by On the twenty-second of May, 1929, at Fort Riley, Kansas, a woman named Zenana Shepard sent her nurse to a closet for a bottle of bootleg whiskey. She said the liquor had a strange smell and a strange taste, and she asked whether enough remained in the bottle to test it for poison. Then she said the sentence that would outlive her by four years and travel all the way to the Supreme Court of the United States: “Dr. Shepard has poisoned me.” The doctor was her husband, Charles A. Shepard, a major in the Army medical corps, and the government would later argue he had salted that whiskey with bichloride of mercury because he loved a woman named Grace Brandon and his wife declined to divorce him. Zenana Shepard died on the fifteenth of June. Her husband was convicted of her murder, and her sentence, the spoken one, did much of the convicting.
Four years later, Justice Benjamin Cardozo threw the sentence out. His reasoning carries the whole strange doctrine this piece is about, and the problem, Cardozo wrote, was hope. When Zenana Shepard accused her husband, she had rallied. Her condition had improved, her recovery was expected, and a fortnight after the accusation, grave as things had turned, she asked one of her physicians, “You will get me well, won’t you?” A woman who asks that question still has hope, and hope, in the law of evidence, is disqualifying. “Fear or even belief that illness will end in death will not avail of itself to make a dying declaration,” Cardozo wrote. There must be a settled hopeless expectation that death is near, and the words must be spoken, in his phrase, “in the hush of its impending presence,” by a patient with “the consciousness of a swift and certain doom.” The Supreme Court of the United States excluded a poisoned woman’s accusation of her poisoner because she had wanted, too visibly, to live.
The doctrine is the dying declaration, one of the oldest exceptions to the rule against hearsay. Hearsay is barred because an absent speaker cannot be cross-examined, and cross-examination is how the Anglo-American trial tests truth. The dying declaration walks through that wall. A murder victim’s final accusation may be repeated in court by whoever heard it, and the jury may weigh those secondhand words as if the dead had taken the stand. The classic statement of why comes from 1789, in Rex v. Woodcock, where William Woodcock stood charged with murdering his wife. Her account of her injuries, given on oath to a magistrate before she died, came into evidence even though she never spoke of her own death; the court decided her wounds must have told her what was coming. Chief Baron Eyre then explained the principle in words every American law student still reads: such declarations are “made in extremity, when the party is at the point of death, and when every hope of this world is gone: when every motive to falsehood is silenced, and the mind is induced by the most powerful considerations to speak the truth; a situation so solemn, and so awful, is considered by the law as creating an obligation equal to that which is imposed by a positive oath administered in a Court of Justice.”
Strip the wig off that sentence and a sermon stands underneath. Eyre’s dying speaker tells the truth because she is about to meet the one judge no perjury statute can reach; the deathbed becomes a courtroom because eternity is the bailiff. The old maxim behind the rule says so outright: nemo moriturus praesumitur mentiri, no one about to die is presumed to lie. That presumption was built for a world in which nearly everyone believed a soul stood in the doorway with an appointment to keep, and the rule has outlived the world that built it. The priest left most American deathbeds decades ago; the drip and the chart moved in; the presumption stayed. Our secular courts still run a rule whose motor is the fear of God.
The modern American version lives in Federal Rule of Evidence 804(b)(2), which admits, in a homicide prosecution or a civil case, a statement by an unavailable declarant who believed death was imminent, concerning the cause or circumstances of that death. States run their own variations, some wider and some narrower, and the federal rule is the spine. The edges of that rule are stranger than the rule. Death itself is optional for the declarant: she has only to believe the end is at hand and then be unavailable to testify. A woman who names her attacker in what she takes for her last breath, then survives into a coma, has made a dying declaration without dying, and the strangeness doubles when the charge is drawn: her assailant now faces trial for attempted murder; an attempted murder is a homicide that failed to finish, and the exception fails along with it. The accusation she spoke in full sincerity at the door of death is barred from the criminal case it created, while the identical words sail unopposed into a civil suit over the wound. Through all of it runs Cardozo’s requirement of despair. The speaker must have given up; hope contaminates the testimony.
