Adaptive Mental Mechanisms
Their Role in a Positive Psychology
George E. Vaillant
Brigham and Women's Hospital
Psychology needs a metric for positive mental health that
would be analogous to the IQ tests that measure above-
average intelligence. The Defensive Function Scale of the
DSM-IV offers a possible metric. In the present article the
author links the transformational qualities of defenses at
the mature end of the Defensive Function Scale---altruism,
suppression, humor, anticipation, and sublimation--to pos-
itive psychology. First, the methodological problems in-
volved in the reliable assessment of defenses are acknowl-
edged. Next, the use of prospective longitudinal study to
overcome such difficulties and to provide more reliable
definition and measurement of defenses is outlined. Evi-
dence is also offered that, unlike many psychological mea-
sures, the maturity of defenses is quite independent of
social class, education, and IQ. Last, evidence is offered to
illustrate the validity of mature defenses and their contri-
bution to positive psychology.
: the days of alchemy, humanity has been fas-
.ted with how to turn lead into gold. People are
gued by the real-life alchemy of the oyster
transforming an irritating grain of sand into a pearl.
Rumpelstiltskin spinning straw into gold is a favorite fairy
tale. In their laudable quest to relieve human suffering,
however, both psychiatry and psychology have been less
interested in positive transformations. Instead, they have
been more concerned with how cold mothers and bad genes
create disease and so turn gold to lead.
In contrast to psychiatry, however, psychology has
made at least some effort to measure the positive as well as
the pathological. Intelligence tests are a good example. In
contrast to intelligence, however, most facets of positive
human behavior--for example, creativity, maturity, and
empathy--are extraordinarily difficult to measure. This
article discusses efforts to conceptualize the mature de-
fenses (aka, involuntary coping mechanisms and "healthy
denial"). I argue that such a schema comprises a facet of
and a possible metric for a positive psychology.
By way of introduction, there are three broad classes of
coping mechanisms. First, there are the ways in which an
individual elicits help from appropriate others: namely, seek-
ing social support. Second, there are conscious cognitive
strategies that people intentionally use to make the best of a
bad situation (Lazarus & Folkman, 1984). Third, there are
involuntary mental mechanisms that distort our perception of
internal and external reality to reduce subjective distress. For
semantic consistency, the Diagnostic and Statistical Manual
of Mental Disorders (4th ed.; DSM-IV; American Psychiatric
Association, 1994) has labeled these mental mechanisms de-
fenses and has organized them in a hierarchical Defensive
Function Scale. Included within the "high adaptive level" of
DSM-IV are the defenses of anticipation, altruism, humor,
sublimation, and suppression. These adaptive mental mecha-
nisms "maximize gratification and allow conscious awareness
of feelings, ideas and their consequences" (American Psychi-
atric Association, 1994, p. 752).
In many ways, the first two classes of coping are
superior to the third. Most important, seeking social sup-
port and cognitive strategies are both under volitional con-
trol and can affect the real world. In three ways, however,
the involuntary defenses or coping processes are superior to
voluntary coping processes. First, as I demonstrate in this
article, involuntary defenses are independent of education
and social privilege. Second, they can regulate people's
perceptions of those internal and external realities that they
are powerless to change. Third, the adaptive defenses can
turn lead into gold. By this I mean such processes can serve
as transformative agents in the real world.
Let me offer an analogy. If a person who cuts a small
artery lacks the cognitive strategies (provided to health
professionals through expensive education) to stop the
hemorrhage and lacks the social supports of access to
physicians (provided to the middle class through expensive
health insurance), the person can still cope with the hem-
orrhage with inborn defenses. He or she can stop the
bleeding through involuntary, transformative, and highly
complex clotting mechanisms. Yet, such clotting mecha-
nisms may be denied to royalty afflicted with hemophilia.
In analogous fashion, when cognitive solutions and social
supports are absent, the psychologically resilient from all
walks of life--achieve similar homeostatic alchemy
through involuntary mental defenses that alter perception
of internal and external reality.
George El. Vaillant, Division of Psychiatry, Department of Medicine,
Brigham and Women's Hospital.
This work was supported by the Division of Psychiatry, Department
of Medicine, Brigham and Women's Hospital; by the Study of Adult
Development, Harvard University Health Services; and by Research Grant
MH 42248 from the National Institute of Mental Health.
Correspondence concerning this article should be addressed to
George E. Vaillant, Brigham and Women's Hospital, 75 Francis Street,
Boston, MA 02115. Electronic mail may be sent to email@example.com.
