Concept
When the Exception Fails
Statement under the belief of imminent death
When the Exception Fails
Rule 804(b)(2) admits a statement only when the declarant spoke while believing death to be imminent, spoke about the cause or circumstances of that death, and is offered in a homicide prosecution or a civil case.1 The exception also sits among those that apply only if the declarant is unavailable as a witness, which adds a fourth condition.2 Rule 804(a) says when a declarant counts as unavailable.3 A proponent can therefore fail at any of four points: the speaker hoped to recover, the words strayed from the death, the case is a prosecution for something other than homicide, or the declarant can testify.
Any one of those failures ends the analysis under Rule 804(b)(2), and under no other rule. The Federal Rules sort the hearsay exceptions into two groups, one for which the declarant's availability is immaterial and one usable only when the declarant is unavailable, and they add a catch-all for circumstances that no other exception covers.4 Each exception is a separate route around the rule against hearsay, with its own conditions and its own reason to trust the words. A statement stopped at one route must still be tested against the others, and an answer that calls it inadmissible at the first failed element stops one step too soon.
The way a statement failed usually shows which route to try next. A speaker who was shaken and badly hurt but never settled into an expectation of death has missed the belief condition, yet the stress that undercuts that claim may supply a different ground for trusting the words. That is the excited-utterance route, and it comes first. A statement that misses by one condition but carries other circumstantial guarantees raises the question whether the residual exception can rescue it, and that route comes second. The third route asks whether the proponent needs the truth of the words at all, because a narrower purpose might admit them, and Shepard v. United States shows why that route fails when the words accuse the defendant of the charged act.
Excited Utterance Fallback
The first alternative to test is Rule 803(2), which admits a statement relating to a startling event or condition that the declarant made while under the stress of excitement the event or condition caused.5 The theory is that excitement temporarily stills the capacity for reflection, so the words come out free of conscious fabrication.6 That guarantee comes from the speaker's emotional state at the moment of speaking, and it needs no belief about death. A person speaking in circumstances that involve the possibility of death is often under just such stress, so statements that fail as dying declarations may often be received as excited utterances.7
Three differences from the dying declaration can each rescue a statement that fails Rule 804(b)(2). The first is availability. A dying declaration stays closed whenever the speaker can testify, but an excited utterance carries no requirement that the declarant be unavailable.8 The stress that stills reflection is a fact about the moment of speaking, and it does not change if the declarant later recovers and takes the stand. The proponent also escapes the showing that Rule 804(a) demands of a declarant who is merely absent, namely that the proponent could not procure the declarant's attendance by process or other reasonable means.9 A victim who spoke in shock, survived, and now sits in the courtroom cannot be heard through Rule 804(b)(2), yet the witness who heard her may repeat her words if Rule 803(2) is satisfied.
The second difference is the kind of case. Mae, a rape victim, tells a medical responder who attacked her, and she believes at that moment that she is dying. Dying declarations are admissible in a criminal trial only if the trial is for homicide, and they are inadmissible in prosecutions for rape, robbery, theft, and the like.10 Rule 804(b)(2) is therefore closed to her words even though her belief and her subject matter fit. Rule 803(2) draws no such line. If the assault was a startling event and Mae was still gripped by its stress when she spoke, her statement relating to the event may be heard.
The third difference is subject matter. A dying declaration is limited to statements about the cause or circumstances of the declarant's impending death, while an excited utterance extends to statements related to the startling event or condition that produced the emotional state.11 The word 'related' reaches well beyond a death. Two cases cited in the commentary show the breadth. One sustained a store clerk's remark, in a slip-and-fall case, that a hazard 'has been on the floor for a couple of hours,' and the other upheld a driver's remark, offered on the issue of agency, that he had to call on a customer and was in a hurry to get home.12 Neither remark concerns the cause of a death, and each relates to the event.
