Night Falls Fast

by Kay Redfield Jamison

  Wars come and go; epidemics come and go; but suicide, thus far, has stayed. Why is this, and what can be done about it? These questions are at the heart of this book: understanding why people kill themselves and determining what doctors, psychologists, schools and universities, parents, and society can do to stop it. The public outrage against war deaths and HIV/AIDS has been far more obvious and effective than the advocacy undertaken on behalf of those who have died from suicide, but the horror and the despair are no less real.

  Suicide is a particularly awful way to die: the mental suffering leading up to it is usually prolonged, intense, and unpalliated. There is no morphine equivalent to ease the acute pain, and death not uncommonly is violent and grisly. The suffering of the suicidal is private and inexpressible, leaving family members, friends, and colleagues to deal with an almost unfathomable kind of loss, as well as guilt. Suicide carries in its aftermath a level of confusion and devastation that is, for the most part, beyond description.

  The same Anglican Prayer Book that excludes from its final rites those who commit suicide speaks elsewhere of a “peace which the world cannot give.” It is this peace that lies beyond the suicidal mind. In The Anatomy of Melancholy, Robert Burton wrote:

  There is … in this [melancholic] humour, the very seeds of fire.… In the day-time they are affrighted still by some terrible object, and torn in pieces with suspicion, fear, sorrow, discontents, cares, shame, anguish, etc., as so many wild horses, that they cannot be quiet an hour, a minute of time, but even against their wills they are intent, and still thinking of it, they cannot forget it, it grinds their souls day and night, they are perpetually tormented.… In the midst of these squalid, ugly, and such irksome days, they seek at last, finding no comfort, no remedy in this wretched life, to be eased of all by death … to be their own butchers, and execute themselves.

  It is possible, with what we now know, to provide both comfort and remedy to stop at least some of the butchery. Most suicides, although by no means all, can be prevented. The breach between what we know and do is lethal.

  CHAPTER 2

  To Measure the Heart’s Turbulence

  —DEFINITIONS AND MAGNITUDES—

  What a job is this, to measure

  lightning with a footrule, the heart’s

  turbulence with a pair of callipers.

  —NORMAN MACCAIG

  ONE WOULD not expect it to be easy to define or classify suicide, and it is not. Death by one’s own hand is far too much a final gathering of unknown motives, complex psychologies, and uncertain circumstance—and it insinuates itself far too corrosively into the rights and fears and despairs of the living—for the definition of suicide to stay locked within the crisp categories chipped out by scientists or for it to adhere to the abstruse elaborations spun out by linguists and philosophers. Yet however permeable its edges, it remains important, as Henry Romilly Fedden observed in 1938, “to discover the elusive boundaries of suicide.” The Indian widow dying on her husband’s pyre is not, as he put it, “to be linked with the lonely individual strung up in his garret.”

  The early Greeks used highly active, decisive language to portray the act of suicide. To kill oneself was “to break up life,” “to grasp or seize death,” “to do violence to oneself,” “to leave the light,” or to commit “self-slaughter.” But the Greeks, in giving words to the act, did not so much define as describe it. Centuries of books and scholarly papers later, the definitions and classification systems for suicide remain diverse, controversial, and subject to ongoing reinvention. Inevitably, a certain intellectual paralysis sets in after reading the hundreds of finely tuned medical, philosophical, and sociological attempts to classify suicide.

  All suicide classification and nomenclature systems are, to a greater or lesser extent, flawed; and all, or almost all, have points that are well or uniquely taken. For clarity’s sake and consistency, I have adopted the criteria developed by the Centers for Disease Control and Prevention (an agency of the United States Public Health Service in Atlanta) for the certification of a death as suicide, criteria that are used by scientists and public health officials, as well as by medical examiners and coroners. Suicide is defined, succinctly, as a “death from injury, poisoning, or suffocation where there is evidence (either explicit or implicit) that the injury was self-inflicted and that the decedent intended to kill himself/herself.” The World Health Organization, whose even simpler definition underlies many of the international studies of suicide, defines suicide as “a suicidal act with a fatal outcome,” where a suicidal act is defined as “self-injury with varying degrees of lethal intent.”

  Society, medical science, and family members require accuracy in determining whether an unnatural death is an accident, murder, or suicide. Families need to know the truth about suicide, as best as that truth can be known, so that they can come to terms with it and so that they can have available to them medical and genetic information that may be important in making treatment decisions about other close blood relatives. Legal and financial issues may be at stake as well, such as property rights, mental competency determinations in disputed estates, life insurance policies, pensions, workers’ compensation, medical malpractice suits, and product liability claims. Accurate suicide statistics are also essential for public health investigators, whose responsibility it is to track trends and correlates of death and disease. (Earlier estimates suggested that suicide rates, based upon information from coroners and medical examiners, were underreported by as much as 25 to 50 percent; more recent studies indicate that the underreporting is now probably less than 10 percent).