January 2000 ° American Psychologist
Copyright 2000 by the American Psychological Association, Inc, (X)O3-066X/00/$5.00
Vol. 55, No. 1. 89-98 DOI: 10.1037//0003-066X 55.1.89
For example, at age 3 l, a suicidal but only partially deaf
musician had written of his loss of heating, "Oh, if I were rid
of this affliction, 1 could embrace the world" (Forbes, 1969, p.
286). At 54, the utterly deaf but no longer suicidal musician
immortalized Schiller's "Ode to Joy" ("Be embraced all ye
millions with a kiss for all the world") in the lyrical, life-
affirming chorus of his Ninth Symphony. But how can psy-
chology differentiate the transformative denial that Beethoven
deployed to overcome depression by writing a hymn to joy
from the mental mechanisms of psychosis such as projection
and psychotic denial? Clearly, the answer is important.
As a start, mature mental health always involves affect
recognition. Beethoven did not totally deny his real depres-
sion, nor was he overwhelmed by it. Thus, we have evidence
that Beethoven's defensive behavior (aka, creative product)
did not reflect complete denial of affect as do less adaptive
defenses. Throughout his composition of the Ninth Symphony,
he remained conscious of his pain. For example, on a draft
version of one instrumental recitative he had scribbled, "No,
this would remind us too much of our despair" (Forbes, 1969,
p. 892). Equally important was Beethoven's defensive use of
sublimation, which not only made him feel subjectively better,
but also was of objective value to the real world.
Adaptive defenses are essential to positive mental health.
Defenses reduce conflict and cognitive dissonance during
sudden changes in internal and external reality. If not modi-
fied, sudden changes result in anxiety and/or depression. First,
defense mechanisms can restore psychological homeostasis
by ignoring or deflecting sudden increases in affect and in-
stinctual press. For example, when the Soviets liberated the
first Nazi death camp, Maidenek, the New York Times denied
its unbearable horror by reporting the news as a Soviet pro-
paganda ploy. Second, defense mechanisms can provide a
mental time out to mitigate changes in reality and self-image
that cannot be immediately integrated--for example, after
major surgery or promotion. Third, defenses transmute unre-
solvable conflict with important people, living or dead. Fi-
nally, defenses soften conflicts of conscience--for example,
after putting a parent in a nursing home. In short, defenses
shield people from sudden changes in affect, reality, relation-
ships, or conscience.
For many years, defense mechanisms have been deserv-
edly unpopular in experimental psychology, because of diffi-
culty in empirical verification. Over the past 20 years, the idea
of involuntary adaptation has re-entered the literature of cog-
nitive psychology under such rubrics as hardiness (Kobasa,
Maddi, & Kahn, 1982), self-deception and emotional coping
(Lazarus & Folkman, 1984), and illusion (Taylor, 1989).
Defense mechanisms are clearly as important in reducing
anxiety from cognitive dissonance as they are in minimizing
anxiety from conflict between conscience and impulse.
In recent years, experimental strategies for studying
defense mechanisms have improved (Cramer, 1991;
Horowitz, 1988; Vaillant, 1992). Building on the work of
Norma Haan (1963) at Berkeley and Elvin Semrad (1967)
at Harvard, I have tried to operationalize defenses and to
demonstrate their predictive validity (Vaillant, 1971,
1993). Over 30 years, such efforts have met with modest
success, and the validity of an adaptive hierarchy of de-
fenses appears clear (Vaillant, 1992). However, as Phoebe
Cramer's (1991) encyclopedic review of the methodology
for identifying and quantifying defenses has illustrated, no
one has yet developed a method for assessing defenses that
meets conventional standards for psychometric reliability.
A second reason that defenses have fallen from favor
in psychology is that there is no commonly accepted lan-
guage. For example, within 50 miles of San Francisco,
there were recently six competing, nonoverlapping nomen-
clatures for involuntary coping mechanisms. Each nomen-
clature was used by a distinguished investigator of stress
(Block & Block, 1980; Haan, 1977; Horowitz, 1988; Laza-
rus & Folkman, 1984; Moos & Billings, 1982; Weinberger,
Schwartz, & Davidson, 1979). Rarely, however, did any
investigator cite the work of his or her neighbors. The
result has been semantic chaos. Recently, the DSM-IV
(American Psychiatric Association, 1994) has offered a
terminology, a glossary, and a tentative diagnostic axis to
provide a common language.
