LINDE, J.

The decision of the Court of Appeals is affirmed.

LINDE, J.

Convicted of theft, defendant obtained a reversal in the Court of Appeals because his trial followed a mistrial brought on by what the court described as "flagrant overreaching" by the prosecutor. 49 Or. App. 415, 418, 619 P.2d 948 (1980). After this court denied review, 290 Or. 551 (1981), the state obtained a writ of certiorari from the Supreme Court of the United States. The Supreme Court reversed the decision insofar as it rested on the double jeopardy and due process clauses of the United States Constitution and remanded the case to the Court of Appeals. Oregon v. Kennedy, 456 U.S. 667, 102 S Ct 2083, 72 L Ed 2d 416 (1982). On remand, the Court of Appeals reconsidered the nature of the prosecutor's misconduct and its consequences under Oregon law and this time affirmed the conviction. 61 Or. App. 469, 657 P.2d 717 (1983). We allowed review to examine the court's assumption that Oregon law concerning retrials after prosecutor-induced mistrials, a question that first reached this court in State v. Rathbun, 287 Or. 421, 600 P.2d 392 (1979), is identical to the view of the federal double jeopardy clause expressed by the majority of the Supreme Court in this case. We conclude that Oregon law is not identical but nevertheless leads to an affirmance of this conviction.

Before reaching the merits, we take up the procedural history that brings the issue before this court.

I. Procedure

The history of this case demonstrates the practical importance of the rule, often repeated in recent decisions, that all questions of state law be considered and disposed of before reaching a claim that this state's law falls short of a standard imposed by the federal constitution on all states. See State ex rel Adult Family Services v. Bradley, 295 Or. 216, 666 P.2d 249 (1983); State v. Davis, 295 Or. 227, 666 P.2d 802 (1983); Suess Builders Co. v. City of Beaverton, 294 Or. 254, 267, 656 P.2d 306 (1982); Cole v. Department of Revenue, 294 Or. 188, 190, 655 P.2d 171 (1982); Hewitt v. SAIF, 294 Or. 33, 41-42, 653 P.2d 970 (1982); State v. Caraher, 293 Or. 741, 653 P.2d 942 (1982); Gale v. Dept. of Rev., 293 Or. 221, 646 P.2d 27 (1982); Portland Police Assn. v. Civil Service Board, 292 Or. 433, 639 P.2d 619 (1982). Like most states, Oregon throughout its history has had a constitutional ban against placing anyone twice in jeopardy for the same offense. Or Const art I, § 12. That guarantee has in the past been given independent interpretation with results that might not correspond to those in other states or in federal law. See State v. Brown, 262 Or. 442, 497 P.2d 1191 (1972), holding that article I, section 12 requires consolidation of charges known to the prosecutor. State v. Rathbun, supra, was decided under article I, section 12 in September 1979, well before the trial or the appeal in this case.

In its initial decision reversing this conviction, the Court of Appeals cited no statutory or constitutional source at all for that result. It quoted from two opinions of the United States Supreme Court to summarize what it called a "general rule" about the permissibility of reprosecution after mistrials. The court then cited its own decision in State v. Rathbun, 37 Or. App. 259, 586 P.2d 1136 (1978), noting only that it had been reversed "on other grounds" by this court. It might not be apparent to a reader that this court in fact had reversed the Court of Appeals on the very point at issue under Oregon's double jeopardy clause.

We denied the state's petition for review of the decision of the Court of Appeals. That, of course, implied nothing as to its correctness. 1000 Friends of Oregon v. Board of County Commissioners, 284 Or. 41, 44, 584 P.2d 1371 (1978). In response to the state's petition to the United States Supreme Court for a writ of certiorari, defendant pointed out that, given the appellate court's cryptic silence on the point, its decision might rest on the Oregon Constitution's double jeopardy clause that was applied by this court in Rathbun. The Supreme Court nevertheless granted the petition, carrying forward its current campaign not to let state or lower federal courts draw more protective constraints from the federal constitution's guarantees in matters of criminal law than the Court itself is prepared to recognize. The possibility that the result might rest on an independent state ground was pursued by the justices both on oral argument and in the opinions. The majority took note of the studied citation by the Court of Appeals of its own opinion in Rathbun, which purported to apply federal law, despite this court's reversal of that decision under Oregon law. Four justices, however, thought the role of state law in this case more complex, as expressed by Justice Stevens:

