{dissenting). No one denies that the crimes precipitating the passage of chapter 980 are among the most heinous afflicting our society. One can readily understand why the legislature, faced with such wrongs, sought redress through the enactment of chapter 980. But much as I might empathize with the legislature and much as I might share the concerns which led to the passage of chapter 980, it is beyond reasonable doubt that chapter 980 is unconstitutional. I join the many judges from Wisconsin1 and other jurisdictions2 who have found that similar statutes create unconstitutional preventive detention based primarily on predictions of dangerousness.
In authorizing the incarceration of individuals on the basis of past crimes for which they have already served their sentences, chapter 980 violates constitutional-provisions against double jeopardy and ex post facto laws. In creating a circularly defined class of "sexually violent persons" who can be committed without evidence of mental illness and who could not be committed under Wisconsin's civil commitment law, chapter 980 violates constitutional guarantees of substantive due process and equal protection.
The state cannot violate individual rights inscribed in the constitutions by creating special classes of individuals whose constitutional rights are *337diminished. Although the end result may seem attractive, under our constitutions the state cannot simply lock people up on the supposition that they will be dangerous in the future when they have already served their sentences for crimes committed in the past.
The legislative, executive and judicial branches have available other, constitutionally valid methods of addressing the dangers posed by violent criminals. These methods include tougher and more stringent supervision of those on parole or conditional release, chapter 51 commitment, more intensive prison treatment programs, longer legislatively enacted sentences for crimes of sexual violence, and prosecutors' advocacy for and judges' imposition of lengthier or consecutive sentences at the time of sentencing. Such responses to the dangers posed by sex offenders can protect the community without eroding the constitutional guarantees that protect all of us. For the reasons set forth, I dissent.3
rH
The issue presented is whether chapter 980's restriction on liberty principally constitutes permissible civil commitment or impermissible punishment. If *338chapter 980 is principally punitive, it violates the ex post facto and double jeopardy clauses of the Wisconsin and federal constitutions.4
This court has explained that "[glovernmental action is punishment under the double jeopardy clause if its principal purpose is punishment, retribution or deterrence. When the principal purpose is nonpunitive, the fact that a punitive motive may also be present does not make the action punishment." State v. Killebrew, 115 Wis. 2d 243, 251 (1983) (emphasis added).
The language of chapter 980 provides insufficient evidence of remedial intent while its legislative history, purpose and effect provide overwhelming evidence of its principally punitive purpose. In determining that chapter 980 passes constitutional muster, however, the majority opinion in State v. Carpenter, 197 Wis. 2d 252, 541 N.W.2d 105 (1995) relies on chapter 980's language and structure while ignoring its legislative history, purpose and effect. This approach misconstrues the very U.S. Supreme Court precedent which, as the majority correctly observes, this court has consistently followed in interpreting the double jeopardy and ex post facto clauses of the Wisconsin and federal constitutions. When correctly applied, the Supreme Court's test clearly reveals that chapter 980 violates the double jeopardy and ex post facto clauses in both constitutions.
*339According to the majority opinion in Carpenter, 197 Wis. 2d 269-70, "we look to the plain language of the statute as evidence of the legislature's intent," Carpenter, (discussing possible double jeopardy violations), and "we must consider the language and structure of the statute to determine whether it serves a legitimate regulatory public purpose," Id. at 273 (discussing possible ex post facto violations). The majority opinion points repeatedly to chapter 980's treatment provisions to conclude that the chapter is remedial rather than punitive. For example, the majority opinion notes that "a person found to be sexually violent is committed to the custody of DHSS for control, care, and treatment, as opposed to the DOC for imprisonment." Id. at 266. The majority opinion thereby concludes that "[t]he emphasis on treatment in ch. 980 is evident from its plain language." Id.
If reference to treatment were sufficient to render a statute civil, however, chapter 302, governing state prisons and jails, would be transmogrified into a civil statute. Arguably the most punitive of all the Wisconsin statutes, chapter 302 nevertheless refers to treatment 30 times; chapter 980 mentions treatment 9 times. Chapter 302 provides for "confinement, treatment, and rehabilitation" in Wisconsin's prisons;5 chapter 980 provides for "control, care, and treatment" of chapter 980 committees.6 One of the purposes of chapter 302 is "to provide a just, humane and efficient program of rehabilitation of offenders."7 Chapter 980 *340contains no comparable statement evincing a purpose to provide treatment.
Looking solely to the plain language of chapter 302, as the majority would have a court do, the court would conclude that chapter 302 manifests great concern with treatment and, applying the majority opinion's reasoning, would conclude that the purpose and effect of the statute governing prisons is remedial. But while both rehabilitation and treatment have long been among the justifications for imprisonment,8 their inclusion in the stated purpose and statutory language of chapter 302 does not alter the fact that the principal purpose of the statute governing prisons and jails is punishment. Statutory language alone, then, cannot resolve the question of whether a statute containing remedial aspects is principally punitive in purpose.
Nothing in the language of chapter 980 refers to the commitment it prescribes as a civil commitment.9 Even if chapter 980 had expressly referred to its commitment procedures as civil, the U.S. Supreme Court has repeatedly warned that a legislature's designation *341of a statute as "civil" or "remedial" rather than "punitive," "retributive" or "deterrent" is not determinative in gauging the principal purpose that statute actually serves. Notwithstanding how a statute is labeled or characterized by the legislature, "a civil as well as a criminal sanction constitutes punishment when the sanction as applied in the individual case serves the goals of punishment... [A] civil sanction that cannot fairly be said solely to serve a remedial purpose, but rather can only be explained as also serving either retributive or deterrent purposes, is punishment, as we have come to understand the term."10 Therefore, a court must look beyond a statute's language and structure and inquire further whether the statutory scheme *342was so punitive either in purpose or effect as to negate the remedial aspects of the statute. United States v. One Assortment of 89 Firearms, 465 U.S. 354, 362-63 (1984) (citing United States v. Ward, 448 U.S. 242, 248 (1980)).
