Divine Providence and Dr. Parkman's Jawbone: The Cultural Construction of Murder as Mystery

by Karen Halttunen

Karen Halttunen is Professor of History at the University of California, Davis. She delivered a lecture on this topic at the National Humanities Center where she was a 1994-95 Fellow. Below: title page from a nineteenth-century trial narrative. All illustrations for this essay have been provided by the American Antiquarian Society, Worcester, Massachusetts.

Photo from nineteenth-century trial narrative When twelve-year-old Hannah Ocuish killed six-year-old Eunice Bolles in New London, Connecticut in 1786, the Rev. Henry Channing preached her execution sermon, warning parents and masters not to neglect the religious instruction of children. Fifty years earlier, this sermon, a classic example of its genre, would probably have been all that was printed concerning Hannah's crime. But in 1786 the editors appended to it a brief account which began: "On the 21st of July, 1786, at about 10 o'clock in the morning, the body of the murdered child was found in the public road leading from New-London to Norwich, lying on its face near to a wall." The document went on to trace the investigation that followed: The neighborhood turned out to hunt for the murderer; Hannah was questioned and claimed that she had seen four boys near the scene of the crime. When a search failed to turn them up, Hannah was interrogated again, and then taken to the Bolles home to be charged with homicide in the presence of the dead child. She burst into tears and confessed. Only at this late point in the narrative is the reader offered a sequential account of the crime. Five weeks earlier, Eunice had reported Hannah for stealing fruit during the strawberry harvest, and Hannah had plotted revenge. Catching sight of her young enemy headed for school one morning, Hannah had lured Eunice from her path with a gift of calico, then beat and choked her to death.

This is a peculiar way to tell a story. It begins with its own endpoint--a dead body--then moves into the past to gather fragments of information, and only at last recounts the events leading up to the murder, coming full circle at its own starting point: a dead body. The volume entitled God Admonishing his People of their Duty, as Parents and Masters marked a literary boundary between the early American execution sermon and a strikingly new way to tell the tale of a murder, based on that "peculiar distortion of more usual narrative conventions" which characterizes detective fiction. Playing on the narratological distinction between recit--the order of events in the text--and histoire--the historical sequence in which those events actually occurred--, the Ocuish narrative, like detective fiction, worked on two levels at once: on the surface openly narrating the story of an investigation, while gradually bringing to light the hidden story of the crime.

But the Ocuish narrative was strikingly different from detective fiction in two significant ways. First, it was not fiction (though no narrative is purely nonfictive). It concerned an actual murder, the nature of which was fully understood at the time of publication; and it focused on a real murderer, whose identity was also known--not just to the volume's editors, but to readers who had just finished reading the execution sermon that preceded this appendix. Second, it set forth no character resembling a detective, no single organizing intelligence to make sense of the crime. In 1786, before the professionalization of law-enforcement in America, the investigators into the Bolles murder were local townspeople.

The term that best captures the nature of the Ocuish account is murder mystery, a narrative that approaches the crime as a mystery, a problem to be solved. The cultural construction of murder-as-mystery was already underway in the nonfictional literature in 1786, over half a century before Edgar Allan Poe's "invention" of the genre with "The Murders in the Rue Morgue" in 1841. The Ocuish account was an early expression of the Gothic cult of mystery, the second major convention which--along with Gothic horror--was reshaping the popular understanding of the crime in the late eighteenth and early nineteenth centuries.

Early American execution sermons--such as that preached for Hannah Ocuish--assigned no mystery to the crime of murder. Their guiding assumption about the crime was a centuries-old English proverb, "Murder will out." Just as the blood of Abel had cried from the ground, calling God to bring Cain to justice for his terrible crime, so did the blood of every murder victim in early New England. As the Rev. William Shurtleff explained in 1739 at the execution of two infanticides, "All our Iniquities are before God, and our secret Sins in the Light of his Countenance, who can when he pleases bring to Light the hidden Things of Darkness." In the sacred cosmos of the execution sermon, the Providential eye witnessed all secret sins, and the Providential hand discovered them, bringing them to light. Early New Englanders had a practical application for this belief: They brought the suspected murderer to touch the corpse, and if blood flowed fresh from it, that person's guilt was proclaimed. (Hannah Ocuish was subjected to a late weak variant of this practice.)

