Suicide
First published Tue 18 May, 2004
Suicide is an enigmatic and disconcerting phenomenon. Because of others' inability to directly occupy the mental world of the suicidal, suicide appears to elude easy explanation. This inexplicability is stunningly captured by Jeffrey Eugenides in his novel The Virgin Suicides. In the novel, the narrator describes the reactions of several teenaged boys to the suicides of five sisters. The boys keep a collection of the dead girls' belongings, repeatedly sifting through them in a vain attempt to understand their deaths.
In the end we had the pieces of the puzzle, but no matter how we put them together, gaps remained, oddly shaped emptinesses mapped by what surrounded them, like countries we couldn't name. (Eugenides 1993, 246)
Undoubtedly, the challenge of simply fathoming suicide accounts for the vast array of attitudes toward suicide found in the history of Western civilization: bafflement, dismissal, heroic glorification, sympathy, anger, moral or religious condemnation. Suicide is now an object of multidisciplinary scientific study, with sociology, anthropology, psychology, and psychiatry each providing important insights into suicide. Particularly promising are the significant advances being made in our scientific understanding of the neurological basis of suicidal behavior (Stoff and Mann 1997) and the mental conditions associated with it. Nonetheless, certain questions about suicide seem to fall at least partially outside the domain of science, and indeed, suicide has been a focus of philosophical examination in the West since at least the time of Plato. For philosophers, suicide raises a host of conceptual, theological, moral, and psychological questions. Among these questions are: What makes a person's behavior suicidal? What motivates such behavior? Is suicide morally permissible, or even morally required in some extraordinary circumstances? Is suicidal behavior rational? This article will examine the main currents of historical and contemporary philosophical thought surrounding these questions.
1. Characterizing Suicide
Surprisingly, philosophical difficulties emerge when we even attempt to characterize suicide precisely, and attempts to do so introduce intricate issues about how to describe and explain human action. In particular, identifying a set of necessary and sufficient conditions for suicide that fits well with our typical usage of the term is especially challenging. A further challenge is that because suicide is strongly colored by negative emotional or moral connotations, efforts to distinguish suicidal behavior from other behavior often clandestinely import moral judgments about the aims or moral worth of such behavior. That is, views about the nature of suicide often incorporate, sometimes unknowingly, views about the prudential or moral justifiability of suicide and are therefore not value-neutral descriptions of suicide. Definitions of suicide are "sometimes dependent on prior judgments about its justifiability." (Lebacqz & Englehardt 1980, 701.) Theorists about suicide often fail to divorce questions about whether an act was suicide from whether its motives were admirable or odious. Hitler, most people contend, was clearly a suicide, but Socrates and Jesus were not. (Though on Socrates, see Frey 1978) Suicide still carries a strongly negative subtext, and on the whole, we exhibit a greater willingness to categorize self-killings intended to avoid one's just deserts as suicides than self-killings intended to benefit others (Beauchamp & Childress 1983, 93-94.) Some go so far as to deny the possibility that an act of self- killing motivated by altruism can count as suicide (Margolis 1980.)
Such conceptual slipperiness befuddles moral arguments about the justifiability of suicide by permitting us to ‘define away’ self-killings we believe are justified as something other than suicide, whereas it would be desirable to identify first a defensible non-normative conception of suicide and then proceed to discuss the moral merits of various acts of suicide (Kupfer 1990.) Some philosophers, on the other hand, have embraced the apparently value-laden character of suicide, suggesting that word ‘suicide’ has as one its functions the ascription of moral responsibility, and insofar as disagreements about the extent to which agents themselves (as opposed to social conditions, medical facts, etc.) are morally responsible for their deaths persist, so too will apparently conceptual disagreements about the nature of suicide persist (Stern-Gillett 1987.)
Supposing, however, that a purely descriptive account of suicide is possible, where should it begin? While it is tempting to say that suicide is any self-caused death, this account is vulnerable to obvious counterexamples. An individual who knows the health risks of smoking or of skydiving, but willfully engages in these behaviors and dies as a result, could be said to be causally responsible for her own death but not to have committed suicide. Similarly, an individual who takes a swig of hydrochloric acid, believing it to be lemonade, and subsequently dies causes her own death but does not engage in suicidal behavior. Moreover, not only are there self-caused deaths that are not suicides, but there are behaviors that result in death and are arguably suicidal in which the agent is not the cause of her own death or is so only at one remove. This can occur when an individual arranges the circumstances for her death. A terminally ill patient who requests that another person inject her with a lethal dose of tranquilizers has, intuitively, committed suicide. Though she is not immediately causally responsible for her death, she appears morally responsible for her death, since she initiates a sequence of events which she intended to culminate in her death, a sequence which cannot be explained without reference to her beliefs and desires. (Such a case might also be an example of voluntary euthanasia.) Likewise, those who commit ‘suicide by cop,’ where an armed crime is committed in order to provoke police into shooting its perpetrator, are responsible for their own deaths despite not being the causes of their deaths. In these kinds of cases, such agents would not die, or would not be at an elevated risk for death, were it not for their initiating such causal sequences. (See Brandt 1975, Tolhurst 1983, Frey 1981, but for a possible objection see Kupfer 1990).