Not surprisingly, in exploring a statute's principal purpose, the Supreme Court has examined legislative history. See, e.g., Kennedy v. Mendoza-Martinez, 372 U.S. 144,169-184 (1963); Flemming v. Nestor, 363 U.S. 603, 618-619 (1960). Quoting Flemming, the majority opinion in Carpenter asserts that courts should not look beyond the language of a statute in determining legislative intent. Carpenter, 197 Wis. 2d at 269. The Flemming Court did, however, look at legislative history in determining Congressional intent. Flemming, 363 U.S. at 619. The Flemming Court refers at length to previous Supreme Court cases in which the Court had relied upon such "Congressional history" or the "Court's first-hand acquaintance with the events and the mood" surrounding passage of a statute in determining that a nominally civil statute was actually punitive. See, e.g., Flemming, 363 U.S. at 615.
Noting that "only the clearest proof could suffice to establish the unconstitutionality of a statute" on the basis of legislative history, Flemming, 363 U.S. at 617, the Flemming Court concluded that the "meagre [legislative] history" available in relation to the statute at issue in that case was insufficient to prove Congress' punitive intent. Flemming, 363 U.S. at 617-619. In contrast, as I explain below, all the legislative history of chapter 980 provides clear proof of its punitive purpose: to reduce the likelihood that sexual predators might reoffend by prolonging their detention past the completion of their prison terms.
*343The context in which a statute is passed assists in determining legislative intent.
It is established practice in American legal processes to consider relevant information concerning the historical background of enactment in making decisions about how a statute is to be construed and applied .... These extrinsic aids may show the circumstances under which the statute was passed, the mischief at which it was aimed and the object it was supposed to achieve.
Norman J. Singer, 2A Sutherland Statutory Construction § 48.03 at 315 (1992) (note omitted).11
The enactment of chapter 980 was preceded by a widely publicized, highly politicized and extremely emotional public debate following the release of the notorious sex offender Gerald Turner.12 In calling a special legislative session to enact chapter 980, Governor Tommy Thompson expressed the hope that "[w]e might be able to use this civil commitment procedure to keep them [i.e., convicted sex offenders] in jail."13 In equating civil commitment with jail, the Governor speaks volumes concerning the primarily punitive nature and purpose of chapter 980's allegedly civil commitment proceedings.
*344Drafting requests and statements made by sponsors of legislation prior to enactment have long been considered authoritative in construing legislative intent.14 The stated views of Representative Lolita Schneiders, a legislator who sponsored chapter 980, make clear that its primary purpose is deterrence, one of "the traditional aims of punishment." Kennedy v. Mendoza-Martinez, 372 U.S. at 168.
In hér drafting request to the Legislative Reference Bureau for the first version of chapter 980, Representative Schneiders stated that the bill "seeks to place further restrictions on the most heinous of repeat sexual offenders" by insuring that "the prison stay [would] be lengthened" for any "predator" who remained "a significant threat to society."15 Representative Schneiders acknowledged in her request that "[t]hese predators are sane, not mentally ill" and opined that they are "highly resistant to change." She sought legislation which would "mak[e] the offender *345face a lifetime of accountability and loss of liberty for engaging in [past] sexually assaultive acts."16
Both the drafting file and the written views of those associated with the drafting process have also long been considered reliable indicia of legislative intent.17 The comments of the principal draftsman of chapter 980, Legislative Reference Bureau attorney Jeffrey Olsen, provide further evidence of the statute's punitive intent. According to the draftsman, he understood that the legislative intent was "to make continued commitment of the person as secure as possible ... ,"18
The events leading up to the passage of chapter 980 therefore confirm the statement of one circuit court judge who held chapter 980 unconstitutional: "[t]o suggest that this law is merely a benign exercise of the State's parens patriae authority without a significant punitive content is to ignore the reality of the political context in which this law was passed and the manner in which it was drafted."19
*346The placement of chapter 980 within the Wisconsin statutes also lends support to the conclusion that its principal purpose is punitive rather than remedial. Chapter 980 is placed squarely within the criminal portion of the Wisconsin statutes. Although the state claims that this placement is not "significant to show the legislature intended to create a criminal statute,"20 Wisconsin case law suggests otherwise. The "position of [a] section [of the statutes] in controversy is very persuasive as to its intent." Montreal Mining Co. v. State, 155 Wis. 245, 248, 144 N.W. 195 (1913); Although not itself dispositive, the fact that the legislature placed an act in a particular section of the statutes can, when supplemented by other evidence, corroborate the impression that placement conveys. State v. Robe, 96 Wis. 2d 48, 73-74, 291 N.W.2d 809 (1980).
Thus the legislative history of chapter 980 clearly demonstrates the extent to which this nominally remedial statute principally evinces a punitive purpose, namely the ongoing incarceration of convicted sex offenders who might otherwise be released.