"Although I am willing to accept the Court's reading of the Oregon Court of Appeals' opinion as having been based on federal law, I find the question somewhat more difficult than does the Court because the Oregon Supreme Court declined to review the case without explaining its reasons. Since the Oregon Supreme Court seems to have interpreted the state constitutional protection against double jeopardy to be broader than the federal provision, see State v. Rathbun, 287 Or. 421, 600 P.2d 392 (1979), it is entirely possible that that court's refusal to review the Court of Appeals' decision was predicated on its view that the decision was sound as a matter of state law regardless of whether it was compelled by federal precedents."

456 US at 681 n. 1 (Stevens, J., concurring, joined by Brennan, Marshall, and Blackmun, JJ.).

This quotation makes clear that a practice of deciding federal claims without attention to possibly decisive state issues can create an untenable position for this state's system of discretionary Supreme Court review. It can also waste a good deal of time and effort of several courts and counsel and needlessly spur pronouncements by the United States Supreme Court on constitutional issues of national importance in a case to whose decision these may be irrelevant. In effect, when this court might reach the same result under the Oregon law that a lower court reaches by citing federal precedents, we would have to allow review at the instance of a losing party objecting only to the federal holding, while the successful party who might prefer a decision on state grounds has no reason to petition us for review. Surely a practice that requires a winning party to seek review solely in order to shift a favorable judgment from federal to state grounds is wholly unreasonable, apart from its logical flaws.

In the present case, we in fact do not reach the same result as the Court of Appeals did in its initial decision. Had that decision given its attention first to the state law precedent of Rathbun and reversed defendant's conviction under article I, section 12, we might have allowed review in order to compare this case with Rathbun. If so, we might not only have decided the state claim against the defendant, as we do today, but also his federal claim, thereby relieving the Supreme Court of concern about a reading of the fifth amendment more expansive than its own. As it is, we reach the issue of Oregon law two and one-half years and hundreds of pages of briefs after it might have been decided in the Oregon courts.

The state contends that we should not reach an Oregon issue at all because it was not adequately argued below. The attorney general concedes that Oregon's article I, section 12, was "dutifully cited" to the circuit court. He argues, however, that it was not urged as a basis "distinct" or "separate" from the federal double jeopardy provision, because defendant cited in support of his claim only cases that were themselves decided on federal grounds. He cites this court's prior admonitions that the specific bases of constitutional claims should be not only quoted but analyzed.

We do not lack sympathy for the state's position, as those citations show. Legal claims raised but not substantially briefed are burdensome to meet and difficult to decide correctly. As the Chief Justice suggested during the oral argument, it might clarify matters if issues of state law were briefed first and federal issues only thereafter. On issues new to this state's law, we may prefer principled arguments to mere citations from other jurisdictions, arguments such as both the state itself and other parties have provided in this and in other cases. Experience suggests, however, that such arguments are more common when a case has no direct analogues in decisions of the United States Supreme Court or other high courts. See, e.g., McCall v. Legislative Assembly, 291 Or. 663, 634 P.2d 223 (1981) (reapportionment); LaGrande/Astoria v. Public Employes Retirement Board, 284 Or. 173, 586 P.2d 765 (1978) (municipal home rule). The reality is that time for original analysis is scarce, particularly in the ordinary criminal case; and particularly at the trial level, lawyers and courts often depend on the shorthand of case citations in preference to scrutinizing statutes and constitutional principles.

