(dissenting). I respectfully dissent. The majority’s decision holds unconstitutional a section of the Paternity Act that was on the books for over forty years, in the process substituting its judgment for that of the 1956 Legislature relative to the appropriate means for carrying out an objective that the majority finds perfectly acceptable. There is nothing in our judicial commissions that makes us better qualified than that Legislature to make such policy decisions, particularly in a situation where the disparate treatment that the majority uses to strike down the offending section could have been, and presuma*298bly often was, obviated by the fairly simple expedient of utilizing language in child support agreements that provided for later modifiability by the parties or by other order of the cognizant court. In my view, therefore, our voyage into judicial activism in this matter is an entirely unnecessary one.
I. STATUTORY PROVISIONS
A. SECTION 3 OF THE PATERNITY ACT
Section 3 of the Paternity Act, MCL 722.713; MSA 25.493 provided1 as follows:
(a) An agreement or compromise made by the mother or child or by some authorized person on their behalf with the father concerning the support and education of the child shall be binding upon the mother and the child only when the court having jurisdiction to compel support and education of the child shall have determined that adequate provision is reasonably secured by payment or otherwise and has approved the agreement or compromise.
(b) The performance of the agreement or compromise, when so approved, shall bar other remedies of the mother or child for the support and education of the child.
As is readily apparent, and as a prior panel of this Court observed, subsection 3(b) of the Paternity Act prevented modification of the support order unless the parties provided for modification in the agreement, either specifically or by using language allowing modification “by other order of the court.” See Crego v Coleman, 201 Mich App 443; 506 NW2d 568 (1993) (Crego I).2 Stated differently, subsection *2993(b) of the Paternity Act created a statutory bar to other remedies, in addition to those contained in the agreement, but was silent concerning what was required to be contained in the agreement itself.* *3
B. THE DIVORCE STATUTES
The Michigan statutes relating to divorce contain at least three separate provisions relating to the modification of divorce and child support orders and decrees. See MCL 552.17(1); MSA 25.97(1), MCL 552.455(1); MSA 25.222(5)(1), MCL 552.517(1); MSA 25.176(17)(1). These provisions, both collectively and individually, explicitly allow for the modification of judgments of divorce and child support orders. Indeed as noted by the majority, in a divorce action the parties may not compromise their child’s right to receive support. Johns v Johns, 178 Mich App 101, 106; 443 NW2d 446 (1989).
*300The majority, however, concludes that children bom outside marriage who were not the subject of a filiation order were statutorily denied the right to seek modification of support orders, a right expressly granted to other children. This is, quite literally, incorrect. Rather, subsection 3(b) of the Paternity Act denied children bom outside marriage who were not the subject of a filiation order the right to seek modification of support orders unless the parties provided for modification in the agreement, either specifically or by using language allowing modification “by other order of the court. ”4 This distinction is, I contend, an extremely significant one: rather than being denied the right of modification, children bom outside marriage were, under these circumstances, granted exactly the same right.
n. STANDARD OF REVIEW: THE THREE LEVELS OF CONSTITUTIONAL SCRUTINY
A. INTRODUCTION: THE GENERAL PRINCIPLES OF EQUAL PROTECTION
The majority, rather too quickly in my view, reaches the conclusion that, “[without question,” classifications based on illegitimacy are subject to the form of intermediate constitutional review known as “ ‘heightened’ ” scrutiny. Ante, p 293. The majority does, however, correctly state that the Equal Protection Clauses of the United States Constitution and the Michigan Constitution provide that no person shall be denied the equal protection of the law. See US Const, Am XIV; Const 1963, art 1, § 2. See also Frame v *301Nehls, 452 Mich 171, 183; 550 NW2d 739 (1996); Spada v Pauley, 149 Mich App 196, 203; 385 NW2d 746 (1986). In this regard, the Michigan and federal Equal Protection Clauses offer similar protection. Doe v Dep’t of Social Services, 439 Mich 650, 670-671; 487 NW2d 166 (1992). This constitutional guarantee requires that persons similarly situated be treated alike. F S Royster Guano Co v Virginia, 253 US 412, 415; 40 S Ct 560; 64 L Ed 989 (1920); El Souri v Dep’t of Social Services, 429 Mich 203, 207; 414 NW2d 679 (1987). Indeed, this Court has held that the equal protection provisions of the federal and state constitutions are coextensive. Neal v Oakwood Hosp Corp, 226 Mich App 701, 716; 575 NW2d 68 (1997); People v McFall, 224 Mich App 403, 413; 569 NW2d 828 (1997).
Conversely, however, the United States Constitution does not require things that are different in fact or opinion to be treated in law as though they were the same. Jefferson v Hackney, 406 US 535, 549; 92 S Ct 1724; 32 L Ed 2d 285 (1972); Reed v Reed, 404 US 71, 75; 92 S Ct 251; 30 L Ed 2d 225 (1971); Tigner v Texas, 310 US 141, 147; 60 S Ct 879; 84 L Ed 1124 (1940). Stated differently, the courts have not interpreted the United States Constitution to require “ ‘absolute equality.’ ” See Doe, supra at 661, citing San Antonio Independent School Dist v Rodriguez, 411 US 1, 24; 93 S Ct 1278; 36 L Ed 2d 16 (1973). 5
*302B. STRICT SCRUTINY
As the majority correctly states, when state legislation creates a classification scheme that is based upon suspect factors, such as race,6 national origin,7 ethnicity or alienage,8 or that affects a fundamental interest,9 courts apply a high standard of review, labeled “strict scrutiny.” People v Pitts, 222 Mich App 260, 273; 564 NW2d 93 (1997). When courts review statutes under this strict standard, they uphold the statutes only “if the state demonstrates that its classification scheme has been precisely tailored to serve a compelling governmental interest.” Doe, supra at 662, citing Plyler v Doe, 457 US 202, 216-217; 102 S Ct 2382; 72 L Ed 2d 786 (1982). Rarely have courts sustained legislation under this standard of review. Manistee Bank & Trust Co v McGowan, 394 Mich 655, 668; 232 NW2d 636 (1975). Justice Brennan, writing for the majority, laid out the rationale underlying the strict scrutiny analysis in Plyler, supra at 216-217, n 14:
*303Several formulations might explain our treatment of certain classifications as “suspect.” Some classifications are more likely than others to reflect deep-seated prejudice rather than legislative rationality in pursuit of some legitimate objective. Legislation predicated on such prejudice is easily recognized as incompatible with the constitutional understanding that each person is to be judged individually and is entitled to equal justice under the law. Classifications treated as suspect tend to be irrelevant to any proper legislative goal. See McLaughlin v Florida, 379 US 184, 192; 85 S Ct 283; 13 L Ed 2d 222 (1964); Hirabayashi v United States, 320 US 81, 100; 63 S Ct 1375; 87 L Ed 1774 (1943). Finally, certain groups, indeed largely the same groups, have historically been “relegated to such a position of political powerlessness as to command extraordinary protection from the majoritarian political process.” San Antonio Independent School Dist v Rodriguez, 411 US 1, 28; 93 S Ct 1278; 36 L Ed 2d 16 (1973); Graham v Richardson, 403 US 365, 372; 91 S Ct 1848; 29 L Ed 2d 534 (1971); see United States v Carotene Products Co, 304 US 144, 152-153, n 4; 58 S Ct 778; 82 L Ed 1234 (1938). The experience of our Nation has shown that prejudice may manifest itself in the treatment of some groups. Our response to that experience is reflected in the Equal Protection Clause of the Fourteenth Amendment. Legislation imposing special disabilities upon groups disfavored by virtue of circumstances beyond their control suggests the kind of “class or caste” treatment that the Fourteenth Amendment was designed to abolish. [10i
*304The majority, again correctly in my view, declines to use the strict scrutiny test; subsection 3(b) of the Paternity Act is not based upon suspect factors, nor does it affect a fundamental interest.