Furthermore, because chapter 980 requires that convicted sex offenders serve their criminal sentences before being committed under its auspices, the statute is inextricably linked to a punitive purpose and effect, notwithstanding its remedial features.21 Why would a *347legislature with a principal interest in treatment create a statute deliberately delaying the promised treatment and thereby exacerbating the alleged ills which it is designed to cure? An individual's need for diagnosis and treatment does not surface only at the end of a prison term. The state's failure to mandate treatment prior to the completion of the punishment phase "strongly suggests that treatment is of secondary, rather than primary, concern." Young v. Weston, 898 P. Supp 744, 753 (D. Wash. 1995).22
*348The majority in State v. Post observes that treatment is already available to sex offenders within the prison setting and that chapter 980 is therefore reserved for those who have not fully availed themselves of previous treatment opportunities or for whom previous treatment has proven ineffective. Majority op. at 308. The statute, the majority continues, is structured "to cover only those demonstrated to be most in need of treatment" and is therefore serious in pursuing the objective of providing treatment. Id.
The limited treatment available in prison belies this observation. According to Raymond Wood, acting chief of the sexually violent person unit at the Wisconsin Department of Corrections' Wisconsin Resource Center, many incarcerated sex offenders currently wait as long as seven years before being transferred to an institution where full treatment might be available.23 Wood's testimony indicated that prison treatment programs are not "nearly as intensive" or *349"broad based" and "don't have the same number of components" as those available following civil commitment. Wood also acknowledged that "there are differences between the way seclusion is used in a mental health facility and the way that segregation is used in a correctional facility" as well as a panoply of differences regarding the rights of the respective populations, the care and treatment owed to the respective populations, and the qualifications and standards expected of the respective staffs.
Notwithstanding these differences, the majority opinion in Carpenter relies upon Allen v. Illinois, 478 U.S. 364 (1986), in claiming that chapter 980's imposition of commitment subsequent to a criminal sentence is not "fatal." Carpenter, 197 Wis. 2d at 270-71. In the Illinois statute under review in Allen, however, commitment was in lieu of rather than in addition to a prison sentence. Hence the Illinois statutory scheme "was focused solely on providing treatment to mentally disordered sex offenders," demonstrating that "Illinois had 'disavowed any interest in punishment.'" Young v. Weston, 898 F. Supp. at 752, (citing Allen, 478 U.S. at 370).
This difference between the Illinois and Wisconsin statutes underscores the remedial nature of the Illinois statute and, by contrast, accents the punitive nature of chapter 980.1 conclude that the Allen decision renders chapter 980 unconstitutional.24
*350To sum up, chapter 980's nominally remedial purpose is belied by a revealing paper trail of legislative history demonstrating its principally punitive purpose and effect. Although one might fairly characterize treatment as one of chapter 980's purposes, careful analysis of the statute establishes that its primary purpose is punitive and therefore unconstitutional. Chapter 980's professed concern with treatment is further compromised by the requirement that those slated for treatment under the statute first serve a full criminal sentence, thereby delaying that treatment, possibly for decades.
According chapter 980 the presumption of constitutionality owing to every legislative enactment, I nevertheless conclude that these indicia of a punitive purpose and effect establish beyond a reasonable doubt that chapter 980 violates the protections against double jeopardy and ex post facto laws incorporated in the Wisconsin and federal constitutions.
The right to substantive due process bars certain arbitrary, wrongful actions 'regardless of the fairness of the procedures used to implement them.'" Foucha v. Louisiana, 504 U.S. 71, 80 (1992) (quoting Zinermon v. Burch, 494 U.S. 113, 125 (1990)). Using a substantive due process analysis, the U.S. Supreme Court has carefully circumscribed those occasions when the state may, for nonpunitive reasons, detain individuals and thereby deprive them of their constitutionally pro*351tected liberty. Youngberg v. Romeo, 457 U.S. 307, 316 (1982); Addington v. Texas, 441 U.S. 418, 425 (1979). According to the cases cited by the majority opinion in Post, a state may not commit any person without clear and convincing evidence that the person is both mentally ill and dangerous.25
Because chapter 980 allows the commitment of individuals who are not both mentally ill and dangerous, I conclude that it violates substantive due process guarantees of the Wisconsin and federal constitutions. Further, because there is no rational basis for authorizing civil commitment according to the substantive standards for commitment under chapter 980 rather than those already available under current civil commitment standards, I also conclude that chapter 980 violates equal protection guarantees inscribed in both constitutions.
A.
The majority opinion in Post acknowledges that "a mental condition component" is a requirement of substantive due process for commitment under chapter 980. Majority op. at 303-07. At the same time, the majority opinion in Post observes that the U.S. Supreme Court has never attempted to establish one constitutionally required definition of "mental illness," but has instead allowed the states some degree of latitude in developing their own definitions. Id. at 304.26
*352But a recognition that mental illness or the neologism "mental condition component" may be defined in more than one way hardly suggests that mental illness can be defined howsoever the state pleases. If the constitutionally prescribed threshold of mental illness has no core meaning and can mean everything, then it means nothing.
The Foucha case teaches that states are not free to define any deviancy they please as a mental illness and thereby commit to mental hospitals anyone who might fit their definition. Were there no limit on a state's substantive power to commit individuals, a state could civilly commit whole categories of criminal offenders such as intoxicated drivers merely by branding them deviant and designating them mentally disordered. The Foucha Court underscored this point in holding that an insanity acquittee with a diagnosed antisocial *353personality disorder could not be confined as mentally ill. Foucha, 504 U.S. at 77-83.
For even as the Foucha Court acknowledged that "psychiatrists widely disagree on what constitutes a mental illness," it nevertheless insisted that there was sufficient consensus regarding a definition to make specific and "reliable" determinations about who can be considered mentally ill for purposes of the constitutionally required threshold for civil commitment. Foucha, 504 U.S. at 76 n.3. If, however, mental illness or a "mental condition component" means whatever a state claims it means, a constitutionally required threshold for deprivation of liberty would be transformed into a meaningless standard signifying whatever state legislatures want it to signify.
