Concept
Proving the Other Act
Other crimes, acts, transactions, and events
Proving the Other Act
Rule 404(b)(2) lets a proponent use another crime, wrong, or act to prove motive, intent, knowledge, identity, absence of mistake, or lack of accident.1 Every one of those purposes rests on a premise the rule never states: the other act happened, and the defendant did it. Earlier dealings in stolen goods undercut a claim of innocent ignorance only if the defendant took part in them, and an earlier attempt on a spouse's life makes an accident story harder to believe only if the attempt occurred. An act that never took place connects the defendant to nothing. Beneath every purpose, then, lies a question of proof: how sure must the court be that the other act happened before the jury hears about it?
The federal answer is, not very sure. Under Huddleston v. United States, the trial judge does not decide whether the act occurred. The judge asks whether there is enough evidence for a reasonable jury to find that the defendant committed it, which is the Rule 104(b) standard, and then applies Rule 403 to the evidence's use on a non-propensity issue. The judge screens, and the jury decides whether the act happened. The first subsection follows Huddleston from question to holding, the second explains how Rule 104(b) screening works, and the third takes up acquittals and later acts. The Rule 403 balance gets its own section below.
Huddleston's Holding
Rule 404(b)(2) is silent on how firmly the proponent must establish the other act before the jury hears about it.2 Huddleston v. United States filled that silence for the federal courts. The government prosecuted Huddleston over Memorex tapes it said were stolen, and it offered testimony from a witness named Toney that Huddleston had sold him televisions. The government's theory was that the televisions were stolen too, so that a series of sales of stolen merchandise from the same suspicious source would be strong evidence that Huddleston knew each item, the tapes included, was stolen.3 The sale of the televisions therefore counted as a similar act only if the televisions were stolen. The district court told the jury to use the similar-acts evidence only on knowledge and not to prove character, and the jury convicted on the possession count only.4
The appeal turned on how much proof the trial court must demand before admitting the evidence. The Sixth Circuit's own panels show the range of answers. One panel reversed because the government had not shown by clear and convincing evidence that the televisions were stolen, and on rehearing, following a decision that adopted a preponderance standard, the court affirmed.5 By the time the case reached the Supreme Court, Huddleston's counsel had abandoned the clear-and-convincing position, conceding at argument that it could not survive Bourjaily v. United States, which held that preliminary factual findings under Rule 104(a) are subject to the preponderance standard.6 What remained was a narrower demand: the jury should not hear about a similar act until the trial judge had found under Rule 104(a) that the defendant committed it. The demand rested on a real worry, that jurors would punish the defendant for the similar act or infer that he is an evil person inclined to violate the law.7
The Court rejected that demand, beginning with the structure of the Rules and the text of Rule 404(b). Rule 403 lets the trial judge exclude relevant evidence when unfair prejudice substantially outweighs its probative value. Rules 404 through 412 single out kinds of evidence that have caused trouble, but they generally do not flatly prohibit it. They limit the purposes for which it may be introduced. Rule 404(b) follows that pattern: it protects against extrinsic act evidence offered solely to prove character, and its text contains no hint that a preliminary showing is needed before the evidence is offered for a proper purpose. Evidence offered for a proper purpose is subject only to the general strictures of Rules 402 and 403.8 Huddleston's reading would have added to the rule a judicial finding that its text does not contain.
The legislative history pointed the same way. The Court read it as stressing admissibility and concluded that Congress was not nearly so concerned with the prejudicial effect of Rule 404(b) evidence as it was with ensuring that restrictions would not be placed on the admission of such evidence.9
Dispensing with a judicial finding does not dispense with proof. Other-act evidence is relevant only if the jury can reasonably conclude that the act occurred and that the defendant was the actor.10 That makes the occurrence of the act a fact on which relevance depends, and Rule 104(b) governs such questions:
“When the relevancy of evidence depends upon the fulfillment of a condition of fact, the court shall admit it upon, or subject to, the introduction of evidence sufficient to support a finding of the fulfillment of the condition.”11
In Huddleston, the sale of the televisions mattered under the government's theory only if the sets were stolen, so the question for the trial court was whether a jury could reasonably find that they were. The government cleared that bar. The record showed a low price per set, a large quantity, no bill of sale, and a link between Huddleston and the stolen tapes and appliances, and a jury could infer from those facts that the televisions were stolen.12 The trial judge did not have to decide for itself that the sets were stolen. It had only to see that a reasonable jury could so find.
Knowledge is the purpose in both Huddleston and this hypothetical, and both involve earlier dealings in suspect goods, so purpose does not separate them. What separates them is whether any evidence stands behind the claim that the earlier act happened and that the defendant did it. The televisions had price, quantity, and source behind them. The whispers have only rumor, which is the unsupported insinuation the relevance requirement keeps out.
