The debate about what to do with a criminal defendant who wants to lie on the witness stand goes back a half-century, to a 1966 Michigan law review article written by a young law professor and former public defender named Monroe Freedman.
This article is reprinted with permission from the March 29, 2013 edition of The Recorder
by Richard Zitrin
Freedman argued that given a lawyer's duties as he saw them — to the client, to the legal system and, in his view, to find out the facts of the case without burying one's head in the sand — the best solution to what he called this "trilemma" was to put on the perjured testimony. Some people in Washington, D.C., where Freedman worked, were so outraged that they tried to get Freedman disciplined for having even suggested this solution.
Then in the early 1970s the American Bar Association approved a set of criminal law standards. Standard 7.7 called upon lawyers faced with a perjurious client to allow the testimony to occur, but only in the "narrative form," as opposed to the usual question-and-answer format. The lawyer was then to refrain from arguing this testimony to the jury. Initially, this method found some favor with institutionalized criminal defense organizations, such as public defender's offices, which were searching for a way to take a position that walked the tightrope between "snitching off" one's own client and permitting outright perjury.
But the problem with the narrative method soon became obvious. It fools absolutely no one, especially the jury. Many were quick to criticize it as tantamount to hanging a sign around the defendant's neck that says "liar." "Ironically," wrote Freedman, "the Standards reject any solution that would involve informing the judge, but then propose a solution that, as a practical matter, succeeds in informing not only the judge but the jury as well."
Chief Justice Warren Burger's opinion in the Supreme Court case of Nix v. Whiteside, 475 U.S. 157 (1986), joined in the criticism, describing the narrative method as "a signal at least to the presiding judge that the attorney considered the testimony to be false and was seeking to disassociate himself from [it]."
When the ABA revised the standards, it abandoned Standard 7.7 and its narrative approach, preferring to rely on new Model Rule 3.3, created in 1983 and amended substantially in 2002, which states that a lawyer shall not knowingly offer evidence "that the lawyer knows to be false." Period.
In 2000 the American Law Institute published its first Restatement of the Law Governing Lawyers. Section 120 of the restatement adopted the prevailing view that "from the unusual format of examination, prosecutor and presiding officer are likely to understand that the accused is offering false testimony, but an unguided jury may be unaware of or confused about its significance. That solution is not consistent with the rule."
Thus, as of 2000, the ABA, the U.S. Supreme Court, the ALI's Restatement, and the leading intellectuals in the field of criminal defense all opposed the "narrative" solution. So what is the majority view of American courts?
That's right, the narrative approach. And leading the band is our great state of California.
In 1998 the court in People v. Johnson, 62 Cal.App.4th 608, engaged in an extensive scholarly discussion of the right of a criminal defendant to testify, and then evaluated six possible solutions to the perjury problem, everything from Freedman's position to refusing the defendant the opportunity to take the stand at all. In all, the Johnson court's analysis was thorough and learned. The court just chose the wrong solution.
In Johnson, after the prosecution rested, defense counsel asked to go in camera with his client and told the judge, "I cannot disclose to the court privileged communications ... but I'm in a position where I am not willing to call Mr. Johnson as a witness despite his desire to testify." "Judge," he went on, "this is not a trial tactic issue. This is an ethical conflict. If it were just a trial tactic ... I always defer to the client's wishes in circumstances like this." This dialogue then occurred:
"THE COURT: What exactly are you trying to tell me ... that you won't examine him if he takes the stand?
"[DEFENSE COUNSEL]: I think under the — based upon the information I have, I would be ethically barred from calling him as a witness under the law as I have come to know it and very specifically researched it regarding this particular issue.
"THE COURT: Based on that ... I'm going to have to abide by it ... I will note, then, that you believe that there are ethical reasons, strong ethical reasons ... [t]hat cause you to come to the conclusion that you cannot under any circumstances call the client. ... And I will note that it's over his objection, and that he desires to testify."
The Johnson appellate court first determined that "the criminal defendant's constitutional right to testify is unlike other matters of trial strategy which are in the control of defense counsel. A criminal defendant has the right to take the stand even over the objections of his trial counsel." In other words, the trial court erred by not allowing Johnson to take the stand.
Then the Johnson court evaluated each of six possible solutions, as follows:
1. "Full cooperation with presenting defendant's testimony even when defendant intends to commit perjury." Johnson noted that despite some intellectual support, no court had sanctioned Freedman's methods, and both the Supreme Court in Whiteside and ABA Rule 3.3., which the court quoted in the absence of a clear California rule on point, expressly forbade perjured testimony.
2. Persuasion, which the court agreed was (quite obviously) "the ideal solution" "when it succeeds."
3. "Withdrawal from representation" as a solution was rejected by the Johnson court, because, quoting from People v. Gadson, 19 Cal.App.4th 1700 (1993), an earlier case that permitted the narrative method, withdrawal "'could trigger an endless cycle of defense continuances and motions to withdraw as the accused informs each new attorney of the intent to testify falsely.'"
4. Disclosure to the court was rejected in Johnson both on the "chance the defendant will change his mind and testify truthfully" and because "disclosure is only a partial solution [because it] will require some additional action, i.e., ...whether the defendant will be permitted to testify and in what form and manner."
5. The narrative approach. And the accepted one; see below.
6. "Refusing to permit the defendant to testify" as violative of the Constitution.
Johnson then concluded: "None of the approaches ... is perfect. Of the various approaches, we believe the narrative approach represents the best accommodation of the competing interests of the defendant's right to testify and the attorney's obligation not to participate in the presentation of perjured testimony since it allows the defendant to tell the jury, in his own words, his version of what occurred, a right which has been described as fundamental, and allows the attorney to play a passive role."
California is the only state that currently, under Johnson, requires lawyers to choose the narrative solution. In addition, the District of Columbia has a rule that lays out a specific protocol to follow for narrative testimony.
But while every state except California has Model Rule 3.3 on its books in one form or another, the courts of at least 10 states other than California — Colorado, Florida, Illinois, Indiana, Kentucky, Mississippi, Massachusetts, New York, Pennsylvania and West Virginia — have expressly endorsed the narrative procedure by ratifying trial courts in after-the-fact appellate decisions.
And Model Rule 3.3 has a loophole that says if the decisional law of any state differs from the rule, practitioners should defer to the state's decisional law.
Almost all commentators agree that the narrative solution is at best only a compromise. Some compromises provide the lowest common denominator, but to me, this one provides the worst of both worlds: the presentation of the false story and the breach of client confidentiality. And when authorities from Chief Justice Burger to professor Freedman and all stops in between agree, it has to be a bad idea.
Can we do better? The Johnson opinion is least persuasive when it criticizes disclosure to the court. This disclosure, coupled with testimony in the ordinary format, is far better than the lose-lose narrative solution. For me, if push came to shove I'd side with Freedman if my strong-arming my client failed, and put on the perjured testimony in the ordinary way, hoping that I won't lose my bar ticket afterwards. Juries are not easily fooled, and know full well that an accused has a motive of to lie. It's partly for this reason that many defense lawyers are reluctant to have their clients testify, regardless of the perjury issue.
Another solution is that with a narrow definition of "knowledge" as virtual certitude, the reality is that many criminal defense lawyers practice the selective ignorance "ostrich" approach. Given the difficulties in finding a reasonable solution, can we blame them?
Richard Zitrin is a professor at UC Hastings College of the Law and of counsel to San Francisco's Carlson, Calladine & Peterson. He is the lead author of three books on legal ethics, including The Moral Compass of the American Lawyer.
The Recorder welcomes submissions to Viewpoint. Contact Vitaly Gashpar at firstname.lastname@example.org.