As both the legislative history of chapter 980 and the records before us reveal, those involved in drafting, enacting and implementing chapter 980 understood very well that the broader, more nebulous notion of "mental disorder" required for chapter 980 differed greatly from the "mental illness" required by the state and federal constitutions.
In her original drafting memorandum to the Legislative Reference Bureau, Representative Schneiders stated that "[t]hese predators are sane, not mentally ill, despite the depraved nature of their crimes."27 The chief draftsman for chapter 980 recognized the constitutional problems inherent in the drafting request. "[A]s I have said before," he warned in raising problems with the term "mental disorder," "I am not confident that the law is being narrowly enough drawn because it *354is impossible to say who should be committed" on the basis of a mental disorder "we are not even sure exists."28
The two psychologists who testified at Carpenter's probable cause hearing for commitment under chapter 980 acknowledged a distinction between the concepts of generic mental disorder and mental illness. Dr. Wood testified that mental illness "may be a subset of that larger group of disorder[s] known as mental disorder" and included within the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (4th ed. 1994) (DSM-IV).29 He also explained that "[m]ental illness is far more incapacitating in terms of reality appreciation, the standard soz~ts of tests that we might apply to determine if somebody was loosely speaking crazy or not."
Greg Van Rybroek, clinical director of the Mendota Mental Health Institute, drew a similar contrast between mental disorders and mental illness, noting that "there is a distinction in terms of definition" and that "mental disorders are the broad big umbrella that all of us could fall under." Among the disorders comprising this broad, big umbrella of mental disorder "that all of us could fall under" and included within the DSM-IV Manual are eating disorders such as anorexia and bulimia; sleeping disorders such as insomnia; caffeine-induced anxiety disorder; and agoraphobia *355(anxiety about being in places or situations from which escape is difficult).30
Finally "mental disorder" is defined in chapter 980 not in terms of mental illness, mental disease or mental defect but in terms of a predisposition to sexual crimes. Under chapter 980 "mental disorder" is "a congenital or acquired condition affecting the emotional or volitional capacity that predisposes a person to engage in acts of sexual violence." Wis. Stat. § 980.01(2). Since every condition is necessarily either congenital or acquired, and since "emotional or volitional capacity" simply describes the decision-making processes affecting how people act, mental disorder under chapter 980 means no more than a predisposition to engage in acts of sexual violence.
Thus chapter 980 attempts to create a mental disorder authorizing lifetime commitment based not on mental illness but on past crimes for which the prospective committee has already served the prescribed sentence. This definition is entirely circular: a prospective committee's "mental disorder" is derived from past sexual offenses which, in turn, are used to establish a predisposition to commit future sexual offenses.31
*356The majority opinions' attempt to uphold the constitutionality of chapter 980 by relying on a circular definition of mental disorder premised on dangerousness reveals that the true purpose of chapter 980 is to lock up those considered dangerous, regardless of whether they are mentally ill. But dangerousness, standing alone, is not constitutionally sufficient to justify a civil commitment. Such a rationale, warned the U.S. Supreme Court, would allow the state to incarcerate any "convicted criminal, even though he has completed his prison term." Foucha, 504 U.S. at 82-83. Indeed, such a rationale would be only "a step away from substituting confinements for dangerousness for our present system which, with only narrow exceptions and aside from permissible confinements for mental illness, incarcerates only those who are proved beyond reasonable doubt to have violated a criminal law." Id. at 83.32
*357Despite this stern admonition, the majority opinion in Post reads Justice O'Connor's concurrence in Foucha and this court's decision last term in State v. Randall, 192 Wis. 2d 800, 532 N.W.2d 94 (1995), as allowing the state to prolong the confinement of potentially dangerous albeit sane individuals, so long as some medical justification for that confinement continues to exist. Majority op. at 315-16. But this reading relying on medical justification overstates both holdings.
Both Foucha and Randall involved insanity acquittees who, but for original diagnoses that they were mentally ill, would have been required to serve prison sentences for the commission of their respective crimes. The relationship between Foucha's and Randall's respective insanity acquittals and the length of time they would have served if they had been found guilty factored heavily in both Justice O'Connor's and this court's assessments of how long they might be held under the aegis of medical justification once they had *358regained their sanity. As Justice O'Connor noted in her Foucha concurrence, "the permissibility of holding an acquittee who is not mentally ill longer than a person convicted of the same crimes could be imprisoned is open to serious question." Foucha, 504 U.S. at 88. Similarly, this court's opinion in Randall, having noted that "[i]t is the determination of guilt which provides the basis for the state to incapacitate and treat the insanity acquittee," held that confinement must be strictly "limited to the maximum term which could have been imposed for the criminal conduct." Randall, 192 Wis. 2d at 833, 841.
A commitment extending beyond the maximum prison term which could have been imposed, then, must meet the constitutional requirement articulated in Addington, Jones, and Foucha: the state must establish that the prospective committee is not only dangerous, but also mentally ill. Although the Supreme Court has not defined mental illness for purposes of commitment, the circular definition of mental disorder in chapter 980 is clearly inadequate; it is not "reliable enough to permit the courts to base civil commitments on clear and convincing medical evidence that a person is mentally ill." Foucha, 504 U.S. at 76 n.3. Instead, chapter 980, in the words of the amicus curiae brief filed by the Wisconsin Psychiatric Association, wraps itself in the aura of science and asks clinicians to "compromise their professional integrity so that a constitutional gloss can be applied to something impermissible." Brief for the Wisconsin Psychiatric Association as Amicus Curiae at 3.