The belief that "Murder will out" drew support as well from the nature of early New England criminal justice. Before the mid-eighteenth century, the Anglo-American criminal trial was not adversarial. It worked rather as an inquisitorial procedure conducted by the judge whose job was to ferret out God's truth, considered to be single and uncontestable. In early New England, the conviction rate in felony trials was quite high--in the neighborhood of 90 percent--and was accompanied by a high rate of confession among those convicted. So there was no mystery here. In the legal world of early New England, indicted criminals were guilty criminals, convicted criminals obligingly confessed their guilt, and the dominant print form for addressing the crime of murder assumed without question that such moral certainty was guaranteed by God's active presence throughout the entire process. The execution sermon never even mentioned the process of criminal justice which had brought the condemned malefactor to the scaffold. Significantly, early New England print culture offered no format for addressing unsolved or unproved murders.

The moral certainty of the execution sermon would soon, however, give way to the moral uncertainty of a new kind of crime literature, the legal discourse of murder which replaced the earlier theological discourse in the late eighteenth and early nineteenth centuries. In the last decades of the eighteenth century, murder trial transcripts began to appear in print, along with hybrid accounts which included the murder trial as one of their central concerns. In the early nineteenth century, a third category of legal narrative emerged: the trial fragment (e.g., the closing for the defense, the judge's charge to the jury) or legal commentary on the trial. Comprising just 15 percent of all murder narratives printed before 1789, this legal literature steadily rose in proportion after 1790, reaching a level of 66 percent in the 1820s and staying there for the duration of the nineteenth century.

The emergence of these legal narratives demonstrates the break-up of the clerical monopoly over the public discourse on murder, an aspect of the larger loss of cultural authority suffered by the clergy in the eighteenth century. Early in the century, the clergy had provided 70 percent of all learned professionals in the colonies and had expressed the leading ideas of their culture through one of its central rituals, the sermon. But eroded by religious disestablishment, denominational competition, and the challenge of other professional groups, clerical authority weakened. By 1800 the clergy had been reduced to only 45 percent of professionals.

The rising generation of intellectuals increasingly embraced law over theology. Still a weak and despised corporate guild early in the eighteenth century, lawyers acquired new social and economic power as agents in the commercializing trans-Atlantic economy, new political authority in their prominent role in the American Revolution, and new intellectual ascendancy in the performance of an impressive number of lawyer-writers including Noah Webster and Charles Brockden Brown. In the early republic, courtroom speech emerged as a crucial form of cultural expression, a development reinforced by the emergence of the modern adversarial format for the Anglo-American criminal trial, which enhanced the power of attorneys for both the prosecution and the defense, at the expense of the judge.

The new adversarial trial made legal truth a matter of formal contestation rather than inquisitorial determination, and thus contributed to the new moral uncertainty arising from the emerging legal discourse on murder. Moral uncertainty first appeared, ironically, in late eighteenth-century execution sermons, in which a waning clergy began to make reference to the process of human justice which had brought their convicted criminals to the scaffold. By and large, they cited the condemned murderer's trial in order to reaffirm his guilt of the crime. But the moment they began to argue the murderer's guilt, they implied the possibility of his innocence, as when the Rev. Moses Baldwin contended that William Shaw had to be guilty because his various alibis had contradicted each other. The Rev. Enoch Huntington in 1796 actually acknowledged that "Human judicatories, and the judgments and executions dependent upon them, may be erroneous"--though he went on to affirm that the trial about which he spoke had been fair.

Such moral uncertainty grew even more pronounced in the emerging legal literature of murder for a particular reason. Whereas the classic execution sermon had assumed the guilt of the convicted murderer, trial reports and other legal accounts made the guilt of the accused murderer their central problematic. After all, some murder trials resulted in acquittals; indeed, one of the earliest printed trial reports, concerning the trial of the eight British soldiers involved in the "Boston Massacre," resulted in the acquittal of six defendants and the conviction of two on a lesser charge. Editors learned to employ terms such as the "alleged" or "supposed" murderer in reporting trials which generated no formal closure on the crime because the defendant was judged "not guilty." Other trials could result in convictions that were then followed by executive reprieves or pardons or, in one early nineteenth-century case, in a retrial based on a legal technicality which resulted in acquittal. Nineteenth-century print culture offered plenty of print formats for addressing unsolved or unproved murders.

