Let’s Break Some Rules: Immodest Proposals for Evidence Reform

In 1975, when the American Bar Association launched Litigation, its editors dedicated the entirety of one of the journal’s first issues to evidence. Authors included distinguished practitioners, respected judges, and academic giants like Margaret A. Berger, Irving Younger, and Jim McElhaney. Subjects varied, but the articles shared a common sensibility: The recently adopted Federal Rules of Evidence were a good thing.

And so they were and so, in general, they remain. They are well organized. They are relatively clear. They hold on to valuable lessons from the common-law tradition while embracing new insights. Their influence has promoted uniformity among the state courts. They get most things right.

I am presently advising an international student whose expansive project involves helping her country build its own set of evidence principles from the ground up. I am struck by how often I find myself recommending the approach taken by our federal rules. They constitute an admirable assemblage of tradition, experimentation, and refinement.

But, to paraphrase Mark Antony, I come neither to praise the Federal Rules of Evidence nor to bury them. I come instead to recommend some much-needed and long-overdue surgery. I have not bothered here with minor cosmetic changes; the operations I suggest require serious cutting and transplants. My proposals may therefore seem immodest and audacious, especially to those who have come to love the patient as is.

Forty years of using these rules in courtrooms, strategizing around them in meeting rooms, teaching them in classrooms, and writing about them here and elsewhere has persuaded me that we can and should do better. Our approach to evidence is mostly sound, but some problems have proven deeply and stubbornly embedded. And we are not going to address them adequately without creating a little wreckage along the way.

So let’s take a deep breath, roll up our sleeves, and break some rules.

The Mistake in Rule 403

Consistent with the grand themes of this anniversary issue of Litigation, let’s start with a rule that is a crowd favorite. It is not an obvious candidate for revision, but it desperately needs repair. I refer here to the familiar and well-worn Rule 403.

Rule 403 gets a lot of use. Lawyers routinely invoke it at trial because it covers a wide range of common concerns, like wasting time and unfairly prejudicing, confusing, or misleading the jury. Rule 403 also comes up often on appeal. Research that I conducted a few years ago identified it as one of the two evidence rules most frequently analyzed in appellate decisions in my state. For most of us litigators, Rule 403 has become an amiable and reliable companion.

That’s the problem. We have become so comfortable with Rule 403 that we have failed to notice that it contains a critical mistake that has potentially grave consequences. We need to fix that.

Rule 403 states: “The court may exclude relevant evidence if its probative value is substantially outweighed by a danger of one or more of the following: unfair prejudice, confusing the issues, misleading the jury, undue delay, wasting time, or needlessly presenting cumulative evidence.” It is widely understood that the word “may” here means that, in deciding Rule 403 questions, the trial court has substantial discretion. Giving the court discretion in deciding these issues makes sense; the rule, however, does not.

Consider this example: In a criminal case, the prosecutor offers a piece of evidence to which the defense objects. The court considers the evidence and finds that it barely passes the relevance test. The court further finds that the minimal relevance of the evidence is substantially outweighed by dangers of unfairness, confusion, and delay. If we read the rule literally, then the trial court has the discretion to admit the evidence despite all of these compelling reasons to keep it out. That’s a bizarre and unjust outcome.

For the rule to make sense, the court’s discretion must lie elsewhere. It must lie in deciding whether the evidence poses any of these dangers in the first place and whether those dangers substantially outweigh its relevance. It is appropriate to say—as the rule should—that a trial court has latitude to figure out whether these concerns exist and how they stack up against the probative value of the evidence. But it defies logic to say—as the rule does—that the trial court remains free to admit evidence that it decides fails the balancing test. This is especially true because that test is heavily weighted in favor of admissibility, so evidence that fails that test must be very bad evidence, indeed.

Rule 403 therefore misstates the proper nature and scope of the trial court’s discretion—an important matter both at trial and on appeal. The rule should say: “The court shall exclude relevant evidence if its probative value is substantially outweighed by a danger of one or more of the following: unfair prejudice, confusing the issues, misleading the jury, undue delay, wasting time, or needlessly presenting cumulative evidence. The trial court has discretion to determine when evidence meets this standard and should therefore be excluded.” Lawyers and judges will likely greet with apprehension any material change to the text of our beloved Rule 403. But having the rule make sense seems like a laudable goal.

