¶ 38. (dissenting). I would deny this habeas petition. This case does not present a retroactivity question in the usual sense, that is, whether a newly declared rule of law applies retroactively to cases still open on direct review or to events or conduct that predate the announcement of the rule. Harper v. Virginia Dep't of Taxation, 509 U.S. 86, 97 (1993). Rather, this case presents the more difficult question of whether a newly declared rule can be applied retroactively in a collateral attack on a judgment already final. Teague v. Lane, 489 U.S. 288, 299 (1989); State ex rel. Schmelzer v. Murphy, 201 Wis. 2d 246, 258-59, 548 N.W.2d 45 (1996).
¶ 39. In Harper, the United States Supreme Court standardized its general approach to retroactivity analysis. It abandoned the multi-factor test of Chevron Oil Co. v. Huson, 404 U.S. 97 (1971), for civil cases, just as it had earlier, in Griffith v. Kentucky, 479 U.S. 314 (1987), abandoned the multi-factor test of Linkletter v. Walker, 381 U.S. 618 (1965), in criminal cases. Harper, 509 U.S. at 90. The Court held that:
[w]hen this Court applies a rule of federal law to the parties before it, that rule is the controlling interpretation of federal law and must be given full retroactive effect in all cases still open on direct review and as to all events, regardless of whether such events predate or postdate our announcement of the rule.
Id. at 97. The federal rules for civil and criminal retroactivity are thus now the same.
¶ 40. Because our retroactivity jurisprudence is based on that of the United States Supreme Court, we will at some point be asked to adopt, overrule Kurtz v. City of Waukesha, 91 Wis. 2d 103, 280 N.W.2d 757 (1979) (in which we adopted the Chevron Oil retroac-tivity test for civil cases), and bring our law into *648conformity with the now-uniform federal rule.1 However, we have not been asked to do so in this case.
¶ 41. As the majority notes, although the the State mentioned Harper in its brief, it declined to take a position on whether we ought to enunciate a uniform standard of retroactivity for civil and criminal cases; Brown did not even cite Harper. Majority op., ¶ 13 n.5. While it is clear that Chevron Oil is no longer good law and Kurtz should therefore be revisited, I agree that the question is better left for another case because the issue was not fully developed here. Without addressing Harper, the court continues to apply Chevron Oil, as it has since Kurtz. See majority op., ¶¶ 13-26 ("Chevron/Kurtz"). Cf. Harper, 509 U.S. at 99-100 ("[T]he Supreme Court of Virginia has simply incorporated into state law the three-pronged analysis of Chevron Oil and the criminal retroactivity cases overruled by Griffith. We reject the department's defense of the decision below.").
¶ 42. In any event, as I have noted, the retroac-tivity question in this case is narrower, and concerns whether a new rule can be applied retroactively in a collateral attack on a judgment already final. In Teague, a plurality of the Supreme Court answered this question "no" (with certain limited exceptions), a position that has since been endorsed by a majority of the Supreme Court. See Schmelzer, 201 Wis. 2d at 257 n.7, (citing Graham v. Collins, 506 U.S. 461, 466-67 (1993)). The Teague plurality was persuaded by Justice Harlan's view, articulated in Mackey v. United States, 401 U.S. 667, 675 (1971) (Harlan, J., concurring in part and *649dissenting in part), that new rules should be applied retroactively to cases on direct review and not yet final, but not to cases on collateral habeas review. Teague, 489 U.S. at 310. This distinction in retroactivity analysis is based on the nature of the judicial process on the one hand (once law is declared by an appellate court it should be applied to cases not yet adjudicated) and the nature of the habeas corpus remedy on the other (it attacks judgments that have already been reviewed and are long-since final, where the interest in repose is strong). Id. at 306.
¶ 43. The exceptions to the rule of nonretroactivity for cases on collateral review are: "First, a new rule should be applied retroactively [on collateral review] if it places 'certain kinds of primary, private individual conduct beyond the power of the criminal law-making authority to proscribe.'" Id. at 307 (quoting Mackey, 401 U.S. at 692 (Harlan, J., concurring in part and dissenting in part)). "Second, a new rule should be applied retroactively [on collateral review] if it requires the observance of 'those procedures that. .. are "implicit in the concept of ordered liberty."'" Id. (quoting Mackey, 401 U.S. at 693 (quoting Palko v. Connecticut, 302 U.S. 319, 325 (1937) (Cardozo, J.)).
¶ 44. In Schmelzer, this court generally adopted the Teague rule of nonretroactivity for cases that have already become final, with a limited modification for claims that, by their nature, can "only be made through a form of collateral relief." Schmelzer, 201 Wis. 2d at 258. In Schmelzer, it was a claim of ineffective assistance of appellate counsel, which can only be pursued by writ of habeas corpus under State v. Knight, 168 Wis. 2d 509, 484 N.W.2d 540 (1992). Id. The court announced a new right to effective assistance of appellate counsel in the preparation of a petition for review and applied it *650to the petitioner Schmelzer. Then, applying Griffith and Teague, the court held that the new rule was applicable to cases in the direct appellate "pipeline" but to cases already final. Schmelzer, 201 Wis. 2d at 258-59. More specifically:
[W]e conclude that we may apply the new rule announced in this case to the defendant, Schmelzer, although, consistent with Teague, we do not apply it retroactively to cases finalized before the issuance of this opinion. This result is somewhat inequitable, in that we have afforded relief to one defendant while not allowing relief to others similarly situated — the result disfavored in Griffith. Nonetheless, we conclude it would be more inequitable, under the special situation posed here, to adopt Teague's, holding entirely and not only deny the benefit of the new rule to this defendant but also to foreclose the possibility of any new rule being created in this type of case.