Defenses, no matter how ingeniously assessed, reflect
value judgments about mental process, as do process con-
cepts in physics (e.g., forward motion and velocity). All
three--velocity, forward motion, and defenses--depend
on the vantage point of the observer and involve processes
rather than static qualities like mass or intelligence. Nev-
ertheless, if people wish to understand their own lives in
time and space, these are judgments worth making.
To overcome relativity, reliability of defense recogni-
tion requires longitudinal study. Before I can assert that the
Ninth Symphony represents the sublimation of Beethoven's
conflict over abusive father figures, I need objective longi-
tudinal evidence. First, I need Beethoven's own contem-
poraneously written diary to document both his despair and
his anger at father figures over decades. Second, for objec-
90 January 2000 • American Psychologist
tive assessment, I need behavioral evidence of his defense:
a symphony (not just a pencil-and-paper response or a
dream report). Finally, I need objective consensus that his
creation was empathic art that others valued, not autistic
lunacy that others mocked. Thus, the documented wild
cheers of a contemporary, musically sophisticated, Vien-
nese concert audience is more convincing than the value
judgment of one 20th century, musically challenged Amer-
ican psychiatrist. Using such triangulation of real symp-
toms, autobiographical report, and contemporaneously as-
sessed biographical fact to measure invisible mental pro-
cess is analogous to surveyors using triangulation to assess
the height of mountains they cannot climb.
As a method to study defenses, I have used three diverse
50-year prospective studies of lives. Using consensus defini-
tions from the literature (Vaillant, 1971), I selected five mech-
anisms-humor, altruism, sublimation, anticipation, and sup-
pression--that, first by hypothesis and then by empirical
study, appeared adaptive in the three samples. The term adap-
tive defense, and its synonym healthy denial, have two con-
notations: The first is transformative (turning lead into gold),
and the second is making the best out of a bad situation.
Whether such a healing response is viewed as miraculous or
merely a patch-up job depends on whether optimal wound
healing is viewed as a scar or as a result of a delicate ballet of
blood clotting and fibroblast migration--neither too much nor
too little. Each adaptive or healthy defense involves the ballet
of keeping idea and affect, subject and object clearly in mind
while simultaneously attenuating the conflict (cognitive
In nonconflictual situations, of course, the putative
defense mechanisms of anticipation, altruism, and suppres-
sion seem quite conscious and voluntary. In highly emo-
tionally charged situations, however, such deployment of
these mechanisms can be seen as both transformative and
making the best of a bad situation. A man with a criminal
record for the first time "counting to ten" (suppression)
while consciously examining his anger, rather than impul-
sively punching a policeman; a mother rehearsing affec-
tively and realistically, rather than denying, the fact that her
child is dying (anticipation); a survivor of child abuse,
rather than abusing her own children, working in a shelter
for survivors of abuse (altruism) are such examples. Such
behaviors emerge with maturation as delicate transforma-
tive mental balancing acts and not as a result of good
advice and self-help cognitive strategies.
The Study of Adult Development
The Study of Adult Development provided the three cohorts
of individuals that were used as a prospective and empirical
means of triangulating and validating defensive behaviors.
Each cohort had been prospectively studied for over half a
century: the "College" sample born about 1920 (Heath, 1945),
the "Core City" sample of inner-city men born about 1930
(Glueck & Glueck, 1950), and the "Terman" sample of gifted
women born about 1910 (Terman, 1925).
For all three samples, the basic methodology of the
Study of Adult Development was to keep raters of psycho-
logical health and prospective behavioral outcome unaware
of defense assessment and to keep raters of defenses un-
aware of evidence of positive mental health and future
adaptation. Taken individually, these three now elderly
Caucasian samples can hardly be viewed as representative
of the general population. However, the three samples have
the virtues of being vastly different from each other and
belonging to historical birth cohorts up to 20 years apart.
Within each sample, there was considerable homogeneity.
Thus, the between-group similarities and the within-group
differences may be generalizable to some other samples.
More important, prospective study permitted defensive al-
truism to be distinguished from simple kindness and de-
fensive projection to be distinguished from the vigilant
recognition of real persecution.
The College Sample
The Grant Study (Heath, 1945; Vaillant, 1977) began at the
Harvard University Health Services in 1938. The study was
underwritten by W. T. Grant because, "Large endowments
have been given and schemes put into effect for the study
of the ill, the mentally and physically handicapped ....