This gloss cannot, in my opinion, save chapter 980. Because chapter 980 allows the indefinite confinement of persons who have not been found to be mentally ill, it *359is beyond a reasonable doubt that chapter 980 violates substantive due process protections.
B.
I turn now to the equal protection challenge. Both the majority opinion in Post and the state observe that for purposes of equal protection analysis, persons committed under chapter 980 are similarly situated to persons committed under chapter 51, Wisconsin's civil commitment statute. Post, majority op. at 318-19; Brief for State in Post at 13. Consequently, the requirements for chapter 51 civil commitment must be harmonized with those for chapter 980 commitment.33 "Equal protection does not require that all persons be dealt with identically, but it does require that a distinction made have some relevance to the purpose for which the classification is made." Baxstrom v. Herold, 383 U.S. 107, 111 (1966).
A state cannot seek a civil commitment under one statute rather than another when the two statutes apply distinct substantive standards for commitment and afford distinct procedural protections for commitment unless those distinctions can be justified by a rational basis and a legitimate purpose. Chapter 980's circular definition of mental disorder is premised on dangerousness rather than on evidence of mental illness. Just as dangerousness alone cannot justify civil commitment, dangerousness alone cannot justify distinct substantive commitment standards. Because the distinctions separating chapter 980 from chapter 51 have no rational basis, I conclude that it is beyond a *360reasonable doubt that chapter 980 violates the equal protection guarantees of both the Wisconsin and federal constitutions.34
Chapters 51 and 980 have similarities, as the majority opinion in Post explains. Both statutes concern persons with mental disorders. Both contemplate the treatability of the individual and the prospect that the individual will prove dangerous to the public or to himself if left untreated. But the "mental disorder" required for a chapter 980 commitment is not equivalent to the types of "mental disorders" readily subsumed under chapter 51. What is the rational basis for this difference? The majority opinion does not answer this fundamental question.
The U.S. Supreme Court has answered it, stating that "there is no conceivable basis for distinguishing the commitment of a person who is nearing the end of a penal term from all other civil commitments." Foucha, 504 U.S. at 79 (quoting Baxstrom, 383 U.S. 107 (1966)). In the statute under review in Baxstrom, New York State allowed a person to be committed at the expiration of a penal sentence without the jury review that was available to all other persons civilly committed. The state contended that the statute created a reasonable classification differentiating between the "criminally and dangerously insane" and the "insane." The Court held that this distinction did not survive even a rational basis equal protection analysis. Baxstrom, 383 U.S. at 111.
*361In Baxstrom the Court made clear that equal protection requires a state to use the same standards and procedures for involuntary civil commitment of incarcerated persons that it uses for nonimprisoned individuals. If at the end of a prison term a prisoner has been freed and "the state then decides to deprive him of liberty and stigmatize him with involuntary hospitalization, the ex-prisoner should be entitled to the same protections granted other citizens."35
The Baxstrom Court was willing to acknowledge that especially dangerous committees might require different treatment once they were committed, but emphasized that dangerousness "has no relevance whatever" in "show[ing] whether a person is mentally ill at all." Baxstrom, 383 U.S. at 111. Hence while post-commitment distinctions between committees with distinct treatment needs might be legitimate, the Bax-strom Court left no doubt that the initial commitment process itself must be applied equally to the entire class of prospective committees unless the state could offer a rational basis and a legitimate purpose for any differences.
The majority opinion in Post does not provide a rational basis for the difference in the commitment standards. Instead, it elides the distinction articulated in Baxstrom between the initial commitment and post-commitment treatment. See, majority op. at 321-22. The majority opinion tries to salvage the statute from an equal protection challenge by stating that the "heightened level of dangerousness and the unique treatment needs of sexually violent persons justify distinct legislative approaches [to chapter 51 commitment of persons with mental illness and chapter 980 commit*362ment of persons with mental disorders] to further the compelling governmental purpose of protection of the public." Majority op. at 322-23. But neither the language and structure of chapter 980 nor the majority opinion reveals why the particular treatment needs of allegedly "mentally disordered" sexually violent persons justify different substantive standards for civil commitment than those currently available under chapter 51.
Because the majority cannot present a rational basis that might explain why chapter 980 adopts different substantive commitment standards than does chapter 51, the majority opinion's justification for the statutory distinctions reduces to no more than the threat of "heightened dangerousness" which chapter 980 sexual offenders allegedly pose — a point the majority underscores repeatedly in its equal protection analysis.36
But as Baxstrom and Foucha make clear, "heightened dangerousness" does not pass muster under equal protection analysis. "The Supreme Court has never upheld a lifetime preventive detention scheme for *363those who are feared dangerous." In re Young, 857 P.2d 989, 1023 (Wash. 1993) (Johnson, J., dissenting), rev'd, Young v. Weston, 898 F. Supp. 744 (D. Wash. 1995).
For the reasons stated, I conclude that chapter 980 violates the equal protection guarantees of the Wisconsin and federal constitutions.
HH HH hH
Although they address distinct constitutional issues, both majority opinions fail to salvage chapter 980 for the same reason: they are unable to demonstrate that chapter 980 is principally concerned with addressing the treatment needs of persons who are both mentally ill and dangerous. But the tension between the majority opinions' respective attempts to demonstrate that chapter 980 meets the crucial constitutional prerequisites for civil commitment (mental illness and dangerousness) cannot be resolved.
In order to surmount ex post facto and double jeopardy challenges, the majority opinion in Carpenter must demonstrate that chapter 980's principal purpose is to provide treatment and that the statute is thereby civil and remedial rather than punitive.