What Rule 803(2) demands in place of the belief of imminent death is stress that lasted until the words were spoken. The proponent of a dying declaration proves the belief because the belief carries the guarantee of truthfulness, and the proponent of an excited utterance proves the stress because the stress carries it. The measure is the duration of the excitement, and elapsed time is only one factor, alongside the nature of the event or condition, the declarant's age and condition, the presence or absence of self-interest, and whether the statement was volunteered or made in response to a question.13 Peter, injured in a crash, gives his account fifteen minutes later. The plaintiff offers evidence that he was still in shock, and the defense offers evidence that he had regained his composure. The judge resolves that dispute and will admit the statement only if it is more likely than not that the stress of the crash still held him when he spoke.14 A finding that his composure had returned defeats the exception even though the crash was real and his words concern it.
Tomas's statement comes in on stress alone, and Ines's comes in on belief alone. Neither exception swallows the other, so a proponent checks both before conceding the hearsay objection.
An exam stem that pairs a hearsay objection with a failed dying-declaration element, such as a non-homicide charge or a speaker who expected to live, tests whether you stop or move on. Move on: look for a startling event and ask about the declarant's condition when speaking. Do not carry the belief of death into that analysis, since Rule 803(2) does not require it, and do not drop the stress requirement, since Rule 803(2) does. If neither belief nor stress can be shown, a statement that nearly fits Rule 804(b)(2) may still have the residual route.
Residual Exception Fallback
The residual exception no longer sits in Rule 804. Rule 804(b)(5), which once held the catch-all, now says only that the provision was transferred to Rule 807.17 Rule 807 may be invoked when a hearsay statement fits none of the specific exceptions in Rules 803 and 804 and the court nonetheless finds it trustworthy.18 It is the second route to test, after the excited utterance, because it applies only when the statement fits no specific exception.
The trustworthiness inquiry looks to circumstances that do the work the speaker's belief in imminent death does under Rule 804(b)(2). Courts consider the disinterestedness of the speaker, the short duration between the event and the declaration, the likelihood that a false statement would be discovered, and reliance on the statement by persons the declarant would not want to deceive.19 Each factor addresses a different way an account goes wrong. A speaker with no stake has little reason to shade the story, a short interval leaves little time to invent or forget, a lie likely to be exposed is costly to the person who tells it, and a statement that others relied on suggests the speaker expected it to be taken seriously by people the speaker would not want to mislead.
A statement that nearly satisfied a specific exception can use this route. The amendment makes Rule 807 apply to hearsay 'not admissible under' the Rule 803 and 804 exceptions, which clarifies that a court assessing guarantees of trustworthiness may consider whether the statement is a near miss of one of those exceptions.20 The Advisory Committee Note adds a caution that shapes the whole route: a court using near-miss analysis should evaluate all relevant guarantees of trustworthiness and also take into account the reasons the hearsay misses the admissibility requirements of the standard exception.21 A court asked to rescue a statement therefore weighs why it fell short before deciding whether the remaining circumstances make up for the gap.
Mae's accusation shows how the argument runs. Her statement to the medical responder met the belief and subject-matter conditions, yet Rule 804(b)(2) stayed closed because a prosecution for rape is not a prosecution for homicide.22 The prosecutor can argue that the words are a near miss with strong guarantees: they came moments after the assault, which left almost no time to fabricate, and a false accusation naming a specific attacker would be likely to come to light. Those points answer the reliability worry. The obstacle is the reason for the miss, since the missing element is the homicide limit. The prosecutor will say the circumstances make up for it, and the defense will say Rule 807 should not erase a limit that the dying-declaration exception sets by its terms. The court decides that contest by weighing the guarantees against the reason for the miss.23
The missing element in the two accusations is the same, the homicide limit, so it cannot be what separates them. What separates them is everything around the words: moments after the assault in one case, and hours of reflection and a motive to implicate in the other. The listed factors describe the circumstances of the statement, and none of them depends on what the speaker expected to happen.
