Concept
Confusion, Delay, and Cumulative Proof
Exclusion for unfair prejudice, confusion, or waste of time
Confusion, Delay, and Cumulative Proof
Rule 403 sets six dangers against the probative value of relevant evidence, and the preceding units examined the first, unfair prejudice. The text names five more: the court may exclude relevant evidence if its probative value is substantially outweighed by a danger of confusing the issues, misleading the jury, undue delay, wasting time, or needlessly presenting cumulative evidence.1 The rule states the dangers in the alternative, so any one of the five can carry an objection without help from the others.
The five divide into two groups with different jobs. Confusing the issues and misleading the jury resemble unfair prejudice in that all three safeguard the factfinding process. Undue delay, wasting time, and needlessly presenting cumulative evidence are time-saving factors that promote judicial efficiency, and they overlap heavily, because each bears on the dispatch and efficiency with which the trial moves forward.2
Both groups work through the same balance, and the balance leans toward admission. Rule 403 codifies the judge’s discretionary power to exclude relevant evidence, but only when a danger substantially outweighs probative value.3 For the time-based dangers, what goes on the probative side of the scale is incremental probative value, the amount an item adds to proof already in the case. Evidence that is merely repetitious or time-consuming may be excluded only if time considerations substantially outweigh that increment.4 The subsections below take the accuracy dangers first and the time dangers second.
Confusing and Misleading the Jury
Confusing the issues and misleading the jury are the dangers closest to unfair prejudice, because each is designed to safeguard the factfinding process.5 Telling the two apart is harder than naming them. The distinction between the terms is unclear in the literature and in the cases, so an objection may invoke both, and the judge runs a single weighing against probative value either way.6 One contrast still works as a shorthand: confusion describes a trial that loses its way, and misleading describes a jury that keeps to the path but misjudges what an item proves.
Confusion most often arrives through a side dispute. A collateral event must be proved like any other fact, so the opponent can contest whether it happened and how closely its conditions matched those of the event on trial. A leading treatise traces what follows: new issues arise about the occurrence of the instances and the similarity of conditions; new witnesses are needed, and their cross-examination and impeachment may lead to further issues; the trial is unduly prolonged; and the jury loses sight of the main issue amid a mass of confused data from which the kernel of the controversy is hard to extract.7 The limiting case is a trial within a trial, in which proving the collateral event becomes a second contest beside the first.
Similar-event evidence shows the danger at work. Suppose a patron falls on a theater’s carpeted stairs, sues the theater, and offers proof that another patron fell on the same stairs two weeks earlier. The earlier fall could bear on whether the stairs were dangerous, whether the theater had notice of the hazard, or whether a single cause produced both falls.8 Each use depends on how closely the two events match. The judge must weigh differences in lighting, in the number of people using the stairs, and in the youth and agility of those who fell, together with competing explanations such as a beverage spilled on the carpet.9 Changed conditions can also drain an earlier event of value: vegetation that blocked a driver’s view of an intersection in July may have been cut back by November, or may simply have lost its leaves, and then it no longer obstructs the view.10
The purpose for which the plaintiff offers the earlier fall sets how much similarity the judge demands. If the fall is offered to prove causation, the earlier event generally must be substantially similar, and if its cause is itself in dispute the evidence is unlikely to survive an objection, because the court will be concerned about a trial within a trial.11 The reason shows in the inference. The step from the earlier fall to a defect in the stairs holds only if that defect caused the earlier fall, so the plaintiff would have to prove it, the theater would answer with its own witnesses on lighting, footwear, and spilled drinks, and the jury would spend its attention on a stranger’s fall instead of the plaintiff’s. Offered only to show notice, the same fall needs no strict similarity, because anything close should have put the theater on a duty of inquiry.12 Notice asks what the owner should have suspected after the earlier mishap, and the jury need not decide what actually caused it, so no second trial arises. One more balance remains: the judge must ask whether the risk that jurors will treat the notice evidence as proof of causation substantially outweighs its probative value on notice.13
Products litigation raises the same sorting on a larger scale. Plaintiffs commonly offer other similar accidents to prove, circumstantially, a product’s dangerous or defective condition, the defendant’s notice of it, or that it caused the plaintiff’s injury, and defendants less often offer the absence of other accidents to prove the opposite.14 Those purposes track the ones in the stairway example, so the same questions govern: what is the evidence offered to prove, how close must the events be for that purpose, and what side dispute would proving the earlier event create?