In order to surmount substantive due process and equal protection challenges, the majority opinion in Post must demonstrate that the prospective committees under chapter 980 are mentally ill. But because chapter 980's circular definition of mental disorder substitutes dangerousness for evidence of mental illness, the majority opinion in Post is compelled to rely heavily on the threat of heightened dangerousness which prospective 980 committees allegedly pose.
To the extent that the majority opinion in Post emphasizes dangerousness to society at large rather than treatment for the mentally ill, it undercuts the *364thrust of the argument advanced in Carpenter that chapter 980 is principally a civil statute advancing the remedial purpose of providing treatment rather than principally a punitive statute advancing the deterrent purpose of preventing harm.
Conversely, the emphasis on treatment in Carpenter makes all the more glaring Post's inability to offer a rational basis for separate chapter 980 and chapter 51 substantive commitment standards and its consequent reliance on dangerousness as the primary justification for chapter 980 civil commitments.
In dividing the task of preserving chapter 980's . constitutionality, the majority opinions have only emphasized the problem intrinsic to chapter 980: Despite its attempt to recast punishment as "treatment for the good of the criminal," chapter 980 punishes rather than treats; its focus is on dangerousness and deterrence rather than on mental illness, mental disorder, or a "mental condition component." And most important, in their approach to the problem posed by violent sex offenders, chapter 980 and the majority foster legal fictions which are in themselves dangerous.
For the reasons set forth, I conclude that it is beyond reasonable doubt that in enacting chapter 980 the legislature has adopted an unconstitutional method to achieve its goals. Accordingly, I dissent.
Approximately one-half of the Wisconsin circuit court judges who have been faced with constitutional challenges to chapter 980 have found the statute unconstitutional.
See, e.g., Young v. Weston, 898 F. Supp. 744 (D. Wash. 1995); In re Blodgett, 510 N.W.2d 910 (Minn. 1994) cert. denied, 115 S. Ct. 146 (1994) (three dissenting justices); In re Young, 857 P.2d 989 (Wash. 1993), rev 'd, Young v. Weston, 898 F. Supp. 744 (D. Wash. 1995) (three dissenting justices).
I dissent from both majority opinions. While State v. Carpenter is primarily addressed to the issues of double jeopardy and the ex post facto clause and State v. Post is primarily addressed to the issues of substantive due process and equal protection, the four respondents do not divide their arguments in this manner. Moreover, the consideration of these four issues together highlights tensions in the respective majority analyses that would not otherwise be apparent. I address these tensions in Part III. This dissent, then, responds to both majority opinions and addresses all four of the constitutional issues which they discuss.
To violate either the double jeopardy or ex post facto clauses, the government action under the statute must constitute punishment or create a criminal proceeding within the meaning of those clauses. Collins v. Youngblood, 497 U.S. 37, 46-52 (1990); United States v. Halper, 490 U.S. 435, 447-51 (1989); State v. Thiel, 188 Wis. 2d 695, 702-03, 524 N.W.2d 641 (1994); State v. Killebrew, 115 Wis. 2d 243, 246-51, 340 N.W.2d 470 (1983).
Wis. Stat. § 302.25(1) (1993-94).
Wis. Stat. § 980.06(1) (1993-94).
Wis. Stat. § 301.001 (1993-94). Chapter 51 (the Mental Health Act), which governs civil commitments, mentions treatment 363 times. The legislative policy in the Mental Health Act is "to assure the provision of a full range of treatment and *340rehabilitation services in the state for all mental disorders and developmental disabilities and for mental illness, alcoholism and other drug abuse."
See, e.g., Kent Greenawalt, Punishment, in 4 Encyclopedia of Crime and Justice 1336-45 (Sanford H. Kadish, ed. 1983); 1 Wayne R. LaFave & Austin W. Scott, Jr., Substantive Criminal Law § 1.5, 32-33 (1986).
The word "civil" appears once in the title and relating clause of the Act creating chapter 980, stating that it is "relating to civil commitment of sexually violent persons." LRB Drafting File for 1993 Act 479 (emphasis added). The word "civil” also appears once in chapter 980 itself, but only with reference to the immunity from civil liability extended to state agency officials under the statute's victim notification provisions. See Wis. Stat. § 980.015(4) (1993-94).
Halper, 490 U.S. at 448 (emphasis added).
In assessing a challenge to the double jeopardy clause, the Halper Court discounted the value of labels, stating as follows:
[T]he labels "criminal" and "civil" are not of paramount importance. It is commonly understood that civil proceedings may advance punitive as well as remedial goals, and, conversely, that both punitive and remedial goals may he served by criminal penalties .... The notion of punishment, as we commonly understand it, cuts across the division between the civil and the criminal law, and for the purposes of assessing whether a given sanction constitutes multiple punishment barred by the Double Jeopardy Clause, we must follow the notion where it leads .... ("[T]he labels affixed either to the proceeding or to the relief imposed are not controlling and will not be allowed to defeat the applicable protections of federal constitutional law").
Halper, 490 U.S. at 447-48 (citations omitted).
See also Collins, 497 U.S. at 46 (how a statute is labeled is not controlling and should not "immunize it from scrutiny" in determining whether the constitutional prohibition against ex post facto laws has been violated, because "[sjubtle ex post facto violations are no more permissible than overt ones,” and the "constitutional prohibition is addressed to laws, 'whatever their form'").
See also Erdman v. Jovoco Inc., 181 Wis. 2d 736, 751, 512 N.W.2d 487 (1994) (relying on fact that statute was passed during the Great Depression in adopting remedial construction).