This reality cannot allow us to accept the state's position that a claim asserted under Oregon law may be disregarded unless it is elaborately briefed in terms separate from parallel claims under federal law. This is so for two reasons. First, an Oregon court should not readily let parties, simply by their choice of issues, force the court into a position to decide that the state's government has fallen below a nationwide constitutional standard, when in fact the state's law, when properly invoked, meets or exceeds that standard. The second reason is that where such parallel claims exist, the discussion of their proper scope and effect is largely found in federal cases. At least this has been so during the past two decades, since the federal courts began to enforce fourth, fifth, sixth and eighth amendment standards as 14th amendment due process. Academic and professional commentary, from which counsel might wish to brief the separate state claim, also has concentrated almost exclusively on federal cases and doctrines.

This court like others has high respect for the opinions of the Supreme Court, particularly when they provide insight into the origins of provisions common to the state and federal bills of rights rather than only a contemporary "balance" of pragmatic considerations about which reasonable people may differ over time and among the several states. It is therefore to be expected that counsel and courts often will refer to federal decisions, or to commentary based on such decisions, even in debating an undecided issue under state law. Lest there be any doubt about it, when this court cites federal opinions in interpreting a provision of Oregon law, it does so because it finds the views there expressed persuasive, not because it considers itself bound to do so by its understanding of federal doctrines. See, e.g., State v. Davis, supra, n. 18; State v. Florance, 270 Or. 169, 527 P.2d 1202 (1974).

On many occasions, this court has decided cases in which a defendant not only failed to brief but wholly failed to raise anything but a federal claim. In recent cases we have taken care to point this out, so that our decision is not misunderstood to foreclose any potential issue of state law for the future. See, e.g., State v. Farber, 295 Or. 199, 666 P.2d 821, n. 10 (1983); State v. Roberti, 293 Or. 59, 644 P.2d 1104, rev'd on other grounds 293 Or. 236, 646 P.2d 1341 (1982); State v. McMurphy, 291 Or. 782, 786, 635 P.2d 372 (1981); State v. Brown, 291 Or. 642, 634 P.2d 212 (1981). When a source in Oregon law has been cited but not briefed or argued, we have asked for additional memoranda or set further argument. See State v. Caraher, supra; State v. Clark, 291 Or. 231, 630 P.2d 810, cert den 454 U.S. 1084 (1981); Sterling v. Cupp, 290 Or. 611, 625 P.2d 123 (1981). When a court is confronted with a mere unexplained citation of an Oregon source tacked on as an "afterthought," in the attorney general's phrase, the court similarly may request counsel either to explain the claim under state law or to abandon it. Although responsible treatment of the state claim is preferable, as stated above, if it is abandoned the court can note that fact so that the decision at least will not be a precedent on the issue.

In the present case, we have in fact received extensive briefs on the issue under article I, section 12, not only from the parties but also from amici curiae. For the foregoing reasons, therefore, we decline the state's invitation to hold that defendant's claim under that section was not raised or preserved.

II. The merits.

Common to State v. Rathbun, supra, and to this case is the question under what circumstances a defendant may not be retried because officers of the state were responsible for the initial mistrial. In Rathbun that officer was a bailiff; here it is the prosecutor. We therefore asked the parties and amici curiae to state and elucidate the rule they consider applicable to this case.

Defendant proposes four "tests," two addressed to the nature of the official's misconduct and two addressed to its effects. He suggests that a retrial is barred if the official's conduct was a flagrant violation of professional standards, if the official knew or should have known that it would seriously interfere with the trial, if it in fact did so and made continuation of the trial highly prejudicial to defendant, and if the trial judge determines that the misconduct was damaging and could not be cured by corrective instructions.

The Oregon Criminal Defense Lawyers Association argues that retrial is barred if the official knew or reasonably should have known that the conduct was improper.

The American Civil Liberties Union's formulation is to bar a retrial if the state official whose conduct caused the mistrial knew or should have known both that the conduct "was improper" and that it would prejudice a fair trial, and if the misconduct in fact did so.

Although these formulations are not identical, each rejects a requirement for the bar that the official must have intended to cause a mistrial, partly for reasons to which we return. First we turn to the state's arguments why we should agree with the position taken by the United States Supreme Court, for the first time in this case, that official misconduct grave enough to require a mistrial should bar a second trial only when it was intended to provoke the defendant into moving for a mistrial.

