the cases upon which appellant relies. In Myers, the Kansas Supreme Court upheld the registration requirements of its sex-offender statute, but considered the public disclosure provisions separately and determined that they constituted a violation of the Ex Post Facto clause. 923 P.2d at 1043. There are, however, several important differences between the Kansas and Texas statutes. The Kansas statute provides specifically that none of the registrants' information is confidential and that all such information is subject to the provisions of the Kansas Open Records Act. Id. at 1031. That required information includes: the registrant's name; place and date of birth; description of offense(s) and conviction(s); sex and age of victim; current address; social security number; identifying characteristics; treatment for any mental abnormality or personality disorder; anticipated future residence; photograph; fingerprints; and school. Kan. Stat. Ann. § 22-4907 (2001). The Texas statute makes registrant information available publicly, but it specifically exempts inter alia the following items from disclosure: (1) the registrant's social security number, driver's license number, or telephone number; or (2) any information that would identify the victim of the offense. Art. 62.08.
The court in Otte specifically distinguished the Tenth Circuit's decision in Femedeer on the ground that the Utah statute examined in Femedeer did not provide for public disclosure of the registrant's workplace. 259 F.3d at 992. Like the Utah statute upheld in Femedeer, Texas's statute does not provide for public disclosure of an employer's address, thereby decreasing the risk that registrants will be "practically unemployable."
However, the Otte court also based a large portion of its excessiveness determination on the fact that Alaska's statute failed to provide for an individualized assessment of future risk. Id. Nor did the statute allow for exemption upon rehabilitation. Id. Texas's statute is of similar effect. Offenders in Texas are required to register and are subjected to the public disclosure provisions of the statute for either a ten-year period or for the remainder of their lives, regardless of whether or not they are found to be at risk of re-offending. After the 1999 amendments, there was no longer a provision for exemption from these requirements for adult sex offenders who could make a showing of rehabilitation. Nonetheless, because appellant specifically challenges only the 1997 amendments, because the scope of the information that Texas makes available to the public is more limited than that found to be excessive in Myers or Otte, and because we must always evaluate the statute in light of its professed nonpunitive aims, we will not categorize the 1997 amendments as excessive. We therefore weigh this factor in favor of a sanction that is not criminal in nature.
The intent of the 1997 amendments to Texas's sex-offender registration statute was civil and remedial in nature. Moreover, weighing all the Kennedy factors, we conclude that the effect of the amendments is not so punitive as to transform the statute into a criminal sanction. Therefore, we affirm the judgment of the Fort Worth Court of Appeals.
KELLER, P.J., concurred and filed a separate opinion.
WOMACK, J., concurred and filed a separate opinion.