concurring in part, dissenting in part.
Petitioner appeals the dismissal of his petition for post-conviction relief. ORS 138.650. That petition, which challenged petitioner’s dangerous offender sentence, alleged, variously, that the sentence was unconstitutional because it was based on facts not pleaded in the indictment and not found by the jury beyond a reasonable doubt, and that he received inadequate assistance of trial counsel because tried counsel failed to argue that the sentence was unconstitutional. The trial court dismissed the petition on three alternative grounds: (1) The petition was untimely filed under *593ORS 138.510(3); (2) petitioner’s allegations were barred by the successive-petition bar of ORS 138.550(3); and (3) even if petitioner’s claims were not procedurally barred, the imposition of petitioner’s dangerous offender sentence comported with principles of law enunciated in Apprendi v. New Jersey, 530 US 466, 120 S Ct 2348, 147 L Ed 2d 435 (2000). For the reasons that follow, I agree with the majority that petitioner was not entitled to post-conviction relief based on inadequate assistance of counsel. However, I conclude that petitioner’s other allegations are not procedurally barred and state a claim for relief under Apprendi. Consequently, I would reverse and remand.
As explained below, I would conclude that petitioner has stated cognizable claims for post-conviction relief pursuant to ORS 138.530(1)(a) and ORS 138.530(1)(c), for the following reasons: (1) The retroactivity test of Teague v. Lane, 489 US 288, 109 S Ct 1060,103 L Ed 2d 334 (1989), does not apply to, or control, the cognizability of claims, based on federal constitutional violations, under Oregon’s post-conviction relief statutes. (2) Even if the retroactivity analysis of Teague v. Lane were applicable, Apprendi represents and announces a “watershed rule” that is subject to retroactive application. (3) With the exception of his claim of ineffective assistance of counsel, petitioner’s other bases for post-conviction relief are not untimely under ORS 138.510(3) or precluded by the successive-petition bar of ORS 138.550(3) and (4). Petitioner has pleaded cognizable claims for relief under Apprendi that the judicial imposition of petitioner’s dangerous offender sentence based on his “future dangerousness” violated federal due process. I address each matter in turn.
I. THE RETROACTIVITY ANALYSIS OF TEAGUE v. LANE DOES NOT CONTROL THE COGNIZABILITY OF FEDERAL CONSTITUTIONALCLAIMS UNDER OREGON’S POST-CONVICTION RELIEF STATUTES
The threshold question here is whether petitioner’s Apprendi-based claims are cognizable in state post-conviction proceedings at all. That, in turn, implicates the subsidiary question of whether the retroactivity analysis of Teague v. Lane applies to federal constitutional claims under Oregon’s post-conviction statutes. For the reasons set forth *594below, I conclude that the Teague retroactivity analysis does not govern whether relief for federal constitutional violations is available in state post-conviction proceedings.
As an initial matter, we have developed a body of nonconstitutional law (discussed more fully below, see 184 Or App at 601-03 & 606-07), concerning when a new constitutional principle that has been articulated between the time of a petitioner’s direct appeal and a post-conviction proceeding can be considered in the post-conviction proceeding although not raised at trial or on direct appeal. See generally Long v. Armenakis, 166 Or App 94, 101, 999 P2d 461, rev den, 330 Or 361 (2000) (discussing standard). The test set forth in Long and other post-conviction cases is not the same as the retro-activity test from Teague. Thus, the question further reduces to whether, as a matter of federal constitutional law, this court is required to apply the Teague retroactivity analysis to an Apprendi-related challenge, or whether we can follow our own law concerning when a newly announced constitutional principle will be applied in a post-conviction proceeding.
To answer that question, we must examine Teague and its precedential underpinnings. See, e.g., Linkletter v. Walker, 381 US 618, 85 S Ct 1731, 14 L Ed 2d 601 (1965); Mackey v. United States, 401 US 667, 91 S Ct 1160, 28 L Ed 2d 404 (1971); Griffith v. Kentucky, 479 US 314, 107 S Ct 708, 93 L Ed 2d 649 (1987). Teague concerned the availability of federal habeas corpus relief for a petitioner who had been convicted in state court of numerous crimes. Teague, 489 US at 292-93. After the petitioner’s original convictions, the United States Supreme Court issued several Sixth Amendment decisions concerning jury selection. The petitioner then sought federal habeas corpus relief, arguing that, under those subsequently decided cases, the jury selection in his trial did not comport with the Sixth Amendment. Id. at 293-94.
The Court, as an initial matter, confronted the question of how and when newly announced constitutional principles should be applied retroactively on collateral review in federal habeas corpus proceedings. The Court began by noting that its case law concerning when new constitutional *595principles should be applied was not entirely consistent. Id. at 300. The Court’s starting point was Linkletter.
In Linkletter, the Court had determined that the retroactivity of an exclusionary principle announced in another case should be determined by examining the purpose of the exclusionary rule, the reliance of the states on prior law, and the effect on the administration of justice of a retroactive application of the exclusionary rule. Teague, 489 US at 302 (citing Linkletter, 381 US at 636-40). Although Linkletter was a habeas corpus case, the principle it announced was applied by the Court “to limit application of certain new rules to cases on direct review,” as well as in other circumstances. Teague, 489 US at 302 (citing cases).
After Linkletter, in Mackey, Justice Harlan, dissenting,1 took issue with the Linkletter retroactivity approach. Justice Harlan began by observing that he found no principled basis for treating one case on direct review that presents the same constitutional issue in a different manner from another case on direct review. Mackey, 401 US at 678-81 (Harlan, J., dissenting). He posited that a new rule of constitutional law should always be applied on direct review. On the question of collateral review where convictions were final, Justice Harlan wrote:
“While, as I have just stated, I think it clear what law should be applied to nonfinal convictions here on direct review, the choice of law problem as it applies to cases here on habeas seems to me a much more difficult one. However, that choice, in my view, is also one that can be responsibly made only by focusing, in the first instance, on the nature, function, and scope of the adjudicatory process in which such cases arise. The relevant frame of reference, in other words, is not the purpose of the new rule whose benefit the petitioner seeks, but instead the purposes for which the writ of habeas corpus is made available.”
*596Id. at 682. Justice Harlan then summarized historical limitations on the availability of the writ of habeas corpus, noting, in particular, the principles of procedural default for failure to raise an issue on direct appeal. Id. at 682-84. He concluded:
“In my view, the issues respectively presented by [the two habeas cases at issue] — whether the new rules of [substantive constitutional law] should be applied ‘retroactively’— must be considered as none other than a problem as to the scope of the habeas writ. We can properly decline to apply the [new substantive constitutional law] to the present cases only if that is consistent with the reasons for the provision, in our federal legal system, of a habeas corpus proceeding to test the validity of an individual’s official confinement.”
Id. at 684. In concluding that it would generally be “sounder” to apply the law as it existed at the time a conviction became final, Justice Harlan relied, in part, on what he perceived as Congress’s intent in crafting federal habeas corpus statutes. Id. at 687-88 n 5. In proposing that the general rule for habeas corpus cases should involve “defining [constitutional] errors according to the law in effect when a conviction became final,” id. at 692, he proposed two exceptions: the first for “substantive due process” issues, id. at 692, and the second for “claims of nonobservance of those procedures that * * * are ‘implicit in the concept of ordered liberty.’ ” Id. at 693 (quoting Palko v. Connecticut, 302 US 319, 325, 58 S Ct 149, 82 L Ed 288 (1937)). He identified as an example of such a “bedrock” principle the right to counsel at trial. Id. at 694.