But even murder trials resulting in convictions could introduce considerable moral uncertainty into tales of the crime, due to the peculiar nature of the courtroom trial. Trials were reconstructive in nature; their task was to piece together fragments of the past in an effort to locate and assign guilt. They were often chronologically non-linear in form, doubling back upon themselves, as when members of the coroner's jury testified about the state of the corpse before any discussion of events leading up to the violent death. Trial narratives were almost invariably incomplete, because most murder cases rested on circumstantial evidence. Gaps opened: The Rev. Ephraim Avery's claim to have been wandering alone around Portsmouth, Rhode Island at the time of Sarah Cornell's death in Tiverton was never corroborated by the two people he said he encountered on his rambles, a man with a gun and a boy with sheep who were never found. Internal conflicts emerged when various witnesses testified at cross purposes: Was Eliza Fales still romantically involved with Jason Fairbanks at the time of her death, as some witnesses testified, or had she lost interest in his suit, as others maintained? Sometimes internal conflicts opened up within the testimony of a single witness, as when James Anthony, charged with murdering Joseph Green in a hatter's shop in Vermont in 1814, offered conflicting alibis concerning his whereabouts on the night of the crime, saying variously that he had been in bed at the shop, that he was out at the time, that he was out and another man was present, and that he was present at the murder but another man had committed it. Even the formal indictment itself sometimes proffered different versions of what had happened: John Francis Knapp, tried for the murder of Captain Joseph White in 1830, was charged with acting alone, committing the murder aided and abetted by George Crownin-shield, aiding and abetting Crowninshield who did the killing, and aiding and abetting a person unknown who did the killing.

Chronologically disjointed, fragmentary and incomplete, internally inconsistent, the new adversarial trial was challenging to follow and difficult to interpret. For that reason, criminal lawyers were developing an important technique for helping their most immediate "readers," the jury, to retain and interpret the chaotic stream of information coming at them. They were learning to construct coherent, "common-sense" narratives to assist jurors in their arrangement of time frames, characters, motives, and means, to achieve narrative clarity and make a decision. But it is the peculiar nature of courtroom storytelling in the modern adversarial trial to be competitive. First the prosecution and then the defense takes up the chaos of details and shapes it into as clear and compelling a narrative as possible, in an effort to sway jurors to their side. Such was the case in the trial of Lucretia Chapman in 1831 for conspiring with her young Cuban lover in the arsenic poisoning of Lucretia's husband William. The trial was a complex and information-packed proceeding involving testimony from servants, boarders, Chapman children, pupils from the Chapmans' school, neighbors, an itinerant bookseller, the family doctor and clergyman, a pharmacist, a tailor, the undertaker, the Mexican consul in Philadelphia, police officers and sundry medical experts. The prosecution took this chaos of information and shaped it into a coherent account of a lustful, matriarchal, avaricious monster of a woman who became first an adulteress and then a murderer so she might marry her foreign lover eleven days after her husband's death. The defense took the same information and crafted it into a tale of a pious and tenderly devoted wife and mother who had fulfilled her weakly feminine nature by falling victim to a diabolically clever confidence man, who had married him out of meek compliance with her dying husband's final wish, and who had had absolutely nothing to do with that final deadly seasoning of her husband's sickroom chicken soup.

The Chapman trial pitted the prosecution and the defense against one another as dueling storytellers. More commonly, the prosecution told a coherent story of the crime, and the defense countered with an anti-narrative, designed not to provide jurors with an alternative story, but to undermine the prosecution's story just enough to generate reasonable doubt. Such was the strategy for the defense of John Boies, on trial for the murder of his wife Jane in Dedham, Massachusetts in 1829. The prosecution called witnesses who testified that the defendant had, for six months before Jane's death, repeatedly battered her, openly claimed his prerogative to beat her, and once threatened her life; and asserted that she had died as a result of several mortal wounds out of some fifteen inflicted on her with an axe on the last night of her life. To this narrative of murder, the defense countered with an anti-narrative: 1) The prosecution had failed to demonstrate that Jane Boies had been murdered: A chronic intemperate subject to fits, she might have died of alcohol-induced apoplexy, or from injuries suffered during a fall from a wagon nine days before her death; 2) Even if she was murdered, the prosecution had failed to demonstrate that John Boies was the murderer; there were no eyewitnesses to the crime and the dying woman had not charged her husband with the assault; 3) Even if she was murdered and John Boies did it, the prosecution had failed to demonstrate that John Boies had murdered her with an axe as cited in the indictment: If she died rather from his extended abuse and neglect, he must be acquitted as formally charged; 4) Even if she was murdered by John Boies with an axe, John might have killed his drunken and ill-tempered wife in self-defense, in which case the jury should convict him not of murder but of manslaughter.

When the sacred narrative of murder was replaced with a secular narrative which made the "truth" of the crime a matter of human determination through legal contest, a new moral uncertainty crept into the popular discourse on the crime. In theological terms, guilt became problematic upon the withdrawal of the all-seeing Providential eye and the Providential hand of discovery. Ironically, moral uncertainty arose not from a poverty of information about the crime, but from a welter of informational detail which lent itself to various interpretive treatments. The problem with the new legal narratives was this: God no longer served as the omniscient narrator of the crime.

