UNITED STATES of America, Plaintiff-Appellee, v. Louise Han PEREZ; Joseph Eclavea Perez; John Velasco Cruz, Defendants-Appellants

KLEINFELD,

Circuit Judge, joined by

HUG, Chief Judge, DAVID R. THOMPSON, TROTT, and T.G. NELSON,

Circuit Judges, concurring in the judgment.

I concur in the result. But I disagree with the majority’s rejection of unreviewability of invited error except where the record shows that counsel knew of the legal entitlement being waived. Traditionally, invited error is unreviewable except in “the most ‘exceptional situation’” where reversal is “necessary to preserve the integrity of the judicial process or to prevent a miscarriage of justice.” United States v. Schaff, 948 F.2d 501, 506 (9th Cir.1991). Defense counsel requested the instruction the judge gave, so invited error doctrine would make the error unre-viewable whether it was prejudicial or not, and whether defense counsel was aware of it or not. I would treat the jury instruction error in this case as waived.

The majority opinion says that “an error is waived and therefore unreviewable,” only if “the defendant has both invited the error, and relinquished a known right.” This is impractical, as I explain below. And it is doetrinally mistaken, because it confuses two levels of rights. When defense counsel asks a judge to proceed a certain way, he may not know of some ease which would entitle him to demand that the judge proceed in a different way. But he does know that he is giving up his client’s right to appeal whatever error might later be identified if the judge proceeds as he requests. Because defense counsel intentionally relinquishes the known right to appeal error in what he himself requests, invited error is a waiver of a known right. The known right relinquished or abandoned is the right to appeal an error in an instruction, not a known right to a particular instruction.

This case concerns only jury instructions, so perhaps we will distinguish it in future cases of invited error not relating to jury instructions. The context is peculiar, in that the error came from our own form book. Defense counsel requested the instruction in our form book, but a case had come down a few months before trial which held that our form incorrectly omitted the “in relation to” element of the crime under 18 U.S.C. § 924 (e)(1). See United States v. Mendoza, 11 F.3d 126 (9th Cir.1993). Perhaps we will limit today’s decision to invited error in jury instructions based on our own form book, but the opinion does not contain express limiting language. It is possible that we will construe today’s opinion broadly, and hold in any future case of any kind of invited error that unless the government can show that the defense knew what error was being made, we will reverse in the same circumstances as for merely forfeited error.

The majority opinion is mistaken in its doctrine and mischievous in its likely consequences. It infers abolition of traditional invited error doctrine from United States v. Olano, 507 U.S. 725, 733-34 , 113 S.Ct. 1770, 1777-78 , 123 L.Ed.2d 508 (1993), but Olano says nothing about “invited error.” In Ola-no, the Supreme Court reversed us, because we had mistakenly reversed a conviction on the basis of an error to which no objection had been made and which had not been shown to be prejudicial. Olano says there is no error at all when a right has been waived, and waiver is the “intentional relinquishment or abandonment of a known right.” Id. at 733 , 113 S.Ct. at 1777 . The majority says that an intentional relinquishment of a known right requires that the person know that there is a right to require something different of the court. Thus, defense counsel cannot, under the majority opinion, waive an error in a jury instruction unless defense counsel knows that the instruction is incorrect and submits it anyway. No ethical de *850 fense lawyer would do such a thing, so it is not clear that any cases will fall into the category of invited error as limited by the majority opinion.

Our decision today is aberrant. The other circuits to address invited error doctrine since Olano have generally held that it survives unchanged. Many have explicitly considered the forfeiture-waiver distinction, and held that invited error should be treated as waiver, reasoning as I have urged above. For example, United States v. Griffin, 84 F.3d 912, 923-24 (7th Cir.1996), holds, citing Olano , that defense counsel’s statement of approval for an instruction “amounts to a waiver of the right to claim error on appeal.” The jury instruction was incorrect, but defense counsel’s express approval “extinguished any right to appellate review” because it amounted to a “waiver” of the right to appeal on that point. This is the analysis we should follow. Likewise, United States v. Ross, 77 F.3d 1525, 1542 (7th Cir.1996), interprets Olano to mean that if defendant requests a particular instruction which is given, “he has voluntarily waived any objection” to the instruction, so “plain error analysis does not apply because there is technically no ‘error’ to correct.” Where defendant submits the instruction given, “we will not review ... even for plain error.” Id. at 1542.