Rule 404: Exceptions and Exclusions

Lawyers, especially prosecutors and criminal defense counsel, have also become accustomed to invoking Rule 404, the character evidence ban. But it, too, includes a nagging source of potential confusion and error—the lengthy list found in Rule 404(b)(2). The time has come to exorcise that problem as well.

To understand what is wrong with Rule 404, we need to recognize the difference between an exception and an exclusion. That requires a bit of discussion. Bear with me, gentle reader. I promise it is worth the effort.

In the case of an exception, a general rule exists but, for some reason, we have concluded that it should not apply in a particular case. We therefore create a carveout from the general rule to address that special circumstance. In contrast, an exclusion involves a case that falls outside of the general rule from the get-go. Those cases require no exception. The rule simply does not apply to them.

An example may help. Let’s say that lawmakers want to adopt a statute that creates an incentive toward the manufacture and sale of fuel-efficient cars. They write a law that imposes a hefty sales tax on all new automobiles but exempts any such vehicles that get more than 40 miles per gallon. That is an exception, because those cars would otherwise fall within the general rule. Because the rule applies only to new automobiles, many things also just fall outside of its reach—they are excluded. The rule plainly does not affect motorcycles, tractors, bicycles, or, for that matter, hedge trimmers or ice cream makers. Indeed, it would seem weird to encounter in such a statute a provision stating that it has no application to electric razors or toasters.

The drafters of the Federal Rules of Evidence occasionally demonstrate a blissful indifference to this distinction. For example, Rule 408(a) prohibits the admissibility of compromises and negotiations when offered to prove specific things, like the validity of a claim. Rule 408(b) then lists what it calls “exceptions” from that prohibition. What Rule 408(b) describes, however, are in fact exclusions, because they involve uses of such evidence for purposes other than the forbidden ones that Rule 408(a) describes. Rule 408(b) evidence requires no “exception” from Rule 408(a) because it does not even fall within that general rule.

With the distinction between an exception and an exclusion in mind, let’s look at the structure of Federal Rule of Evidence 404. Rule 404(a)(1) states the general principle: Evidence of a person’s character is barred when it is offered to prove that on the occasion in question the person acted in a manner consistent with that character. So, for example, a prosecutor cannot offer evidence in her case-in-chief that the defendant has a violent character in order to prove that the defendant assaulted the victim on the occasion as charged.

Rule 404(a)(2) recognizes a number of exceptions to this general rule. For example, it allows the criminal defendant in that assault case to offer evidence of his character for peacefulness in order to prove that he did not initiate the altercation. If the defendant does come forward with such evidence, then that triggers another exception under which the prosecutor can in rebuttal offer proof of the defendant’s violent character. The provisions of Rule 404(a)(2) constitute true exceptions. The evidence described there fits within the general rule barring character evidence. The evidence could not come in if not for the carveouts contained in Rule 404(a)(2). So far, so good.

When we arrive at Rule 404(b)(1), however, things start to get complicated. That provision states: “Evidence of any other crime, wrong, or act is not permissible to prove a person’s character in order to show that on a particular occasion the person acted in accordance with that character.” This language does not seem to add anything to what Rule 404(a)(1) already told us. Instead, it appears to emphasize the tautological proposition that if you offer evidence of a specific act as character evidence, well, then, it is character evidence.

Despite its apparent redundancy, we might conclude that Rule 404(b)(1) serves a worthwhile clarifying purpose. Evidence of other crimes, wrongs, or acts usually does not come to us wearing a badge that says: “Hello, I’m character evidence.” Rule 404(b)(1) alerts us to the possibility that it may nevertheless be so if it is offered to prove what the general principle forbids. To borrow from Walter Reuther, if it walks, talks, and quacks like character evidence, then that’s what it is.

When we come to Rule 404(b)(2), however, we find ourselves in much more serious trouble. The rule states: “This evidence [i.e., evidence of other crimes, wrongs, and acts] may be admissible for another purpose, such as proving motive, opportunity, intent, preparation, plan, knowledge, identity, absence of mistake, or lack of accident.” Rule 404(b)(2) gets a lot of airtime at criminal trials, with prosecutors and defense counsel regularly wrangling over its meaning and application. And rulings on Rule 404(b)(2) issues have provided fodder for countless appeals.