Evidence of a later redesign shows how a neighboring rule can leave Rule 403 as the only filter. Rule 407 does not apply to remedial measures taken by a non-party, so a court errs when it excludes a non-party’s redesign under that rule.15 The redesign may still fail on its own merits. In Grenada Steel Industries, Inc. v. Alabama Oxygen Co., the court upheld the exclusion of an alternative design developed by someone else years after the product was made, because the design lacked sufficient probative value and injected the dangers of confusion and misleading the jury.16 A later decision applying Grenada Steel came out the other way on its facts. The redesign there was a bumper welded onto a road widener, a measure undisputedly available when the machine was made, and it was adopted solely in response to the plaintiff’s accident. No gap in time was left for jurors to overlook and no rival reason for the change to discount, so excluding the evidence as confusing was an abuse of discretion.17
Misleading the jury concerns weight. A jury can follow the case perfectly well and still give an item more force than its logic supports. One way to frame the question asks what probability a wholly rational trier of fact would assign after receiving the evidence, and whether excluding the item would bring the real jury closer to that figure. Exclusion improves accuracy only when it does, and because of the role the system assigns the jury, only a rather large disparity justifies it.19
Exhibits that re-create an event present the sharpest form of the danger. Since seeing is believing, real and demonstrative evidence, such as a model or an animation, can seem to possess an immediacy and apparent reality that gives it particularly persuasive effect. The judge may exclude it when it raises significant dangers for the jury’s reasoning that substantially outweigh its probative value.20 The usual question is whether the replication is substantially similar to what it purports to replicate. Without that similarity, jurors may take the exhibit as more faithful than it is, or time may go to explaining the differences.21 A dissimilarity matters most when it is likely to make a significant difference that jurors would find hard to detect or assess.22
Digital evidence is now offered in virtually every motion, hearing, or trial, from emails and cellphone records to digitally enhanced photographs and computer-generated versions of disputed events.23 Bowoto v. Chevron Corp. shows a computer-generated exhibit failing the similarity test. The plaintiffs offered a three-dimensional model of the barge where the disputed incident occurred. A viewer could move to a spot on the deck and see what a person standing there would see, which let jurors test the vantage points of the trial witnesses. Aerial photographs taken around the time of the incident showed a deck cluttered with large objects, while the model showed a clean deck, and the court excluded the model under Rule 403 because it did not accurately represent conditions at the time.24
The inaccuracy struck the model’s central function. A viewer placed at a witness’s position in a clean model sees sight lines that the clutter may have blocked in life, so the exhibit could make a witness’s claimed view look easier than it was. A tool that puts jurors on the deck can leave a strong impression, and nothing in the model tells them where it departs from the real deck. The decision does not demand perfection, however. Rule 403 carries a presumption of admissibility, so an exhibit need not match the facts exactly; the question is whether its departures are significant and hard for jurors to detect.25
Computer simulations add a reliability question that photographs do not. A simulation is the work product of many out-of-court actors whose knowledge and beliefs affect its reliability, including the sources of the factual basis, the sources of the scientific and technical formulae and the assumptions the programmers made, and the design, testing, selection, and operation of the program.26 Those issues are akin to hearsay, and they sit alongside the danger of misleading the jury, because a realistic animation gives jurors no visible hint of where its weak points lie.
The line in both scenes follows the test above: a dissimilarity that changes a quantity the jury must decide and escapes detection supports exclusion, and one that is trivial or easily exposed does not. A later section of this chapter returns to these exhibits in practice.
Undue Delay and Wasted Time
Undue delay and wasting time ask a different question from confusion and misleading: not whether the jury will go wrong, but what the evidence costs the trial. Both are time-saving factors that promote judicial efficiency, and they overlap because each bears on the dispatch with which the trial moves forward.28 The weighing is the same for both. The judge asks how much the item adds to what the jury already has, its incremental probative value, and excludes only if time considerations substantially outweigh that addition.29
Time can justify exclusion where nothing else does. Exclusion on accuracy grounds makes sense only if it would bring the real jury closer to what a rational factfinder would conclude.30 An item of slight force that the jury would weigh about right fails that test, since excluding it gains nothing in accuracy, so whatever case remains for exclusion rests on the time it consumes.