The departure of moral certainty from the American murder narrative did not go unnoticed. Trial lawyers, who had taken over from clergymen the major responsibility for crafting stories of murder in nineteenth-century American culture, made a major effort to restore to their legal narratives the moral certainty of the earlier theological narrative of the crime by affirming the Providential nature of circumstantial evidence. Divine Providence, they repeatedly and ritualistically asserted, was responsible for the trail of evidence that had enabled them to reconstruct the murder. The general formula, repeated in trial after trial, was this: "Providence has so ordered the affairs of men, that it most frequently happens that great crimes committed in secret leave behind them some traces or are accompanied by some circumstances which lead to the discovery and punishment of the offender. Therefore the law has wisely provided that you need not have, in cases of this kind, direct proof." The "finger of Providence was visible" in James Eldredge's accidental dropping of the key to the satchel where he stashed his arsenic; "the mysterious ways of divine Providence, which lead to the detection of secret murderers" were at work in the discovery of the letter which exposed the murderers of Joseph White; the "hand of Providence was visible" in Anna Ayer's daughter's survival of the attack which caused her mother's death; the clot of blood found in Maria Hendrickson's heart during the autopsy was "a Providence!" Within their legalistic version of "Murder will out," prosecution lawyers assured jurors that Providence--the all-seeing and all-powerful divine agency--was present in every small link in the chain of circumstantial or "moral" evidence (as it was alternatively called). This was a revealing remnant of sacred story-telling, rattling around within the larger secular narrative of the crime, a legal fiction which provided an argument, not just for the guilt of any given defendant, but for the legitimacy of circumstantial evidence itself, and in the largest possible sense, for the authoritative legitimacy of legal narrative.

Even so, lawyers could not fully close the gaps that opened up in their circumstantial narratives of criminal guilt. They were not, finally, themselves Providential in their powers of vision and agency; their narratives of murder remained incomplete. No dropped key, no incriminating note, no clot of blood could take the place of the Providential eye. Narrative gaps and puzzles and inconsistencies remained, and when the lawyers came upon them, they resorted to a language of mystery. Sometimes prosecutors invoked the term in strategic concession that their own stories of the defendant's guilt remained incomplete: Regarding a private conversation between Lucretia Chapman and her alleged accomplice following William Chapman's death, the prosecution admitted, "That interview . . . is among the unexplained mysteries of this singular history." More commonly, defense attorneys pronounced the crime utterly mysterious, to emphasize the inadequacy of the prosecution's narrative: The death of Maria Bickford "may remain in mystery till the heavens are no more," intoned Rufus Choate in closing his defense for Albert Tirrell; "the government has not reached the bottom of this transaction." Whatever the immediate legal purpose of an attorney's acknowledgment of, or insistence upon, the final element of undispellable mystery in the crime, it read as a statement of the crime's ultimate unknowability, its fundamental resistance to authoritative, omniscient narration within the new legal framework.

The discourse of the law led to the cultural construction of murder as mystery, which spilled over from formally legal narratives of the crime into the larger popular literature. The murder-mystery narrative echoed the structure of the trial report. It began with the corpus delicti, the fact of the crime, usually established by the dead body (in the trial itself, the coroner's report); then moved through the investigation, citing the fragments of circumstantial evidence uncovered by that investigation (trial testimony and the introduction of physical evidence); and pulled those fragments together within a narrative that usually remained incomplete (the prosecution's closing). Only at the end did it definitively identify the murder (the verdict). In the early nineteenth century, the murder-mystery narrative, the first example of which was the account of Hannah Ocuish's crime, grew in popularity. Significantly, editorial prefaces to printed trial reports sometimes used the format, opening with lines such as: "On the evening of the sixth of April 1830, Mr. Joseph White, a rich and respectable citizen of Salem, Massachusetts, retired to bed at his usual hour of about nine o'clock . . . . In the morning he was found dead in his bed . . . . The motives of the murderer were wrapped in mystery." A growing attention to the "mysterious" nature of murder was evident as well in the titles of nineteenth-century murder narratives, which introduced such phrases as "Mysterious Murder," "Mysterious Disappearance," and "Mysterious Affair," and labeled certain cases as "the Burdell Mystery" or "the Udderzook Mystery!"