C. THE “TRADITIONAL” OR “RATIONAL BASIS” TEST
The “traditional” or “rational basis” test represents the other polar extreme. Under this standard, courts will not strike down a statute if the classification scheme it creates is rationally related to a legitimate governmental purpose. Justice Stewart articulated the test in Dandridge v Williams, 397 US 471, 485; 90 S Ct 1153; 25 L Ed 2d 491 (1970):
If the classification has some “reasonable basis,” it does not offend the Constitution simply because the classification “is not made with mathematical nicety or because in practice it results in some inequality.” Lindsley v Natural Carbonic Gas Co, 220 US 61, 78; 31 S Ct 337; 55 L Ed 369 (1911). “The problems of government are practical ones and may justify, if they do not require, rough accommodations — illogical, it may be, and unscientific.” Metropolis Theatre Co v City of Chicago, 228 US 61, 69-70; 33 S Ct 441; 57 L Ed 730 (1913). “A statutory discrimination will not be set aside if any state of facts reasonably may be conceived to justify it.” McGowan v Maryland, 366 US 420, 426; 81 S Ct 1101; 6 L Ed 2d 393 (1961).
*305In Michigan, courts have applied the rational basis test principally to economic and social11 legislation. See Manistee Bank, supra at 668. Under the traditional or rational basis test, a classification will stand unless it is shown to be “ ‘essentially arbitrary.’ ” Id., quoting Lindsley, supra. Stated differently, one who attacks an enactment must show that it is “arbitrary and wholly unrelated in a rational way to the objective of the statute.” Smith v Employment Security Comm, 410 Mich 231, 271; 301 NW2d 285 (1981). See also In re Kasuba Estate, 401 Mich 560, 568-569; 258 NW2d 731 (1977); McAvoy v H B Sherman Co, 401 Mich 419, 453-454; 258 NW2d 414 (1977). “Few statutes have been found so wanting in ‘rationality’ as to fail to satisfy the ‘essentially arbitrary’ test.” Manistee Bank, supra at 668. Stated positively, the test is that courts must uphold a statutory classification where it is rationally related to a legitimate government purpose. Doe, supra, 439 Mich 662; American States Ins Co v Dep’t of Treasury, 220 Mich App 586, 592; 560 NW2d 644 (1996). The rational basis test “ ‘reflects the judiciary’s awareness that “it is up to legislatures, not courts, to decide on the wisdom and utility of legislation.” ’ ” Id. at 597, quoting Gronne v Abrams, 793 F2d 74, 77 (CA 2, 1986) (further citation omitted).
D. HEIGHTENED SCRUTINY
The United States Supreme Court formally adopted “intermediate” or “heightened” scrutiny for “illegiti*306macy” classifications in 1988 in Clark v Jeter, 486 US 456; 108 S Ct 1910; 100 L Ed 2d 465 (1988). In that case, Justice O’Connor, writing for a unanimous court, stated:
To withstand intermediate scrutiny, a statutory classification must be substantially related to an important governmental objective. Consequently we have invalidated classifications that burden illegitimate children for the sake of punishing the illicit relations of their parents, because “visiting this condemnation on the head of an infant is illogical and uiyust.” Weber v Aetna Casualty & Surety Co, 406 US 164, 175; 92 S Ct 1400; 31 L Ed 2d 768 (1972). Yet, in the seminal case concerning the child’s right to support, this Court acknowledged that it might be appropriate to treat illegitimate children differently in the support context because of “lurking problems with respect to proof of paternity.” Gomez v Perez, 409 US 535, 538; 93 S Ct 872; 35 L Ed 2d 56 (1973). [Clark v Jeter, supra at 461.]12
*307Gomez involved facts considerably different than those involved in this matter. Gomez arose in Texas where, both at common law and under the statutes of the state, the natural father had “a continuing and primary duty to support his legitimate children.” Gomez, supra at 536. However, there was no counterpart duty, either at common law or under the state stat*308utes, for natural fathers to support their “illegitimate children.” Id. at 536-537. Thus, but for the fact that the child in Gomez was bom outside marriage, she would have been entitled to support from her natural father. Id. at 537. Under such circumstances, the Court held:
We therefore hold that once a State posits a judicially enforceable right on behalf of children to needed support from their natural fathers there is no constitutionally sufficient justification for denying such an essential right to a child simply because its natural father has not married its mother. For a State to do so is “illogical and unjust.” [Weber, supra] at 175. We recognize the lurking problems with respect to proof of paternity. Those problems are not to be lightly brushed aside, but neither can they be made into an impenetrable barrier that works to shield otherwise invidious discrimination.113] [Gomez, supra at 538.]
*309Here, the statutory scheme is considerably different. Section 3 of the Paternity Act did not deny children bom outside marriage the right of child support from their natural fathers; rather, it provided a mechanism for obtaining such support. The issue here relates to the modifiability of child support orders under subsection 3(b), an issue that the Court in Gomez did not come close to addressing. Thus, while perhaps “seminal” in the overall sense, the holding in Gomez is not particularly helpful in dealing with this issue.
*310The situation in Michigan is similarly somewhat mixed. The Michigan Supreme Court clearly has adopted the United States Supreme Court’s definition of heightened scrutiny. See Dep’t of Civil Rights ex rel Forton v Waterford Twp Dep’t of Parks & Recreation, 425 Mich 173, 191; 387 NW2d 821 (1986):
Under this level of scrutiny, there are two determinations that must be made. The first question is whether the classification serves an important governmental interest. The second question is whether the classification is substantially related to the achievement of the important governmental objective. Craig v Boren, 429 US 190, 197; 97 S Ct 451; 50 L Ed 2d 397 (1976), reh den 429 US 1124 (1977). [Emphasis in original.]