Likewise United States v. Mitchell, 85 F.3d 800 (1st Cir.1996) holds that where defense counsel affirmatively said he had no objection to an arguably erroneous procedure, “[t]hat action raises his later silence from mere forfeiture to waiver.” Id. at 808 . United States v. Balter, 91 F.3d 427, 434 (3d Cir.1996) (“we would not find that the district court erred even if Cutler could show that he was prejudiced.”).

In United States v. Tandon, 111 F.3d 482, 487-89 (6th Cir.1997), the Sixth Circuit distinguished our en banc decision in United States v. Keys, 95 F.3d 874 (9th Cir.1996), on the ground that the error in Keys was merely forfeited, but the error claimed in the case at bar was invited. Because the “requested instruction invited the error,” the Sixth Circuit held that plain error review was inapplicable. Id. at 489.

The Tenth Circuit, applying Olano , likewise held that “[a] defendant cannot invite a ruling and then have it set aside on appeal.” United States v. Hardwell, 80 F.3d 1471, 1487 (10th Cir.1996). Hardwell holds that invited error “cannot be reviewed for plain error” because it is waived.

I disagree with the majority’s reading of the authorities in Footnote 9 of the majority opinion. The eases, as explained above, do expressly consider Olano in connection with invited error, and conclude that invited error should be treated as waived, not forfeited. The quoted and summarized material from the cases shows that to be so. Footnote 9 attempts to distinguish the eases I have cited on the ground that the defendants had waived and not merely forfeited the error in those cases. But the majority’s reasoning is circular. The reason, the only reason, the exact, expressly stated reason, why the error was treated as waived, was that it was invited. Our sister circuits reasoned in those cases that if error is invited, it should be ' treated as waived, not merely forfeited.

Invited error has generally been unreviewable. See United States v. Miguel, 111 F.3d 666, 673-74 (9th Cir.1997); United States v. Baldwin, 987 F.2d 1432, 1437 (9th Cir.1993); United States v. Schaff, 948 F.2d 501, 506 (9th Cir.1991); Guam v. Alvarez, 763 F.2d 1036, 1037-38 (9th Cir.1985); United States v. Alexander, 695 F.2d 398, 402 (9th Cir.1982); Sherwin v. United States, 320 F.2d 137, 147 (9th Cir.1963). The exception to this rule has been narrowly limited to “the most ‘exceptional situation’ ” where reversal is “necessary to preserve the integrity of the judicial process or to prevent a miscarriage of justice.” United States v. Schaff, 948 F.2d 501, 506 (9th Cir.1991).

Invited error doctrine has been around for too long to suppose that the Supreme Court would have tossed it away without even mentioning it. See, e.g., Philadelphia, Wilmington and Baltimore Railroad Company v. Howard, 54 U.S. (13 How.) 307, 343-44 , 14 L.Ed. 157 (1851) (in upholding instruction attacked by defendants, “it is ... conclusive” that defendants requested substantially similar instruction); Wells v. United States, 257 F. 605, 619-20 (9th Cir.1919) (party may not *851 complain of instruction it invited). Courts have long recognized its “wide and varied application” throughout a trial, Alabama Great Southern R. Co. v. Johnson, 140 F.2d 968, 971 (5th Cir.1944), when the pressures on judges to make quick decisions are great.

The majority decision confuses knowledge that one is waiving a particular right with knowledge that one is waiving the right to appeal error, known or unknown. All of us who have practiced on the civil side have used releases with the standard language releasing “any and all claims, known or unknown.” Invited error is likewise an abandonment of all appellate claims of error, known or unknown, affected by the invitation. In civil cases, a plaintiff needs the power to release unknown claims in order to get the most money for the claim he knows about, because otherwise he cannot sell the defendant legal insulation from claims he may think of or discover later. Likewise, a criminal defendant needs the ability to have his lawyer say to the judge, “if you do what we request, you will not be reversed because of it,” as an inducement to persuade the judge to comply with requests the correctness of which the judge does not have time to research.