Very few have thought it pertinent to make a systematic
inquiry into the kinds of people who are well and do well"
(Heath, 1945). Sixty years ago, then, the Grant Study
anticipated the need for a positive psychology.
In the selection process, about 40% of each Harvard
class was arbitrarily excluded for academic reasons. The
health service records of the remaining 60% of each class
were then screened, and half were excluded because of evi-
dence of physical or psychological disturbance. The college
deans then selected one third of the remaining 300 men who
they thought would do well. Between 1939 and 1942, 268
sophomores were selected for study. For half a century, all but
20 of the men have continued to participate in this study of
positive psychology with remarkable loyalty. They have re-
ceived questionnaires about every 2 years, physical examina-
tions every 5 years, and interviews about every 15 years.
Socioeconomically, the College sample men were
drawn from a privileged group but not exclusively so.
Although one third of the men's fathers had some profes-
sional training, one half of the men's parents never grad-
uated from college. Although one half of the men had some
private education, half of the men were on scholarship
and/or had to work during the academic term to earn
tuition. In adult life, the College sample enjoyed the in-
come and social status of corporate managers, yet they
drove the battered cars and pursued the hobbies, politics,
and lifestyle of college professors.
The Core City Sample
These 456 men represent a very different cohort but one
also chosen for relative mental health. In junior high
school, they were selected as nondelinquent controls for a
prospective study of juvenile delinquency. The study was
conducted by Sheldon and Eleanor Glueck at Harvard Law
School and led to their landmark book Unraveling Juvenile
Delinquency (Glueck & Glueck, 1950, 1968). Like the
College men, the Core City men were studied originally by
a multidisciplinary team of physicians, psychologists, psy-
January 2000 • American Psychologist 91
chiatrists, social investigators, and physical anthropolo-
gists. The Core City men were interviewed at ages 14, 25,
32, and 47 (Vaillant, 1995).
The Core City sample came from the 60% of Boston
census tracts with the highest rates of juvenile delinquency.
The boys' average IQ was 95, and 61% of their parents were
foreign born. In childhood, half of the Core City men had
lived in clearly blighted slum neighborhoods. Half came from
families known to five or more social agencies, and more than
two thirds of their families had recently been on welfare. Over
the years, however, this group has experienced marked up-
ward social mobility (Long & Vaillant, 1984).
The Terman Women Sample
Through the cooperation of Robert Sears and Albert Hastorf,
I obtained access to a Stanford University (Terman women)
cohort of gifted women studied since 1920. The 90 women
that make up the current study sample are a representative
subsample of the 672 women in Terman's original cohort of
gifted California public school children (Holahan & Sears,
1995; Terman, 1925; Terman & Oden, 1959).
The high intelligence of the Terman women--mean
IQ of 151--was a social asset. Their mental health was
demonstrably better than that of their California class-
mates. They showed significantly more humor, common
sense, perseverance, leadership, and even popularity than
their school peers. Up to the age of 78, the mortality of the
Terman women has been only half of what would be
expected for White American women in their birth cohort.
Investigators followed the Terman sample by ques-
tionnaire every five years and by personal interview in
1940 and 1950. In 1987, Vaillant and Vaillant (1990a)
selected a representative subsample. Of the 90 women
selected, 29 had died and 21 of the surviving women
refused to interview, some because of poor health. We
reinterviewed the remaining 40 women.
Adaptive or Mature Defenses
Adaptive or mature defenses (altruism, sublimation, sup-
pression, humor, anticipation) are common among the
mentally healthy and become more salient as individuals
mature from adolescence to midlife (Vaillant, 1977). In
keeping with the conceptualization of positive psychology,
the association of mature defenses with mental health re-
mains whether health is measured by subjective happiness,
psychosocial maturity, occupational success, richness and
stability of relationships, or absence of psychopathology
(Vaillant, 1992). Individuals with brain damage (e.g., al-
cohol dependence, schizophrenic relapse, multiple sclero-
sis) replace adaptive defenses with more maladaptive
mechanisms, most notably projection.