This broader cultural construction of murder as mystery was evident as well in the special attention given to certain murders which proved particularly elusive to comprehend, such as the slaying of Mary Cecilia Rogers in New York in 1841. Her body was found in the Hudson River near Hoboken in July 1841, three days after she had disappeared from her mother's boarding house on Nassau Street. The coroner's inquest found that she had been bound and gagged, raped by several men, then strangled. But public rumors circulated that "the Beautiful Cigar Girl" (she had tended counter at a popular tobacco shop frequented by the men of New York's sporting culture) had taken her own life, or died of complications following an abortion, or had been strangled by a single male companion with whom she had been seen at a New Jersey roadhouse. One suspect after another was taken into custody and questioned, and their depositions were printed in the New York papers. But no one was ever charged with the crime, a fact that a number of writers found irresistible. Multiple and competing fictional narratives posed a range of solutions, from Poe's armchair-detective argument in "The Mystery of Marie Roget" that she had been raped and killed by a young naval officer, to Ned Buntline's laying of the crime to a "hag" and "she devil" abortionist modeled on the notorious Madame Restell. But as one hack author acknowledged, "Time passed without unveiling the mystery connected with the fate of the Beautiful Cigar Girl." If the printed trial report provided a kind of scaffolding for the cultural construction of murder as mystery, the Mary Rogers case demonstrated that the scaffolding might be taken down and the construction of murder as mystery would remain standing.

The growing concern for mystery emerging from the new legal narrative of murder led crime writers to borrow liberally from Gothic fiction a range of motifs and conventions used to enhance the "mysterious" nature of the crime. There was actually little real mystery involved in Henrietta Robinson's murder of her neighborhood grocer in Troy, New York in 1853. But popular attention was drawn to the insane defendant's insistence on veiling her face at her trial, over the protests of the judge and the jury and against advice of her own counsel. She became known as the "veiled murderess," and the one independently printed treatment of her trial offered readers a frontis illustration of her fully clothed and veiled, with her hair drawn into a bun, and a backcover portrait entitled "THE VEILED MURDERESS UNVEILED" which showed her, implausibly beautiful, hair flowing over her naked shoulders, clothed only in loose drapery which barely covered her breasts. In 1869, in the sensational Richardson-McFarland case, a veiled witness was similarly memorialized in an account entitled The Veiled Lady: or, the Mysterious Witness in the McFarland Trial.

Missing corpses usually prompted special attention to the "mystery" of certain suspected murders, as in Edward Ruloff's suspected killing of his wife and child at Ithaca. Editors of narrative treatments of his crime invoked such phrases as "shrouds of mystery" and "veils of secrecy." Even the temporary disappearance of a corpse could prompt treatment of a case as mystery: The "barrel mystery" of 1859 and the "great trunk mystery" of 1871 were both named after the murderer's method of corpse-disposal. Stephen and Jesse Boorn were convicted of the murder of Russell Colvin even though the corpus delicti consisted of just a few dubious bone fragments and items of Colvin's clothing, and the dramatic return of the alleged victim on the very eve of Stephen's scheduled execution prompted an account of the crime entitled Mystery Developed. Alternatively, the discovery of dead persons in suspicious circumstances that nonetheless offered no clear evidence of foul play were constructed as murder mysteries: Mysterious and Terrible Tragedy in the Merrimack House concerned the discovery of two disreputable women found dead in a hotel room in Newburyport, Massachusetts in 1854.

Another narrative detail frequently invoked to heighten the mystery of murder concerned the intrusion of the supernatural. When Amos Boorn received a ghostly visitation from alleged murder victim Russell Colvin that prompted the trial of the Boorn brothers, Mystery Developed made much of the incident (as did English mystery-writer Wilkie Collins half a century later in his fictionalization of the case, published in Boston in 1874). A fictionalized biography of murder-victim Helen Jewett told of her haunting by a childhood friend who had died of consumption in a New York brothel; a similar account of the life of Maria Bickford related her preternatural visitation by a raven upon her first arrival at a Boston brothel. Confessed murderers offered first-person accounts of their own guilt-wracked hauntings: Wife-killer John Lechler saw a terrific flying thing just after his crime; Andress Hall saw "a bird of warning" just before he slaughtered the Smith family; Adam Horn, who killed and dismembered his wife, saw a preternatural light near her grave, panicked, and left town, thus exposing his guilt. In a largely naturalistic nineteenth-century cosmology, such intrusions of the supernatural were uncanny, preternatural, fantastic (in the sense explored by Tzvetan Todorov)--a source of open-ended mystery. No such intrusions had occurred in earlier execution sermons. Within their supernatural cosmology, Providential activity was structured into every worldly occurrence, and posed no mystery demanding explanation.

All these conventions--veiled ladies, disappearing persons and corpses, the return of the presumed dead, terrifying intrusions of the supernatural--were drawn from Gothic fiction to lend meaning and coherence to nineteenth-century murder narratives. But it was a peculiar kind of coherence that arose from the exigencies of narrative incoherence and moral uncertainty stemming from the new legal discourse of the crime.