Waterford Twp, however, involved an allegation of gender discrimination, as did Craig v Boren. The most recent Michigan Supreme Court case involving an allegation of discrimination based on illegitimacy is Frame, supra. One of the questions presented in that case was whether § 7b, MCL 722.27b; MSA 25.312(7b), of the Child Custody Act, MCL 722.21 et seq.) MSA 25.312(1) et seq., unconstitutionally deprived the paternal grandfather of a child bom outside marriage equal protection of the law. (The Court referred to § 7b as the “grandparent visitation statute.”) In Frame, the mother of the child filed a complaint (apparently under the Paternity Act) to determine paternity. Frame, supra at 175. The father admitted paternity, and the circuit court entered an order of filiation, awarding the mother custody and awarding the father regular visitation. Id. The child’s paternal grandfather then petitioned the circuit court for grandparent visitation under the grandparent visitation statute. Id. The mother moved for summary *311disposition (rejecting the petition), and the circuit court granted that motion. The grandfather appealed and this Court reversed. Frame v Nehls, 208 Mich App 412; 528 NW2d 773 (1995). This Court specifically applied heightened scrutiny to classifications based on illegitimacy:
Classifications based on illegitimacy are subject to intermediate or “heightened” scrutiny under the Equal Protection Clause of the Fourteenth Amendment. Clark [supra, 486 US 461]; and see Gerhardt v Estate of Moore, 150 Wis 2d 563; 441 NW2d 734 (1989) (statutory scheme similar to Michigan’s statute invalidated on equal protection grounds). [208 Mich App 415.]
The Michigan Supreme Court, however, reversed. Justice Boyle, writing for the majority, stated that “[t]he classification created involves neither a suspect class, [sic] nor a fundamental right, is rationally related to a legitimate government purpose, and does not violate equal protection.” Frame, supra, 452 Mich 174-175 (emphasis supplied). In responding to the grandfather’s assertion (relying on Clark, supra, and Dones v Thomas, 210 Mich App 674; 534 NW2d 221 [1995]) that heightened scrutiny must be used, the majority found, after reviewing the statute as a whole, that “the Legislature did not create classifications using legitimacy as a factor.” Frame, supra, 452 Mich 185. Citing Califano v Boles, 443 US 282, 296; 99 S Ct 2767; 61 L Ed 2d 541 (1979),14 the majority held that the *312incidental effect on children bom outside marriage as a class “ ‘is not sufficient’ ” to warrant heightened scrutiny analysis. Frame, supra, 452 Mich 189. The majority then explicitly used a rational basis test to find the grandparent visitation statute constitutional:
The classifications created under the grandparent visitation statute are grandparents whose child is deceased or involved in a custody dispute (who may seek visitation) and grandparents whose child is alive or not involved in a custody dispute (who may not seek visitation). Because there is no fundamental right or suspect classification involved, a rational basis test is used. Under this test, the grandparent visitation statute will be upheld as long as the classifications therein are rationally related to a legitimate governmental purpose. Doe, supra, at 662. [Frame, supra at 189 (emphasis supplied).]
Thus, at least in Frame, the highest court in this state has declined to use heightened scrutiny in a case involving the allegation of statutory denial of equal protection based on illegitimacy. Panels of this Court have, however, not been so reluctant to apply the heightened scrutiny test. In two cases that this Court decided before the formal adoption by the United States Supreme Court of heightened scrutiny for illegitimacy classifications in Clark, supra, this Court relied on the reasoning of pre-Clark United States. Supreme Court cases — in particular, Gomez — in reviewing statutory classifications alleged to be based *313on illegitimacy.15 In addition, the Dones panel16 and the panel in Crego v Coleman, 226 Mich App 815; 573 NW2d 291 (1997), (Crego II)17 used the heightened scrutiny hallmark language of whether the challenged classification is “substantially related to permissible state interests.”
On the other hand, this Court’s approach has not been uniform on this point. In Hisaw v Hayes, 133 Mich App 639, 644-645; 350 NW2d 302 (1984), the panel, after discussing contractual considerations, addressed the equal protection aspects of the Paternity Act:
In Gomez [supra], the Court held that a statutory scheme which required natural fathers to support their legitimate *314children but not their illegitimate children amounted to a denial of equal protection. However, equal protection does not require that things different in fact be treated as though they were the same. Tigner [supra at 147]. The difference between a paternity case and a case involving support for legitimate children is the necessity of proving paternity in the former case. Due process requires that a litigant be afforded a fair trial of the issues involved in the controversy and a determination of disputed questions of fact on the basis of evidence. . . . The right of an illegitimate child to equal protection of law does not justify depriving the alleged father of the right to a trial of a disputed question of paternity. We decline to follow Boyles [v Brown, 69 Mich App 480; 245 NW2d 100 (1976),] to the extent that, in a paternity settlement like that at issue here, it would permit a court to increase an alleged father’s support obligation, albeit leaving him bound by his agreement to surrender his right to a judicial determination of paternity. Such a settlement cannot be modified, the only judicial remedy being recission.
And, of course, in Crego /, the panel declined to follow Boyles and followed Hisaw, thereby at least implicitly refusing to utilize the heightened scrutiny test.
Clearly, then,-the situation with respect to the application in Michigan of heightened scrutiny to classifications alleged to be based on illegitimacy is anything but clear. I am considerably troubled by the fact that the majority has declined, without meaningful explanation,18 to follow the example set in the most *315recent Michigan Supreme Court case, Frame, supra, in which classifications touching on illegitimacy were a significant issue and in which that Court explicitly applied the rational basis test. My concern is heightened by the fact that the majority pays no attention whatever to fundamental principles of judicial deference.
E. THE PRTNCTPT.ES OF JUDICIAL DEFERENCE
There is almost universal agreement that the power of the Legislature is not without limits. See Marbury v Madison, 5 US (1 Cranch) 137; 2 L Ed 60 (1803): “[T]hat those limits may not be mistaken, or forgotten, the Constitution is written.” And, as the Michigan Supreme Court stated in Manistee Bank, supra at 666: “[T]hat those limits not be exceeded, the courts are entrusted with the responsibility to review and *316the power to nullify legislative acts which are repugnant to the constitution.”
