The phenomena of suicide-by-cop and murder-suicide raise the
more general issue of method choice in suicide. From the current
perspective, choice of an especially lethal means of suicide should of
course be more common among those who have acquired the ability
to enact lethal self-injury than others. There is some evidence that
this is the case. For instance, people who had engaged in repeated
self-injury in the past reported that their current episode of self-
harm was more aggressive and more lethal than first-time self-injury
patients.26 Similarly, among mood-disordered and borderline per-
sonality–disordered patients, number of previous suicide attempts is
a strong predictor of the extent of medical damage resulting from the
most serious lifetime suicide attempt.27 Additionally, regarding per-
ceived burdensomeness, my colleagues and I found that it was associ-
ated with more lethal (e.g., self-inflicted gunshot) as opposed to less
lethal (e.g., overdose) means of completed suicide, consistent with
the current theory.28
There are two ambiguities regarding suicide method, and indeed,
the study on perceived burdensomeness illustrates one of them. In
one of the two studies in that report, all participants had died by
What Do We Mean by Suicide? ● 151
suicide, yet some did so by relatively less lethal means. The associa-
tion between method choice and intent to die is thus complex, and
in fact, some studies document a low association between intent
and lethality of method.29 A persuasive theory of suicide should be
able to explicate, at least in part, the interrelations between method
choice, intent to die, and death by suicide, as well as allow for com-
plexities inherent in these interrelations; the current theory does so.
A point prediction derived from the current theory is that those who
choose especially lethal means of self-injury will be characterized
by the theory’s parameters (i.e., acquired ability, perceived burden-
someness, thwarted belongingness) regardless of whether they die or
suffer extensive medical damage; those who choose a low lethality
method will be characterized by the theory’s parameters only if they
die or suffer extensive medical damage.
The second complexity regarding method choice involves marked
cross-cultural differences. In the United States, a country with high
gun ownership, 55 percent of those who die by suicide do so by self-
inflicted gunshot wound; the second leading method is hanging and
other forms of suffocation (20 percent); and the third is poisoning
(17 percent).30 By contrast, poisoning is a more common means of
suicide in England and China than in the United States.31 The theory
presented in this book has few predictions regarding which method
would be chosen in which culture; rather, the theory attempts to ex-
plain which individuals will develop serious intentions to die by sui-
cide, and which ones will act on them in serious ways (often by lethal
means, but not always, because lethality of means differs cross-cul-
turally and is not always a clear expression of intent to die). However,
it is of interest that the most rare forms of suicide methods are also
the most fearsome—for example, falls account for 2 percent of all
deaths by suicide and fire, less than 1 percent. Thus, even among
those who die by suicide and therefore by definition had developed the
152 ● WHY PEOPLE DIE BY SUICIDE
ability to enact lethal self-injury, the fearsome quality of some meth-
ods as compared to others seems to play a role in method choice.
Prevalence
In one sense, death by suicide is not rare. Around 30,000 people die
by suicide in the United States each year, which translates to approxi-
mately eighty deaths by suicide every day—one person every eigh-
teen minutes. Worldwide, approximately a million people die by sui-
cide each year—one person every forty seconds. This rate means that
more people are dying by suicide than in all of the world’s armed
conflicts combined; about as many people die by suicide as die in
traffic accidents. Suicide is a leading cause of death.32
In another sense, however, suicide is a relatively rare occurrence.
Expanding on the data presented in Chapter 1, in a U.S. city of
100,000 people, around ten will die by suicide each year (indeed, the
usual metric for death rates is deaths per 100,000, and the U.S. rate is
approximately 10 per 100,000 per year). Of all deaths in the United
States, a little over 1 percent are due to suicide. Given that a person
has died, the chance that the cause was heart disease or cancer is 52
percent. Whereas approximately eighty people per day die by suicide
in the United States, approximately 1,900 U.S. people die from heart
disease every day.
Many explanations for suicide fail fairly obviously in the face of
prevalence rates. If suicide is due to factor X, it must be explained
why this factor is fairly common and suicide less so. For example,
mental illness is commonly invoked as an explanation for suicide. As
will be detailed in the next chapter, there is no doubt whatsoever that
mental illnesses play a role in suicidal behavior. But mental illness
alone does not provide a satisfying explanation for suicide, because
mental illness is much more common than suicide. How should we
explain all those people with mental illness who do not die by sui-
What Do We Mean by Suicide? ● 153
cide? Moreover, the absence of our hypothetical X factor in some sui-
cides needs to be explained. Though it is rare for people without
mental illness to die by suicide, it occasionally happens—a fact that
an explanation centered around mental illness cannot account for.