Nothing shows the rule’s grip like the modern Confrontation Clause. In 2004, in Crawford v. Washington, the Supreme Court rebuilt the Sixth Amendment around a hard principle: a testimonial accusation reaches a jury only if the accused had a chance to cross-examine the accuser. Whole categories of prosecution evidence fell. The dying declaration survived in a footnote. Justice Scalia, citing Woodcock by name, acknowledged authority for admitting even plainly testimonial dying declarations, and wrote that if the exception must be accepted on historical grounds, “it is sui generis,” a thing of its own kind. Two decades on, the deathbed and forfeiture by wrongdoing remain the only founding-era doors the Court has acknowledged through its confrontation wall. Cross-examination, our courts like to say after Wigmore, is the greatest legal engine ever invented for the discovery of truth. The dying declaration is that engine’s one permitted silence. The accuser can never be questioned, and the law’s answer is that death already questioned her.
A physician standing in the lawyer’s place sees a different room. Evidence teachers reduce every witness to four risks: did she perceive the event, does she remember it, can she put it into words, does she mean to tell the truth. Those four risks are the reason hearsay gets barred and cross-examination gets prized, and the deathbed aggravates every one of them, because the dying brain is an organ in failure. Oxygen falls, carbon dioxide climbs, the liver and kidneys stop scrubbing the blood, infection flares, and the opioids that make dying bearable fog whatever remains. Medicine has a name for the result, terminal delirium, and medicine has counted it. In a prospective study of advanced cancer patients admitted to a palliative care unit, published in the Archives of Internal Medicine in 2000, Lawlor and colleagues found delirium in 42 percent of patients on arrival and in 88 percent of those who died on the unit, in the hours and days before death. Nine dying people in ten. A homicide’s deathbed usually runs faster than a cancer ward’s, and speed is no cure: hemorrhage starves the brain that tumors never touched, shock narrows perception to a tunnel, morphine arrives with the paramedics, and the questioner leaning over the gurney is often a police officer holding a theory. The condition the law selects for, the settled hopeless expectation, the swift and impending doom, names the clinical window in which confusion peaks. A courtroom would never seat a witness in that state. The law does one better: it seats her after she can no longer be questioned at all, and calls her reliable because of the hour.
And yet. Medicine has also recorded the opposite, and honesty requires it here, because the opposite is the strangest fact in this entire subject. Some dying people clear. A patient mute with dementia for years sits up, knows her daughter’s face, speaks in whole sentences, says goodbye, and dies within the day. Reports of this reach back across two and a half centuries of medical writing, and in 2009 the biologist Michael Nahm and the psychiatrist Bruce Greyson gave the phenomenon its name, terminal lucidity. Their 2012 review with Emily Williams Kelly and Erlendur Haraldsson gathered eighty-three cases, and in the collected reports most patients died within a week of the lucid episode, nearly half within a day. No one can explain it, no one can predict it, and no one can produce it on demand, which means it can rescue a family and can never rescue a legal rule; a presumption applied to every dying declarant cannot stand on a marvel that visits a few. But I will grant the old judges this much: the folk observation beneath their doctrine was a real observation. People have watched the dying come back to themselves at the last hour for as long as people have kept watch. Eyre built a courtroom on it. He built on the exception.
So the law believes the dying, and medicine mostly doubts them. For a third opinion, ask the theatre, which has been taking depositions at deathbeds for twenty-five centuries and writing down what it heard. The stage delivered its verdict on the dying declaration long before Woodcock, and the verdict sits in the fifth act of Othello, where Shakespeare stages two of them in one bedroom, minutes apart, and lets them cross-examine each other.
Desdemona goes first. Othello has strangled her on their bed, in the grip of an adultery that never happened, and as Emilia beats on the door the dying woman surfaces for a final breath. Emilia asks the question a coroner would ask: “O, who hath done this deed?” And Desdemona, with every hope of this world gone, with every motive to falsehood silenced, answers: “Nobody; I myself. Farewell. Commend me to my kind lord.” Then she dies. It is a lie. She did no such thing, and she spends her last breath on earth saying she did, to shield the husband who killed her. Some readers soften the line into self-blame, a wife hauling the guilt onto her own shoulders, and the softening changes nothing at law: the statement is false on its face, and false in the one direction that matters, away from the man who did it. Eyre’s whole theory detonates on that bed. The point of death did induce her to speak from the deepest place in her, and the deepest place in her held love, and the love produced a falsehood. Othello, standing over her body, catches the lie at once: “She’s like a liar gone to burning hell: ’twas I that killed her.” Give the man his due. It is his one clean piece of evidence-reading in the entire play, and he uses it to slander her corpse. Scholars have called her lie a gift, a devotion, a small martyrdom, and they are right about its beauty. Its beauty is the problem. A courtroom cannot cash beauty. Run her sentence through the federal rule: a declarant who believes death imminent, speaking on the cause and circumstances of that death, in a homicide case, unavailable forever. Admissible. The machinery built on the premise that the dying speak true would carry her loving lie straight to the jury, stamped with the force of a positive oath.