Nevertheless, courts are to use this authority sparingly. “[U]nder established rules of statutory construction, statutes are presumed constitutional, and courts have a duty to construe a statute as constitutional unless unconstitutionality is clearly apparent.” Mahaffey v Attorney General, 222 Mich App 325, 344; 564 NW2d 104 (1997). As the Michigan Supreme Court stated in Council of Organizations & Others for Education About Parochiaid, Inc v Governor, 455 Mich 557, 568; 566 NW2d 208 (1997), citing United States v Raines, 362 US 17, 21; 80 S Ct 519; 4 L Ed 2d 524 (1960), “When compelled to make a constitutional pronouncement, the court must do so with great circumspection and trepidation, with language carefully tailored to be no broader than that demanded by the particular facts of the case rendering such a pronouncement necessary.” Further, in General Motors Corp v Attorney General, 294 Mich 558, 568; 293 NW2d 751 (1990), the Court stated:
The court will not go out of its way to test the operation of a law under every conceivable set of circumstances. The court can only determine the validity of an act in the light of the facts before it. Constitutional questions are not to be dealt with in the abstract.
Indeed, the party challenging the facial constitutionality of an act “ ‘must establish that no set of circumstances exists under which the [a]ct would be valid. The fact that the . . . [a]ct might operate unconstitutionally under some conceivable set of circumstances is insufficient ....’” Council of Organizations, *317supra at 568, quoting United States v Salerno, 481 US 739, 745; 107 S Ct 2095; 95 L Ed 2d 697 (1987).
Thus, this Court should presume § 3 of the Paternity Act to be constitutional unless its unconstitutionality is clearly apparent. Further, in ruling concerning this constitutional challenge, this Court should carefully tailor its language to the particular facts of this case. This Court should not go out of its way to test the operation of § 3 under every conceivable set of circumstances, nor should it deal with the constitutional question in the abstract. Rather, this Court should recognize that the fact that § 3 might operate unconstitutionally under some conceivable set of circumstances is insufficient,. In my view, the majority opinion violates each of these precepts, and indeed does so without acknowledging their very existence.
For example, the majority asserts that children bom outside marriage and not subject to a filiation order “may be foreclosed from future modification of child support, regardless of need.” Ante at 295 (emphasis supplied). It is equally tme, however, that such children bom outside marriage were not foreclosed from future modification of child support if the parties provided for modification in the agreement, either specifically or by using language allowing modification “by other order of the court.” The majority characterizes this distinction as a “narrow . . . thread.” Ante at 295, n 6. I regard it as central. The majority uses the fact that subsection 3(b) of the Paternity Act may have operated discriminatorily in some circumstances as the peg upon which to hang its conclusion that it was therefore unconstitutional in all circumstances on equal protection grounds. It seems to me that, under well-recognized principles of *318judicial deference to legislative enactments, we should take exactly the opposite approach: if subsection 3(b) of the Paternity Act operated exactly the same as the counteipart provisions in the divorce statutes in some circumstances, it was therefore constitutional on equal protection grounds. The majority’s approach, in my view, allows the constitutionality of an enacted statute to hinge on the tenacity — or the lack thereof — of a party in a paternity proceeding or the skill — or the lack thereof — of that party’s lawyer in negotiating a support order that provides for later modification. This, indeed, is a “narrow thread.”
m. THE CONSTITUTIONALITY OF § 3 OF THE PATERNITY ACT
A. TREATMENT OF THE TWO CLASSIFICATIONS
Plaintiff’s basic position here is that § 3 of the Paternity Act created an impermissible classification based on whether the mother of a child was married to the father at the time of the child’s conception or birth. She thus claims that she, and others similarly situated in the same classification under § 3 of the Paternity Act, are treated differently than mothers of children within marriage who seek modification of child support under the divorce statutes, the second classification. The majority agrees. I do not.
In my view, § 3 of the Paternity Act did not create such an impermissible classification as it did not require that the mother of a child bom outside marriage, such as plaintiff, be treated differently from the mother of a child bom within marriage when seeking modification of child support. First, the Paternity Act did not require plaintiff to enter any paternity agreement with defendant whatsoever. Moreover, plaintiff could have insisted on the inclusion of a term in the *319paternity agreement allowing for modification of support, either specifically or by reference to “other order of the court,” as a precondition to entering the agreement. By its silence, subsection 3(b) of the Paternity Act did not bar such an agreement and, as I have previously noted, we have judicially recognized such agreements. Had the parties entered into such an agreement here and had the circuit court approved it, plaintiff would have been in exactly the same position with respect to the ability to seek modification of the amount of child support as if she and defendant had conceived their child within marriage.19
Although involving a different context, the equal protection issue in McFall, supra, is instructive. In that case, a jury convicted the defendant of assault with intent to commit criminal sexual conduct involving sexual penetration20 and fourth-degree criminal sexual conduct.21 The defendant argued that the trial court violated his constitutional right to equal protection when it read CJI2d 20.25,22 which informed the jury that evidence other than the complainant’s testimony was not necessary to prove a charge if that testimony proved guilt beyond a reasonable doubt. McFall, supra at 413. The defendant argued that this violated the right to equal protection of defendants in cases involving charges of criminal sexual conduct *320“by exalting the complainants [in criminal sexual conduct cases] above other crime victims by instructing jurors that their testimony alone can justify conviction.” Id. The McFall panel rejected the defendant’s equal protection argument because the instruction could have been given in cases that did not involve criminal sexual conduct charges:
Defendant’s argument must fail because his premise, that CJI2d 20.25 is given only in cases involving charges of criminal sexual conduct, is without factual support. This Court is aware of no authority, nor has defendant cited any, establishing that this instruction cannot be given in cases involving other offenses. Moreover, it is well established that trial courts are not required to use the Michigan Criminal Jury Instructions, which do not have the official sanction of the Michigan Supreme Court. People v Petrella, 424 Mich 221, 277; 380 NW2d 11 (1985).
Defendant has failed to demonstrate that the circuit court treated him less favorably than it would have treated a similarly situated defendant not charged with criminal sexual conduct. Accordingly, defendant has not established that the court’s use of CJI2d 20.25 violated his right to equal protection under the laws. [McFall, supra at 413-414.]
McFall differs from this case: McFall involved the “standard” criminal jury instructions, which the trial courts are not required to use, while this case involves a statute that had the force of the law. Nevertheless, this case is analogous to McFall in that § 3 of the Paternity Act did not require that plaintiff, as the mother of a child bom outside marriage seeking modification of a child support agreement, be treated differently from the mother of a child bom within marriage seeking modification of a child support agreement. In essence, the inclusion of CJI2d 20.25 only in the “Sex Crimes” chapter of the Michigan *321Criminal Jury Instructions created a “default situation” in which it might have required extra effort or thought for a prosecutor to obtain a “no corroboration required” instruction in a case that did not involve a criminal sexual conduct charge. Similarly, § 3 of the Paternity Act created a “default situation” in which the mother of a child bom outside marriage who entered a child support agreement could not later seek modification of the level of child support in the agreement unless the mother insisted that the agreement include a specific provision allowing for modification or the “other order of the court” language. Thus, § 3 of the Paternity Act did not violate plaintiff’s constitutional right to equal protection, because it did not require that she be treated differently than the mother of a child bom within marriage.