By contrast, my theory is compatible with the epidemiology of
suicide. Relatively few people have the experiences or opportunities
to acquire the capability for lethal self-injury. Even among people
who do have these experiences or opportunities, not all will necessar-
ily fully acquire the capability, because some may have relatively mild
experiences that do not lead to habituation. This capability is thus
relatively rare and difficult to obtain. The other factors—perceived
burdensomeness and failed belongingness—are relatively rare too.
The confluence of these three factors, which according to my model
is required for serious suicidal behavior, is more rare still. The cur-
rent framework explains—indeed predicts—that death by suicide
will be a relatively rare event.
Suicide rates are not evenly distributed across geographic regions,
and a penetrating theory of suicidal behavior should have something
to say about this. In certain regions of the United States, cultures of
honor appear to reign. In these regions, the creed seems to be “give
me honor or give me death.” In fact, all fifteen of the states with
the highest suicide rates are culture-of-honor states. A comprehen-
sive theory of suicide thus must be able to incorporate the role of
thwarted honor. The current theory does so by empha
sizing one’s
standing vis-à-vis others, and proposing that if one’s standing falls to
the degree that one perceives oneself as a burden on others, risk for
serious suicidal behavior is increased.
If suicide rates differ by state, such that culture-of-honor states
have higher rates, rates may also differ within states, with higher rates
occurring in rural counties where cultures of honor might be more
influential than in urban areas. I examined this in my home state of
Florida by randomly picking three rural counties and comparing
154 ● WHY PEOPLE DIE BY SUICIDE
them to counties in the urban areas of Miami and Tampa. The rural
counties were Lake County (which is in the middle of the pen-
insula, northeast of Tampa, 2002 population 233,835), Calhoun
County (which is in the panhandle west of Tallahassee, 2002 popula-
tion 12,567), and Suwanee County (which is near where the pan-
handle and peninsula intersect, west of Jacksonville, 2002 popula-
tion 36,121). The urban counties were Broward County (Miami,
2002 population 1,709,118), Dade County (Miami, 2002 population
2,332,599), and Hillsborough County (Tampa, 2002 population
1,053,864). The suicide rate per 100,000 in the rural counties was
over fifteen. The comparable rate in the three counties in Miami and
Tampa was just over nine.
These findings converge with those of a recent study that found
that, from 1990 to 1997, the suicide rate in men was around 27 per
100,000 in rural U.S. counties, as compared to around 17 per 100,000
in urban areas.33 A roughly similar pattern emerged for women as
well. When interviewed about this finding by the Atlanta Journal-
Constitution, 34 the study’s lead author specifically invoked low
belongingness. He said, “The usual explanations are that there are
physical isolation and limited social interactions in rural areas. You
have limited opportunity for social interaction and networks.”
Indeed, a variety of explanations could account for this pattern,
including more economic stress and less ethnic diversity in the rural
counties (as will be discussed later in this chapter, white people in the
United States have higher rates of suicide than nonwhite people).
Nevertheless, it is an intriguing speculation that cultures of honor
reign more in rural Florida than in places like Miami and Tampa,
and in rural versus urban areas in general, and that this plays some
role in the substantial difference in the suicide rate between rural and
urban areas of the state.
As of 2001, the five countries with the highest suicide rates were
Lithuania, Russia, Belarus, Latvia, and Ukraine. These countries are
What Do We Mean by Suicide? ● 155
contiguous with each other, and all were formerly part of the Soviet
Union. The suicide rates in these countries are staggering: In all five,
the rate per 100,000 men is over fifty, and in Lithuania and Russia,
the rate per 100,000 men is over seventy. For comparison’s sake, the
rate per 100,000 men in the United States is approximately sixteen.
As many women die by suicide in Lithuania as men do in the United
States.
It is not difficult to imagine these countries as typifying cultures of
honor. Moreover, these countries have long histories of hardship (in-
cluding in some cases brutal Soviet repression) and have undergone
extremely fraught transitions from Soviet communism to indepen-
dence with attendant crises of national identity and economy. That
the acquired ability to enact lethal self-injury (developed through
habituation to various hardships), perceived burdensomeness (due
in part to economic stress), and failed belongingness (due in part to
strains on societal integration) would occur at high rates in these
countries provides one way of understanding their extremely high
suicide rates.