One of the most important characteristics of the murder mystery was the new demands it placed on the reader. The execution sermon had demanded no heuristic participation by its hearers or readers, requiring only that they accept on clerical (and ultimately divine) authority the nature of the sin, the identity of the sinner, and the proper spiritual use of the crime and its punishment. The new legal narrative of murder, by contrast, required that readers play an active role in shaping the narrative and assigning it meaning. If trial attorneys tended to treat jurors as readers whose active selection of one courtroom story over its competitors would determine the trial's outcome, printed trial reports and the various other murder-mystery narratives it generated tended to treat readers as jurors who must actively participate in the crafting of legal truth. At the simplest level, the legal narrative implicitly invited the reader to cast a vote for the guilt or innocence of the defendant, as did the unknown contemporary reader of convicted murderer James Eldredge's self-exculpation who wrote in the margin, "This statement is universally regarded as a lie from beginning to end."

But in requiring readers to make a juror-like decision concerning the guilt of the accused killer, nineteenth-century murder mysteries actually demanded much more: that readers actively shape the narrative, sifting through evidence, piecing the fragments together, tracking, detecting, probing, "unveiling"--effectively taking upon themselves the responsibility for crafting the master narrative of the event after the withdrawal of the Providential Narrator. To this end, the murder-mystery narrative form pulled readers into the investigation of the crime, from the initial discovery of the corpus delicti. Trial reports often included illustrations of key evidence in a case, such as the bludgeon used to murder Joseph White, or the cut on the hand of accused murderer Lucian Hall; one publication on the Sarah Cornell slaying consisted entirely of facsimile copies of eight letters from accused murderer Ephraim Avery to his alleged victim. Following several particularly sensational trials, key pieces of physical evidence were actually placed on public display. The hatchet used on Helen Jewett was exhibited in a window of the Police Gazette office; an expensive dress worn by murder victim Maria Bickford (whose extravagance was said to have driven her lover to take her life) was displayed on her life-size image by a Boston wax museum; and one alleged murder victim, Russell Colvin, enjoyed a brief engagement alive and in the flesh at the Albany Museum immediately after his surprise return to Manchester, Vermont cleared the Boorn brothers.

Murder-mystery narratives also paid growing attention to the spatial--and more specifically the architectural--dimensions of a murder, offering both verbal descriptions and graphic illustrations of the scene of the crime. Visual illustrations included drawings of architectural facades (the Conrad Garber farmhouse where two women were murdered, the Boston boarding house where Maria Bickford was killed) and interior scenes (Ursula Newman's parlor, Joseph White's bedchamber, the staircase of the boarding-house where Joel Clough slew Mary Hamilton). Even more important were diagrams aimed at assisting the reader in reconstructing the crime in three-dimensional space. A cartography of murder appeared. A map of Salem accompanied the account of the murder trial of John Francis Knapp in 1830; that same year, one of Bristol, Rhode Island and environs was included in the trial report for Ephraim Avery. By 1844, a far more detailed, fold-out map of Meriden, Connecticut charted the country between the residence of accused murderer Lucian Hall and his victim Lavinia Bacon; the purpose here, clearly, was to enable readers to situate spatially the murderer's movements to and from the scene of the crime. By the 1850s, detailed maps of outdoor murders were including the relative locations of the victims' bodies and various pieces of physical evidence, and citing the exact distances (e.g., 54 rods between the bodies of the two Lester boys killed by Reuben Dunbar) between them.

Even more than maps, architectural floor plans provided readers with information aimed at assisting their reconstruction of the murder. From early on, these floor plans emphasized the murderer's mode of access to the victim--rear window, unlocked door--and in some cases tracked his path to his victim with a dotted line. And with increasingly frequency over time, these floor plans showed the location and position in which the victim's body was found--Lavinia Bacon stretched out on the floor of her front room, Ruth Fyler in the doorway between the sitting room and the bedroom she had shared with her murderer-husband. The editor of the Fyler trial report explained that the purpose of this diagram was to "enable the reader to perfectly understand the location and relative positions of the several apartments, and to better comprehend the testimony in which allusion is made to the house and the terrible deed of which it was the scene." In this new visual-spatial reconstruction of the murder, the narrative responsibilities of the Providential eye were effectively passed on to the "private eyes" of the ordinary reader, who was required to penetrate visually the secret spaces where the crime had taken place to try to achieve narrative mastery.