The state argues that when the Oregon Constitution employs terms "substantially identical" to those in the constitutions of the United States or of other states, the framers of the Oregon Constitution should be presumed to have sought to achieve the same objectives. That they sought the same objectives is generally true in the absence of contrary evidence. This does not, however, say much toward demonstrating the correct application of such a constitutional text. In particular, the proposition does not support the non sequitur that the United States Supreme Court's decisions under such a text not only deserve respect but presumptively fix its correct meaning also in state constitutions.

The state finds some difficulty in explaining why this should be so. If state guarantees are presumptively bound to interpretations of the federal Bill of Rights merely because they are federal, the argument extends similar force to lower federal court decisions when the Supreme Court has not spoken. If the argument is only that the federal guarantees are older than the Oregon Constitution, the fact, of course, is that they were adopted in order to bind the federal government to guarantees already established in the existing states. See generally Schwartz, The Bill of Rights: A Documentary History, Vols. I and II (1971); Perry, Richard L., Sources of Our Liberties (1959). If priority in time governs, the United States Supreme Court would be bound to follow decisions from Virginia, Massachusetts, Delaware, or in the case of double jeopardy, those from New Hampshire, whenever the first Congress is not shown to have intended a different rule. Diversity in rules derived from a common historical source is familiar in what, despite the diversity, is still called "common law." Cf. Norwest v. Presbyterian Intercommunity Hospital, 293 Or. 543, 652 P.2d 318 (1982). Diversity is the price of a decentralized legal system, or its justification, and guidance on common issues may be found in the decisions of other state courts as well as in those of the United States Supreme Court. See, e.g., State v. Kessler, 289 Or. 359, 614 P.2d 94 (1980) (right to bear arms); State v. Haynes, 288 Or. 59, 602 P.2d 272 (1979) (access to counsel).

The state argues, correctly, that diversity does not necessarily mean that state constitutional guarantees always are more stringent than decisions of the Supreme Court under their federal counterparts. A state's view of its own guarantee may indeed be less stringent, in which case the state remains bound to whatever is the contemporary federal rule. Or it may be the same as the federal rule at the time of the state court's decision, which of course does not prevent that the state's guarantee will again differ when the United States Supreme Court revises its interpretation of the federal counterpart. The point is not that a state's constitutional guarantees are more or less protective in particular applications, but that they were meant to be and remain genuine guarantees against misuse of the state's governmental powers, truly independent of the rising and falling tides of federal case law both in method and in specifics. State courts cannot abdicate their responsibility for these independent guarantees, at least not unless the people of the state themselves choose to abandon them and entrust their rights entirely to federal law.

The state also argues that resort to state constitutional provisions may frustrate the United States Supreme Court's "institutional goal of responsible law development" and "abandon the stability" of that Court's precedents. The Supreme Court itself does not share that view. The Court or its individual members have repeatedly drawn the state courts' attention to their independent responsibility for their states' constitutional law. In a recent case, three Justices explained that they withheld votes sufficient to have allowed a writ of certiorari on an important issue because some state courts were dealing with that issue, and their experience would help to inform the Supreme Court's own views. Moreover, the state acknowledges that the federal rule to which it asks us to defer is the rule that the state persuaded a bare majority of the Supreme Court to adopt for the first time in this very case.

The disputed difference between one or another formulation of the double jeopardy bar after a mistrial actually is quite narrow. It is only whether such a bar requires the court to find that the prosecutor intentionally provoked the mistrial. That is the view taken by the Supreme Court majority and defended here by the state. The sole issue is whether, as four Justices maintained, there is room for a double jeopardy bar beyond the case of an intentionally provoked mistrial when a prosecutor "harasses" the defendant with what the prosecutor knows to be prejudicial error.