A false presupposition underlies the majority’s analysis. The majority imagines a lawyer who, if competent, knows of every decision bearing on any point of law arising in his case. There is no such lawyer. Lawyers make judgments, not only which known rights to assert, but also which rights are worth looking for. The majority opinion says that “there is no evidence that Cruz and Perez considered submitting” a correct instruction and decided against it for tactical reasons, and “the record reveals that” the defendants were unaware of Mendoza , so the right to a correct instruction cannot be treated as waived. This reasoning implies that the defense could not waive its right to appeal omission of the “during and in relation to” language from the instruction it invited, unless defense counsel had found and read Mendoza , so that they would know of the right to “during and in relation to” language.

The law is so vast that all of us are necessarily ignorant of most of it at any particular time. Were it not for the essential help of counsel, supplemented by our law clerks’ research, even my learned colleagues might not know of every decision bearing on each issue coming before us, and I know that I would not. Preparation for trial is potentially infinite, but time is most assuredly finite. A lawyer’s most critical decisions are often how to use the very limited time available for preparation of a defense. Defense preparation may include trips to the jail for client interviews, reading whatever discovery the defense can get, examining real evidence such as audio and videotapes, finding and interviewing witnesses, researching legal questions likely to be at issue, drafting motions and oppositions, drafting jury instructions, preparing real evidence for use at trial, cajoling or compelling witnesses to appear at trial, negotiating toward a plea, and preparing outlines for direct and cross examinations.

Lawyers do not research every possible issue of law in every case. Nor should they. A lawyer necessarily and properly exercises professional judgment about how to allocate the limited time for preparation in a way likely to produce the most benefit for the client. These time allocation decisions are by logical necessity made in partial or complete ignorance of what would be accomplished if time were allocated differently. Sometimes researching the law is a waste of time, while finding and talking to a witness would produce a defense bonanza. Often there is not enough time to do both to the maximum possible extent. Experienced lawyers usually know what they are doing and are acting wisely for their clients, when they make their decisions about what to do, and what need not be done, to prepare the case.

In the case at bar, the lawyers did not know about Mendoza , so far as the record shows. The majority’s conclusion that Cruz and Perez were not prejudiced by omission of the “in relation to” element shows that the lawyers were entirely correct in not wasting their time finding out about Mendoza . A couple of hours in the law library finding Mendoza , reading it, and drafting a revised instruction, would have been a waste of time, because as the majority concludes, an in *852 struction expanding upon the “in relation to” element would not have gotten their clients acquitted. Obviously a jury was not going to acquit defendants who sold heroin for guns, carried guns to narcotics sales, and sat on a gun or chambered a round during their narcotics arrests, on the theory that perhaps their possession and carrying were independent of, and not “during and in relation to,” their drug crimes. The majority’s conclusion, that we should affirm despite the error, shows that the defense lawyers were right in not allocating the time necessary to find Mendoza .

Though the majority is plainly correct that there was no prejudice, the majority underestimates the sagacity of defense counsel. Defense counsel probably were ignorant of Mendoza , but knew enough about the case to know that their clients would not be prejudiced by omission of the research which would have uncovered it. Defense counsel doubtless knew just what we have concluded, that “in relation to” did not matter to this case. Had it mattered, defense counsel would have wondered whether some sort of instruction could be drafted that would enable a jury to acquit based on this, and would have spent some time in the law library and found Mendoza . By concluding that we should affirm despite the error, we conclude some years after the lawyers decided how to prepare the defense that it would have made no difference to add “during and in relation to” language to the instruction. So the lawyers were right — the research that would have turned up Mendoza would have been waste of time. Scarce time.

We mistakenly say “the lawyer didn’t know about Mendoza , so he did not intentionally relinquish a known right.” The lawyer made his important decision earlier, when he elected not to spend time in the law library researching an issue that would not affect the outcome. Then when he submitted the instruction, the lawyer abandoned whatever appellate claim might derive from rights, known or unknown, to a different instruction.

Our addition of a new element to invited error doctrine today, that the record show knowledge by the defense of the error in its own invitation, is inconsistent with the “strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.” See Strickland v. Washington, 466 U.S. 668, 689 , 104 S.Ct. 2052, 2065 , 80 L.Ed.2d 674 (1984). The consequence is likely to be weakening of defense counsel’s position at trial, as well as unreasonable appellate reversals of guilty defendants’ convictions. Invited error doctrine is not only a tool that helps the prosecution hold onto convictions on appeal despite errors. At trial, invited error doctrine works for the defense.