Table 1 schematizes the defenses discussed in this article
within the adaptive levels suggested by DSM-1V. The table
provides an oversimplified schema for the mutually exclusive
definitions that contrast the five adaptive defenses listed above
with less adaptive mechanisms. Each defense has been char-
acterized by the extent to which it denies or distorts subject
and object and idea and affect in the experience of and
expression of impulse. For example, defense mechanisms can
allow a person to ignore the affect (isolation, intellectualiza-
tion), to ignore the cognitive representation of the affect (re-
pression), to reverse the direction of an impulse (make the self
the object; projection), or to make the object the self (suicide
or passive aggression). Each defense has also been character-
ized by the way in which it modifies the four lodestars of
conflict: affect, reality, conscience, and relationships. The
high-adaptive-level defenses provide the most balanced re-
sponse to such involuntary homeostatic distortions of inner
and outer reality.
To the beholder, adaptive mechanisms appear as con-
venient virtues, and there is rarely a therapeutic reason to
alter them. Although closer to consciousness than mecha-
nisms like projection and repression, mature mechanisms
cannot be voluntarily deployed. No one is more transparent
than someone trying to use humor or altruism; No one is
more angry looking than someone consciously suppressing
rage; and when depressed just try writing Beethoven's
Ninth Symphony on purpose.
In keeping with positive psychology, adaptive defenses
often appear as moral to the observer as maladaptive defenses
appear immoral. The prejudice of projection and the tantrums
of acting out appear to others as sins. In contrast, doing as one
would be done by (altruism), a stiff upper lip (suppression),
planning for the future (anticipation), the ability not to take
one's self too seriously (humor), and "turning lemons into
lemonade" (sublimation) are the very stuff of which a positive
psychology should be concerned.
Let me elaborate on the ~ansformative nature of each of
five mature mental mechanisms schematically defined in Table 1.
When used to transform conflict, altruism involves getting
pleasure from giving to others what people would themselves
like to receive. For example, victims of childhood sexual
abuse often pathologically cut themselves (turning anger
against the self), abuse children (acting out), or use "neurotic"
compromises such as becoming frigid or joining convents
(reaction formation). Alternatively, and transformatively, al-
truistic victims of child abuse might work in shelters for
battered women and in support groups or hotfines for abuse
victims. Often almaism is an adaptive outgrowth of the de-
fense of reaction formation, a mechanism that can maladap-
tively make the person's desires all bad and the needs of
others all good. Using reaction formation, an ex-drinker who
suddenly declares drinking as a filthy habit annoys his friends.
Using altruism, the ex-alcoholic who serves as a sponsor to a
new Alcoholics Anonymous member achieves a transforma-
tive process enjoyed by giver and receiver.
My wife, five months pregnant, was interviewing a
couple from the Core City sample to whom our study
offered no compensation. The greatest pain in their life was
having lost six children through Rh incompatibility. As my
wife got up to leave, the childless wife, whose grief and
envy can only be imagined, gave my wife a handsome,
handmade baby sweater. The lives of everyone in the room
had been suddenly enriched.
92 January 2000 • American Psychologist
-~ 0 c t-
"-0 --~ 0..
"11 I I
I I I I I~ ~ I I I ~ $
NO0 0 ~0
I I I I I
° I ~1 I~ I~ I I I~ I I
-(3 ~ -0
°"~ ~ I I
~, "0 0 ..0'~
~ ~"c "1 I N
o~ "~ --',: "~ "E
"-O -O "-O "-~ -O "-O
~, I~ e-o
-o ~ -o
-oe o~ I I N ~-0 N O-N I I
O') > ~ ,--
6 o • "~
~ ,~ ~,~ ~'~ o
o~ "Z ~ •
~ ~.o ~_~
o o._O ~--~.-._"
~.2 E :~.~
._._~ ~'~ _~
The sign of a successful defense is neither careful cost
accounting nor shrewd compromise, but rather psychic
alchemy. Upon an inanimate Attic vase John Keats discov-
ered-and shared--an attenuated yet passionate sexuality.
In "Ode on a Grecian Urn," Keats conveyed a miraculous
concept: "More happy, happy love! / Forever warm and
still to be enjoyed. / Forever panting and forever young."
With marvelous control of language Keats turned lust,
perhaps even imminent rape--"What maidens loth? / What
mad pursuit? What struggle to escape?"--into more happy
love. A less poetic member of the study wrote, "I have
twice the sex drive of my wife. We adjust ourselves by
varying our sex play to suit each other. We believe that
lovemaking should be practiced as an art!"