The underlying premises of history and policy were ably briefed here as well as in the Supreme Court and are summarized in that Court's past and present opinions and in our opinion in State v. Rathbun, supra. We agree with the state that a bar against prosecution must be derived from the constitutional objective to protect defendants against "the harassment, embarrassment and risk of successive prosecutions for the same offense," and that "[i]t is not a sanction to be applied for the punishment of prosecutorial or judicial error." But to see the double jeopardy guarantee as protection rather than as sanction has implications for the rule against reprosecutions. It bears on defining how far the rule depends on the culpability of the official whose conduct prevented a fair trial. If the rule were viewed as a penalty against the state, it might reasonably be limited to penalizing only intentional misconduct. From the standpoint of a defendant forced to choose between accepting prejudicial errors or undergoing a second trial, the precise degree of the official's mens rea is a matter of indifference. Whether the prosecutor deliberately pursues an improper course of conduct because he means to goad a defendant into demanding a mistrial or because he is willing to accept a mistrial and start over is a distinction without a difference, when the guarantee is viewed as protection against "the harassment, embarrassment and risk of successive prosecutions," as the state correctly says. We agree with the state, however, that a guarantee against "harassment" implies a requirement of some conscious choice of prejudicial action before the guarantee bars correction of the error by a new trial. Negligent error, "gross" or otherwise, is not enough.

In this case, the Court of Appeals initially believed that the kind of conduct "motivated by bad faith or undertaken to harass or prejudice" the defendant extended to prosecutorial "overreaching." 49 Or App at 417-18. In the United States Supreme Court, the majority was concerned that "more general standards" for barring retrials "than one merely based on intent [would] offer virtually no standards for their application" and that this was particularly true of "overreaching." 456 US at 674. The four concurring Justices, on the other hand, rejected a requirement that the prosecutor must actually intend to force the defendant to demand a mistrial. They maintained that "the rationale for the exception extends beyond the situation in which the prosecutor intends to provoke a mistrial," applying as well when a prosecutor's objectives are served by pressing his improper conduct regardless whether the defendant chooses to demand a mistrial or to endure the prejudicial misconduct. 456 US at 689 (Stevens, J., concurring). While the majority was concerned about the difficulty of administering the "broad and somewhat amorphous standard adopted by the Oregon Court of Appeals," 456 US at 676, Justice Stevens's concurring opinion expressed concern about the difficulty of proving the prosecutor's intention. 456 US at 688. The majority and Justice Powell, concurring, in turn acknowledged the practical difficulties of proving intent but emphasized that intent could be inferred from the objective circumstances of the conduct at issue. 456 US at 675, 680.

As we have stated, the difference is a narrow one. We believe that the acknowledged objective of the double jeopardy guarantee can be served by a rule that avoids the indefiniteness of "overreaching" and yet extends beyond intentional provocation to cover other possible abuses. It is common ground that when a prosecutor or other responsible official intentionally provokes the defendant to demand a mistrial, the double jeopardy guarantee precludes a further prosecution for the same offense. It also is agreed that a court may infer from the character and the circumstances of the prejudicial conduct that it was so intended without having to obtain an admission to that effect. When a court draws that inference, a retrial is barred.

A test limited to intentional provocation of mistrials, however, has two shortcomings besides the question of proof. First, the Supreme Court's analysis focused on prosecutorial misconduct, as in this case. The Court adopted that test in part because, within the limits of professional ethics and the state's other overriding values, prosecutors are expected to strive for convictions, and "overreaching" could be asserted of every prejudicial error that may require a retrial. But prosecutors are not the only officials whose conduct may cause a mistrial or a reversal. State v. Rathbun, supra, involved a bailiff, who discussed the charge with jurors in terms that were found to be motivated by her intention, not to cause a mistrial, but to assist the state in securing a conviction. That, of course, is not the function of a judge, bailiff, or other courthouse officials, and the difficulty of defining when error caused by zeal shades into "overreaching" does not apply to them. In holding that the bailiff's misconduct barred a retrial, this court stated: "The state put this officer of the court in the position to wreak havoc and must bear the same burden as when its prosecutor or judge in like manner offends." 287 Or at 433.