Demographics
Gender
In the United States, men are approximately four times more likely
than women to die by suicide, whereas women are approximately
three times more likely than men to attempt suicide.35 This pattern
holds in most countries, with a ratio of at least four male suicides
to every one female suicide. In the five countries with the highest
suicide rates—Lithuania, Russia, Belarus, Latvia, and Ukraine—this
ratio ranges from 4.69 (Lithuania) to 6.69 (Belarus). This pattern
of male lethality is partly related to the tendency toward violent be-
havior in general, which is, of course, more common in men than
women. Women’s attempts are more frequent but less violent; vice-
156 ● WHY PEOPLE DIE BY SUICIDE
versa for men. For every three male suicide victims in the United
States, two die by firearm, as compared to one of three for women.
The most common method for U.S. female victims is overdose/poi-
soning.36
I believe that men are more lethal regarding suicide in part be-
cause they have acquired more of the ability to enact lethal self-in-
jury. Men, more so than women, acquire this capability through an
array of means. They have more exposure to guns, to physical fights,
to violent sports like boxing and football, and to self-injecting drug
use. They are, on average, more likely to be physicians. Moreover,
men, more so than women, may struggle with belongingness. In this
context, it has been suggested that women may die by suicide less
frequently than men because women are less likely to abandon rela-
tional values that form part of their identities.37 Perceived burden-
someness may also be more of an issue for men than women, given
that traditional male gender roles include providing for others. Frus-
tration of the breadwinner role may contribute to feelings of per-
ceived burdensomeness in men, somewhat more so than in women.
A study comparing two models of the relation of previous to sub-
sequent suicide attempt may also be relevant to gender differences.
One model was called the “trait model,” meaning that suicide at-
tempts are, in a sense, predetermined by enduring dispositions—or
traits—and are uninfluenced by intervening occurrences of suicidal
behavior. The other model was called the “crescendo model,” mean-
ing that each occurrence of suicidal behavior increases the subse-
quent likelihood of suicidal behavior. The crescendo model bears
similarities to my idea that various painful and provocative experi-
ences increase the ability to enact lethal self-injury.
Data were equally consistent with both models.38 The theory
articulated here would more accurately be captured by an amended
crescendo model, in which occurrences of suicidal behavior increase
the likelihood of subsequent suicidal behavior only if the original be-
What Do We Mean by Suicide? ● 157
havior contains elements that approach or exceed previous worst-
point suicidality. When this happens, habituation is furthered and
opponent processes engaged. Relatively mild suicidality, on the other
hand, may not increase the subsequent likelihood of severe or esca-
; lated suicidality. This may in part explain the empirical fact that
women are more likely than men to experience mild suicidality, but
less likely than men to escalate to completed suicide.
An important exception to the rule of more completed suicides in
men than women occurs in China, where recent data show the rate
for women to be 14.8 per 100,000 and the rate for men to be 13.0 per
100,000. The role of Confucianism in Chinese society and its view of
the inferior position of women has been emphasized as one explana-
tion,39 one that is consistent with the current emphasis on effective-
ness as a buffer against suicide.
An interesting speculation on the phenomenon of high suicide
rates in Chinese women, consistent with the current theory, involves
sports. In contrast to countries like the United States, Chinese
women have performed considerably better than Chinese men in
international sport competitions. It has been suggested that sport
achievement contributes to an ethic among Chinese women of
physicality, masculinity, and aggression, and that this may contrib-
ute to more lethality in women’s suicidal behavior (perhaps via the
acquired ability to enact lethal self-injury).40 A survey of over 4,000
U.S. college students found evidence consistent with the possibility
that sport-related masculinity and aggression in women may con-
fer higher suicide risk. Women who engaged in vigorous athletic ac-
tivity were at greater odds of reporting suicidal behavior than other
women.41 There is evidence, incidentally, of increased pain tolerance
in women athletes,42 which would also facilitate their acquisition of
the ability to enact lethal self-injury.
A practice common in China from the 900s to the 1900s was de-
scribed by M. Roach in her 2003 book Stiff: “adult children . . . were
158 ● WHY PEOPLE DIE BY SUICIDE
obliged to demonstrate filial piety by hacking off a piece of them-
selves and preparing it as a restorative elixir.”43 More often than not,
the adult child was a daughter-in-law and the recipient her mother-
in-law. One is tempted to speculate that traditions like these may
have set in place a culture in which women habituate to pain and in-
jury, including deliberate self-harm.
I pointed out earlier that the usual male:female ratio for suicide
rates is at least 4:1 in most countries, often higher (for example, it is
Why People Die By Suicide Page 20