Thus, sublimation allows an indirect resolution of
conflict with neither adverse consequences nor marked loss
of pleasure. Unlike the autistic fantasy of the child and the
schizophrenic, artists can peddle their most private dreams
to others. In contrast, the mechanism of acting out--rape--
dissipates the torrent of our unmodulated affect on strang-
ers, and reaction formation dams such affect expression
Finally, sublimation does more than make affect accept-
able; it also makes ideas exciting. In terms of their Harvard
grades and tested intellectual aptitudes, the men in the College
sample with brilliant teaching careers at Stanford and Harvard
were not more gifted than fellow study members teaching
joylessly at mediocre institutions. Too often the less success-
ful professors in the College sample used displacement and
isolation so compulsively that their cognitive interests were
stripped of affect and passion. In every facet of their lives--
not just their teaching and publishing--the successful profes-
sors were more comfortable in coloring their ideas with the
pigment of emotion (Vaillant, 1977).
Suppression (~toicism) is not as elegant as sublimation be-
cause suppression always sacrifices beauty for truth. Suppres-
sion has none of the humanity of altruism or humor, and
suppression is often regarded by psychotherapists as a vice,
not a virtue. When used effectively, however, suppression is
analogous to a well-trimmed sail; every restriction is precisely
calculated to exploit, not to hide, the winds of passion.
Suppression involves the semiconscious decision to
postpone paying attention to a conscious impulse and/or con-
flict. A critical difference between suppression and repression,
between suppression and isolation, and between stoical sup-
pression and Spartan reaction formation is the degree to which
suppression allows all the components of conflict to exist at
least partially in consciousness (cf. Table 1). The distinction
between suppression and Pollyanna's dissociation or "neu-
rotic denial" is more complex. Both the stoic person and the
person behaving like Pollyanna note that clouds have silver
linings, but Pollyanna leaves her umbrella at home. Evidence
that suppression is not a conscious cognitive strategy as many
believe is provided by the fact that jails would empty if
delinquents could learn to just say no.
As an example of suppression, the normal life tempo
of one highly energetic College sample man was to work a
60-hour week as chief executive officer of two large cor-
porations and then run for six miles on Sunday to relax.
However, he described a navy diving accident that took
place during World War II in the following manner: He
was 40 feet underwater; his air valve was jammed; his radio
did not work; and he knew that only eight minutes of air
were left in his diving helmet. He recognized that there was
nothing that he could do for himself. "I thought my end had
come.., struggling would not have helped and used
maybe three times as much air. I didn't pray. I merely sat,
very much like an old cow, and waited for help--very
unhappy." He knew his feelings, and he knew they would
not help, so he suppressed them until he was rescued. The
delicate mental balance involved in successful suppression
is as voluntary and as involuntary as walking on a tight-
rope. Such balance seems easy for the accomplished, co-
ordinated acrobat and seems utterly impossible and anxiety
provoking for everyone else.
As with altruism, the use of anticipation is often voluntary and
independent of conflict resolution. Rather, it is in cases of "hot
cognition" that anticipation becomes an involuntary coping
skill. If suppression reflects the capacity to keep current im-
pulse in mind and control it, anticipation is the capacity to
keep affective response to an unbearable future in mind.
The defense of anticipation reflects the capacity to
perceive future danger affectively as well as cognitively
and by this means to master conflict in small steps. In the
1950s, as scientists began the deliberate study of healthy
adaptation, Irving Janis (1958) discovered that moderate
amounts of anxiety before surgery promoted adaptation. At
the National Institute of Mental Health, David Hamburg
and his colleagues (Friedman, Chodoff, Mason, & Ham-
burg, 1963) noted the value of anticipatory mourning in
parents of children with leukemia. Psychiatrists responsible
for preparing Peace Corps volunteers noted that volunteers'
capacity to anticipate future affective difficulty better pre-
dicted subsequent adaptation than did their apparent emo-
tional stability on psychological tests (Ezekiel, 1968).
Anticipation differs in an important way from using
isolation and intellectualization to make soothing "lists."
Anticipation involves more than just the ideational work of
cognitive planning. Anticipation involves both thinking
and feeling about the future. For example, consider leg-
endary aviators, like Charles Lindbergh and Chuck Yeager.
They calmly survived exciting flying careers by dealing
with stress as Mithradates did with poison--taking a little
at a time. To have underestimated danger would have been
fatal. To have exaggerated danger would have been emo-
tionally incapacitating. Thus, they worried in advance, they
made lists, and they practiced. Then, appreciating that they
had prepared as well as they could, they relaxed. Like
suppression and altruism, anticipation is so easy to pre-
scribe but so difficult to do.