The Court's answer to the prejudice worry is that other safeguards operate whenever other-act evidence is offered. Four of them matter:
- The evidence must be offered for a proper purpose, which means it must be probative of a material issue other than character.13
- The evidence must be relevant, and Rule 104(b) enforces that requirement by asking whether the jury could reasonably find that the act occurred and that the defendant was the actor.14
- The evidence must survive Rule 403, so its probative value on the non-propensity issue cannot be substantially outweighed by the danger of unfair prejudice.
- The opponent may request a limiting instruction under Rule 105 that confines the jury to the permitted purpose, as the trial judge gave in Huddleston when she told the jury to use the evidence only on knowledge and not to prove character.15
A common misreading takes the holding to mean that other-act evidence comes in whenever the prosecutor names a permitted purpose. Huddleston removed only one requirement, a separate judicial finding that the act occurred. Proper purpose, relevance under Rule 104(b), Rule 403, and a limiting instruction all remain. Weak proof that the act occurred is also fair material for the Rule 403 argument, because evidence that barely clears the sufficiency screen has little probative value to set against its prejudice.
The Rule 104(b) Standard
Rule 104(b) applies whenever the relevance of one item of evidence depends on another fact, and other-act evidence is the standard example. Every purpose in Rule 404(b)(2) assumes that the earlier act happened and that the defendant did it, so the evidence is relevant only if the jury can reasonably conclude both.16 Rule 104(b) makes a prima facie showing of the conditioning fact a prerequisite to admission. The showing is met by evidence that, viewed in the light most favorable to the proponent, would permit a reasonable juror to find the condition more probably true than not.17 The rest of this subsection explains who decides what, how the judge's screen works, and why the screen is set low.
Suppose a defendant is charged with passing a forged check and says she believed it genuine. To show knowledge, the government offers a witness who says he watched her pass a different forged check last year. Whether she passed that check is a question for the jury, using a more-likely-than-not standard. The judge asks only whether a reasonable juror could find that the earlier act occurred more likely than not.18 The judge's private doubts about the witness do not decide the matter, because the proponent need not convince the judge. It is enough that a reasonable juror could believe the act occurred.
A defendant who objects that the jury should hear nothing about the earlier check until the judge is personally satisfied that it was passed is borrowing the model of Rule 104(a). Under Rule 104(a) the court decides preliminary questions itself, and Huddleston's own counsel conceded that such findings follow the preponderance standard.19 Rule 104(b) reverses that allocation for conditional relevance. The jury decides whether the fact exists, and the judge decides only whether there is minimally sufficient evidence to support a finding of it.20 The reason is that if the court alone resolved such questions, the function of the jury would be unduly restricted.21 Rule 104 sorts its questions against a background of distrust, since the Advisory Committee's note repeats the orthodox view that the exclusionary rules grew largely out of doubts about the capacity of lay jurors.22 Rule 104(b) is where the rules trust the jury anyway, and the judge's screen keeps from it evidence that has nothing behind it. In running that screen, the court may consider only the evidence the jury will have before it, evidence admitted under the rules of evidence.23
Within other-act evidence, the preliminary questions therefore divide between judge and jury. The judge decides whether the evidence is offered for a proper purpose, which Huddleston called the threshold inquiry, and whether its probative value is substantially outweighed by the danger of unfair prejudice under Rule 403.24 The jury decides whether the defendant committed the act.25 The line is not self-evident, because the question whether the defendant was the actor became one of conditional relevance largely because the Supreme Court so characterized it.26 The allocation is worth memorizing rather than deriving.
The second step matters because the screen looks only at what the jury will hear, and the third explains why the forged-check witness carries the proponent through it: credibility belongs to the jury, not to the judge. The word “together” in the second step carries weight as well, because the standard is met by the whole body of proof and no single item has to settle the matter. Huddleston shows how proof accumulates. No one fact established that the televisions were stolen, but the low price, the large quantity, the missing bill of sale, and the link to the stolen tapes and appliances together let a jury reasonably infer that they were.27
The order of proof is largely within the trial judge's discretion, which is why a promise to connect up is acceptable at all, but the promise carries a price. Evidence admitted on that promise will be excluded on a motion to strike if the connection is never established.28 The screen also continues after admission. If, considering all the evidence, including what the opponent later introduces, the jury could no longer reasonably find that the condition was fulfilled, the court should instruct the jury to disregard the evidence.29 Candidates err in two directions here. Some treat the screen as the judge's own factfinding, which it is not. Others treat it as no inquiry at all, which overlooks that rumor and insinuation with nothing behind them cannot support a finding and fail the screen.