What does this provision mean? Again, an example may help. Assume that the defendant is charged with making and detonating the pipe bomb that killed the victim. The prosecutor wants to offer evidence that the defendant has made pipe bombs in the past. If the prosecutor wishes to admit that evidence to show that the defendant is a bad and violent person, she cannot do so. Rule 404(a)(1) prohibits it.

But if the prosecutor wants to admit the same evidence to show something other than a propensity to act in conformity with a certain character, Rule 404(a)(1) does not apply. For example, that rule is not implicated if the prosecutor wants to use the evidence to show that the defendant had the knowledge and expertise essential to make the sort of pipe bomb that killed the victim. In essence, Rule 404(b)(2) simply underscores that point by listing some noncharacter uses of other-acts evidence and declaring that the character evidence ban does not apply to them.

Rule 404(b)(2) is unnecessary and should go. If that provision did not exist, the meaning of Rule 404 would not change at all. Think about it: The only thing Rule 404 bars is character evidence, and the things listed in Rule 404(b)(2) are not that.

All the uses of other-acts evidence inventoried in Rule 404(b)(2)fall outside of the scope of the general rule. It lists things that are not affected by the rule because, by definition, they are not within it. Circling back to my hypothetical fuel-efficiency statute, Rule 404(b)(2) is the functional equivalent of having a rule that says, “By the way, a toaster is not an automobile.”

Rule 404(b)(2) is not worrisome just because it is surplusage, although we do try to avoid baking meaningless verbiage into our laws. Rather, it gives rise to at least three much more significant problems. And, in my view, those problems outweigh the benefits, if any, that this provision offers.

First, judges and lawyers often talk about Rule 404(b)(2) as if it lists exceptions to the character evidence ban. But, as I have explained, that’s wrong: They are exclusions. This error has persuaded courts that they must closely scrutinize and interpret each word in the Rule 404(b)(2) list to determine the precise parameters of the carveouts that it creates. That’s a fundamental mistake, because exclusions do not carve out anything at all—they just fall outside the scope of the rule.

Second, the outsized energy that judges have devoted to figuring out the exact meaning of each of these words cuts against the ostensible goal of Rule 404(b)(2). On its face, this rule underscores the flexibility that prosecutors have in admitting other-acts evidence when they are not offering it for the forbidden purpose. But the judicial glosses on each of the words in Rule 404(b)(2) have yielded a vast body of case law that does more to constrain prosecutors than to liberate them. For example, decisions under Rule 404(b)(2) have created elaborate multipart tests for determining when evidence qualifies as showing “identity” or a “common scheme or plan.” Those demanding standards are inconsistent with the concept of an exclusion and appear nowhere in the rule itself. Nevertheless, prosecutors cannot use other-acts evidence for those permissible purposes unless they satisfy these additional judicially imposed tests.

Third, and perhaps paradoxically, Rule 404(b)(2) also gives too much latitude to prosecutors, providing them with opportunities for manipulation and abuse. It invites prosecutors to go to the list, choose some label from the inventory, and stick it on the evidence they are offering—even if they are really offering that evidence for the forbidden character purpose. Some courts have proven much better at spotting this ruse than others.

Rule 404(b)(2) therefore comes at a high cost. And it is an unjustifiable one, given that the rule is redundant and expendable and it effectively does nothing more than tell us that a car is not a toaster. It is way more trouble than it’s worth. The time has come to thank the rule for its service and send it packing.

Rule 412 Needs Repair

We encounter a sharp contrast in Rule 412, the provision that protects victims of sexual assault from inquiry into their sexual history and predispositions. That rule is not going anywhere, and it should not. It has a vital role in evidence doctrine, especially in the states, where most sexual assault cases are prosecuted. Prior to the adoption of this rule, survivors of sexual assault were routinely subjected to abusive examinations that chilled the reporting of crimes and revictimized the witness.