After Mackey but before Teague, the Court, in Griffith, adopted the first prong of Justice Harlan’s proposed retroactivity analysis from the Mackey dissent — that newly announced constitutional principles would be applied on direct review. 479 US at 322-23, 328. Two years later in Teague, the Court adopted the second prong of Justice Harlan’s proposed analysis that, “[u]nless they fall within an exception to the general rule, new constitutional rules of criminal procedure will not be applicable to those cases which have become final before the new rules are announced.” Teague, 489 US at 310. The Court also adopted Justice Harlan’s proposed exception for substantive due process, and *597a variation on his “implicit in the concept of ordered liberty’ exception. Id. at 311 (internal quotation marks omitted). The Court elaborated on the latter exception, stating that it is “reserved for watershed rules of criminal procedure.” Id. (emphasis added). The Court indicated that this second exception applied not only to procedures “implicit in the concept of ordered liberty” that implicate fundamental fairness, id. at 311 (internal quotation marks omitted), but also to “ ‘ “new” constitutional rules which significantly improve the pre-existing fact-finding procedures.’ ” Id. at 312 (quoting Desist v. United States, 394 US 244, 262, 89 S Ct 1030, 22 L Ed 2d 248 (1969) (Harlan, J.)).
In Teague, the Court also elaborated on the reasons for treating the retroactivity question differently depending on whether the issue arose on direct appeal or on collateral review in habeas. The Court noted, as had Justice Harlan in Mackey, the historic limitations on the writ of habeas corpus, observed that procedural default rules limited a petitioner’s federal habeas corpus claims in numerous ways, and “recognized that interests of comity and finality must also be considered in determining the proper scope of habeas review.” Teague, 489 US at 308. The Court further observed that retroactive application of new rules of constitutional law in habeas corpus proceedings imposed significant costs upon the states and was, in some ways, more intrusive than enjoining criminal prosecutions because it “continually forces the States to marshal resources in order to keep in prison defendants whose trials and appeals conformed to then-existing constitutional standards.” Id. at 310 (emphasis in original).
The question in this case is whether Teague expresses a general federal constitutional rule of nonretroactivity for cases on collateral review that is applicable in a context other than the federal habeas corpus context, e.g., state post-conviction proceedings. I conclude that it does not. Although Teague refers generically to “collateral review” at a number of points, and that term could be read to encompass review in state post-conviction proceedings, there is no suggestion that the general rule of nonretroactivity applied in federal habeas corpus cases is constitutionally mandated to be applied in any other context.
*598One thing is entirely clear from Teague, and from Justice Harlan’s dissent in Mackey, which was ultimately adopted: The retroactivity of a newly announced rule of constitutional criminal procedure does not depend primarily on the nature or magnitude of the constitutional right involved, but on the procedural posture of the case in which the issue arises. If the case is before the court on direct review (and, thus, the underlying judgment is nonfinal), then any newly announced rule of constitutional criminal procedure will be applicable, regardless of its nature or magnitude. On the other hand, if the case before the court is a federal habeas corpus case, then the general rule is that a newly announced rule of constitutional criminal procedure will not be applied to undo a conviction that became final before the new rule was announced. Although exceptions to that general rule of nonretroactivity may depend on the nature of the constitutional rights involved, the basic premise of current retroactivity law from the Supreme Court is that retroactivity depends on the type of proceeding, and is not part and parcel of the underlying substantive constitutional right.
Nor is there any suggestion in Teague itself (or in Justice Harlan’s dissent in Mackey, which Teague adopted) that the rules of direct appeal retroactivity and collateral review nonretroactivity are themselves based on the requirements of the Due Process Clause or on any other constitutional provision that is made applicable to the states through the Fourteenth Amendment. Rather, to the extent that the Court expressed concerns of a constitutional magnitude regarding retroactivity on collateral review, those concerns hinged on federalism and comity — namely, concerns that constantly relitigating constitutional issues based on evolving law imposed a significant burden on states. Teague, 489 US at 310.
Those concerns simply do not transfer, either as a matter of policy or a matter of federal constitutional law, to state post-conviction proceedings. State post-conviction relief provides remedies defined by the legislature that enacted the post-conviction statutes.2 Thus, the state legislature, in *599enacting post-conviction statutes, defines the level of burden imposed on the state in terms of relitigating issues in criminal cases after convictions are final. The legislature has, in fact, placed some limitations on when claims will not be considered despite asserted errors of constitutional magnitude. See, e.g., ORS 138.510(3); ORS 138.550(3) (establishingtimelines on post-conviction claims and limiting successive petitions, as discussed below). Those limitations are — and are properly — the product of a state legislative choice, and not a federal constitutional command. Consistently with long-established principles of federalism and comity, those legislative choices are not, and cannot be, supplanted or overridden by judicial decisions addressing the availability of federal habeas corpus relief.
In sum, the analysis set forth in Teague does not apply as a sort of Supremacy Clause-mandated overlay to determine whether claims of due process violations based on Apprendi are cognizable in a state post-conviction proceeding. Rather, our state’s post-conviction statutes themselves, and related Oregon appellate decisions, govern when, or whether, such “retroactive” relief is available in Oregon courts.
The majority concludes, nevertheless, that Oregon case law requires us to follow federal precedent on retroactivity. I disagree.
The majority relies, in part, on State v. Fair, 263 Or 383, 502 P2d 1150 (1972), for its conclusion that it must adhere to federal precedent in analyzing the retroactivity of newly announced federal constitutional rules asserted as grounds for collateral relief from criminal conviction. 184 Or App at 585-87. That reliance is misplaced. Fair was a direct criminal appeal, decided when the United States Supreme Court was still following the Linkletter standard that it disavowed in Griffith. See 184 Or App at 594-97) (discussing history of Court’s retroactivity jurisprudence). That type of *600analysis, as noted above, treated the question of retroactivity as part and parcel of the substantive constitutional right at issue and did not distinguish between direct appeals and collateral challenges.
The Fair court noted that it had “closely followed the retroactivity rules adopted by the United States Supreme Court.” 263 Or at 385. It further noted, however, that it did not always choose to do so. Id. at 387. It stated that it was “free to choose the degree of retroactivity or prospectivity which we believe appropriate to the particular rule under consideration, so long as we give federal constitutional rights at least as broad a scope as the United States Supreme Court requires.” Id. at 387-88. The court observed that “[t]he decisions of the United States Supreme Court are not binding on us, but we may look to those cases for guidance.” Id. at 388. The court then went on to apply a variation on the Linkletter standard to an issue of Oregon constitutional law. Id. at 388-90.
The majority’s suggestion that the court’s decision in Fair somehow forces us to apply the federal habeas corpus retroactivity rule announced in Teague is, frankly, puzzling. The court in Fair did not “adopt” a federal retroactivity analysis. It noted its freedom to apply or not apply that analysis as it chose, and emphasized that it was looking to the Court’s cases “for guidance.” Fair, 263 Or at 388. It then proceeded to apply a variation on federal retroactivity analysis that bears no analytical resemblance to Teague whatsoever. It is difficult to see how Fair can be read as a mandate that we apply Teague, which obviously had not been decided at that point. Moreover, Teague simply has no application to the issue presented in Fair, i.e., whether or not a newly announced constitutional principle should be applied in the context of a direct appeal. Griffith answers that question, and answers it in a manner quite different from the manner in which the question was answered in Fair. Thus, to the extent that a federal constitutional question is presented on direct appeal, Fair clearly is no longer good law because Griffith decrees that newly announced federal constitutional principles do apply on direct appeals. Bluntly: Fair was not a case *601about federal constitutional rights, was not a case about collateral challenges, and furthermore has no remaining vitality and force insofar as federal constitutional retroactivity analysis on direct appeal is concerned. In sum, the majority’s attempt to bootstrap the Teague standard into Oregon law by invoking Fair is ultimately unavailing.3
Nor does our own prior case law mandate that we apply the federal habeas corpus standard enunciated in Teague. The majority states that “we are to adhere to federal precedent in analyzing the retroactivity of newly announced federal constitutional rules asserted as grounds” for post-conviction relief. 184 Or App at 586. However, the cases to which the majority points in support of that proposition predate the United States Supreme Court’s decisions in Griffith and Teague, in which the Court changed its approach to the question of retroactivity and determined that retroactivity depended on the type of proceeding rather than on the nature of the constitutional right.4
*602For example, in Kellotat v. Cupp, 78 Or App 61, 714 P2d 1074 (1986), we addressed the propriety of retroactive application of a newly announced rule of law under Article I, section 20, of the Oregon Constitution. We reiterated the language from Fair that we would look to the case law of the United States Supreme Court “only for guidance,” 78 Or App at 66, and ultimately concluded that the right in question was not one to be applied retroactively, id. at 67. But Kellotat, as well as the many cases on which the majority relies, all rest on an underlying assumption that the retroactive application of a new rule of constitutional law depends on the nature of the substantive right involved. As noted, the United States Supreme Court abandoned that approach in Griffith and Teague in favor of a model of analysis that is radically different. While Fair, Kellotat, and the various cases on which the majority relies do stand for the proposition that we looked for guidance from the Court’s “substantive” retroactivity decisions, they simply do not write a “blank check” that we would, in the future, adhere to any radically different approach to retroactivity jurisprudence that the United States Supreme Court might later espouse.