B. THE APPLICATION OF THE RATIONAL BASIS TEST
If this Court were to follow the lead of the Michigan Supreme Court in Frame, we could certainly conclude that, with regard to its effect on mothers of children bom outside marriage, such as plaintiff, we may review § 3 of the Paternity Act under the “rational basis” test. Under this test, this Court should uphold § 3 of the Paternity Act as long it bears a rational relationship to a legitimate governmental purpose. See Frame, supra, 452 Mich 189.
In applying a rational basis review, the Court should recognize that “[t]he lines [were] drawn where they [were] because the Legislature put them there. As long as these lines survive rational basis review . . . this Court must defer to the legislative policy choices.” Id. at 191, n 27. In short, this Court should review the constitutionality of legislation, not its wisdom.*32223 By allowing parties to agree voluntarily to a scheme precluding modification of a child support order, § 3 of the Paternity Act could have increased the financial incentive for a putative father to acknowledge paternity. Also, by providing this option, the Legislature may reasonably have sought to facilitate settlement agreements between the mother and the putative father that would avoid the intense animosity often associated with litigation and thereby encourage the development of a satisfactory father-child relationship. While the majority emphasizes technological developments regarding the determination of paternity as a reason to hold § 3 of the Paternity Act unconstitutional, such technological developments did not eviscerate the reasonableness of § 3 of the Paternity Act in facilitating the relatively amicable settlement of paternity disputes. If the rational basis test were to be applied, I would conclude that § 3 of the Paternity Act was constitutional as applied in this case because it bore a rational relationship to a legitimate government objective.
C. THE APPLICATION OF THE HEIGHTENED SCRUTINY TEST
If, however, this Court chooses to depart from the Michigan Supreme Court’s lead in Frame and to look *323more broadly at subsection 3(b) of the Paternity Act so as to conclude that it does, although indirectly but nevertheless in fact, affect the equal protection rights of the child bom outside marriage, this Court then arguably is compelled to apply the heightened scrutiny test. Under this test, a statutory classification must be substantially related to an important governmental objective. Here, discerning the objective of the Paternity Act is relatively easy. Clearly, the objective of the Paternity Act is to provide a mechanism for assuring support for children bom outside marriage. Indeed, this Court has stated that “[t]he purpose of the Paternity Act is to provide for the support of an illegitimate child.” Smith v Robbins, 91 Mich App 284, 289; 283 NW2d 725 (1979). See also, id., citing Boyles, supra, to the effect that “[t]he announced public policy of this state is to treat children bom out of wedlock as no less deserving of support than those bom in wedlock.”
There is also very little question that this objective is an important one. The issue under the heightened scrutiny test, then, is whether the statutory mechanism in § 3 of the Paternity Act was “substantially related” to the important governmental objective of providing for the support of children bom outside marriage. I would hold that it was.
I first note, however, as did Justice Rehnquist dissenting in Trimble v Gordon, 430 US 762, 783-784; 97 S Ct 1459; 52 L Ed 2d 31 (1977), that this type of review puts a court in the position of finding a legitimate, important objective that the Legislature in its wisdom adopted and then considering whether the very same Legislature was astute enough to adopt an *324appropriate mechanism to achieve that objective. As Justice Rehnquist observed:
If the great difficulties ... [of ascertaining legislative intent] are surmounted, this Court then takes it upon itself to inquire into whether the Act in question accomplished the “purpose” which the Court first determines the legislature had in mind. It should be apparent that litigants who wish to succeed in invalidating a law under the Equal Protection Clause must have a certain schizophrenia if they are to be successful in their advocacy: They must first convince this Court that the legislature had a particular purpose in mind in enacting the law, and then convince it that the law was not at all suited to the accomplishment of that purpose.
But a graver defect than this in the Court’s analysis is that it also requires a conscious second-guessing of legislative judgment in an area where this Court has no special expertise whatever. Even assuming that a court has properly accomplished the difficult task of identifying the “purpose” which a statute seeks to serve, it then sits in judgment to consider the so-called “fit” between that “purpose” and the statutory means adopted to achieve it. In most cases, and all but invariably if the Court insists on singling out a unitary “purpose,” the “fit” will involve a greater or lesser degree of imperfection. Then the Court asks itself: How much “imperfection” between means and ends is permissible? In making this judgment it must throw into the judicial hopper the whole range of factors which were first thrown into the legislative hopper. What alternatives were reasonably available? What reasons are there for the legislature to accomplish this “purpose” in the way it did? What obstacles stood in the way of other solutions?
The fundamental flaw, to me, in this approach is that there is absolutely nothing to be implied from the fact that we hold judicial commissions that would enable us to answer any one of these questions better than the legislators to whose initial decision they were committed. Without any antecedent constitutional mandate, we have created on the premises of the Equal Protection Clause a school for legislators, whereby opinions of this Court are written to *325instruct them in a better understanding of how to accomplish their ordinary legislative tasks. [Id,.]
Here, the majority engages in exactly this process. It finds the overriding purpose of the Paternity Act to be assuring that minor children bom outside marriage receive support and care. Clearly, then, the majority believes the 1956 Legislature to have been wise enough to perceive a legitimate need and concerned enough to pass legislation to address that need. However, in the next breath, the majority condemns the 1956 Legislature for being so dim as to adopt mechanisms that provided for “disparate treatment” of the very children that same Legislature sought to benefit. To use Justice Rehnquist’s analogy, the majority first takes the 1956 Legislature to school and then administers a rather public judicial spanking by substituting its judgment concerning the “fit” between objectives and means. As did Justice Rehnquist, I find nothing in our judicial commissions that gives us superior qualifications to those of the 1956 Legislature to effect such a substitution.
I also note that the circumstances present when a child was bom within marriage and, thus, with a presumptive biological father were objectively different in substance from the situation where a child is bom outside marriage with no individual male readily apparent as the presumptive father. The majority disagrees with this assertion and states that “[c]hildren, whether bom within or without marriage, are the same with regard to their rights to support.” Ante at 296.
While this is an attractive statement rhetorically— and as a truism it is certainly beyond debate — it *326misses the point. When subsection 3(b) was enacted in 1956, dna testing was not even a gleam in a scientist’s eye and identifying the putative father was not a matter of science but of largely anecdotal evidence. While in hindsight this seems quaint, the fact remains that, objectively, a substantive difference did exist between the factual circumstances surrounding the birth of a child within marriage and the birth of a child outside marriage. Standing alone, this is not a basis for upholding disparate treatment. Nevertheless we cannot avert our eyes from the circumstances that existed at the time the statute was enacted. In any event, it certainly will not do to assert a generalized equal right to child support payments as the deciding principle in this case without also showing a statutory violation of that right. Here, as I have repeatedly emphasized throughout this dissent, subsection 3(b) of the Paternity Act treated children bom outside marriage exactly the same as children bom within marriage under circumstances where the agreement allowed for modification of support, either specifically or by reference to “other order of the court.”