Whatever its virtues, however, Rule 412 needs work. Its current disrepair is signaled by a strange and singular exception that the drafters included in the rule, which states that it does not apply to “evidence whose exclusion would violate the defendant’s constitutional rights.” Of course, all evidence rules must comply with constitutional requirements, so this provision does nothing to correct any infirmities that exist within Rule 412. And it is worrisome that the drafters so clearly anticipated constitutional problems with respect to this rule that they felt they had to say something about it.

Indeed, one does not have to struggle to come up with examples of instances where Rule 412 bars evidence that should be admitted. George Fisher’s excellent casebook on evidence includes the example of a defendant who wants to explain the presence of his fingerprints in a rape survivor’s bedroom by disclosing that on a prior occasion the couple had engaged in consensual sex. The literal terms of Rule 412 would prevent the defendant from offering this critical exonerative evidence. Nor does that evidence fall within any of the rule’s specific exceptions. Leaving this defendant with nothing but a fair trial or due process argument to try to get the evidence admitted seems gravely unfair.

It is true that Rule 412 has an essential place in our evidence law and that victims of sexual assault deserved better than the abusive tactics defense lawyers regularly used prior to its adoption. But it is also true that criminal defendants, who enjoy the presumption of innocence, deserve better than a rule that is subject to exceptions so poorly crafted that its drafters felt compelled to note that the Constitution of the United States might need to come to the rescue. A detailed rewriting of Rule 412 and its exceptions is a significant undertaking that extends beyond the scope of this article and this author’s expertise. But it is sorely needed.

The Hearsay Rules Are Too Complicated

The rules governing hearsay are so complicated and Delphic that we might rename them the Evidence Law Professor Full Employment Act. Scholars have devoted thousands of pages to the analysis and clarification of hearsay doctrine, but achieving a firm grasp on it continues to elude many lawyers and more than a few judges. I have listened patiently while an earnest jurist has held forth on some aspect of hearsay law and gotten almost everything completely wrong.

Hearsay doctrine is, at least in part, complex by necessity. It engages with questions that border on the metaphysical: What is an assertion? When is an assertion offered for its truth? When is it offered for another purpose? When is it sufficiently reliable that we should except it from the hearsay ban? A certain amount of impenetrability inheres in questions of this nature. That makes it even more important that we simplify hearsay doctrine where we can. So let’s do that.

Begin by remembering the close connection that exists between hearsay and cross-examination. Judicial opinions and evidence books routinely cite the trial of Sir Walter Raleigh as the paradigmatic example. The accusations of treason against Raleigh came from declarants who made their statements outside of court—principally one Lord Cobham, who, while being held in the Tower of London, signed a written confession implicating Raleigh in a plot to kill King James. Raleigh demanded that the court produce Cobham at trial so he could confront and examine his accuser, but the court refused.

Raleigh was convicted (and ultimately executed) on the strength of hearsay—statements made outside of court and offered in court for their truth. He had no opportunity to test the credibility of Cobham’s accusation by cross-examining him. And it appears Raleigh could have done so effectively: He had in his possession another document signed by Cobham in which Cobham recanted his confession, prime material for impeachment.

The close connection between hearsay and cross-examination suggests that if the latter is available, then the former is not a problem. Many of my students sensibly infer that our hearsay concerns vanish when the person who made the out-of-court statement appears as a witness at trial and can be cross-examined. I have run into a fair number of lawyers and judges who think the same thing and who believe that the rule so holds. They are wrong.

Under our hearsay rules, the fact that the declarant of the out-of-court statement appears as a witness at trial does not necessarily dispose of the hearsay issue. Rule 801(c) defines “hearsay” as a “statement that the declarant does not make while testifying at the current trial or hearing and a party offers in evidence to prove the truth of the matter asserted in the statement.” That definition does not distinguish statements made by declarants who do testify at trial from statements made by those who do not.

Imagine that Lord Cobham had appeared at Raleigh’s trial and said: “Well, as I told the inquisitors in the Tower of London, Sir Walter Raleigh planned to assassinate King James.” Under Rule 801(c), that testimony would fit the definition of “hearsay” even though Cobham was present as a witness and Raleigh could ask him all the questions he wanted. If the hearsay doctrine exists to address circumstances where cross-examination of the declarant is unavailable, then that outcome does not make much sense.