The jury view shows how the weighing runs. A pedestrian struck by a car sues the driver, and the plaintiff’s lawyer asks the court to take the jury to the accident scene so the members can see the surroundings, although the scene lies forty miles from the courthouse. If testimony, photographs, diagrams, or other means can depict the surroundings accurately, the judge would probably deny the request as a waste of time or an undue delay.31 The view’s only task is to convey information about the scene, so when other means convey the same information the view adds almost nothing, and a trip of that length costs a great deal. Exhibits raise a related logistical cost, because courts are built to receive witnesses, and presenting real evidence or demonstrative aids may require the court to move or unwieldy objects to be brought into the courtroom.32
The distance is the same in both scenes, so the difference lies in what the alternatives can supply. Where they match the view, its incremental value is close to nothing; where they cannot, the view carries real probative value that the time can justify. Because the ruling turns on that comparison, it stays with the trial judge, who can see how much the rest of the proof already shows.
Delay also comes from side disputes that one more witness would open. State v. Oswalt shows the pattern. A defendant charged with a Seattle robbery presented a Portland restaurant owner who said the defendant was in the restaurant every day for the two months before the robbery. The prosecutor offered a police detective to testify that the defendant had admitted being in Seattle on a date a month earlier, which would have shown the owner to be wrong. The Washington Supreme Court held that the detective’s testimony should not have been allowed, because the issue was collateral: it had no real bearing on the robbery and served only to contradict another witness. Admitting it would have set off a chain of answering witnesses, with the defense calling others to place the defendant in Portland on that date and the prosecution rebutting them, until the trial became a contest over one witness’s credibility on an unrelated point. The Federal Rules contain no express collateral-issue rule; the judge’s Rule 403 power to exclude for confusion, undue delay, or waste of time does similar work.
A different result followed in United States v. Stevens. The defendant offered “reverse 404(b)” evidence of a robbery of a man named Mitchell, portraying parallels between that crime and the one charged, and estimated that the presentation would take fifteen minutes. The court found that even if the estimate was overly optimistic, there was no appreciable risk that the evidence would degenerate into a mini-trial over whether Stevens had robbed Mitchell, and no real danger that it would obstruct the orderly progress of the trial or distract the jurors from the real issues. The government did not seriously dispute those points, and the thrust of its challenge rested on Rule 401, not Rule 403, so the probative value of the Mitchell robbery was not substantially outweighed by undue delay or confusion.33 The two cases differ in what the evidence would have set in motion. Answering the detective in Oswalt required new witnesses on a point that mattered to nothing else, while in Stevens the court saw no appreciable risk of a second contest, so a doubtful time estimate could not carry the objection.
A party’s offer to concede a point feeds into the same weighing, although it leaves the evidence relevant. Rule 401 requires only that the evidence bear on a fact of consequence, and nothing requires that the fact be in dispute, so a defendant who admits negligence and contests only damages cannot exclude proof of negligence as irrelevant. Proof of background facts, even undisputed ones, can help the jury understand the motivations and actions of the parties. The concession supports instead a Rule 403 objection: the evidence goes to a fact not in dispute, so its probative value is low, and that value is outweighed by the waste of time.34
How far the objection carries depends on what the concession covers and on the proponent’s freedom to refuse it. Suppose a defendant offers to stipulate to the killing and contests only his state of mind. The judge could save time by confining the prosecution to what is really in dispute, yet a fundamental tenet of the adversary system is that a party may prove its case in the way it wants.35 Courts therefore consider an offer to stipulate, or a failure to dispute the proposition for which the evidence is offered, when they measure incremental value, but they give great weight to the proponent’s freedom to decline the stipulation and present otherwise admissible evidence.36 A time objection on a conceded point is therefore strongest when the evidence bears only on the conceded fact, and weakest when it also bears on something still contested.
Old Chief shows that a concession can decide the matter when the danger is unfair prejudice. The stipulation there supplied everything that could properly be proved, the defendant’s felony status, and it was just as probative of that fact as the conviction record, so the conviction’s minimized probative value was substantially outweighed by the risk of unfair prejudice.37 A later section develops that holding.