But significantly, the reader's effort to dispel the mystery through such reconstruction was doomed to failure. The classic case illustrating the murder mystery's new requirements of the reader, especially with respect to spatial construction, as well as the reader's inevitable failure to dispel moral uncertainty, was the murder of Dr. George Parkman by Professor John Webster at the Harvard Medical College in 1849. Parkman, a wealthy real-estate speculator (and Webster's creditor), disappeared on the afternoon of Friday, November 23, 1849 (you see here my own historical complicity in the murder-mystery narrative). Boston's newly formed professional police force launched a search, which gradually zoned in on the Harvard Medical College where Parkman was last seen heading for an appointment with chemistry professor John Webster. A quick and careless police search of the building turned up nothing suspicious. But five days after Parkman's disappearance, Ephraim Littlefield, the college janitor, began his own private search for the corpse. Convinced that Webster had killed Parkman, and that he must have disposed of the body in the vault below his private privy at the College, Littlefield used a trapdoor in the basement to enter a crawl space leading to the outer wall of the privy vault, then tunneled through five courses of brickwork to reveal a dismembered human pelvis, right thigh, and lower left leg. Webster was promptly arrested and brought by police to the Medical College where, in a faint echo of the corpse-touching test, he was shown the discovered body fragments arranged in proper relationship to each other on a board. The subsequent police search turned up additional body parts secreted in various places in Webster's lab--false teeth and jawbone in Webster's laboratory furnace, a human trunk and left thigh in a tea chest nearby--which witnesses identified as having belonged to Parkman.

The enormous popular attention to the Webster-Parkman murder owed much to the high social station of the two major participants, and to the grisly horror of dismemberment. But its strongest appeal concerned the mystery of the crime, with special attention to the Gothic precincts of the Harvard Medical College, with its basement laboratories and dissection vaults, its trapdoors and subterranean passageways, its mysteriously locked doors, roaring furnaces, and bottled blood used for experiments. The peculiarly spatial qualities of both the murder and its investigation were dramatized throughout the eleven-day trial, at which the prosecution--completely ignorant of Webster's method in killing Parkman--focused instead on the discovery of the dismembered body in an effort to induce jurors to reconstruct mentally just how his various body parts came to be deposited throughout the Medical College. It was a highly visual strategy. Prosecution witnesses produced a map of the area around the Medical College so jurors could follow the victim's route to Webster's lab; various interior sketches and floor plans of the different levels of the College to explain the searches and discoveries within; a still-life sketch of physical evidence, items used in the dismemberment and storage of the body; and a three-dimensional wooden model of the College with moveable walls, stairs and furnishings constructed so that it could be taken apart during the course of the trial--which was said to "excite much attention in court." It then remained for the prosecution to impress upon jurors exactly what had been discovered within that three-dimensional space. Medical witnesses provided several anatomical drawings including sketches of the major body parts that had been found (laid out in their proper relation to one another) and of jaw fragments and fused mineral teeth alongside the dental molds brought to court by Parkman's dentist.

But the piece de resistance among all these visual aids was the life-size drawing of a human skeleton (sneakily drawn to resemble Dr. Parkman with his vigorous pedestrian stride and his protruding lower jaw), with its parts variously shaded and keyed to the architectural floor plan of Webster's laboratory, so that viewers could see at a glance which portions of Parkman were found in the furnace (marked black), which were found in the privy vault and tea chest (shaded)--and which parts of Dr. Parkman were never found at all (white). Laid side by side, the anatomical sketch and the architectural drawing provided jurors and readers of the nine printed versions of the Webster trial with an imaginative do-it-yourself-kit for the two layers of narrative involved in all murder mysteries. When the viewer's eye moved first to the skeleton, imaginatively pulling it apart and distributing its pieces in the various parts of the College identified by the floor plan, the viewer reconstructed the murder itself. And when the reader began instead with the floor plan, locating the various body fragments and reassembling them into the skeleton, the reader reenacted the murder investigation. Peter Brooks's phrase "storied bodies" proves unusually applicable to the Webster-Parkman murder narrative, in which the reader moved through different layers of story even as she moved through the different stories or floors of the Harvard Medical College.

So I arrive at the relationship between Divine Providence and Dr. Parkman's jawbone. Lawyers for the prosecution in this case explicitly observed that the Providential hand had been at work in bringing to light the various fragments of Dr. Parkman's body, including his distinctively shaped jawbone. In the absence of a clear corpus delicti (for the physical identification of a headless corpse rested on shaky legal grounds), they directed attention to the black and shaded portions of the skeleton--pieces that were found. The focus of the murder-mystery narrative, however, was on the white spaces of the diagram which marked the gaps in the narrative left by Providential absence--the empty spaces into which flowed the cult of mystery. For the fundamental mystery of the Webster-Parkman murder did survive John Webster's conviction for the crime. The jury's decision, and indeed the entire conduct of the trial, were greeted with a public outcry (focused on the absence of that clear corpus delicti) and with a number of specific legal challenges (to Chief Justice Lemuel Shaw's treatment of burden of proof concerning corpus delicti, his presentation of the law of circumstantial evidence, and the extreme bias of his summation of evidence in his charge to the jury). Then, after Webster had been sentenced to die for his crime, he wrote a "confession" which acknowledged that he had killed Parkman but claimed (very plausibly) that he had done so in a burst of anger, without premeditation; he was guilty, in other words, not of first-degree murder but of manslaughter. Hundreds of letters to the governor pleaded for a reprieve, but he ignored them, and Webster was duly executed. But the moral uncertainty surrounding his crime survived him; this nineteenth-century "confession," unlike its early eighteenth-century counterpart, did not bring narrative closure to the case but rather kept it open.