In addition, it seems clear to me that the substantial relationship of § 3 of the Paternity Act to the objective of providing for the support of children bom outside marriage was that it facilitated and regulated the formation of child support agreements. Thus, the mechanism that § 3 provided was as substantially related to the objective of the Paternity Act as the mechanisms provided in MCL 552.17(1); MSA 25.97, MCL 552.455(1); MSA 25.222(B)(1), and MCL 552.517(1); MSA 25.176(17)(1) are to the same or similar objectives of the divorce statutes. Thus, in my view, § 3 of the Paternity Act was constitutional even *327when we examine it under the more rigorous standard of the heightened scrutiny test.
D. CONCLUSION
I would find that § 3 of the Paternity Act was not clearly unconstitutional in its application. I do not view the fact that, under the law of this case, plaintiff could not modify the agreed-upon support provision to be dispositive, for we are not required to test the operation of § 3 under every set of circumstances. Rather, mothers of children bom outside marriage were treated exactly the same as mothers of children bom within marriage under one set of circumstances: where the child support agreement contained a specific provision allowing the modification or the “other order of the court” language. Further, there was a substantial justification for the enactment of this mechanism for the support of a child bom outside marriage, given the objective differences at the time the Legislature adopted the statute between such a situation and one in which a child was bom within marriage. Overall, the Legislature’s action in adopting § 3 of the Paternity Act was reasonable. In making this assertion, I do not deal with the constitutional question in the abstract, but would rather adhere to our duty to construe the statute as constitutional unless its unconstitutionality is clearly apparent.
IV. RETROACTIVITY
The majority makes no holding concerning— indeed, it makes no reference to — the issue of the retroactive*32824 effect of its decision. Inasmuch as the Paternity Act was enacted in 1956, I think it safe to assume that literally thousands of cases were brought under its provisions and that a fair number resulted in nonmodifiable child support agreements under § 3. The effect of the majority’s decision is to allow these agreements to be reopened. Presumably, it is the majority’s intent to allow women, including plaintiff, who are the mothers of children who are covered by these agreements to reopen their cases in order to obtain modification of the agreements. However, presumably, defendant and other similarly situated putative fathers may in response contest the issue of paternity.25 What, then, if the putative father, using modem DNA testing methods, is now “cleared” of paternity? Will such an individual be able to obtain reimbursement of the total amount of child support he paid for a child bom outside marriage from the mother of that child?26
*329In this regard, I note that in some — and perhaps many — cases, the mother may have had a good-faith basis for believing that the man involved was the biological father, although she was honestly mistaken. Requiring such a woman to make “restitution” of years of child support payments could impose a tremendous hardship, literally to the point of bankruptcy. Also, putative fathers may in many cases still be making child support payments pursuant to child support agreements. If in a significant number of these cases the men involved are now able to disprove paternity and renounce those agreements, will the children bom outside marriage be henceforward deprived of any “paternal” child payments whatsoever?
A practical answer to these troubling questions would be to give the decision in this matter “limited retroactive effect.” We have typically used this phrase to refer to a holding that articulates a rule of law that is to be applied to (1) future cases or circumstances (a circumstance not involved here) and (2) to only certain cases or circumstances that were initiated before the opinion, i.e., to cases in which direct appellate review is still pending. See, e.g., Jahner v Dep’t of Corrections, 197 Mich App 111, 115; 495 NW2d 168 (1992); Seder v Peoples Community Hosp Authority, 169 Mich App 238, 241; 425 NW2d 775 (1988). However, the majority articulates no such limiting principle in its opinion. I think it fair to predict that while the resulting litigation may not particularly endear us to the trial bench, it certainly will to the family law bar.
V. CONCLUSION
*330I would vacate the circuit court’s order granting plaintiff’s motion for an increase in defendant’s agreed-upon child support payments on the basis that § 3 of the Paternity Act was constitutional. However, if I agreed with the majority’s resolution of that issue, I would give today’s decision limited retroactive effect.
The Legislature repealed § 3 of the Paternity Act effective June 1, 1997.
I, therefore and quite obviously, do not read Crego I as having held that § 3 of the Paternity Act precluded the parties from providing for mod*299ification of the agreement, despite the reference to the lack of authority for the circuit court to “modify the parties’ agreement by entering a support order that was inconsistent with the parties’ settlement of the paternity action that had previously been approved by the court.” Crego I, supra at 447. Rather, I read this language, particularly in light of Judge Griffin’s dissent, to mean that a majority of the Crego I panel believed that the circuit court could not, sua sponte and without the agreement of the parties, insert the phrase “or until further order of the court” in a support order. I regard this reading of Crego I as being the law of the case with respect to the modifiability of the support order.
I observe, however, that subsection 3(b) of the Paternity Act required that the circuit court approve the agreement and that subsection 3(a) required that the circuit court make a determination that “adequate provision [for the child’s support and education] is reasonably secured by payment or otherwise.” Clearly, therefore, the statute required that the circuit court scrutinize the agreement independently to assure, at a minimum, that provision for the child’s support and welfare was (a) “adequate” and (b) “reasonably secured by payment or otherwise.” Beyond these minimum standards, subsection 3(b) of the Paternity Act was silent with regard to the contents of the agreement.
We have judicially recognized such agreements. See Crego I, supra-, Hisaw v Hayes, 133 Mich App 639, 643; 360 NW2d 302 (1984).
Similarly, it is well established that the equal protection guarantee is not a source of substantive rights or liberties; rather, it is a measure of our constitutions’ tolerance of government classification schemes. Doe, supra at 661, citing San Antonio Independent School Dist, supra, 411 US 58 [sic, p 59] (Stewart, J., concurring).
See American States Ins Co v Dep’t of Treasury, 220 Mich App 586, 592-594; 560 NW2d 644 (1996). See also Yick Wo v Hopkins, 118 US 356; 6 S Ct 1064; 30 L Ed 220 (1886); Strauder v West Virginia, 100 US 303; 25 L Ed 664 (1879); Bickel, The original understanding and the segregation decision, 69 Harv L R 1 (1955).
See Oyama v California, 332 US 633; 68 S Ct 269; 92 L Ed 249 (1948); American States, n 6, supra at 592-594.