Rather, to the contrary, Palmer v. State of Oregon, 318 Or 352, 357, 867 P2d 1368 (1994), indicates that post-conviction relief is, in fact, available “where the right subsequently sought to be asserted was not generally recognized to be in existence at the time of trial.” (Internal quotation marks omitted.) In Long, which I discuss at length below, 184 Or App at 607, we explicitly stated that “when a new constitutional principle has been articulated between the time of a petitioner’s direct appeal and the post-conviction proceeding, *603a claim based on the new constitutional principle will be considered in the post-conviction proceeding.” Long, 166 Or App at 101 (emphasis added). That principle was not novel — and, indeed, reiterated our commitment to a long line of precedent. See, e.g., Twitty v. Maass, 96 Or App 631, 633, 773 P2d 1336 (1989) (“when a new constitutional principle is articulated between the time of a petitioner’s direct appeal and the petition for post-conviction relief, a claim based on the new constitutional principle will be considered in the post-conviction proceeding”); Myers v. Cupp, 49 Or App 691, 695, 621 P2d 579 (1980), rev den, 290 Or 491 (1981) (“where a new constitutional principle is recognized between the time of a petitioner’s direct appeal and his petition for post-conviction relief and where he could not have reasonably asserted his claim based on this principle on appeal, it will subsequently be considered on a petition for post-conviction relief’).
Long is the law of this court. The majority does not propose to overrule Long and its antecedents or offer any persuasive reason for abandoning that line of cases in favor of some federally announced, and potentially variable, standard of collateral cognizability of federal claims. Long, not Teague, controls — and, as described below, 184 Or App at 606-18, under Long, petitioner’s Apprendi-based claims are cognizable under ORS 138.530(1).
II. EVEN IF TEAGUE v. LANE WERE APPLICABLE, APPRENDI ANNOUNCES A ‘WATERSHED” RULE THAT IS SUBJECT TO RETROACTIVE APPLICATION IN COLLATERAL PROCEEDINGS
As noted, 184 Or App at 596-97, in Teague, the Court carved out an exception to its general rule of nonretroactivity. A new constitutional rule of criminal procedure will be given retroactive application on collateral review where that rule “requires the observance of those procedures that are implicit in the concept of ordered liberty.” Teague, 489 US at 307 (internal quotation marks and ellipses omitted). That is, retroactive application is required where the newly announced principle represents a “watershed rule[ ] fundamental to the integrity of the criminal proceeding.” Sawyer v. Smith, 497 US 227, 229, 110 S Ct 2822, 111 L Ed 2d 193 (1990). A new rule of law constitutes such a “watershed” *604when it “alter[s] our understanding of the bedrock procedural elements that must be found to vitiate the fairness of a particular conviction.” Teague, 489 US at 311 (internal quotation marks omitted; emphasis in original).
Thus, the essential quality of a “watershed” new rule for Teague purposes is that it fundamentally implicates and enhances the accuracy of fact-finding procedures in the criminal trial process. See Teague, 489 US at 313 (“watershed rule[s]” refer to “those new procedures without which the likelihood of an accurate conviction is seriously diminished”).
Here, the majority implicitly, and correctly, acknowledges that the Apprendi rule is a “new” rule for Teague retroactivity purposes. 184 Or App at 587. However, the majority concludes that Apprendi’s fundamental feature — requiring that sentencing enhancement facts that had previously been found by judges by a preponderance of the evidence must be specifically determined by a jury beyond a reasonable doubt — does not constitute a “watershed rule” within Teague’s exception. 184 Or App at 591-92.
Even assuming that Teague controls the cognizability of Apprendi-based claims under Oregon’s post-conviction relief statutes, I disagree: Apprendi radically altered deeply grounded premises about the respective functions of court and jury in criminal sentencing. Moreover, at least with respect to the determination of “future dangerousness” for sentencing enhancement purposes, the new procedures Apprendi mandates qualitatively enhance the accuracy of the criminal fact-finding process.
My assessment oí Apprendi as announcing a “watershed rule” is inextricably intertwined with my consideration, below, of the merits of petitioner’s argument that certain features of Oregon’s dangerous offender sentencing statutes cannot be reconciled with Apprendi. I defer much of my substantive discussion of Apprendi to my analysis of the merits. See 184 Or App at 619-22. However, for present purposes, I emphasize two salient considerations:
First, under Apprendi, an offender’s future dangerousness must be determined by a jury, beyond a reasonable *605doubt, rather than by a judge, by a preponderance of the evidence. Thus, Apprendi alters not only who makes that determination — which can result in an additional 10 years’ incarceration — but also the quantum of proof required to establish that fact. See Apprendi, 530 US at 476 (noting that “[a]t stake in this case are constitutional protections of surpassing importance”).
Second, and in a related sense, that change in the law qualitatively alters and enhances the accuracy of the fact-finding process. See In re Winship, 397 US 358, 363, 90 S Ct 1068, 25 L Ed 2d 368 (1970) (characterizing “reasonable doubt” standard as “a prime instrument for reducing the risk of convictions resting on factual error”).
In sum, Apprendi transforms the traditional landscape of criminal sentencing and goes to the very constitutional “bedrock” of a defendant’s right to have a jury determine beyond a reasonable doubt a fact that will result in the imposition of additional years of incarceration beyond the statutory maximum. I acknowledge, as the majority emphasizes, 184 Or App at 588-90, that none of the federal courts that has considered the question has ultimately deemed Apprendi to pronounce a “watershed rule.” But, with respect, we are not bound by those holdings, and I do not find them persuasive, at least with respect to the question before us.5
The Supreme Court has not yet spoken on Apprendi's retroactivity. Pending the Court’s direction, I would conclude that, as applied to the determination of *606“future dangerousness” under Oregon’s dangerous offender statutes, Apprendi announces a “watershed rule.”
III. PROCEDURAL BARS: PETITIONER’S CLAIM FOR INEFFECTIVE ASSISTANCE OF COUNSEL IS TIME-BARRED. PETITIONER’S REMAINING CLAIMS ARE NEITHER TIME-BARRED NOR PRECLUDED UNDER THE “SUCCESSIVE-PETITION” BAR
I turn, then, to whether petitioner’s present claims are procedurally barred under Oregon’s post-conviction relief statutes.6 Specifically, are petitioner’s claims time-barred under ORS 138.510(3) or barred under ORS 138.550(3) because petitioner previously had sought post-conviction relief? ORS 138.510(3) provides:
“A petition [for post-conviction relief] must be filed within two years of the following, unless the court on hearing a subsequent petition finds grounds for relief asserted which could not reasonably have been raised in the original or amended petition:
“(a) If no appeal is taken, the date the judgment or order on the conviction was entered in the register.