The Webster-Parkman case captured the core dynamic of the new murder-mystery narrative. On the one hand, it placed the burden of moral knowledge on the reader, transferring narrative responsibility from the Providential eye to the readers' own private eyes, requiring the reader to reconstruct the crime and to assign guilt for it. But at the same time, the murder-mystery ensured that there would be no moral certainty about either the nature of the crime or the assignment of guilt. There could be no clear narrative closure, no omniscient account, because legal truth was problematic, contestable, and reasonable doubt worked like an acid on coherent narrative. This transfer of the burden of knowledge from the Providential eye to the private eye helps explain the deep power of detective fiction in modern secular culture. Once we understand that the cultural construction of murder-as-mystery preceded the "invention" of detective fiction, the way is cleared to comprehending how detective fiction offered a fantasized solution to the problem of moral uncertainty in the world of true crime. For the Heroic Detective of detective fiction did exercise Providential powers of vision and agency. His French archetype, Eugene-Francois Vidocq, boasted in his heavily fictionalized memoirs in 1829, "Nothing escaped me--I knew all that was passing or projecting." His American prototype, Auguste Dupin, solved the mystery of Marie Roget without ever leaving his armchair in his fantastically gloomy house. The English Inspector Bucket, in Charles Dickens's Bleak House, entered rooms magically, read people's minds (as did Dupin), and seemed to be, in the words of Jo the crossing-sweeper, "in all manner of places, all at wanst." Omnipresent, omniscient, all-powerful--a clear line of descent ran from the Providential eye of the execution sermon to the "private eye" of detective fiction (whose name came from the advertising logo of the Pinkerton Detective Agency: a single large eyeball surrounded by the words "The eye that never sleeps"). The Heroic Detective was equal to the task of constructing a full and authoritative narrative of the crime, thus removing the narrative burden from the shoulders of the ordinary reader and restoring moral certainty to the world.

But the Heroic Detective was, after all, just a fiction. Auguste Dupin did not really solve the mystery of Mary Rogers (Poe's attempted solution was proved wrong even before his story went to print). True crime persists and the nonfictional murder mystery offers no moral certainty. This moral uncertainty has shaped ever since the nature of true-crime literature: a genre obsessive about details, insistent in its imperative need to reenact, endlessly fascinated with physical evidence, time tables, crucial gaps and missing pieces, and the spatial reconstruction of the crime. And moral uncertainty accounts for the quantity of true-crime literature, which is seemingly infinite. Most early American murders received one printed treatment, the sacred narrative of the execution sermon: one truth, God's truth, authoritatively narrated just once. But many nineteenth-century murders generated multiple narratives, multiple interpretations, because their truth was contingent, fluctuating, uncertain. And some nineteenth-century murders received new narrative attention decades after their occurrence, including those of Mary Rogers, Lizzie Borden, and of course Professor Webster (which has generated at least three accounts in the last quarter century, including one by Simon Schama). Why? Because in the murder-mystery, truth is open-ended; there is no closure--which makes crime literature an ideal commodity for a consumer-oriented print culture in which hunger grows from what it feeds upon.

Finally, the cultural construction of murder-as-mystery has important implications for the modern liberal understanding of the problem of human evil. For early New Englanders committed to a doctrine of universal depravity, evil was no mystery. The Gothic understanding of evil that was emerging in late eighteenth- and early nineteenth-century America, by contrast, rested on the elusiveness of evil within an Enlightened liberal culture which treated human nature as innocent unless proved guilty. The murder-mystery stressed the moral incomprehensibility of this radical act of transgression, and the question "How did this happen?" shaded readily into the question "How could this possibly have happened?" Just as the new cult of horror made murder unspeakable, inexplicable, the cult of mystery made murder infinitely incomprehensible. Together, Gothic horror and Gothic mystery joined to make the liberal confrontation with the crime of murder an endless ritual, without narrative closure and without moral comprehension.


Return to Ideas Menu.
Return to National Humanities Center Home Page.

Ideas is published twice a year. Editor: Jean Anne Leuchtenburg.
Copyright © 1996 by the National Humanities Center.
Comments to: lmorgan@ga.unc.edu
Revised: September 1996
nationalhumanitiescenter.org