See Nyquist v Mauclet, 432 US 1, 8, n 9; 97 S Ct 2120; 53 L Ed 2d 63 (1977); El Souri, supra; Chan v City of Troy, 220 Mich App 376; 559 NW2d 374 (1996). While Congress has broad power to regulate immigration and the status of aliens, “[t]he States enjoy no power with respect to the classification of aliens.” Plyler v Doe, 457 US 202, 225; 102 S Ct 2382; 72 L Ed 2d 786 (1982). Only rarely is alienage a matter “relevant to legislation by a State.” Id.
See Harper v Virginia Bd of Elections, 383 US 663, 672; 86 S Ct 1079; 16 L Ed 2d 169 (1966); Neal, supra at 717-718.
There has not been unanimity on the United States Supreme Court that anything other than a “traditional” “rational basis” equal protection analysis should be used. See the dissents of Justice Harlan in Katzenbach, v Morgan, 384 US 641, 660-661; 86 S Ct 1717; 16 L Ed 2d 828 (1966) (reflecting his view that requiring a state to show a need greater than mere rational policy to justify classifications in the area of voting rights is not “consistent with the purposes of the Equal Protection Clause, with the overwhelming weight of authority, or with well-established principles of federalism which underlie the Equal Protection Clause”) and in Shapiro v Thompson, 394 US 618, 658-663; 89 S Ct 1322; 22 L Ed 2d 600 (1969) (reflecting his insistence on judicial restraint lest the Court become a “super-legislature”). Justice Harlan summarized his view in his concur*304rence in Dandridge v Williams, 397 US 471, 489; 90 S Ct 1153; 25 L Ed 2d 491 (1970), when he said:
Except with respect to racial classifications, to which unique historical considerations apply, see Shapiro, at 659,1 believe the constitutional provisions assuring equal protection of the laws impose a standard of rationality of classification, long applied in the decisions of this Court, that does not depend upon the nature of the classification or interest involved.
In Dcmdridge, Justice Stewart, writing for the Court, noted that many of the previous United States Supreme Court cases applying the traditional or rational basis test had dealt with business and industry but then stated that there was no basis for applying a different constitutional standard to cases involving social welfare. Dcmdridge, supra at 485.
As Justice Levin noted in his dissenting opinion in Frame, supra at 207-209, it can easily be argued that at least some of the United States Supreme Court’s pre-1988 decisions regarding “illegitimacy” classifications are consistent with an intermediate form of constitutional review. See, e.g., Picket v Brown, 462 US 1, 8; 103 S Ct 2199; 76 L Ed 2d 372 (1983) (“In view of the history of treating illegitimate children less favorably than legitimate ones, we have subjected statutory classifications based on illegitimacy to a heightened level of scrutiny.”). See also Nowak & Rotunda, Constitutional Law (4th ed), § 14.14, p 719:
This standard of intermediate scrutiny, which falls between the rational relationship test and the strict scrutiny test in terms of the strictness of the judicial review of classification, was not formally adopted for illegitimacy classifications until 1988. Nevertheless, the Supreme Court’s pre-1988 decisions are consistent with this form of intermediate standard of review.
But see Glona v American Guarantee & Liability Ins Co, 391 US 73, 75; 88 S Ct 1515; 20 L Ed 2d 441 (1968), where Justice Douglas, writing for the majority, arguably used the rational basis test in holding that a wrongful death statute that authorized actions by mothers of legitimate children but did not authorize actions by mothers of children bom outside marriage denied the equal protection of the laws to the mothers in the latter class. See also the dissent by Justice Harlan, joined by Justices Black and *307Stewart, id. at 76, describing the decision in Gloria, and the companion case of Levy v Louisiana, 391 US 68; 88 S Ct 1609; 20 L Ed 2d 436 (1968), as “constitutional curiosities” in which the majority reached the answer to the question presented “by a process that can only be described as brute force.” See also the acerbic dissent of Justice Rehnquist in Trimble v Gordon, 430 US 762, 777; 97 S Ct 1459; 52 L Ed 2d 31 (1977), in which, in words echoing the earlier concerns voiced by Justice Harlan, he questioned the entire concept of taking equal protection analysis beyond a rational basis test in cases other than those involving race or national origin:
The Fourteenth Amendment’s prohibition against “any State . . . denypng] to any person . . . the equal protection of the laws” is undoubtedly one of the majestic generalities of the Constitution. If, during the period of more than a century since its adoption, this Court had developed a consistent body of doctrine which could reasonably be said to expound the intent of those who drafted and adopted that Clause of the Amendment, there would be no cause for judicial complaint, however unwise or incapable of effective administration one might find those intentions. If, on the other hand, recognizing that those who drafted and adopted this language had rather imprecise notions about what it meant, the Court had evolved a body of doctrine which both was consistent and served some arguably useful purpose, there would likewise be little cause for great dissatisfaction with the existing state of the law.
Unfortunately, more than a century of decisions under this Clause of the Fourteenth Amendment have produced neither of these results. They have instead produced a syndrome wherein this Court seems to regard the Equal Protection Clause as a cat-o’-nine-tails to be kept in the judicial closet as a threat to legislatures which may, in the view of the judiciary, get out of hand and pass “arbitrary,” “illogical,” or “unreasonable” laws. Except in the area of the law in which the Framers obviously meant it to apply — classifications based on race or on national origin, the first cousin of race — the Court’s decisions can fairly be described as an endless tinkering with legislative judgments, a series of conclusions unsupported by any central guiding principle.
Although not central to my view in this matter, I note in passing that both the panel in Dones v Thomas, 210 Mich App 674, 677-679; 534 NW2d 221 (1995), and the panel in Crego v Coleman, 226 Mich App 815, 821; 573 NW2d 291 (1997) (Crego II), placed considerable reliance on recent scientific innovations in the area of testing for paternity as part of their rationale for finding § 3 of the Paternity Act to be unconstitutional. I first note that subsection 6(5) of the Paternity Act, MCL 722.716(5); MSA 25.496(5), creates a presumption of paternity when the “probability of paternity” after testing by dna profile or other methods is ninety-nine percent or higher, but that a party may always rebut that presumption. Secondly, today’s scientific “fact” may become tomorrow’s heresy. See Conley, The First Principle of Real Reform: The Role of Science in Constitutional Jurisprudence, 65 NC L R 935, 942-943 (1987):
I want to consider what guidance cases such as these can offer for future interactions between science and the law. To my mind, the fatal flaw in these cases is the misperception of science as a well oiled machine for generating truth, a process somehow less susceptible to human passion and error than is law. When science suggests the resolution of a legal problem, a lawmaker with this view may take comfort in rising above the admittedly human and inherently fallible process of traditional lawmaking — he may believe, in Holmes’ words, that he is ‘approaching the first principle of real reform.’ [Referring to a letter by Justice Oliver Wendell *309Holmes mentioning Buck v Bell, 274 US 200; 47 S Ct 584; 71 L Ed 1000 (1927), containing Justice Holmes’ notorious statement that “[t]hree generations of imbeciles are enough.” Id. at 207. Conley’s premise is that this statement was based upon a belief that there was in 1927 an “irrefutable scientific basis for upholding the compulsory sterilization of a mentally retarded woman.” Conley, supra at 935.]