  It is not always difficult to determine that suicide has occurred. Many instances are unequivocal: a gun is nearby, distinctive powder marks are found, a note has been written, and a psychiatric history or previous suicide attempt is documented. At other times, however, evidence must be pieced together—from autopsy findings, toxicology studies, psychological investigations, and statements from the deceased’s family members, or witnesses to the death—to establish that death was, in fact, self-inflicted. Intent must also be established.

  Most medical examiners and coroners use scientific and public health guidelines that specify that there must be evidence that the decedent intended to kill himself or herself or wished to die. The evidence may be explicit, that is, verbal or nonverbal expression of intent to kill oneself, or it may be implicit or indirect, such as “preparations for death inappropriate to or unexpected in the context of the decedent’s life; an expression of farewell or the desire to die or an acknowledgment of impending death; expression of hopelessness; expression of great emotional or physical pain or distress; effort to procure or learn about means of death or to rehearse fatal behavior; precautions to avoid rescue; evidence that decedent recognized high potential lethality of means of death; previous suicide attempt; previous suicide threat; stressful events or significant losses (actual or threatened); serious depression or mental disorder.”

  Medical and psychological criteria can go only so far, of course, and many other factors influence the accuracy of suicide statistics. The reporting officials, for example, may be medical examiners who, in addition to being physicians, also have extensive specialist training in forensic medicine, or they may be coroners or elected officials; the latter, in particular, may be more swayed by a family’s religious concerns, potential stigmatization or blame by the community, and possible financial repercussions to the survivors. Religious background may also influence the decisions reached by coroners and medical examiners. (In Canadian studies, for example, fewer suicide judgments were rendered by Catholic medical examiners than by non-Catholics, suggesting the possibility that religious views and sanctions continue to play a role in determining whether some unnatural deaths are designated as suicides or accidents.)

  Cultural attitudes and practices also have an impact. In one investigation, Danish coroners, using the same case material as their English counterparts, assigned a much higher proportion of equ
ivocal deaths to suicide. Perhaps, as the investigators noted, at least some of the differences were due to the fact that the determination of cause of death in Denmark is carried out in a medical rather than a legal context and that suicide, which was a criminal act in England until 1961, has not been a criminal offense in Denmark since 1866. Too, the Danes believe that they attach less stigma than the English to the primary mental illnesses underlying suicide—depression, manic-depression, and schizophrenia—and therefore to suicide as well.

  The method of dying is also important. Coroners and medical examiners, for instance, tend to view hanging as an almost certain indicator of suicide; this is also true when death has resulted from carbon monoxide poisoning from the exhaust pipe of an automobile, plastic bag asphyxiations, and fatal wrist cuttings or throat slashings. Drowning deaths are far more debatable, however, as they may result from suicide, accident, or murder. Indeed, most drownings are accidental, but investigations may be problematic, as sociologist Maxwell Atkinson makes clear in this law enforcement version of “Hearts,” where the corpse is played as the queen of spades and shuffled from jurisprudence to jurisprudence:

  For just as it is difficult to imagine ways in which people might end up hanged other than as a result of their own actions, it is very easy to envisage persons slipping, falling or even being pushed into the water from which their body is ultimately retrieved. The consequent difficulty in conducting the investigation and arriving at a definite verdict may be one of the reasons for a practice reported to me by a policeman who worked on one bank of a tidal river, which also marked the border between two police forces. According to him, it was not uncommon for policemen finding a body washed up on their side of the river to push it back into the water so that the tide would wash it up on the other side “so that the other force would have to deal with it.” The other force, however, presumably with similar thoughts in mind used to do the same thing, so that a body might float backwards and forwards several times before it was finally taken in and investigated.

  Deaths in single-car accidents, or head-on collisions with a large weight disparity between vehicles, also lend themselves to equivocal interpretations about the cause of death, as do some types of pedestrian deaths and deaths resulting from falls from high places. The most important cause of uncertainty in death certification, however, because of the large number of cases involved and the many confounding issues, is self-poisoning or drug overdose. Self-poisoning deaths—unlike cases of strangulation, drowning, gunshot wounds, or falls from heights—may not even look like unnatural deaths; as a result, they may not come to the attention of a coroner or medical examiner. Unless a death is seen as unexpected (for example, in a young person), an overdose death may be mistaken for a natural one.

  Intent to die is ambiguous in many drug overdose deaths. Mental states may be clouded by mental illness, especially depression, leading some to take too many pills by accident; others, not fully intending to die, may underestimate the lethal strength of a drug or misgauge its potency in combination with alcohol or other drugs. In these situations, as in other equivocal circumstances surrounding death, a retrospective examination of the life and death of the decedent, a so-called psychological autopsy, can provide critical information about intent and state of mind.