The drafters of Rule 801(c) could have solved the problem by including a provision that says: “A statement is not hearsay if the declarant testifies and is subject to cross-examination about a prior statement.” And, indeed, they did so in Rule 801(d)(1). But then they went on to say, in Rule 801(d)(1)(A), (B), and (C), that this exemption holds true only under certain circumstances.

Those three subsections have a baroque complexity that invites lots of questions. Rule 801(d)(1)(A) allows for the admissions of prior statements that are inconsistent with the witness’s testimony. But how inconsistent do those statements need to be? Is it inconsistent for a witness to testify that the witness does not remember something that he or she remembered in the past? Is the proponent of evidence that is admissible under Rule 801(d)(1)(A) entitled to an instruction that the jury can consider the prior statement for its truth and not just for its inconsistency?

Rule 801(d)(1)(B) allows for the admission of prior consistent statements. For years, the rule permitted such evidence only to rebut an accusation that the witness had recently fabricated his or her story. But then the rule was amended to allow prior consistent statements whenever the witness’s credibility has been attacked. Given how frequently that happens with witnesses, as a practical matter does the current version of the rule really exclude any prior consistent statements at all?

Rule 801(d)(1)(C) allows a witness to repeat an identification of a person that the witness made at an earlier time. This provision obviously applies to lineups and other similarly formal statements made to law enforcement authorities. But does it apply to less formal ones made to others, such as a statement made to a friend, like “Hey, I know that guy running out of the alley. That’s Tom!”?

In short, Rule 801(d)(1) makes a bad trade-off. Granted, it does prevent the admission of some out-of-court statements of limited relevance and reliability. But the primary rationale behind the preclusion of such evidence—that the declarant is not subject to cross-examination—does not apply to those statements. If we do indeed have reason to view those statements with suspicion, the opposing counsel can probe the matter by questioning the witness.

Further, this exclusionary rule comes at a substantial cost. It gives us a Byzantine collection of principles that are difficult to understand and to navigate. It increases the risks of confusion and error. And at least some of the evidence it excludes would probably be objectionable under other rules anyway, like our old friend Rule 403.

The time has come to scrape these barnacles off the hearsay rule. Rule 801(d)(1) should be revised to say that a statement is not hearsay if the declarant testifies and is subject to crossexamination—period. This would bring the rule in line with the reason we have a hearsay principle in the first place. And it would tidy up a rule that—with dozens of other subparts, exclusions, and exceptions—can use all the tidying it can get.

The “Best Evidence” Rules Are Obsolete

My final, and most ambitious, recommendation is that we at long last rid ourselves of the so-called “best evidence rule,” embodied in Federal Rules of Evidence 1001–1008. The impulse underlying the rule—that we should prefer the original of a document over a copy of it or testimony about it—may have made sense at one time in our history. But changes in how we do business and in technology have undercut its rationale. These days, we tend to treat copies as if they are as reliable as originals in the transactions we conduct outside of court. And, given the myriad ways in which documents and images are now created, it has become increasingly difficult to tell what qualifies as an “original.”

The Federal Rules of Evidence have desperately clung to the best evidence principle while trying to accommodate those changes, with bizarre results. So Rule 1002 tells us that an original of a document is required to prove its contents; but then Rule 1003 tells us that a duplicate is equally admissible unless for some reason it is unfair to use it. The rules thus create a universe in which a document can simultaneously function as both an original and a duplicate, an achievement of abstraction equal to physicist Erwin Schrödinger’s proof that his cat was both alive and dead.

These rules add considerable complexity to our evidence principles without commensurate value. The little good that they do to help ensure the reliability of evidence can almost always be achieved by applying other principles, such as authentication (Fed. R. Evid. 901), conditional relevance (Fed. R. Evid. 104(b)), and unfair prejudice (Fed. R. Evid. 403). In my view, the evidence rules would be better evidence rules without the best evidence rules.

We have lived with the Federal Rules of Evidence for half a century since this journal printed its first issue dedicated to them. The rules have undergone changes during those five decades, mostly for the better. The vast majority of those changes have been gradual, cautious, and incremental, and that approach has generally served us well. My immodest proposals intend no disrespect to the precision tools we have used to date. But we do need to remember that sometimes we have a bigger job at hand. And those bigger jobs may call for a chainsaw, a sledgehammer, and a little skillfully employed dynamite.