Greg Rosenberg, The Legislative History and Implementation of Chapter 980, Wisconsin Defender, June-August 1995, at 4; Erich C. Straub & James E. Kachelski, The Constitutionality of Wisconsin's Sexual Predator Law, Wisconsin Lawyer, July 1995, at 15.
Sexual predator bill sparks session call: Offenders would be kept in jail, Milwaukee Sentinel, May 18,1994, at A-ll.
Norman J. Singer, 2A Sutherland Statutory Construction § 48.15 at 364 (1992); Bartus v. DHSS, 176 Wis. 2d 1063, 1075-76, 501 N.W.2d 419 (1993) (drafting request of legislative sponsor indicative of legislative intent); Kelley Co., Inc. v. Marquardt, 172 Wis. 2d 234, 248-49, 493 N.W.2d 68 (1992) (statements by bill's sponsor comprise "legislative history" revealing purpose of statute); Foerster, Inc. v. Atlas Metal Parts Co., 105 Wis. 2d 17, 24, 313 N.W.2d 60 (1981) (statements by bill's sponsor, including a press release regarding the bill, provide evidence of legislative intent).
Drafting Request Memo from Representative Lolita Schneiders to Bruce Feustel, Assistant Chief Counsel, Legislative Reference Bureau, LRB Drafting File for 1993 AB 955 (March 15,1993).
/<i. See also Lolita Schneiders, Putting a Stop to Sex Offenders, Milwaukee Journal, November 16,1993, at A-15.
Bartus, 167 Wis. 2d at 1075-76; Robert Hansen Trucking, Inc. v. LIRC, 126 Wis. 2d 323, 336,377 N.W.2d 151 (1985) ("this court has given weight to the written comments of those involved in drafting the legislation"); State v. Barkdoll, 99 Wis. 2d 163, 176, 298 N.W.2d 539 (1980) (citations omitted) (written views of those involved with the drafting process "can properly be considered as an authoritative statement of legislative intention"); Bendorf v. City of Darlington, 31 Wis. 2d 570, 579, 143 N.W.2d 449 (1966) (memo in drafting file by drafter of bill represents appropriate source of legislative history in determining meaning of bill).
Drafter's note to 2975/1 at 1 (October 25,1993).
State v. Carpenter, No. 94-CF-1216 (Dane Co. July 22, 1994).
Brief for State in State v. Carpenter at 16.
Although the majority opinion in State v. Carpenter, 197 Wis. 2d 252, 541 N.W.2d 105 (1995), claims that "the mere fact that a prior conviction is a predicate of the current sanction does not render the current sanction punishment for the past offense," Carpenter, 197 Wis. 2d at 274, the U.S. Supreme Court has explained that conditioning the restraint of liberty on the commission of a crime is "significant of penal and prohibitory intent." Dep't of Revenue of Montana v. Kurth Ranch, 114 S. Ct. *3471937, 1947 (1994) (quoting United States v. Constantine, 296 U.S. 287, 295 (1935)).
See also In re Young, 857 P.2d at 1024 (Johnson, J., dissenting) (when treatment for sex offenders follows rather than substitutes for prison sentences, this "timing alone is a strong indication that the legislature was less interested in treatment than in confinement" and demonstrates that while "the Statute provides for treatment, this goal is completely subordinated to punishment"); State v. Carpenter, No. 94-CF-1216 (Dane Co. July 22, 1994) ("The fact that treatment is not offered until the end of an underlying prison sentence which may be many years after the last sexual offense strongly suggests that treatment is virtually an afterthought in this legislative scheme. Further, the fact that there is no requirement for a finding of amenability to treatment as required in Chapter 51 commitments bolsters this conclusion"); State v. Oldakowski and Post, Nos. 94-CF-1200-01, slip op. at 14 (Dane Co. Sept. 2,1994) (suggesting that treatment is "an afterthought masking the real concern for keeping predators out of the community," since medical treatment models suggest that treatment is more effective when provided earlier); State v. Watson, No. 94-CF-2377 (Dane Co. April 7,1995) (chapter 980's definition of "mental disorder" is a "characterological" description of persons whose potential to commit future sexually violent acts is based on past crimes rather than mental illness).
One commentator asserts that such delays in treatment can reduce the prospect that treatment will succeed, because they allow the offender to implement defense mechanisms and cognitive distortions which, in turn, make it more difficult for the offender to accept responsibility for what he has done. The passage of time also increases the risk of memory loss of events which are often poorly recalled to start with because of alcohol or substance abuse. Robert M. Wettstein, A Psychiatric Perspective on Washington's Sexually Violent Predators Statute, 15 U. Puget Sound L. Rev. 597,617 (1992). Finally, even when prisons themselves offer treatment programs, the prison milieu reduces an offender's ability to benefit from treatment because prisons socialize an inmate "to avoid disclosing personal weakness or vulnerability, avoid taking responsibility for his crime, or reveal himself to be a sex offender for fear of retaliation." Id. See also Stephen J. Morse, Mentally Disordered Offenders, in 3 Encyclopedia of Crime and Justice, supra, at 1046, 1048 (treatment is *349minimal in prisons and in hospitals that house the criminally insane).
In response to Allen, the second draft of chapter 980 required the state to choose, within 60 days of a conviction or a finding of not guilty by reason of mental insanity, whether to pursue sentencing through a criminal proceeding or to file a petition for a civil commitment. Though the legislature was *350advised that this change had been made in an effort to insulate the proposed law from a possible double jeopardy challenge, the legislature nevertheless instructed the draftsman to redraft the bill so that after a sex offender had completed his prison term, the state could seek a chapter 980 commitment.