94 January 2000 • American Psychologist
We all recognize that humor makes life easier. As S. Freud
(1905/1960) suggested, "Humor can be regarded as the high-
est of these defensive processes," for humor "scorns to with-
draw the ideational content bearing the distressing affect from
conscious attention, as repression does, and thus surmounts
the automatism of defense" (p. 233). Humor permits the
expression of emotion without individual discomfort and
without unpleasant effects on others. Humor, like anticipation
and suppression, is such a sensible coping device that it ought
to be conscious, but almost by definition, humor always
surprises people. Like the other mature defenses, humor re-
quires the same delicacy as building a house of cards--timing
is everything. The safety of humor, like the safety of dreams
during REM sleep, depends on cataplexy. People see all and
feel much, but they do not act.
Humor keeps both idea and affect in mind. Mature
humor allows people to look directly at what is painful,
whereas dissociation and slapstick distract people so that
they look somewhere else. Much of humor is lost in the
retelling. Thus, unlike Beethoven's sublimation, humor is
difficult to illustrate. Humor, like a rainbow, is real but
forever evades our grasp.
Adaptive or Health-Promoting
Defenses May Be Inde_pendent of
Social Class and Gender
The study assessed defenses of the Core City men and of
the College men at age 47 and of the Terman women at age
77. Defenses were identified by behavioral vignette (Vail-
lant, 1992). Then, the ratio of adaptive level defenses to
less adaptive defenses was calculated, and the ratio was
converted to a 1-to-9 scale. Table 2 illustrates that such
quantification of adaptiveness of defenses was relatively
independent of years of education, IQ, and parental social
class. Admittedly, within each cohort the range of socio-
economic status was truncated; nevertheless, within the
Core City cohort both IQ and education predicted future
occupational prestige and social class (p < .001)--just not
the adaptiveness of defenses.
Correlation of Social Antecedents With Adaptiveness
Adaptiveness of defenses
(n = 154 a)
(n = 189 °)
In = 40) Antecedent
Years of education
Parental social class
° Sample size is reduced. To control confounders, men with IQs less than 86,
depression, alcohol dependence, and schizophrenia were excluded.
* p < .05. (Spearman rank correlation coefficient was used.)
If the three samples are contrasted with each other, 34%
of the Terman women, 25% of the privileged College men,
and 22% of the less socially and intellectually privileged Core
City men manifested defenses predominantly at the adaptive
level. These differences were not significant and can be at-
tributed to differences in the original rules for selection.
Longitudinal study not only facilitated rater reliability in
the identification of defenses (Vaillant, 1992), but also
facilitated the demonstration of predictive validity. Thus,
roughly 20 years after the relative adaptiveness of defenses
was rated, the physical health and the psychosocial adjust-
ment of the study men was assessed by raters unaware of
the conditions of the participants' lives before age 50.
Separate ratings were obtained for evidence of subjective
and objective mental and physical health.
Table 3 illustrates the power of scaled adaptiveness of
defenses to predict multiple facets of positive health. For
contrast, Table 3 also presents the power of attained social
class (measured by years of education) and the power of trait
neuroticism measured contemporaneously at age 65 by the
NEO Personality Inventory (Costa & McCrae, 1985; McCrae
& Costa, 1985) to predict the same variables. Because mental
illness can lead both to maladaptive defenses and to poor
future health, men with alcohol dependence, major depressive
disorder, schizophrenia, or IQs less than 85 were omitted from
the tests of predictive validity in Table 3. Had mentally ill
participants been included, the predictive power of defense
choice would have been greater.
As shown in Table 3, psychosocial adjustment (objec-
tive), social supports (objective), and marital satisfaction
(subjective) were assessed by independent raters who inte-
grated data from seven questionnaires from the College and
Core City men and two questionnaires from the wives of
the College men. Psychosocial adjustment from age 50 to
age 65 was assessed by evidence of job promotions and
enjoyment, marital stability, games with others, and no use
of psychiatrists or tranquilizers (Vaillant, Meyer, Muka-
real, & Soldz, 1998; Vaillant & Vaillant, 1990b). Social
supports from ages 50 to 70 were assessed by evidence of
close relations with wives, children, siblings, and social
network, as well as by strength of religious affiliation, the
presence or absence of a confidante, and games with
friends (Vaillant et al., 1998).