Traditional invited error doctrine lets trial judges give defense counsel leeway to manage their cases as they and their clients think best. In response to many defense requests before and during trial, the judge says to himself, “I’m not sure defense counsel is right, but I should let the defense run the defense.” Judges freely allow defense counsel to run their cases, because there has traditionally been no risk of reversal for invited error.

If judgments can be routinely reversed for invited error, judges are not likely to take many chances on defense counsel’s proposed instruction language. They will just use their own boilerplate, which is often harder for juries to understand because it is so general.

If today’s decision applies more broadly, as to examination of witnesses, the effect on trials will be dramatic. Judges will call bench conferences if they think defense counsel is making a mistake, to demand justification for apparently failing to exercise a right. Prosecutors will “ask to oversee defense counsel’s conduct at trial — to ensure against reversal.” United States v. Decoster, 624 F.2d 196, 208 (D.C.Cir.1976).

If the government were required to prove that its adversary defense counsel was adequate, it would be strongly motivated and well advised during a criminal trial, in order to protect the prospect of guilty verdicts, to oversee the major decisions and activities of defense counsel and the accused that affect trial. Performing this function would, as a practical matter, require the prosecution to probe what has *853 heretofore been a sacrosanct area-the highly confidential relationship between a criminal defendant and his lawyer.

Id. at 228 (concurring opinion). This reasoning applies with equal force to the invited error context. This can be very destructive of legitimate and important defense tactics. Defendants usually do not have the FBI, DEA, offers of immunity and lenient sentences, and the other powerful tools of the prosecution to work with, nor do they get much discovery in federal proceedings. If prosecution witnesses lie, defense counsel often need the leeway to ask what sound like obtuse questions, but are really intended to surprise the witnesses, in order to expose the lies. If defense counsel are interrupted and forced to justify what they are doing, much truth will not come out, and more innocent people will be convicted.

“Judicial scrutiny of counsel’s performance must be highly deferential.” Strickland, 466 U.S. at 689 , 104 S.Ct. at 2065 . But it cannot be deferential, if on .appeal we are going to look for evidence that the defense knew of the right being waived, and then waived it intentionally. Worse, our inquiry into whether defense counsel intentionally waived a known right is by its very nature corrosive. Cf. Strickland, 466 U.S. at 690 , 104 S.Ct. at 2065-66 . At trial, judges will have to ask defense counsel if they realize they are giving up rights their clients have, an inquiry sure to reduce defendant’s confidence in his lawyer, and make it harder for the lawyer to manage the case effectively on his client’s behalf. On appeal, defense lawyers will feel compelled to assert that they, or predecessor counsel, gave up rights in ignorance, leaving us to decide whether they are lying.

How will judges make that determination? Shall judges favor regulars, such as local assistant United States attorneys and federal public defenders? Shall judges cross-examine lawyers to see how honest they seem? Shall we ruin lawyers’ reputations by making express determinations that they lied about their purported ignorance? Shall defense lawyers be told at continuing legal education programs that effective representation requires them to plant some error in the record as insurance against convictions? See United States v. Boyd, 86 F.3d 719, 721-22 (7th Cir.1996) (“Many a defendant would like to plant an error and grow a risk-free trial”); Alabama Great Southern, 140 F.2d at 971 (invited error doctrine “prevents a litigant from speculating on a verdict, and then, when the speculation turns out badly, escaping the consequences of having done so”). All we need to do to avoid this corrosiveness is to leave invited error doctrine alone.

The moral principle of estoppel is at the heart of invited error doctrine. Invited error doctrine has been called “estoppel to allege error.” Id. at 971 . It is wrong for the defense to ask a trial judge to do something, and then ask an appellate court to reverse because the trial judge did what was requested. Lawyers usually are very intelligent and capable people. They necessarily know more about their eases than the judges trying them, because the judges do not have secret discussions with the defendants, or find and interview witnesses. Defendants should be bound by their lawyers’ acts on their behalf, because judges properly rely on defense counsel’s superior knowledge of the defense’s best interests and duty to pursue them. Cf. Boyd, 86 F.3d at 721-24 . There is no moral justification for imposing on society the risk of unpunished or repeated crime, inevitable in some percentage of reversed convictions, because of errors which defendants invited trial judges to make.