Foucha v. Louisiana, 504 U.S. 71, 75-76 (1992); Jones v. United States, 463 U.S. 354,368 (1983); Addington v. Texas, 441 U.S. 418, 426 (1979); O'Connor v. Donaldson, 422 U.S. 563 (1975); Jackson v. Indiana, 406 U.S. 715 (1972).
The two cases cited by the majority in discussing the states' power to define mental illness do not support the majority's broad assertion concerning a state's power to define mental *352illness for purposes of commitment. See Addington, 441 U.S. 418 (1979); Jones, 463 U.S. 354 (1983).
First, the committees in both cases had been diagnosed as paranoid and schizophrenic, conditions universally associated with mental illness.
Second, the issue in Addington is the standard of proof required in a civil commitment by the Fourteenth Amendment. The decision does not discuss the definition of mental illness.
Finally, in Jones as well, the Court does not address whether the committee is mentally ill. Jones, 463 U.S. at 363 n.11. The Court upheld the legislative determination of procedures accompanying civil commitment in a context where the committee "himself advances insanity as a defense and proves that his criminal act was a product of mental illness." Jones, 354 U.S. at 367. The sentence quoted by the majority opinion, State v. Post, majority op. at 304, for the proposition that courts should defer to legislative judgments is followed by a caveat relating such deference to cases involving the insanity defense. Jones, 463 U.S. at 370.
Drafting Request Memo from Representative Lolita Schneiders to Bruce Feustel, Assistant Chief Counsel, Legislative Reference Bureau, LRB Drafting File for 1993 AB 955 (March 15,1993).
Drafter's Note to 2975/1 at 1 (October 25,1993).
The disorders incorporated within DSM-IV include the antisocial personality disorder with which both the acquittee in Foucha, 563 So. 2d 1138,1141 n.2 (La. 1990), as well as three of the four prospective chapter 980 committees whose cases we now review were diagnosed.
DSM-IV, 213, 396, 439, 539-557.
Wettstein, supra; J. Christopher Rideout, So What's in A Name? A Rhetorical Reading of Washington's Sexually Violent Predators Act, 15 U. Puget Sound L. Rev. 781, 793 (1991-92).
See also Young v. Weston, 898 F. Supp. at 750 (finding that the Washington State statutory definition of "mental abnormality," which, like the definition of "mental disorder" under chapter 980, requires proof of "a congenital or acquired condition affecting the emotional or volitional capacity which predisposes the person to the commission of criminal sexual acts," creates "an unacceptable tautology: a sexually violent predator suffers from a mental condition that predisposes him *356or her to commit acts of sexual violence;" also finding that the term "personality disorder" "evokes a circular definitional structure in which the only observed characteristic of the disorder is the predisposition to commit sex crimes"); In re Young, 857 P.2d at 1021 (Johnson, J., dissenting) (definition of mental abnormality under the Washington statute is "circular" because "abnormality" "will be derived from the person's past sexual behavior, and this in turn will be used to establish the person's predisposition to future dangerous sexual behavior"); State v. Carpenter, No. 94-CF-1216 (Dane Co.) (chapter 980 deploys "a watered down version of the classically accepted definition of mental illness, us[ing] a circular definition that is an invitation to arbitrary and erroneous interpretation").
One of those "narrow exceptions," the pretrial detention of dangerous arrestees permitted by the Bail Reform Act of 1984 (Act), was upheld in United States v. Salerno, 481 U.S. 739 (1987). But the majority's reliance on this case for the proposition that danger-reducing confinement can justify *357constitutional violations, State v. Post, majority op. in Post at 317, is misplaced. The Salerno Court upheld the Act because its legislative history evinced a regulatory rather than punitive purpose and because "[t]he Bail Reform Act carefully limits the circumstances under which detention may be sought," "[t]he arrestee is entitled to a prompt detention hearing," and "the maximum length of pretrial detention is limited by the stringent time limitations of the Speedy Trial Act." Id. at 747.
Having catalogued these features of the Act, the Foucha Court rejected Louisiana's reliance on Salerno to justify its continued confinement of an individual whom doctors had assessed as still dangerous but who was no longer mentally ill. Foucha, 504 U.S. at 81-82. Neither, then, can Salerno rescue chapter 980 which, in contrast to the Act, has a legislative history evincing punitive intent and which allows for potential lifetime incarceration rather than stringently limited pretrial detention.
See also State ex rel. Farrell v. Stovall, 59 Wis. 2d 148, 207 N.W.2d 809 (1973) (chapter 51 civil commitments and chapter 975 sex crime offender commitments deal with similarly situated classes).
Because I conclude that chapter 980 does not meet a rational basis standard, I join the majority in reserving for another day the question of which standard of constitutional review is appropriate when applying an equal protection analysis to a non-suspect class.
Stephen J. Morse, Mentally Disordered Offenders, in 3 Encyclopedia of Crime and Justice, supra, at 1049.
In responding to arguments advanced by Post and Oldakowski, the majority itself refutes other possible bases for distinguishing chapter 51 committees from chapter 980 committees. As the majority points out, for example, Wis. Stat. § 51.20(l)(ar) already waives its general requirement that those committed evince dangerousness through a recent overt act if the prospective committee, like every potential chapter 980 committee, is currently imprisoned. Post, majority op. at 324. And as the majority also points out, Wisconsin case law allows the commitment under chapter 51 of even those who, like many potential committees under chapter 980, might be unamenable or hostile to treatment. C.J. v. State, 120 Wis. 2d 355, 354 N.W.2d 219 (Ct. App. 1984); Post, majority op. at 323-25.