“(b) If an appeal is taken, the date the appeal is final in the Oregon appellate courts.”
(Emphasis added.) Petitioner’s petition for post-conviction relief was filed in March 2000, more than two years after the date of his conviction, after appeal to the Oregon appellate courts, became final. Thus, his petition would be barred under ORS 138.510(3)(b) unless he asserted grounds for relief “which could not reasonably have been raised in the original or amended petition.”
In a similar manner, ORS 138.550(3) provides, in part:
“All grounds for relief claimed by petitioner in a petition [for post-conviction relief] must be asserted in the original or amended petition, and any grounds not so asserted are deemed waived unless the court on hearing a subsequent petition finds grounds for relief asserted therein which could *607not reasonably have been raised in the original or amended petition.”
(Emphasis added.) Because petitioner previously had sought post-conviction relief before seeking relief in the present case, his present petition is barred unless it asserts grounds for relief “which could not reasonably have been raised in the original or amended petition.” Id.
The question, then, regardless of whether ORS 138.510(3) or ORS 138.550(3) applies, is whether the grounds for relief that petitioner asserts in the present case “could not reasonably have been raised in the original or amended petition.” In Long, 166 Or App at 101, we interpreted identical language from ORS 138.510(2) (1991), as follows:
“[W]hen a new constitutional principle has been articulated between the time of a petitioner’s direct appeal and the post-conviction proceeding, a claim based on the new constitutional principle will be considered in the post-conviction proceeding even though it was not raised at trial or on appeal. The same result does not necessarily follow where the constitutional principle is an acknowledged one, and the uncertainty is in its scope or application to a particular circumstance. The touchstone is not whether a particular question is settled, but whether it reasonably is to be anticipated so that it can be raised and settled accordingly. The more settled and familiar a constitutional or other principle on which a claim is based, the more likely the claim reasonably should have been anticipated and raised. Conversely, if the constitutional principle is a new one, or if its extension to a particular statute, circumstance, or setting is novel, unprecedented, or surprising, then the more likely the conclusion that the claim reasonably could not have been raised.”
(Emphasis in original; citations omitted.)
Petitioner asserts that the post-conviction court erred in concluding that his present claims should reasonably have been raised in his original petition because the rule of law that was foreshadowed in Jones v. United States, 526 US 227, 119 S Ct 1215, 143 L Ed 2d 311 (1999), and announced in Apprendi, was a new constitutional principle. Defendant responds that the trial court correctly concluded *608that petitioner’s present claims are barred, and that petitioner’s claims reasonably could have been raised earlier, because the constitutional principle enunciated by the Court in Jones and Apprendi is, according to defendant, “almost identical” to the principle enunciated by the Oregon Supreme Court in State v. Quinn, 290 Or 383, 623 P2d 630 (1981), and State v. Wedge, 293 Or 598, 652 P2d 773 (1982).
As noted above, petitioner asserted two bases for his post-conviction claims. First, he asserted that, because the Due Process Clause requires that the facts upon which an enhanced sentence beyond the statutory maximum is based be pleaded in the indictment and found by the jury, his sentence was unconstitutional.7 Second, he asserted that he received constitutionally inadequate assistance of trial counsel because trial counsel faded to argue that the factors supporting the enhanced dangerous offender sentence should have been pleaded in the indictment and found by the jury.8
With respect to inadequate assistance, the majority concludes that petitioner’s complaint fails because his Apprendi-based claims are sufficiently “new” that trial counsel cannot reasonably be faulted for “not anticipating] that the law would develop as it did.” 184 Or App at 592.1 agree. Consequently, I concur in that aspect of the majority’s disposition.
Petitioner’s other claim, that he is entitled to post-conviction relief on the ground that his sentence was unconstitutionally excessive, ORS 138.530(1)(c), requires a very different analysis. Because of the nature of the escape clause provisions of ORS 138.510(3) and ORS 138.550(3), a determination of whether petitioner is excused from the time constraints provided by those statutes necessarily requires some analysis of the substance of his claim.
*609Defendant was sentenced as a dangerous offender pursuant to ORS 161.725(1) (1987), which allowed the court to impose an indeterminate sentence of 30 years if “[t]he defendant is being sentenced for a Class A felony, and the court finds that the defendant is suffering from a severe personality disorder indicating a propensity toward criminal activity.” ORS 161.735 (1987) required a court, before imposing a dangerous offender sentence pursuant to ORS 161.725 (1987), to obtain and consider a presentence investigation report and an examination by a psychiatrist or psychologist. The statutory scheme clearly indicated that the sentencing court, and not a jury, would make the determination whether the defendant was suffering from a severe personality disorder indicating a propensity for criminal activity.
Here, defendant argues — and the post-conviction court agreed — that petitioner could and should have made a constitutional challenge. to that aspect of the dangerous offender sentencing scheme in a timely petition for post-conviction relief, because Apprendi and Jones did not establish a new principle of constitutional law. Defendant argues that those cases were, in fact, substantially similar to Oregon decisions that predated petitioner’s conviction. In so arguing, defendant first invokes Quinn.
In Quinn, the defendant challenged the constitutionality of a murder sentencing scheme that permitted the jury to convict him of murder based on a specified mental state, but allowed the sentencing court to impose a sentence of death if the court found that the murder “was committed with a greater culpable mental state than that found by the jury.” 290 Or at 404. The court concluded that the statutory scheme was unconstitutional under Article I, section 11, of the Oregon Constitution:
“We have upheld other enhanced penalty statutes even though they required additional post-trial findings by the court as a basis for a greater sentence. In particular, we have upheld the former Habitual Criminal Act and sexually dangerous offender statutes over challenges that the procedures violated the right to trial by jury of the facts upon which enhanced punishment was to be based. The difference between those statutes and [the murder sentencing scheme at issue], however, is found in the simple principle *610that the facts which constitute the crime are for the jury and those which characterize the defendant are for the sentencing court.”
Quinn, 290 Or at 405 (emphasis added). The court indicated that, while factors concerning the “kind and character” of a defendant may properly be considered by the sentencing court, factors that “go to the criminal acts for which defendant is to be punished” must be decided by a jury. Id. at 406 (internal quotation marks omitted). The court further indicated that the result was the same under either Article I, section 11, of the Oregon Constitution or the Due Process Clause of the United States Constitution. Id. at 406 n 9.
Wedge addressed a similar issue, which concerned a judicially imposed minimum sentence for use of a firearm during the commission of the crime for which the defendant was sentenced. There, it was unclear from the indictment whether the state was alleging that the defendant used a knife or a gun in the course of a first-degree robbery. 293 Or at 602-03. While the use of either weapon would support the jury’s robbery verdict, the sentencing court imposed a gun minimum based on its own finding that the defendant had used a firearm. Id. at 603. The court reiterated verbatim its test from Quinn that “facts which constitute the crime are for the jury and those which characterize the defendant are for the sentencing judge.” Id. at 607. It held: “The use or threatened use of a firearm is a finding that goes to the criminal act for which this defendant is punished, and thus is closer to an element of the crime than to a characterization of the defendant.” Id.9
In State v. Mitchell, 84 Or App 452, 734 P2d 379, rev den, 303 Or 590 (1987), we applied the Supreme Court’s test from Quinn and Wedge to a portion of the dangerous offender sentencing scheme. In Mitchell, the defendant had *611been sentenced as a dangerous offender under a different provision than the one at issue in the present case. The provision at issue in Mitchell permitted a sentencing court to impose a dangerous offender sentence if
“[t]he defendant is being sentenced for a felony that seriously endangered the life or safety of another, has been previously convicted of a felony not related to the instant crime as a single criminal episode, and the court finds that the defendant is suffering from a severe personality disorder indicating a propensity toward criminal activity.”
