Why People Die By Suicide
Page 16
larly, for participants who were excluded during “cyberball” and re-
ceived the psychological pain stimulus of feeling socially excluded,
the brain scan detected activity in the anterior cingulate cortex. The
researchers argued that evolutionary processes had recruited the phys-
ical pain system to warn us of something potentially as dangerous as
physical pain—namely, social exclusion or ostracism (which is like
death, figuratively and often literally, for highly social animals). The
need to belong is psychologically fundamental, and may have been
fundamental in conferring evolutionary benefits to our ancestors.
My view is that this need to belong is so powerful that, when satis-
fied, it can prevent suicide even when perceived burdensomeness
and the acquired ability to enact lethal self-injury are in place. By
the same token, when the need is thwarted, risk for suicide is in-
creased. This perspective, incidentally, is similar to that of Durkheim,
who proposed that suicide results, in part, from failure of social inte-
gration.
The Desire for Death ● 119
Some of the Stoic philosophers were proponents of rational sui-
cide, but even they could not overcome the need to belong. For ex-
ample, Seneca said, “Does life please you? Live on. Does it not? Go
from whence you came. No vast wound is necessary; a mere punc-
ture will secure your liberty. It is a bad thing (you say) to be under
the necessity of living; but there is no necessity in the case. Thanks
be to the gods, nobody can be compelled to live.” But when Seneca
became seriously ill and desired suicide, he could not carry through,
specifically because he could not bear to think of how his father
would react. His connection to his father prevented his suicide. (I
believe he also underestimated the ease with which suicide is
completed—his statement that “No vast wound is necessary; a mere
puncture will secure your liberty” contrasts with statements in Chap-
ter 2 about the extreme difficulty of death by suicide.) Notably, Sen-
eca later died by suicide, but well after his father’s death.
This same sentiment is often expressed in suicide risk assessments.
When asked about the likelihood of suicide, many patients respond
that though they have thought of suicide, their connection to a loved
one makes it impossible (e.g., “I couldn’t do that to so-and-so”). This
is of course no guarantee that someone will not attempt suicide, but,
as noted below, this clinical anecdote has some empirical support.
Women with numerous children may be less prone to suicide than
women with no or few children, for example.61
In his work The Metaphysics of Ethics, Immanuel Kant writes, “To
dispose of one’s life for some fancied end is to degrade the humanity
subsisting in his person, and entrusted to him that he might uphold
and preserve it.” Kant misses the perspective of the truly suicidal in-
dividual, whose belongingness is so thwarted that she or he does not
feel connected to humanity, and who feels that living life, not dying, degrades humanity.
Sylvia Plath, the poet who died by suicide at the age of thirty,
wrote, “So daddy, I’m finally through. / The black telephone’s off at
120 ● WHY PEOPLE DIE BY SUICIDE
the root, / The voices just can’t worm through.” I think these lines
convey that thwarted belongingness is more than just loneliness;
rather, it is the sense that sustaining connections are obliterated
(“off at the root”). These lines also conflate belongingness and death
(voices worming), which can be seen as an instance of the pattern
mentioned previously, in which imminently suicidal people fuse
death and life.
Shneidman’s Ariel recalled that on the day of her self-immolation
“I had various friends that I did know and it just seemed like they re-
ally just didn’t have time for me.” Just before her self-immolation,
Ariel described going to her friends’ house to return a borrowed
toaster; she wrote, “I remember just walking in and walking through
the house and by this time I was sobbing again. And not one word
was said to me by these people . . . And I just walked through the
house, put the toaster on the kitchen table and walked right out. And
nobody touched my arm, nobody asked what’s wrong, nobody even
gestured, and it upset me even more that this was sort of the end.”
In his 2003 New Yorker article on suicide at the Golden Gate
Bridge, Tad Friend quoted psychiatrist Jerome Motto on the suicide
that affected him most. Motto said, “I went to this guy’s apartment
afterward with the assistant medical examiner. The guy was in his
thirties, lived alone, pretty bare apartment. He’d written a note and
left it on his bureau. It said, ‘I’m going to walk to the bridge. If one
person smiles at me on the way, I will not jump.’”