Second, a finding that a prosecutor initially pursued a course of prejudicial misconduct for the purpose of forcing a mistrial is a grave matter. Such behavior is a contempt of court. ORS 33.010 (c), (d). It also is a violation of professional standards that can lead to disbarment or other discipline, and perhaps of federal civil rights statutes. See DR 1-102(A)(5), 7-102(A)(1); 18 U.S.C. § 242. A judge prepared to make such a finding properly would not only declare a mistrial without possibility of reprosecution but also report the episode to the Oregon State Bar. Code of Judicial Conduct, Canon 3(B)(3). But we do not think that impermissible double jeopardy for the defendant is limited to the few situations in which a judge is sufficiently convinced of a prosecutor's improper intentions to invoke those penalties. That places too heavy a burden on the inference that a defendant must ask a judge to draw from the objective conduct and circumstances. To repeat, punishment of the errant official is not the object of the guarantee against placing the defendant again in jeopardy for the same offense. Whether or not the official's error rises to that level of culpability, for purposes of the guarantee we think it suffices that he has consciously chosen to engage in prejudicial misconduct, whatever the motive.

We therefore conclude that a retrial is barred by article I, section 12, of the Oregon Constitution when improper official conduct is so prejudicial to the defendant that it cannot be cured by means short of a mistrial, and if the official knows that the conduct is improper and prejudicial and either intends or is indifferent to the resulting mistrial or reversal. When this occurs, it is clear that the burden of a second trial is not attributable to the defendant's preference for a new trial over completing the trial infected by an error. Rather, it results from the state's readiness, though perhaps not calculated intent, to force the defendant to such a choice.

This formulation differs in some respect from those proposed to us and described above. It differs from that proposed by defendant because the "flagrant" nature of the misconduct and its irremediable character are not independent criteria for the constitutional bar on retrials. These are factors in determining whether a mistrial should be ordered, if the error is not an invariable reason for a mistrial, as stated in State v. Rathbun, supra, 287 Or at 425-26. See also State v. Kristich, 226 Or. 240, 359 P.2d 1106 (1961). Each also is a fact from which the court may infer whether a prosecutor knew that his or her conduct was improper and was indifferent to the result of forcing the defendant to a likely second trial on the same charge. Our formulation differs also from that proposed by the amici curiae insofar as it holds the state only to the consequence of what its official knew to be prejudicial misconduct, not what he or she should have known. A court facing the issue may, of course, infer that the prosecutor, bailiff, or other official in fact knew what anyone in his or her position should know. But in ruling on an asserted bar against reprosecution at the time of a motion for mistrial or before an attempted retrial, the court must find whether the official in fact had that knowledge. Incompetence, thoughtlessness, or excitability of the state's officers may lead to a mistrial, but it does not reflect a willingness to risk placing the defendant repeatedly in jeopardy for the same offense.

In the present case, there was no finding that the prosecutor knowingly acted in an improper and prejudicial manner, indifferent to the mistrial that could be expected to result. Nor can we infer from the two circuit judges' rulings that either would have made such a finding if requested. The judge presiding at the first trial declared a mistrial after the prosecutor asked a witness whether the reason he had never done business with defendant was "because he was a crook." In the colloquy with the court, the prosecutor defended her question as proper because the possible bias of the witness against defendant had been introduced in crossexamination by defendant's counsel. The judge rightly ruled that the question went beyond permissible bounds, but nothing in the colloquy suggests that he thought it was conscious misconduct.

Similarly, at the time of the motion to dismiss the second prosecution, the judge ruling on that motion concluded that the prosecutor's question had been improper but that it did not reflect "bad faith" or an "intentional impropriety," or even "gross negligence." Although this finding is not phrased in the terms of "knowing" misconduct coupled with indifference toward the probable risk of a mistrial that we have stated above, nothing in the record offers a basis for speculation that this difference in formulation would have led either judge to a different conclusion. There was, for instance, no persistence in pressing a line of questioning or argument after an objection to it had been sustained, nor any suggestion that the prosecutor on previous occasions had been warned against similar transgressions. On this record, therefore, the criteria for imposing a constitutional bar against a second trial were not met.

The decision of the Court of Appeals is affirmed.