ORS 161.725(2) (1987). The issue in Mitchell was whether the defendant was entitled to a jury determination that the crime for which he was being sentenced was one that “seriously endangered the life or safety of another.” 84 Or App at 454. Applying Quinn and Wedge, we concluded that the defendant was entitled to a jury determination of that issue: “The critical fact does not relate to the offender’s status or character, which the court is privileged to find, but relates particularly to the crime for which the offender is being sentenced.” Id. at 457.
Even before Mitchell, we had held on two occasions that the Due Process Clause did not require proof beyond a reasonable doubt that a criminal defendant had a “severe personality disorder indicating a propensity toward criminal activity.” State v. Hunter, 58 Or App 99, 109, 647 P2d 943 (1982), rev den, 294 Or 391 (1983); State v. Sanders, 35 Or App 503, 506-09, 582 P2d 22 (1978), rev den, 285 Or 195 (1979). Mitchell contained no suggestion that those earlier cases were in any way called into question by the court’s application of the principle announced in Quinn and Wedge to the dangerous offender statute.
In sum, at the time petitioner’s sentence was imposed — and at the time his initial post-conviction petition was timely filed — Oregon courts had held that the Due Process Clause did not require that the sentence-enhancing fact that a defendant suffered a “severe personality disorder indicating a propensity toward criminal activity” be proved beyond a reasonable doubt. However, our cases did hold that Article I, section 11, as well as the Due Process Clause, *612required that the finding that the crime for which the defendant was being sentenced “seriously endangered the life or safety of another” be found by a jury beyond a reasonable doubt, because the latter “relates particularly to the crime for which the offender is being sentenced.” Mitchell, 84 Or App at 457. Thus, Oregon case law — the Supreme Court’s decisions in Quinn and Wedge, and our decisions in Sanders, Hunter, and Mitchell — drew a clear distinction between facts that constitute the crime for which the defendant is being sentenced, for which a jury finding is required, and facts that “characterize the defendant,” which may be found by the sentencing court.
Given the contemporary consensus of Oregon appellate precedent, a post-conviction petitioner could reasonably have viewed the argument that he was entitled to a jury’s determination as to whether he suffered from a severe personality disorder indicating a propensity toward crimes that seriously endanger the life or safety of another as foreclosed by settled precedent. As noted, Hunter and Sanders rejected similar due process arguments. Mitchell, Quinn, and Wedge, which concerned only facts that related to the crime itself, did not call into question our conclusion in Hunter and Sanders. Rather, they implicitly, but necessarily, confirmed the correctness of our holdings by drawing, and reiterating, the distinction between circumstances of the crime and the personal characteristics of the offender.
Defendant does not suggest that, regardless of how the Apprendi rule squares with the analyses of Quinn and Wedge, petitioner could have “reasonably anticipated” the Apprendi rule based on federal precedent. Defendant’s reticence in that regard is unsurprising. Given the state of then-contemporaneous federal law, the Apprendi rule could not have been reasonably anticipated until the late 1990s. That conclusion comports with the lengthy discussions of United States Supreme Court precedent (or the lack thereof) in Apprendi itself, as well as its immediate antecedents, Almendarez-Torres v. United States, 523 US 224, 118 S Ct 1219, 140 L Ed 2d 350 (1998), and Jones, 526 US at 231-52, which I describe at length below. That understanding of the pre-Apprendi state of federal law is also corroborated by postApprendi federal appellate decisions addressing the Teague *613rule of retroactivity for federal habeas corpus cases. As noted above, federal habeas corpus relief may be granted under limited circumstances, even if the petitioner failed to raise and litigate the issue in the underlying criminal proceeding, if the issue concerns a newly announced constitutional principle. Teague, 489 US at 311.
Courts determining whether a constitutional principle qualifies as “newly announced” for purposes of Teague have looked at factors similar to those we consider in. determining whether a rule of law reasonably could have been anticipated. For the most part, federal courts have concluded that the rule announced in Apprendi and foreshadowed in Jones, was, in fact, a newly announced principle of constitutional law. In McCoy v. United States, 266 F3d 1245, 1256 (11th Cir 2001), cert den, _ US _ , 153 L Ed 2d 183 (2002), the court stated:
“Apprendi established a new rule of criminal procedure, one that was not dictated by precedent existing before the Apprendi decision was released. Until it was announced, all circuits had been upholding sentences that were greater than the otherwise applicable maximum sentences based on drug quantity not charged in the indictment, submitted to the jury, and proved beyond a reasonable doubt.”
See also United States v. Moss, 252 F3d 993, 997-98 (8th Cir 2001), cert den, 534 US 1097 (2002) (Apprendi rule concerning statutory maximum was “obviously” a new constitutional principle because the result was not dictated by precedent existing at the time the defendant’s conviction became final); Jones v. Smith, 231 F3d 1227, 1236 (9th Cir 2000) (Apprendi announced a new rule for purposes of Teague because it broke new ground and imposed new obligations on the government).10
I return to Apprendi’s immediate antecedents. The first of those cases was Almendarez-Torres, decided in 1998, *614in which the Court considered whether a statute that provided that a person who had been convicted of an aggravated felony and who had returned to the United States after having been deported was subject to up to 20 years’ imprisonment described a separate crime from merely returning to the country after an initial deportation (which otherwise was punishable by a sentence of no more than two years) — or whether consideration of the defendant’s commission of an aggravated felony was only a sentence enhancement factor. 523 US at 226. The Court, in a 5-4 decision, relied on McMillan v. Pennsylvania, 477 US 79, 87, 106 S Ct 2411, 91 L Ed 2d 67 (1986) (see 184 Or App at 610 n 9), and upheld the statute as merely concerning a penalty enhancement rather than creating a separate crime requiring proof of an additional element. 523 US at 242-47.
In reaching its conclusion, the Almendarez-Torres majority relied heavily on Patterson v. New York, 432 US 197, 97 S Ct 2319, 53 L Ed 2d 281 (1977), for the conclusion that “the Constitution requires scarcely any sentencing factors to be treated [as elements of a crime]. ” Almendarez- Torres, 523 US at 241. The decision in Almendarez-Torres drew a sharp dissent that pointed out the arbitrariness with which a legislative body could choose to define “elements” of crimes versus “sentencing enhancement factors.” Id. at 252-53 (Scalia, J., dissenting). It was the Almendarez-Torres dissent that first suggested that a distinction noted in McMillan (see 184 Or App at 610 n 9), but not decided in that case — that, perhaps, a fact that increased the maximum penalty for a crime should need to be proved beyond a reasonable doubt — should be the dispositive question. Almendarez-Torres, 523 US at 254 (Scalia, J., dissenting). The Almendarez-Torres majority, however, explicitly rejected the notion that McMillan could be read to stand for that proposition, noting that “McMillan said that the petitioners’ argument in that case would have had ‘more superficial appeal’ if the sentencing fact ‘exposed them to greater or additional punishment,’ ” and concluded that the greater additional punishment distinction was not determinative. Almendarez-Torres, 523 US at 245 (quoting McMillan, 477 US at 88) (emphasis in Almendarez-Torres).
In Jones, decided the following year, the Court again split in a 5-4 decision. However, the four dissenters from *615Almendarez-Torres were in the majority, joined by one other justice. The question in Jones was whether a federal carjacking statute described one crime, or three different crimes with three different maximum penalties: 15 years for carjacking; 25 years if serious bodily injury results; up to life imprisonment if death results. 526 US at 229-30. The indictment in Jones did not allege which section of the statute would apply, and the jury was instructed only as to the elements for carjacking, with no mention of serious bodily injury or death. Id. at 230-31. After the jury returned a verdict of guilty, the court at sentencing made a finding that the crime had involved serious bodily injury and imposed a 25-year sentence. Id. The Court observed that both due process and the Sixth Amendment require that “any fact (other than prior conviction) that increases the maximum penalty for a crime must be charged in an indictment, submitted to a jury, and proven beyond a reasonable doubt.” Id. at 243 n 6. The Court, suggesting that there were serious constitutional concerns with construing the carjacking statute merely as a sentencing enhancement statute, avoided a decision on the constitutional question by construing the statute as creating three separate offenses. Id. at 251-52.