Shneidman noted several other poignant examples of failed be-
longingness in suicide: “I haven’t the love I want so bad there is noth-
ing left” (from a forty-five-year-old married woman who died by
overdose); “I really thought that you and little Joe were going to
come back into my life but you didn’t” (from a twenty-year-old mar-
ried man who died by hanging); “I just cannot live without you. I
might as well be dead . . . I have this empty feeling inside me that is
The Desire for Death ● 121
killing me . . . When you left me I died inside” (from a thirty-one-
year-old separated man who died by hanging).62
David Reimer, the man described in Chapter 2 who was born a
boy, raised as a girl, and then changed back to a man in adolescence,
felt this inability to belong. A few years before his death by suicide, he
described some of his past experiences in relating to others, and said,
“There’s no belonging. So you’re an outcast. It doesn’t change.”63
When asked how he had felt as a girl watching his classmates pair off
romantically, he said, “These people looked like they knew where
they belonged. There was no place for me to feel comfortable with
anybody or anything.”64
On January 5, 2002, fifteen-year-old Charles Bishop stole a small
single-engine plane and crashed it into the Bank of America building
in Tampa, Florida. An article from the June 14, 2004 Tampa Bay On-
line described the final report on the boy’s death (classified as a sui-
cide). The last line of the article read, “Bishop’s mother said he had
no neighborhood friends, and she had not met any of his friends
from school.”
The empirical literature also affirms a connection between failed
belongingness and feeling suicidal. In the sections below, this work is
summarized, starting with research on the general connection be-
tween depressive symptoms (one of which is suicidality) and experi-
ences of disconnection from others.
Behavioral Features of Depression Indicating Low Social Connection
Connection to others can be seen in basic behavior, like eye contact
and harmony between one person’s and another’s f
acial expressions
or gestures. Several studies have demonstrated that depressed people
engage in less eye contact than do nondepressed people.65 Similar
findings have emerged with regard to non-verbal gestures. For exam-
ple, as compared to others, depressed people may engage in less
122 ● WHY PEOPLE DIE BY SUICIDE
head-nodding during conversation; head-nodding is a gesture af-
firming connection that communication partners find rewarding.66
Depressed and suicidal people have trouble engaging in the subtle
back-and-forth dance of nonverbal communication—they often do
not return eye contact, do not display animated facial expression in
reactions to others, and do not use gestures like head-nodding that
others find affirming and engaging.
Research on Social Isolation, Disconnection, and Suicidal Behavior
In the last section on perceived burdensomeness, I noted that rela-
tively few studies had empirically assessed the connection between
burdensomeness and suicidality, though all studies were supportive
of the link. The situation is different with regard to failed belonging-
ness: The fact that those who die by suicide experience isolation and
withdrawal before their deaths is among the clearest in all the litera-
ture on suicide.
An intriguing example involves language use by poets who died by
suicide compared to nonsuicidal poets as the poets’ deaths neared.67
These researchers used the Linguistic Inquiry and Word Count
(LIWC), mentioned earlier, to analyze text into its components—
for example, tendency to use action verbs, words denoting negative
emotion, and so on. Their results suggested escalating interpersonal
disconnection in the suicidal poets but not the poets who died by
other means. Specifically, as the suicidal poets’ deaths approached,
their use of interpersonal pronouns (e.g., “we”) decreased noticeably.
Similarly, Shneidman reports on a young man who had survived a
self-inflicted gunshot wound and later wrote, “Those around me
were as shadows, bare apparitions, but I was not actually conscious
of them, only aware of myself and my plight. Death swallowed me
long before I pulled the trigger. I was locked within myself.”68
I recently conducted a study using the LIWC software to examine
psychological variables associated with suicidal behavior by analyz-
The Desire for Death ● 123
ing differences in linguistic patterns in two literary characters in Wil-
liam Faulkner’s novel The Sound and the Fury. These characters, the brothers Quentin and Jason Compson, each have a section in the
novel that is written from a first person, stream-of-consciousness
point of view. Quentin’s section is written the day before he dies by
suicide. Jason’s section is written approximately ten years later. Each
section of the novel reveals characters’ inner thoughts and cognitive
processes as they are occurring. Accordingly, an analysis of psycho-
logical variables over time as written by Faulkner for Quentin and Ja-
son seemed feasible.
I wondered whether Faulkner was skilled enough to accurately
portray psychological aspects of approaching suicide, phenomena
that were not well elucidated by those studying suicide at the time
Faulkner was writing The Sound and the Fury (published in 1932).
Indeed, we found that Quentin’s use of social words decreased as his
death by suicide approached; Jason’s use of social words did not
change over time.69 Faulkner accurately portrayed relatively poorly
understood, intense, and rare psychological processes—still more in-
dication of his literary genius.