Minutes later, on that bed, Emilia answers her. Stabbed by her own husband for exposing the handkerchief plot, she asks to be laid beside her mistress and swears her dying testimony in the plain form of the doctrine: “Moor, she was chaste; she loved thee, cruel Moor; so come my soul to bliss as I speak true.” Here stands Eyre’s ideal declarant, arrived nearly two centuries early, a woman staking her soul on the accuracy of her last words, whose last words happen to be accurate. One scene, two dying declarations. Emilia performs the legal theory to the letter, Desdemona refutes it, and the refutation comes wrapped in the play’s most tender act. Shakespeare hands the doctrine its proof and its counterexample in a single room and declines to referee. Eyre’s premise never made it onto the boards at all. A last line, in the theatre, reveals the speaker, and whether it also reveals the facts depends on who is doing the dying.
King Lear supplies the third witness. Edmund, cut down by the brother he wronged, lies dying among the wreckage of his own plots and listens to Edgar’s account of their father’s death. Something in him turns. “I pant for life,” he says. “Some good I mean to do, despite of mine own nature.” The good is a confession: he has ordered Cordelia hanged in her cell, and the order can still be countermanded. “Nay, send in time,” he begs, and here Shakespeare turns the legal fantasy inside out, because this dying declaration is true, urgent, complete, and useless. The officer runs. Lear enters carrying his daughter’s body. The truth kept to Edmund’s schedule, and Edmund’s schedule had run out. Even his honesty arrives in character; he confesses in the language of a bargain, one good deed itemized against the ledger of his own nature. Mercutio dies punning, a grave man by tomorrow. Emilia dies loyal and loud, Desdemona dies shielding, and Edmund dies bargaining. The deathbed, on the evidence of the stage, is a distillery: what comes out is the speaker, reduced and concentrated, and the words are evidence of that before they are evidence of anything else. A dying declaration is testimony about who died. Whether it is also testimony about what happened depends on the character of the dying, and the law has no rule for character.
Why does the belief persist, then, against the medicine and against the drama? Because we need it. Last words carry a scarcity value no other words possess; they close an account no one can reopen, and the mind mistakes the unrevisable for the true. Grief wants a verdict, and the dead have stopped producing evidence, so we promote their final sentence to a ruling. Religious residue clings to it too, even among the unchurched, the sense that a person at the threshold has one foot in a country where lies have stopped mattering, and so has stopped telling them. And there is the plain mercy of it. Telling a jury to distrust a dead woman’s last accusation feels like killing her twice. The rule survives on that feeling, and Cardozo, of all people, proved the feeling has limits, because the feeling did nothing for Zenana Shepard. She told the truth as the jury found it: mercury in her organs at the autopsy, a husband in love elsewhere, a conviction for her murder. Her accusation was as accurate as Emilia’s, and the law threw it out anyway, for the crime of hope. A doctrine that promises to honor the dying honored its own machinery instead, and the machinery ran her over.
Every courtroom keeps a chair for witnesses, set where the jury can watch a face under pressure, because our trials trust faces under pressure over paper, memory, or grief. In a dying-declaration case the chair stays empty. Twelve strangers lean instead toward a bed that has long since been stripped and remade, in a room none of them will ever see, listening for a sentence nobody can call back for one more question, spoken by someone who was, in her final minutes, whoever she had always been. The law swears that sentence in. The theatre writes it down as an exit line and knows the difference between the two. I side with the theatre, and I hold a wary respect for the law’s old wager even so, because somewhere beneath it sits a hope of our own, the living court’s dying declaration: that when our turn comes, at the point of death, with every hope of this world going, someone will lean in, and listen, and believe us. The chair stays empty. The room leans in.
No One Dies Lying: Last Words on Trial is available now from David Boles Books. The Kindle edition and the 365-page paperback are at Amazon, and a free PDF edition, fully designed, with the complete index, table of cases, and glossary, is at bolesbooks.com. Read it in whatever form suits you. The dying declaration has waited more than two centuries for its cross-examination; it can wait the extra minute while you choose.
Leave a Reply