In 2000, the Supreme Court decided Apprendi. There, the defendant was convicted of, inter alia, possession of a firearm for an unlawful purpose, which, under New Jersey law, carried a maximum sentence of 10 years. 530 US at 468. The sentencing court found, by a preponderance of the evidence, that the defendant committed the crime because of racial bias. Id. at 471. Under New Jersey law, the defendant was eligible for an extended term of imprisonment of up to 20 years based on the court’s finding of racial bias. Id. at 469. The defendant received an actual sentence of 12 years for the crime. Id. at 471. The question before the Court was “whether [the defendant] had a constitutional right to have a jury find such bias on the basis of proof beyond a reasonable doubt.” Id. at 475-76. The Court announced the following rule of law:
“Other than the fact of a prior conviction, any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury, and proved beyond a reasonable doubt.”
*616Id. at 490 (emphasis added).
We recently summarized the Court’s Apprendi decision as follows:
“The Court noted that, at the time of the nation’s founding, ‘[a]ny possible distinction between an “element” of a felony offense and a “sentencing factor” was unknown’ and that, ‘[a]s a general rule, criminal proceedings were submitted to a jury after being initiated by an indictment containing all the facts and circumstances which constitute the offensef.]’ [530 US] at 478 (internal quotation omitted). The Court explained that, as a result of those established practices and principles, the substantive criminal law ‘tended to be sanction-specific,’ id. at 479, and, although applicable statutes might prescribe a range of penalties, trial judges otherwise typically have had little discretion in imposing the sentence for a particular offense. Id. at 479-80.
“The Court stated that the resulting ‘historic link’ between a conviction and the sentence for that conviction ‘highlight[s] the novelty of a legislative scheme that removes the jury from the determination of a fact that, if found, exposes the criminal defendant to a penalty exceeding the maximum he would receive if punished according to the facts reflected in the jury verdict alone.’ Id. at 482-83 (emphasis in original). After reviewing its earlier case law relating to facts affecting the severity of a defendant’s sentence — particularly Jones v. United States, 526 US 227, 119 S Ct 1215, 143 L Ed 2d 311 (1999)—the Court concluded that, ‘[o]ther than the fact of a prior conviction, any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury, and proved beyond a reasonable doubt.’Id. at 490.”
State v. Crain, 177 Or App 627, 633-34, 33 P3d 1050 (2001), rev den, 334 Or 76 (2002) (footnote omitted; emphasis in original).
In sum, Apprendi held as a matter of constitutional law what the Court had suggested in the nonconstitutional Jones case a year earlier and what the Almendarez-Torres dissent had proposed two years earlier: Facts, other than facts of prior conviction, must be decided by a jury under the “reasonable doubt” standard if the existence of such facts are *617used to increase the penalty for a crime beyond the prescribed statutory maximum. See also Ring v. Arizona, 536 US 584, 122 S Ct 2428, 153 L Ed 2d 556, 572-77 (2002) (reiterating holding from Apprendi and applying it to invalidate sentence of death imposed based on aggravating circumstances found by a court rather than a jury).
I return to this case. The question before us is whether, given the state of the law at the time petitioner timely filed his first petition for post-conviction relief, the constitutionality of his dangerous offender sentence enhancement could “reasonably have been raised.” ORS 138.510(3); ORS 138.550(3). As noted above, the answer depends on whether the constitutional principle at issue “is a new one, or if its extension to a particular statute, circumstance, or setting is novel, unprecedented, or surprising.” Long, 166 Or App at 101.
Here, the constitutional principle underlying petitioner’s Apprendi-based claim — viz., that due process requires proof beyond a reasonable doubt as to every fact necessary to constitute the crime — is not new. See Winship, 397 US at 364. While that basic principle is well established, its application to “sentencing enhancement factors” as opposed to “facts constituting a crime,” was not obvious — nor were the Court’s conclusions about this subject in Jones and Apprendi foreshadowed in its earlier case law.11 Indeed, the Oregon cases discussed above provided more helpful guidance as to this issue than did the cases of the United States Supreme Court. Moreover, those Oregon decisions had settled the question adversely to petitioner.
*618As recounted above, the rule of law from Quinn, Wedge, and Mitchell established that (1) sentence enhancement factors that related to the commission of the crime itself must be decided by the jury, but (2) sentence enhancement factors “which characterize [a] defendant are for the sentencing court.” Quinn, 290 Or at 405. A finding under the dangerous offender statute that a defendant has a “severe personality disorder indicating a propensity toward criminal activity” does not relate to the circumstances of the crime itself. Rather, it is, literally, a characterization of the defendant being sentenced — that is, of the defendant’s personal characteristics. Moreover, as explained above, we specifically held, both in Hunter and in Sanders, that the “severe personality disorder indicating a propensity toward criminal activity’ factor used in the dangerous offender statute did not need to be proved beyond a reasonable doubt. Hunter, 58 Or App at 109; Sanders, 35 Or App at 507.
In sum, a post-conviction petitioner in petitioner’s position could not, and would not, have “reasonably’ anticipated petitioner’s present constitutional claims before Jones and Apprendi. See Long, 166 Or App at 101. Sanders and Hunter expressly rejected such a claim, and Quinn, Wedge, and Mitchell corroborated that rejection. Nothing in the Oregon case law — and defendant points to no contemporaneous federal case law — suggested that such a claim had even the slightest likelihood of success. Contrary to defendant’s present position, the rule of Apprendi is hardly “almost identical” to the analysis of Quinn, Wedge, et al. With the single explicit exception of an offender’s prior convictions, see Apprendi, 530 US at 490, nothing in Apprendi endorses Quinn and Wedge’s sweeping differentiation between the circumstances of the crime and the characteristics of the offender. Indeed, Apprendi’s bedrock holding explodes any such distinction.
I thus conclude that the grounds for relief asserted in petitioner’s petition for post-conviction relief “could not reasonably have been raised in the original or amended petition.” ORS 138.510(3); ORS 138.550(3). Consequently, the trial court should not have dismissed petitioner’s petition on the grounds that it was barred under ORS 138.510(3) and ORS 138.550(3).
*619IV. THE MERITS: PETITIONER HAS STATED LEGALLY COGNIZABLE CLAIMS FOR POST-CONVICTION RELIEF UNDER APPRENDI
The trial court here not only granted the state’s motion to dismiss on procedural grounds, but also ruled in the alternative:
“Even if petitioner’s filing is deemed timely, neither the decision in Apprendi or Jones supports petitioner’s allegations. Petitioner was convicted of a Class A felony, and the sentencing court did not condition the dangerous offender sentence on any findings that could be characterized as facts of the crime. The decisions in Apprendi and Jones are therefore inapplicable in the present case.”
Thus, the court reasoned that petitioner’s claim failed on the merits under Apprendi, apparently based on the conclusion that Apprendi drew the same distinction drawn by the courts in Quinn, Wedge, and Mitchell, i.e., that “facts which constitute the crime are for the jury and those which characterize the defendant are for the sentencing court.” Quinn, 290 Or at 405. As discussed above, however, Apprendi is not essentially the same as Quinn, Wedge, and Mitchell.