... The problem is that often the human frailty of science can be detected and exposed only by the trained insider. The outsider, even one as astute as Holmes, may confront a wall of superficial unanimity, and see no cause to inquire further.
So what is the law to do? In my view, it should respond with extreme care when science demands that new legal doctrine be created or established doctrine be altered radically. The more strident the demand, the more cautious it should be. Above all, judges and legislators should look skeptically at the scientific process, remembering that such terms as ‘fact,’ ‘truth’ and ‘law’ are, in most cases, as relative for scientists as they are for the rest of us.
. . . The findings of science sometimes have appealed to lawmakers as a simple substitute for the complex interaction among precedent, facts, and values that comprise legal decisonmaking, and as a source of certainty in an otherwise uncertain world. But the appeal of science as a first principle of real reform has been often illusory, and sometimes dangerous. If the Constitution is ‘to endure for ages to come’ in the face of technological challenge, real reform must continue to be inspired by the document itself and the eminently human processes it has spawned.
Califano v Boles, 443 US 282; 99 S Ct 2767; 61 L Ed 2d 541 (1979), holds that a provision of the federal Social Security Act, 42 USC 402(g)(1), making “mother’s insurance benefits” payable only to divorced spouses and surviving spouses of wage earners, but not to unwed mothers, is not a classification on the basis of illegitimacy because any effect on children bom outside marriage is incidental and, to a large degree, speculative. *312(Although the language of 42 USC 402[g][l] as then in effect limited eligibility for “mother’s insurance benefits” to females, the gender-based exclusion of males from eligibility for these benefits had been held to violate the equal protection component of the Fifth Amendment’s Due Process Clause in Weinberger v Wiesenfeld, 420 US 636; 95 S Ct 1225; 43 L Ed 2d 514 [1975].)
See Boyles v Brown, 69 Mich App 480, 483; 245 NW2d 100 (1976) (“[m]ore recently, the United States Supreme Court, in Gomez [supra] found disparate statutory treatment between legitimate and illegitimate children to be constitutionally invalid”) and Spada, supra at 203 (“Both article 1, § 2 of the Michigan Constitution and the Fourteenth Amendment to the United States Constitution provide that no person shall be denied the equal protection of the law. Classifications based on illegitimacy are invalid under the Fourteenth Amendment if they are not substantially related to permissible state interests,” citing Pickett, supra, and Gomez, supra at 538.)
See Dones, supra at 677:
Classifications based upon illegitimacy violate the Equal Protection Clauses of US Const, Am XIV, and Const 1963, art 1, § 2, unless they are substantially related to permissible state interests. Spada [supra at 203.] The question then is whether the different treatment of legitimate and illegitimate children with respect to the modifiability of an award of support is substantially related to a permissible state interest. We are not persuaded that it is.
See Crego II, supra at 818:
Classifications based on illegitimacy are subjected to intermediate or “heightened scrutiny” and are unconstitutional unless “substantially related to permissible state interests.” Spada, supra at 203.
In a footnote, the majority refers to the dissent’s “earnest but strained” attempt to distinguish Gomes and Frame. Ante at 293, n 4. While (perhaps) being complimentary, I believe the majority misses the point of Frame. Certainly, the majority in Frame found that the Legislature did not create classifications using legitimacy as a factor. Frame, supra at 185. However, the majority’s contention that “only then” did the Court in Frame proceed to use the rational basis test is incorrect. Ante at 293, n 4. *315Rather, the Court had previously determined to use the rational basis test, id. at 174-175, and made its comments about the classifications using legitimacy as a factor in response to the grandfather’s assertion that heightened scrutiny must be used. (Certainly, the Court again later in its opinion referred to its use of the rational basis test: “Because there is no fundamental right or suspect classification involved, a rational basis test is used.” Id. at 189. However, as the quotation itself makes clear, the Court rather clearly was of the opinion that the choice of tests under the facts presented was between strict scrutiny and rational basis.) I also note that the only parties in the case before us are the mother, plaintiff, and the acknowledged father, defendant. The child of the parties is not a party, and the increased child support payments being sought would not be paid directly to the child but rather to plaintiff. Accordingly, one hypothetical situation that we arguably need not deal with is whether application of § 3 of the Paternity Act to bar an action by a child bom outside marriage in the child’s own right for increased or back child support would violate the child’s equal protection rights. One could therefore narrowly frame the question before us, in parallel with the Michigan Supreme Court’s framing of the question in Frame, as whether § 3 of the Paternity Act violated the equal protection rights of the mother of a child bom outside marriage.
If defendant had declined to enter a child support agreement with such a provision allowing for modification by the circuit court of the level of child support, then plaintiff, like the mother of a child bom inside marriage, could have sought a judicial award of child support without defendant’s consent.
MCL 750.520g; MSA 28.788(7).
MCL 750.520e(l)(a); MSA 28.788(5)(l)(a).
An instruction from the “Sex Crimes” chapter of the second edition of the Michigan Criminal Jury Instructions.
In this regard, I note that developments in technology used to provide evidence of paternity, or perhaps concerns about fairness to children bom outside marriage and their mothers, may have influenced the Legislature’s decision to repeal § 3 of the Paternity Act effective June 1, 1997. My view that § 3 of the Paternity Act had a rational basis certainly does not reflect any disapproval of the Legislature’s decision to repeal that statute, but merely is a recognition that, within constitutional limits, the Legislature has broad power to determine matters of public policy. With the repeal of § 3 of the Paternity Act, it is clear that the default prohibition on modification of child support orders has now been prospectively removed by the Legislature.
The decision can have no prospective effect because the Legislature repealed § 3 of the Paternity Act effective June 1, 1997.
Such a result seems inevitable in light of the fact that, presumably, a defendant father in a paternity suit gave up his right to contest paternity in return for a nonmodifiable support agreement. See Hisaw, supra at 644-645. If the defendant father now loses nonmodifiability, he certainly regains the right to contest paternity.
To push the point one step further, there is nothing in the majority’s decision that limits a putative father, absent an action by the mother to reopen the child support agreement, from independently commencing an action to reopen that agreement in an attempt to disprove his paternity. Given modem dna testing methods and assuming a certain amount of prudence on behalf of that putative father, one would suspect that such an action would, if commenced, be successful. The same troubling questions would then arise. Would the mother then be required to make “restitution” of past child support payments? Would the child then be deprived of the benefit of future child support payments? The majority’s decision leaves the answer to these questions for another day. Given the litigious nature of our society, I suspect that such a day will not be long in coming.