Needlessly Cumulative Evidence
The last danger on Rule 403’s list is needlessly presenting cumulative evidence, proof that repeats a point the jury already has, and the court may exclude it when its probative value is substantially outweighed by that danger.39 The adverb needlessly does the sorting. Repetition on a contested point can be legitimate, as the robbery example below shows, while repetition on a minor or undisputed point is the usual target.40 The measure is the same incremental one used for delay and wasted time: the judge excludes only if the time the repeated proof would consume substantially outweighs the little it adds.41
An objection of cumulativeness concedes relevance, and the contest moves to the weighing. Relevance asks whether each inference linking the evidence to a fact of consequence rests on a valid assumption, and the inferences need not be equally strong.42 A fifth witness to an established fact can pass that test and still add almost nothing to what the jury has, so it is the balance, not the threshold, that decides the question.
Return to the pedestrian struck by the defendant’s car. Suppose the parties do not seriously dispute that the defendant was driving a European sports car capable of high speeds or that the accident happened at noon on a clear day, and the plaintiff proposes to call several witnesses on those facts. The trial court may decline to hear them.43 Once the point is established, each further witness adds time and little else. If the defendant instead denied that the car could reach high speeds, the point would no longer be undisputed, and the objection would weaken.
A defendant in a criminal trial who denies involvement in a charged robbery presents the opposite picture. The trial court would likely allow several or more prosecution witnesses to testify that they saw the defendant at the scene, and several or more defense witnesses to support the alibi.44 Both settings involve many witnesses to one fact, so the count does not separate them. In the robbery trial the fact is the defendant’s presence at the scene, the central matter in dispute, and each account can add something on the very question the jury must decide. The clear-sky point is neither important nor contested, so repetition has nothing left to accomplish.
Even on a contested point the allowance has a ceiling, because each added witness contributes less than the one before. A trial judge may stop a party from calling five experts of the same kind to testify to the same conclusion.45 The rule fixes no number. The question is what the next witness adds to what the jury already has.
Cumulation is measured against everything the jury already has, including what jurors can work out for themselves. In the eyewitness-identification case discussed above, the defendant offered an expert, Dr. Penrod, to comment on the suggestiveness of a wanted board. The court called the point, and a second one about the defendant’s “relation-back” argument, rather pedestrian. Each could be elucidated without specialized scientific knowledge, and counsel could develop it through probing cross-examination and arguments pitched to the common sense of the jury. Because the jury clearly could have understood the argument without an expert’s insight, exclusion was justified by the same Rule 403 considerations the court had invoked earlier.46 The expert would not have repeated another witness. The testimony would have repeated what jurors could grasp unaided and what counsel could argue, and testimony that tells jurors only that adds little but time. Testimony that supplies something jurors cannot work out for themselves keeps real incremental value.
Photographs show how cumulativeness joins the other dangers. In State v. Chapple, the prosecution offered photographs of the victim, Varnes. That Varnes was killed, the medical cause of his death, and what was done with his body afterward were not in controversy. The defense did not dispute the testimony of the detective and Dr. Jarvis on those subjects and even offered to stipulate to the cause of death. The only issue to be tried was whether two witnesses were correct in identifying the defendant as “Dee.” The court accepted that the prosecution’s arguments established relevancy but found that, on these facts, the photographs had little probative value.47 The facts the photographs illustrated were uncontested and the photographs said nothing about who the killer was, which is the pattern in which the likelihood of exclusion rises: the probative value is slight, or the fact sought to be proved is unimportant or undisputed.48
Chapple does not make photographs cumulative whenever testimony or a stipulation covers the same ground. Photographs in criminal prosecutions do not necessarily become cumulative and inadmissible merely because oral testimony or a stipulation addresses the same general subject. Even when the defendant offers to stipulate to each fact for which a photograph is relevant, the photograph may come in on the theory that the prosecution may prove every element of its case, or for the moral weight of the evidence.49 Admission is the norm, given the importance of photographs to the trier of fact and the trier’s ability to resist punishing a defendant for gruesomeness alone.50 What separates the two results is whether the photograph bears on something still in dispute. A later section returns to photographs in practice.
The third step does the real work. The usual error is to treat any repetition as objectionable, and a second witness on a contested point is not needless. The opposite error is to answer the objection by arguing that the evidence is relevant, since cumulative evidence is relevant by definition and the contest takes place in the Rule 403 weighing.