  The psychological autopsy is carried out by either an individual or a “Suicide Team” that conducts extensive interviews with family members, friends, physicians, and colleagues of the decedent in order to clarify the intent to die (if any) and the degree to which the death was self-inflicted. This technique was first used, in a more open-ended and less systematic format, by Gregory Zilboorg in an early psychoanalytic study of New York City police officers who had committed suicide. Although developed and most extensively used in the United States, the psychological autopsy is also utilized by researchers in Europe, South America, Australia, and Asia. Eli Robins, a psychiatrist at the Washington University School of Medicine, developed a more standardized interview format in the 1950s and used it to conduct a retrospective, community-based investigation of 134 consecutive cases of suicide in St. Louis. The study, now a classic in psychiatry, remains one of the clearest demonstrations of the almost ubiquitous presence of mental illness in those who commit suicide.

  The psychological autopsy as a method of clinical and scientific investigation was most vigorously pursued and developed in the late 1950s and early 1960s by Norman Farberow, Robert Litman, and Edwin Shneidman of the Los Angeles Suicide Prevention Center, who worked in collaboration with Theodore Curphey, then chief medical examiner for Los Angeles County. Their “death investigation,” or psychological autopsy, set as its goal the re-creation of the mental state of the victim during the time leading up to his or her death. Members of the Suicide Team interview friends, family members, and doctors of the victim, covering a comprehensive range of topics: the cause or method of death; the victim’s medical and psychiatric history; family background; the personality and lifestyle of the victim, as well as his or her typical patterns of reaction to stress, emotional upheavals, and “periods of disequilibrium”; upsets, pressures, tensions, or anticipation of trouble during the days, weeks, and months preceding death; the role of alcohol or drugs in the lifestyle and death of the victim; the nature of the victim’s personal relationships; fantasies, dreams, thoughts, or premonitions relating to death or suicide that the victim may have expressed; changes in personal or work habits and eating or sexual behavior; information about upswings, successes, or plans; an assessment of intention; a rating of the severity of suicidal thinking and behavior; and reactions of those interviewed to the victim’s death.

  From this information and a detailed analysis of the death itself, the Suicide Team puts together a description of the victim’s last days and then presents its findings to the coroner or medical examiner. Often, in seemingly equivocal cases, the recommendation is a persuasive one for a verdict of suicide; in other instances, however, the evidence leads to a decision for accident. The following case, illustrative of the kinds of questions asked and the investigative work involved, is from the files of the Los Angeles Suicide Prevention Center. When first presented to the Suicide Team, the death appeared to be a suicide; after completion of the psychological autopsy, however, the team recommended that it be certified as an accident:

  In practically any coroner’s office, a death that results from playing Russian roulette would automatically be certified as suicide. Indeed, there is now legal precedent for such certifications. Because of a special interest on the part of the Suicide Team in this type of death, this case was turned over to them by the coroner for investigation, with, as it turned out, extremely surprising results. On the basis of interviews it was ascertained that the victim, a 28-year-old male, was an Army veteran who had a collection of revolvers, which he kept in perfect operating condition. It was determined from his best friend that the victim’s favorite activity at parties was to play Russian roulette (following the usual rules of the game by having one chamber of the cylinder loaded) and that he had done this literally dozens of times in the preceding few years. At this point, the Suicide Team wondered about the psychology of a man who would behave in such a fashion: was he psychotic or was he intent on killing himself? Interviews with the widow clarified the situation: the victim had told her that there was no possibility of his hurting himself, as he always glanced at the gun to ascertain that the bullet was in a nonlethal position before he pulled the trigger. If the bullet was one notch to the left of the barrel, he would spin the cylinder again. There had been no suicidal ideation and no evidence of depression, psychosis, or morbid content of thought. What had happened? The Suicide Team knew that the death had occurred in someone else’s home. Interviews developed the information that he shot himself with a revolver that was not his own but belonged to his host of the evening. What seemed most important was the fact that, whereas his collection consisted entirely of Smith and Wesson revolvers, he had killed himself with a Colt revolver. The act
ions of the two guns are different; that is, the cylinder of the Smith and Wesson revolves clockwise. It was believed that the victim, checking and seeing the bullet one space to the right of the barrel, thought that he could not possibly kill himself, whereas in reality pulling the trigger put the bullet in the lethal position and he died immediately.

  In the absence of any indication of suicidal affect or any indications of suicidal ideation, and with the additional information about the two types of revolvers, the Suicide Team recommended that this death be considered as accidental. One member of the Suicide Team labeled this case as one of Soviet Roulette, that is, Russian roulette in which one cheats.

  The psychological autopsy, in modified and different standardized forms, has been widely used in suicide research, as well as in coroners’ and medical examiners’ offices. It has proven especially useful in understanding the extent of the link between psychopathology and suicide.

  SUICIDE IS the anchor point on a continuum of suicidal thoughts and behaviors. This continuum is one that ranges from risk-taking behaviors at one end, extends through different degrees and types of suicidal thinking, and ends with suicide attempts and suicide. Suicide attempts include not only those acts where there is a clear or likely intent to die but those where there is no intent to die (for instance, acts where the individual wishes to use the appearance of intending to kill himself or herself in order to obtain some other end).