Id. at 258. The court also indicated in a footnote that persons whose Knight habeas petitions were filed but not yet decided by the court were "nonfinal" for purposes of retroactivity. Id. at 259 n.9.
¶ 45. As the majority notes, Brown's case was final a year before this court decided State ex rel. Nichols v. Litscher, 2001 WI 119, 247 Wis. 2d 1013, 635 N.W.2d 292. Majority op., ¶ 28. Furthermore, the rule announced in Nichols does not fall within either of the Teague exceptions. The tolling rule is intended to compensate for the vagaries of prison mail systems and is applicable to a layer of appellate review that is permissive rather than as-of-right. It concerns neither primary, individual conduct, nor procedures that are implicit in the concept of ordered liberty.
¶ 46. Accordingly, applying Teague and Schmelzer, Brown cannot seek retroactive application *651of the Nichols tolling rule in this collateral habeas attack. Only Nichols himself, and others coming after him, may benefit from the new tolling rule.
¶ 47. The majority reaches a substantially similar conclusion in Part II, although it does so by a different route. Applying the Chevron/Kurtz test, the majority concludes that "neither a prospective nor a fully retroactive application of the tolling rule we adopted in Nichols is warranted." Majority op., ¶ 26. The majority decides that "limited retroactive application" of Nichols is appropriate: retroactive application to cases pending on direct appeal and not yet final, and to "pro se prisoners who had raised the issue in habeas petitions that were still pending" at the time Nichols was decided. Majority op., ¶¶ 26-27.
¶ 48. This is basically the same as applying the Harper/Griffith retroactivity rule, and adding any pro se prisoners with habeas petitions pending at the time of Nichols, pursuant to the Schmelzer footnote. In any event, Brown does not qualify for the limited retroac-tivity adopted by the majority, which the majority readily acknowledges. Majority op., ¶ 28.
¶ 49. The majority then invokes Harmann v. Hadley, 128 Wis. 2d 371, 382 N.W.2d 673 (1986), and applies Nichols retroactively anyway. In Harmann, this court made an exception to the prospective-only adoption of social host liability that it had announced a year before. In Koback v. Crook, 123 Wis. 2d 259, 277, 366 N.W.2d 857 (1985), this court held that social hosts can be liable for injury caused by minors to whom they negligently furnish alcohol, and employed the technique of "sunbursting" to limit the effect of this new liability rule, applying it to the parties in Koback but otherwise making it prospective only from a specified *652date four months after the date of decision. See Harmann, 128 Wis. 2d at 373-74.
¶ 50. "Sunbursting" is an exception to the general rule of retroactivity, intended to ameliorate the "inequities [that] may occur when a court departs from precedent and announces a new rule of law." Id. at 377-78. The Harmann case was pending in the court of appeals at the time Koback was decided (this court having denied the Harmanns' petition to bypass), but because of the "sunbursting," the Koback rule did not apply retroactively, as ordinarily would have been the case. This court allowed the Harmanns to take advantage of the new liability rule. Id. at 386.
¶ 51. In other words, Harmann is an exception to an exception, intended to restore retroactivity where it otherwise would have existed but for the court's "sun-bursting" of the new rule. Harmann does not address the issue of retroactivity in the context of a collateral attack on a judgment already final; it only allowed a still-pending, non-final case to move forward under the newly created rule. It is not surprising, then, that neither Brown nor the State found the Harmann case relevant enough to warrant citation here.
¶ 52. I find it ironic that the majority declines to address Harper, which at least was cited although not expounded.upon by the State, yet it grants habeas relief based on a case not cited by either party. This approach —permitting retroactivity simply because Brown is "similarly situated" to Nichols — undercuts Schmelzer's adoption of the Teague general rule of nonretroactivity in cases on collateral habeas review. I would not extend Harmann to final judgments, especially not without subjecting the issue to the adversarial process, and *653especially not without giving the State a fair opportunity to be heard.2 Therefore, I respectfully dissent.
In State v. Koch, 175 Wis. 2d 684, 694, 499 N.W.2d 152 (1993), this court followed Griffith v. Kentucky, 479 U.S. 314 (1987), and abandoned the multi-factor retroactivity test in criminal cases.
Footnote 9 of the majority opinion misses the point entirely. I do not quarrel with the majority's conclusion that the petitioner in State ex rel. Nichols v. Litscher, 2001 WI 119, 247 Wis. 2d 1013, 635 N.W.2d 292, and the petitioner in this case may have been similarly situated at the time their respective petitions for review were denied. That does not mean that Harmann v. Hadley, 128 Wis. 2d 371, 382 N.W.2d 673 (1986), applies. Harmann and the case whose rule it applied, Koback v. Crook, 123 Wis. 2d 259, 366 N.W.2d 857 (1985), were direct review cases; nothing in Harmann authorizes the reopening of judgments already final for purposes of retroactive application of a new rule of law. Collateral review cases are governed by State ex rel. Schmelzer v. Murphy, 201 Wis. 2d 246, 548 N.W.2d 45 (1996), and Teague v. Lane, 489 U.S. 288 (1989). The majority has sua sponte extended Harmann to cases on collateral review.