Work on nonfictional people reaches similar conclusions. In a sur-
vey on several hundred community participants, as well as on five
high-suicide-risk groups (e.g., general psychiatric patients, incarcer-
ated psychiatric patients), social isolation stood out as a correlate of
suicidal ideation (perceived burdensomeness toward family was also
a strong correlate).70 Similarly, in a study of psychiatrists’ reports on
their patients’ suicides, three variables were seen as frequently pres-
ent in the month preceding suicide: feeling a burden on others; social
withdrawal; and help negation.71 Help negation (the tendency to
thwart help, especially therapeutic help) has been viewed as a process
of interpersonal disconnection, and as such, may represent an in-
stance of thwarted belongingness.72
Conner and colleagues assessed men with alcohol dependence
124 ● WHY PEOPLE DIE BY SUICIDE
who died by suicide. Among the risk factors for completed suicide
were living alone and loss of a partner within the last month or two
before death.73 Similarly, a comparison of those who died by suicide
and those who died by other means revealed that those who died by
suicide were more likely to have been recently separated and living
alone.74 Significant others of people who had recently attempted sui-
cide pointed to loneliness in the patients as an important factor in
their suicide attempt.75
A study of African-American women examined reasons for the as-
sociation between types of childhood maltreatment and suicidal be-
havior. Of various factors examined, alienation (defined as inability
to establish basic trust and achieve stable and satisfying relation-
ships) was the most robust, fully explaining the link between all
forms of childhood maltreatment and later suicidal behavior.76
Marital Status, Parenting, and Suicidality
As the studies on social isolation might suggest, nonmarried status
is a demographic risk factor for suicide. The majority of deaths by
suicide among Native Americans of the Apache, Navajo, and Pueblo
tribes, for instance, were of single people.77 Statistics indicate the
following suicide rates in the United States in 1999: divorced—32.7
per 100,000; widowed—19.7 per 100,000; single—17.8 per 100,000;
married—10.6 per 100,000.78 These national statistics are of course
open to many interpretations, but they are consistent with the view
that belongingness (as indicated by married status) is a suicide
buffer, whereas thwarted belongingness (as indicated by nonmarried
status) is a risk for death by suicide. This is particularly evident
with regard to divorce (which confers a threefold increase in risk rel-
ative to married status). In context of the model proposed here, it is
tempting to speculate that suicide rates among divorced people are
particularly high because divorce can affect both basic feelings of ef-
fectiveness (e.g., feeling a failure as a spouse) and basic feelings of
The Desire for Death ● 125
connectedness (losing social contact not only with a spouse, but po-
tentially with the spouse’s family, with children, and with friends
previously shared with the spouse).
These statistics on marital status converge with a literature in-
spired by Durkheim’s emphasis on failures of social integration as a
source for suicide.79 For example, a study of suicidality and family
and parental functionin
g in over 4,000 high school students in Ice-
land concluded that those adolescents who were well integrated into
their families thereby derived protection from suicide; the indices re-
lated to family integration (cf. belongingness) wielded stronger in-
fluence on suicidality than did indices related to how the parents
were functioning.80
With regard to connections with children, there is evidence that
having large numbers of children protects against suicide. In a study
of nearly a million women in Norway, over 1,000 died by suicide
during a fifteen-year follow-up. Women with six or more children
had one-fifth the risk of death by suicide as compared to other
women.81 The suicide rates in Canada’s provinces are associated with
birth rates, such that more births correspond with fewer suicides,
consistent with the possibility that ties to new children buffer against
suicidality.82 In a very persuasive study on this point on over 18,000
Danish people who died by suicide and over 370,000 matched con-
trols, having children, especially young children, was protective
against suicide, even when accounting for powerful suicide-related
variables like marital status and psychiatric disorder.83 To my knowl-
edge, no studies like this have been conducted specifically on fathers,
parenthood, and suicide.
The result on mothers, parenthood, and suicide may even extend
to pregnancy. Marzuk and colleagues examined the autopsy reports
for all women who died by suicide in New York City from 1990 to
1993 and compared them to overall mortality statistics in age- and
race-matched women in New York during the same time period.
126 ● WHY PEOPLE DIE BY SUICIDE
During the study, there were 315 women who died by suicide in New
York City, six of whom were pregnant, which was one-third the
number that was expected given New York City female population
rates.84 These researchers concluded that pregnancy conferred pro-
tection from suicide; I would suggest that the protective influence in-
volved feelings of connection to the baby, as well as feeling needed by
the baby and thus not a burden.
Pregnancy, by itself, is no solution to longstanding feelings of dis-
connection and perceived burdensomeness, however. In fact, consis-
tent with this assertion, my colleagues and I found that initially pes-