The standard draws a fair criticism, which is that it looks lenient. From the defendant's side it seems to lower the standard for conviction from proof beyond a reasonable doubt to mere sufficiency. The better view is that the problem lies in the evidence and not in the admission rule, because once admitted, evidence of prior criminal activity has very strong probative value in cases like Huddleston, where knowledge is the contested issue.30 Nor can the answer be to raise the threshold. Individual items of evidence need not meet the standard that governs the case as a whole, and if each item had to be proved beyond a reasonable doubt, the prosecution would find it virtually impossible to obtain a conviction.31 The prosecutor therefore need not prove the earlier act by clear and convincing evidence, and certainly not beyond a reasonable doubt.32
The federal allocation interprets the Federal Rules of Evidence, and state courts applying their own versions of Rule 404(b) have gone different ways. Vermont followed the federal approach in State v. Wheel. Colorado, in People v. Garner, has the judge decide the question under Rule 104(a) by a preponderance. New Jersey, in State v. Hernandez, and Arizona, in State v. Terrazas, go further and require clear and convincing proof that the defendant committed the act.33 A federal evidence problem is answered under Rule 104(b). A problem expressly set under a state rule requiring a judicial finding is answered under that rule, because the jurisdiction's own choice controls.
Acquittals and Later Acts
Rule 404(b) reaches evidence of “any other crime, wrong, or act,” and the phrase asks nothing about how the earlier event ended in court or whether it came before or after the charged crime.35 The rule regulates the use the proponent makes of the event, so any act that reaches a disputed fact without passing through the defendant's character can serve a purpose listed in Rule 404(b)(2).36 Huddleston supplies the proof requirement: the jury must be able to reasonably find that the defendant committed the act, and the evidence must still survive the Rule 403 balance. Two follow-on questions test that framework at its edges. What happens when a jury has already acquitted the defendant of the other act? And can an act that came after the charged crime be an “other” act at all?
An acquittal looks at first like the end of the matter, since a jury has already declined to find that the defendant committed the act, and using the act again feels like punishing what the first verdict forgave. The objection loses force once you ask what an acquittal decides. A not-guilty verdict means the prosecution failed to prove guilt beyond a reasonable doubt, which leaves open that the defendant did the act and the proof simply fell short of that demanding standard. The standard for other-act evidence is far lower: the court asks only whether the jury could reasonably find that the defendant committed the act, and the prosecutor need not meet a clear-and-convincing burden, much less proof beyond a reasonable doubt.37 One body of facts can therefore leave a reasonable doubt in the first trial and still be enough for a different jury to find the act more likely than not, and the two outcomes do not contradict each other. A rule of automatic exclusion would treat a failure of proof beyond a reasonable doubt as though it were a finding that the act never occurred.
The leading illustration is Dowling v. United States, a trial for a bank robbery in which the robber wore a ski mask and carried a pistol. The trial court let a woman testify that the defendant and a man named Christian had entered her home two weeks after the robbery, when the defendant wore a knit mask with cutout eyes and carried a handgun. She could identify the defendant because the intruder was unmasked in a struggle. The defendant had been acquitted of that home invasion, yet the government offered her testimony to strengthen the identification of the defendant as the robber and to link him with Christian, whom other evidence tied to the robbery.38
The Supreme Court held that the acquittal did not keep her testimony out. Under Huddleston, the prosecutor had to prove the earlier act only to the lower standard, and the Court rejected the argument that admitting her testimony unfairly forced the defendant to spend time and money relitigating points already resolved.39 The cost of defending an old charge a second time is real, but a side trial over the earlier event is a concern for the Rule 403 balance and does not justify a categorical bar. The acquittal also gives the proponent nothing in return. A prosecutor with no theory of relevance cannot borrow one from the fact that an acquittal fails to exclude.
Dowling differs from the handbag scene in what the other act offered. The masks, the handgun, and the link to Christian were facts tied to who committed the robbery, while the handbags were tied to nothing but the defendant's supposed type. The federal approach to acquittals also reflects the federal rules. As with the proof standard, a stem expressly governed by a state rule calls for that jurisdiction's rule, and a stem silent on the point calls for the federal approach.
The second follow-on point concerns timing. Rule 404(b)(1) speaks of “any other crime, wrong, or act” and says nothing about sequence, so an act that came after the charged crime is as much an other act as one that came before it.40 Admissibility depends on the route from the act to a disputed fact, and a later act can reach a fact as directly as an earlier one. Dowling is itself a later-act case, because the home invasion followed the robbery by two weeks, and the testimony was still admissible for the purpose the government offered.41 The same proof standard governs, so the jury must be able to reasonably find that the defendant committed the later act.
The later acts in the first scene come in because they bear on one specific question, and what they show about the defendant's character plays no part in the reasoning. The forgery says nothing on that question. Distance in time still matters, because a later act far removed from the charged crime loses probative force, but that is a point for the Rule 403 discussion that follows.