Furthermore, many philosophers (Fairbairn 1995, chapter 5) doubt whether an act's actually resulting in death is essential to suicide at all. It is common to speak of ‘attempted’ or ‘failed’ suicides, instances where because of agents' false beliefs (about the lethality of their behavior, for example), unforeseen factual circumstances, others' interventions, etc., an act which might have resulted in an agent's death does not.
Hence, suicidal behavior need not result in death, nor must the condition that hastens death be self-caused. It follows, therefore that, first, a correct account of suicide (contra Durkheim 1897) must emphasize the non-accidental relationship between suicidal behavior and death (i.e., death is in some respect the aim of suicidal behavior). Second, what appears essential for a behavior to count as suicide is that the person in question chooses to die. Suicide is an attempt to inflict death upon oneself and is "intentional rather than consequential in nature." (Fairbairn 1995, 58) These conclusions imply that suicide must rest upon an individual's intentions (where an intention implicates an individual's beliefs and desires about her action. (See Brandt 1975, Tolhurst 1983, Frey 1978, O'Keefee 1981) One intention-based account of suicide (similar to Graber 1981, 57) would say, roughly, that
- A person S's behavior B is suicidal iff
- S believed that B, or some causal consequence of B, would make her death at least highly likely, and
- S intended to die by engaging in B.
This account renders the notion of suicide as self-inflicted attempted death more precise, but it is not without its shortcomings.
Condition (a) is a doxastic condition, and is meant to rule out as suicides deaths (or increased risks for death) caused by an individual's behavior where the individual causes these outcomes but does so out of ignorance of the relevant risks of her behavior, as when an individual accidentally takes a lethal dose of a prescription drug. At the same time, (a) accounts for cases such as the aforementioned terminally ill patient whose death is caused only indirectly by her request to die. Condition (a) does not require that S know that B will put her at a significantly greater risk for death, nor even that S's beliefs about B's lethality be true or even justified. Suicidal individuals often have false beliefs about the lethality of their chosen suicide methods, greatly overestimating the lethality of over the counter painkillers while underestimating the lethality of handguns, for instance. An individual could believe falsely, or on the basis of inadequate evidence, that placing one's head in an electric oven significantly increases one's chances of dying, but that behavior is nonetheless suicidal. The demand that S believe that B makes death highly likely is admittedly inexact, but it permits us to navigate between two extreme and mistaken views. On the one hand, it rules out as suicidal behavior that which is in fact only marginally more likely to cause a person's death (you are more likely to die in your car than in your living room) and is rarely utilized as a suicide method anyway. On the other hand, to demand that S believe that B certainly or almost certainly will cause S's death is too strict, since it will rarely be the case (given the possibility of intervening conditions, etc.) that B will necessarily cause S's death, and in fact, many suicidal individuals are ambivalent about their actions, an ambivalence which is turn reflected in their selecting suicide methods that are far from certain to cause death. It also allows us to distinguish genuinely suicidal behavior from suicidal gestures, in which individuals engage in behavior they believe is not likely to cause their death but is nonetheless associated with suicide attempts, while in fact having some other intention (e.g., gaining others' sympathy) in mind.
Condition (b), however, is far more knotty. For what is it to intend by one's behavior that death result? There are examples in which condition (a) is met, but whether (b) is met is more problematic. For instance, does a soldier who leaps upon a live grenade tossed into a foxhole in order to save his comrades engage in suicidal behavior? Many, especially partisans of the doctrine of double effect, would answer ‘no’: Despite the fact that the solider knew his behavior would likely cause him to die, his intention was to absorb the blast so as to save the other soldiers, whereas his death was only a foreseen outcome of his action. Needless to say, whether a clear and non-manipulable divide exists between foreseen and intended outcomes is controversial (Glover 1990, ch. 6) (It is of course possible that whether death is foreseen or intended has no bearing on whether an act counts as suicide but still bears on whether that suicide is justified.) Some would argue that given the near certainty of his dying by jumping on the grenade, his death was at least weakly intended, in Alvin Goldman's sense (Tolhurst 1983.) At the same time, cases that are commonly viewed as suicide do not exhibit an full-fledged intention to die. Current psychiatric theory holds that many examples of suicidal behavior do not aim at death but are "cries for help." In such cases, the person does not wish to die, but intends to gain others' attention in such a fashion that holds out the possibility of death. However, it seems correct to say that when a person who issues a cry for help does die, despite not intending to die, their death is neither foreseen, since the person actually intends not to die, nor wholly accidental, since the person knowingly engaged in behavior that she believed will make her death significantly more likely, making her death in an obvious sense self-inflicted. (But see Graber 1981, 58) Such a case might indicate the need for a third category besides intentional suicide and accidental death, call it unintentional death or unintended suicide.
In the Holy Love of God I am your brother in Christ and my name is Royal