The residual route is therefore demanding, and it is no easy substitute for the dying-declaration conditions. A court must find real guarantees of trustworthiness and must weigh why the statement missed the specific exception.24 An answer that reaches for Rule 807 as the first reply to a failed element has the order reversed, because Rule 807 is invoked when a statement fits no specific exception.25 Test the specific exceptions first, including the excited-utterance route, and turn to the residual exception after naming the element that failed and the circumstances that make up for it. Rule 807 carries further conditions that belong to the full treatment of the residual exception. Even a successful residual argument asks the factfinder to treat the words as true, which leaves a different question: whether the proponent needs the truth of the words at all.
Limited-Purpose Theories
Shepard shows what happens when the government, unable to rely on the dying-declaration exception, offers the same accusation under a narrower label. The defense had called witnesses who testified that Mrs. Shepard said more than once before her illness that she had no wish to live and had nothing to live for, and on one occasion that she expected some day to make an end to her life.29 That testimony gave the defense a suicide theory. The government argued that it also opened the door to rebuttal declarations that her husband had poisoned her, offered not for their truth but as showing a state of mind inconsistent with suicidal intent.30 The argument has surface appeal, because an out-of-court statement used for a purpose other than its truth raises no hearsay problem and needs no exception. The Court rejected it in two steps, and each rests on a different reason.
The first step looked at the record. The testimony was received by the trial judge and offered by the government with the plain understanding that it would be used for an illegitimate and gravely prejudicial purpose.31 In fact the government used it as proof of an act committed by someone else, as evidence that Mrs. Shepard was dying of poison given by her husband.32 A defendant who watches evidence offered and accepted on that footing has no reason to ask for a ruling that confines it to some other use. A trial becomes unfair if testimony thus accepted may be used in an appellate court as though admitted for a different purpose, unavowed and unsuspected. The result is the same when the purpose held in reserve is so obscure and artificial that it would be unlikely to occur to uninstructed jurors and, if it did, would be swallowed up in the purpose disclosed.33 The Court added that the case would differ if the purpose had been left at large, without identifying token, since the defendant might then have been expected to demand an explanatory ruling. Here the course of the trial put the defendant off his guard.34
The second step assumes the narrow purpose had been avowed and rejects the theory anyway. The Court said the defendant would have no grievance if the rebuttal had been narrowed to that point, the existence of a vital urge.35 What the government put in evidence was something very different. It used the declarations as proof that someone else committed an act, and that fact the government was free to prove, but not by hearsay declarations.36
The Court's reason is practical. A limited-purpose label works only when jurors can use evidence for the narrow purpose and put it out of mind for the broad one. An accusation that names the defendant as the poisoner defeats that sequence, because the same words that supposedly show a will to live also charge another person with the death. It will not do to say that the jury might accept the declarations for any light they cast on a vital urge and reject them to the extent they charged the death to someone else.37 'It is for ordinary minds, and not for psychoanalysts, that our rules of evidence are framed,' and when the risk of confusion is so great as to upset the balance of advantage, the evidence goes out.38
The decision therefore draws a line inside a single set of facts. Statements showing that Mrs. Shepard wanted to go on living, offered to answer the defense's showing that she wanted to die, speak to her own thoughts and feelings, and rebuttal narrowed to that point would have left the defendant no complaint.39 Her accusation that her husband had poisoned her asserts what another person did. Once the government used it to show that he had done so, it served as proof of the charged act and could not be rescued as state-of-mind evidence.40
The typical exam setup is a victim's accusation that has just been held inadmissible under Rule 804(b)(2) and is then offered again for a narrower reason, most often to show the victim's state of mind in answer to a suicide defense. Two questions test the move. The first asks whether the narrow purpose was the one on which the evidence was actually offered and received at trial, because a party cannot rely later on a purpose it never avowed. The second asks what the jury will do with the statement once it hears it. If the statement names the accused as the person who committed the charged act, ordinary jurors will take it as proof of that act, and the limited-purpose theory fails. The accusation then stays out unless another route, such as the excited-utterance or residual exception, actually fits, since the failure of a narrow label closes only that one door.