Defendant asserts that Apprendi must be understood to establish a right to “a jury trial on a disputed fact, and consequently to have the fact alleged in the indictment, only if the fact both serves to aggravate the crime, by requiring or authorizing the imposition of an enhanced sentence, and relates to the manner or circumstances in which [the criminal defendant] committed the crime.” (Emphasis in original.) In making that assertion, defendant notes that Apprendi and Jones both concerned circumstances of the crime and suggests that the Court’s holding must be limited to such circumstances. The only language in the Apprendi opinion to which defendant points in support of his proposed rule of law is the following:
“New Jersey’s reliance on Almendarez-Torres is also unavailing. The reasons supporting an exception from the general rule for the statute construed in that case do not apply to the New Jersey statute. Whereas recidivism ‘does not relate to the commission of the offense’ itself, 523 US at 230, 244, New Jersey’s biased purpose inquiry goes precisely *620to what happened in the ‘commission of the offense. ’ Moreover, there is a vast difference between accepting the validity of a prior judgment of conviction entered in a proceeding in which the defendant had the right to a jury trial and the right to require the prosecutor to prove guilt beyond a reasonable doubt, and allowing the judge to find the required fact under a lesser standard of proof.”
Apprendi, 530 US at 496 (emphasis added). In essence, defendant is positing that the Court’s broad directive that, “[o]ther than the fact of a prior conviction, any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury, and proved beyond a reasonable doubt,” id. at 490 (emphasis added), does not mean what it says — and that “any fact” really means “any fact relating to the commission of an offense.”
I disagree. First, whatever the Court might otherwise have intended to express regarding New Jersey’s misreliance on Almendarez-Torres, “any fact,” other than prior convictions, means “any fact.” Not “some facts.” Not “any fact relating solely to the commission of the offense.” We are not free to engage in revisionist reconstruction of the Court’s plain language. Nor are we free to deprecate that explicit directive as dictum. Rather, as confirmed by the Court’s recent decision in Ring, we must assume that the Court meant what it said in Apprendi.
Further, the Court’s “any fact” language comports contextually with the totality of its analysis. The Court discussed at length the history of sentencing practices and sentencing enhancements, as well as its prior case law that touched on the subject. A significant focus of the Court’s discussion was on whether the “sentence enhancement” elevated the crime beyond the prescribed statutory maximum. For example, after reviewing that history, the Court concluded, “nothing * * * suggests that it is impermissible for judges to exercise discretion — taking into consideration various factors relating both to offense and offender — in imposing a judgment within the range prescribed by statute.” Apprendi, 530 US at 481 (emphasis in original).
Finally, the excerpt on which defendant relies concerned the Court’s treatment of Almendarez-Torres, and the *621Court had a number of other things to say about that decision. In Apprendi, the Court described Almendarez-Torres as an “exceptional departure” from the practice of requiring proof beyond a reasonable doubt of facts used to enhance penalties. Id. at 487. It described the case at length, and stated:
“Both the certainty that procedural safeguards attached to any ‘fact’ of prior conviction, and the reality that [the defendant] did not challenge the accuracy of that ‘fact’ in his case, mitigated the due process and Sixth Amendment concerns otherwise implicated in allowing a judge to determine a ‘fact’ increasing punishment beyond the maximum of the statutory range.”
Id. at 488 (footnote omitted). The Court concluded:
“Even though it is arguable that Almendarez-Torres was incorrectly decided, and that a logical application of our reasoning today should apply if the recidivist issue were contested, [the defendant] does not contest the decision’s validity and we need not revisit it for purposes of our decision today to treat the case as a narrow exception to the general rule we recalled at the outset.”
Id. at 489-90 (footnote omitted; emphasis added).
The Court, thus, entertained doubts about the vitality of Almendarez-Torres in light of its broad holding, but maintained the rule announced in that case concerning “facts” relating to recidivism as the “narrow exception” to the broad rule it was announcing. In short, while AlmendarezTorres may have made something of the distinction between facts that relate to the commission of an offense and other types of facts, the Apprendi Court did not. It discussed the “fact” of recidivism — which clearly does not relate to the commission of the offense being sentenced — and maintained a “narrow exception” to its broad rule on the ground that the “fact” of the previous conviction was one for which the defendant had been entitled to a jury trial, and proof beyond a reasonable doubt. To put it bluntly, had the Court intended to announce the ride of law posited by defendant here — that the rule applies only where a sentence beyond the statutory maximum is being imposed and the fact on which the enhanced sentence is based relates to the circumstances of the crime — then it would have had no reason to engage in a *622lengthy discussion of Almendarez-Torres. The Court could simply have stated that its new rule had no application to the facts of Almendarez-Torres because the recidivism at issue in that case did not relate to the circumstances of the crime. But it did not do so.
I take the Court’s announcement in Apprendi at face value: “Other than the fact of a prior conviction, any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury, and proved beyond a reasonable doubt.” 530 US at 490 (emphasis added). In the underlying criminal proceeding, the trial court found, by a preponderance of the evidence, that petitioner was suffering from a severe personality disorder indicating a propensity toward criminal activity, and, as a consequence of that finding, sentenced petitioner to a maximum indeterminate sentence of 30 years, which was 10 years more than the maximum sentence that petitioner could have received in the absence of such a finding. That finding concerned a fact other than the fact of a prior conviction; it also “increase[d] the penalty for [the] crime beyond the prescribed statutory maximum.” Id.12
In closing, I acknowledge that application of the rule of law announced in Apprendi is fraught with potential difficulties — and potential irony. That rule may confer a mixed blessing, at best, for criminal defendants. As human and legal experience often cautions, “Be careful of what you ask for; you might get it.”
Justice Breyér, in his dissent in Apprendi, highlights one glaring potential incongruity: Requiring a criminal defendant to defend against aggravating factors at trial “could easily place the defendant in the awkward (and conceivably unfair) position of having to deny he committed the crime yet offer proof about how he committed it, e.g., T did not sell drugs, but I sold no more than 500 grams.’ ” 530 US at *623557 (Breyer, J., dissenting); see also Nancy J. King and Susan R. Klein, Essential Elements, 54 Vand L Rev 1467, 1495 (2001) (suggesting that, unless bifurcated trials are authorized for a jury’s consideration of sentence enhancement factors, a jury that hears evidence of prejudicial aggravating circumstances may be more inclined to convict a defendant).
The dangerous offender sentencing scheme at issue here exemplifies the difficulty posed by having a jury decide sentence-enhancement facts. Regardless of the nature of the felony for which a defendant is being tried, the dangerous offender statute requires a finding that the defendant is suffering from a severe personality disorder indicating a propensity toward crimes that seriously endanger the life or safety of another. In the normal course of things, such “propensity” evidence is not admissible and is considered highly prejudicial. See, e.g., OEC 404(3) (generally, evidence of “other crimes, wrongs or acts is not admissible to prove the character of a person in order to show that the person acted in conformity therewith”). Thus, under Apprendi, a defendant facing a dangerous offender sentence could look forward to having the jury consider not only evidence pertaining to commission of the underlying crime, but also evidence of his or her psycho-social history bearing on propensity to commit crimes.
In sum, Apprendi seems to sanction the presentation of potentially highly inflammatory evidence pertaining to a defendant’s personal characteristics before the jury ever renders a verdict as to guilt. Whether, or how, that can be squared with the “fundamental fairness” requirements of the Due Process Clause awaits another case. Cf. Ring, 536 US at _ , 153 L Ed 2d at 579 (Scalia, J., concurring) (suggesting that, although a jury must find the facts necessary to sentencing enhancement factors, it could do so either in a separate penalty-phase proceeding or “by placing the aggravating-factor determination (where it logically belongs anyway) in the guilt phase”).
I thus conclude that the post-conviction trial court erred in determining that petitioner’s claims, other than those alleging ineffective assistance of counsel, failed to state *624a claim for relief under Apprendi. I respectfully dissent from the majority’s affirmance of the trial court’s dismissal of those claims.
In his opinion beginning at 401 US at 675, Justice Harlan concurs in the judgment in Mackey, but also dissents from "Williams v. United States, 401 US 646, 91 S Ct 1148, 28 L Ed 2d 388 (1970), and concurs in the judgment in Elkanich u. United States, decided in the same opinion with Williams. For ease of reference, we refer to Justice Harlan’s opinion in those cases as a “dissent.” All citations to Mackey in this opinion are to Justice Harlan’s opinion as described above.
That is not to say that state post-conviction statutes have no constitutional underpinnings. Article I, section 23, of the Oregon Constitution provides that“[t]he *599privilege of the writ of habeas corpus shall not be suspended unless in case of rebellion, or invasion the public safety require it.” ORS 34.330(3), in turn, provides that a person who is eligible to obtain post-conviction relief pursuant to ORS 138.510 to 138.680 may not prosecute a writ of habeas corpus. Thus, post-conviction proceedings do indirectly provide remedies that are mandated by the state constitution.
The majority also does not explain why, if Fair is controlling, it is not using the Linkletter standard followed in that case, rather than the Teague standard that came much later. The majority offers no support for its implicit proposition that Oregon appellate courts, when choosing to follow nonbinding United States Supreme Court precedent, necessarily abandon those analyses when the Court does, and somehow prospectively adopt the Court’s subsequent analyses on a subject. Yet that is exactly what the majority does here, in concluding that case law that predates Teague decrees that we apply the test announced in Teague.
The majority cites Stewart v. Cupp, 12 Or App 167, 506 P2d 503 (1973), for the proposition that we have adhered to federal precedent in analyzing the retro-activity of newly announced federal constitutional rules in post-conviction cases. 184 Or App at 586 n 6. It is true that Stewart did cite Fair in a footnote, observing that one of the post-conviction petitioner’s arguments was meritless and that, in any event, it was based on a constitutional right held (in direct appeal cases including Fair) not to be retroactive. Id. at 170 n 1. However, while relying only on that isolated footnote, the majority does not undertake to discuss the actual holding of Stewart. In Stewart, the post-conviction petitioner asserted that he was entitled to the benefits of a plea bargain and that his prosecution was in violation of the plea bargain. He relied on Santobello v. New York, 404 US 257, 92 S Ct 495, 30 L Ed 2d 427 (1971), which had been decided after his conviction. We agreed with the petitioner’s argument that he was entitled to the benefit of his bargain and held that the petitioner was entitled to post-conviction relief. Stewart, 12 Or App at 173. We did not undertake an analysis of whether the constitutional rule of law announced in Santobello was to be applied retroactively; we must have implicitly assumed that it did apply retroactively, since we applied it retroactively. The case could hardly be said to provide a definitive analysis on retroactive application of federal constitutional principles in state post-conviction proceedings, given that it failed to engage in any retroactivity analysis at all, but it certainly did not adhere to federal precedent in analyzing retroactivity, as the majority suggests.
*602The majority’s reliance on Cornell v. Cupp, 25 Or App 805, 550 P2d 1386 (1976), similarly does not support its assertion that we follow federal precedent in analyzing retroactivity in post-conviction cases, 184 Or App at 586 n 6, because Cornell was not a post-conviction case. Cornell was a habeas corpus “conditions of confinement” case that did not involve a collateral attack on an underlying conviction. Rather, it concerned whether an inmate was entitled to expungement of institutional records concerning his confinement under the due process rule announced in Bekins v. Cupp, 274 Or 115, 545 P2d 861 (1976) (which also was a “conditions of confinement” habeas corpus case). It is true that in Cornell the court cited Fair for the proposition that we generally give newly announced constitutional rights the same level of retroactive effect that the United States Supreme Court does. 25 Or App at 808. However, it certainly cannot be said to hold anything about the availability of post-conviction relief, given that it did not involve post-conviction relief.
Virtually all of the federal decisions address Apprendi’s application to sentence enhancement for drug crimes based on the quantity of drugs involved. See, e.g., United States v. Mora, 293 F3d 1213 (10th Cir 2002), cert den, _ US _, 154 L Ed 2d 315 (2002); San-Miguel v. Dove, 291 F3d 257 (4th Cir 2002), cert den, _ US _ , 154 L Ed 2d 242 (2002); United States v. Sanchez-Cervantes, 282 F3d 664 (9th Cir 2002), cert den, _ US _ , 154 L Ed 2d 243 (2002); Murphy v. United States, 268 F3d 599 (8th Cir 2001), cert den, 534 US 1169 (2002); McCoy v. United States, 266 F3d 1245 (11th Cir 2001), cert den, _ US _, 153 L Ed 2d 183 (2002). At the very least, those cases are distinguishable in that the “judge vs. jury” and “preponderance vs. beyond a reasonable doubt” features of Apprendi have little, if any, effect on the accuracy of determination of the critical fact in those cases, viz., the (objectively verifiable) quantity of drugs. The fact-finding accuracy implications with respect to the determination of future dangerousness — i.e., whether a particular person has “a severe personality disorder indicating a propensity towards criminal activity” — are qualitatively different.
Given the majority’s analysis, it does not need to reach and resolve this question. I must.
ORS 138.530(1)(c) provides that a petitioner may obtain post-conviction relief if subjected to a “[sjentence in excess of, or otherwise not in accordance with, the sentence authorized by law for the crime of which petitioner was convicted; or unconstitutionality of such sentence.”
ORS 138.530(1)(a) provides that a petitioner may obtain post-conviction relief if there has been a “substantial denial in the proceedings resulting in petitioner’s conviction * * * of petitioner’s rights under the Constitution of the United States, or under the Constitution of the State of Oregon, or both, and which denial rendered the conviction void.”
But see McMillan v. Pennsylvania, 477 US 79, 84-91, 106 S Ct 2411, 91 L Ed 2d 67 (1986) (due process did not require that, where use of a gun was a factor required for imposition of a mandatory minimum sentence, such use must be found beyond a reasonable doubt; however, the Court suggested that the result might be different if the factor was used to enhance a sentence beyond what otherwise would be the statutory maximum). See also Harris v. United States, 536 US 545, 122 S Ct 2406, 2417-18, 153 L Ed 2d 524, 541-45 (2002) (McMillan was neither implicitly nor explicitly overruled or modified by Apprendi).
Similarly, state courts that have addressed whether Apprendi issues can be raised in state collateral proceedings have treated it as announcing a new rule of law. See State v. Sepulveda, 201 Ariz 158, 159, 32 P3d 1085 (Ariz App 2001) (“Apprendi constitutes a significant change in federal constitutional law”); People v. Beachem, 317 Ill App 3d 693, 700-02, 740 NE 2d 389 (2000) (recognizing that Apprendi announced a “new rule” of constitutional law).
As the Eleventh Circuit Court of Appeals noted in a recent decision, all federal circuit courts that had addressed the issue “determined that Apprendi did constitute a new rule of criminal procedure by requiring that any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to the jury and proved beyond a reasonable doubt.” McCoy v. United States, 266 F3d 1245, 1256 (11th Cir 2001), cert den, _ US _ , 153 L Ed 2d 183 (2002) (citing cases) (emphasis in original). See also Apprendi, 530 US at 525 (O’Connor, J., dissenting) (“the Court cannot identify a single instance, in the over 200 years since the ratification of the Bill of Rights, that our Court has applied, as a constitutional requirement, the rule it announces today” (emphasis in original)).
Defendant expressly acknowledges, and I agree, that the dangerous offender statute authorizes a court to impose a sentence that is longer than the statutory maximum otherwise prescribed for the underlying offense. That fact distinguishes this case from State v. Dilts, 179 Or App 238, 39 P3d 276 (2002), rev allowed, 335 Or 42 (2002) (holding that, for Apprendi purposes, guidelines departure sentences do not involve imposition of sentences that are longer than the statutorily prescribed máximums).