Joy in living (subjective) was quantified by summing
each man's satisfaction over the past 20 years (on a 5-point
scale) in four life areas (marriage, children, job, and friends)
and by then adding his best score from one of four additional
areas (hobbies, sports, community activities, or religion)
(Vaillant, 1999). Physical functioning (subjective) for the Col-
lege men from ages 70 to 75 and for the Core City men from
ages 60 to 65 was assessed by repeatedly monitoring instru-
mental actMties of living (e.g., ability to climb stairs, walk
two miles, carry suitcases, and drive at night; Vaillant, Orav,
Meyer, McCullough, & Roston, 1996).
Finally, Table 3 illustrates that adaptive defenses
transform only the perception of reality, not reality itself.
Thus, adaptive defenses predicted the absence of subjective
January 2000 • American Psychologist 95
Late Life Consequences of Adaptive Defenses at Ages 20-47
Core City (n = 137 °)
College (n = 154 °}
of defenses Evidence Neuroticism Neuroticism
Psychosocial ad ustment (ages 50-65)
(Vai ant & Vai ant, 1990b)
Social supports b (Vaillant et al., 1998)
Joy in living b
Marital satisfaction (midlife)
Subjective physical functioning b
Objective physical health c
Objective physical health decline
.25** .25** -.08
.37 . . . .
.30 . . . .
.32" * *
.09 .14 -.09
.05 .28 . . . . .08
.34 . . . .
.35 . . . .
.01 .04 -.13
° Sample size is reduced because men who died before age 65 are excluded,
men. c 1 = well, 2 - minor irreversible illness, 3 = chronic illness, 4 = disabling illness, 5 = dead (Vaillant, 1979). Measured at age 60 for Core City men and
at age 70 for College men.
* p < .05. ** p < .01. *** p < .001. (Spearman rank correlation coefficient was used.)
Neuroticism was measured with the NEe Personality Inventory.
b Measured at age 65 for the Core City men and measured at age 75 for the College
physical disability--up to 30 years later--but such de-
fenses did not predict physical health decline (objective)
assessed by an independent internist.
The point of Table 3 is that the relative adaptiveness
of defenses (measurement described in Vaillant, 1992,
1993, and schematized in the DSM-IV) may offer as good
a metric for positive mental health (Vaillant & Schnurr,
1988) as there is. For two very socioeconomically diverse
samples of men, income, objective psychosocial adjust-
ment, social supports, marital satisfaction, subjective phys-
ical functioning, and joy in living were more highly corre-
lated with adaptive defenses measured 20 years earlier than
with either education or neuroticism.
Psychology needs to know more not only about the mea-
surement of positive mental health but also about how
people exposed to severe risk factors maintain positive
mental health. I address four major risk factors: childhood
poverty, the physical limitations of old age, stressful life
events, and severe combat. First, the 70 Core City men who
manifested the most adaptive defenses were just as likely to
have come from welfare families in Social Class V (Hol-
lingshead & Redlich, 1958) as were the 73 men with the
least adaptive defenses. In contrast, as adults only 1% of
men with the most mature defenses but 21% of men with
the least mature defenses were in Social Class V. In short,
adaptive defenses may catalyze escape from poverty.
Second, Figure 1 depicts the subjective physical func-
tioning at age 65 of those Core City men who were still in
good physical health at age 50. In other words, the figure
includes only those men whose defense levels could not have
been impaired by prior poor health. The more dominant their
use of adaptive defenses between ages 20 and 47, the more
likely they were at 65 to report being able to climb stairs, walk
long distances, and engage in vigorous physical activities that
they enjoyed. As Table 3 shows, however, their objective
physical health was uncorrelated with defense level.
Core City Men Without Disability at Age 50 Who
Were Still Without Disability 15 Years After the Initial
0-1 2 3 4 5
Use of Adaptive Defenses
subjective disability score of 10 through 14. This meant that the men had not
given up any major activity and were still able to move heavy furniture and/or
chop wood, walk two miles, and climb two flights of stairs without resting, albeit
sometimes more slowly. Use of adaptive defenses was rated on a scale from 0
(unimportant or absent) to 5 (style dominant).
The percentage of men with no significant disability was based on a
96 January 2000 ° American Psychologist
Likelihood of Depression Covaried With Total Life
Stress and With Adaptiveness of Defenses
I-O- Men with total life stress scores > 20 I n = 4
--x- M___en with total life stress scores < 21 Ii
X )( X X
n=15 n=14 n=15 n=7 n=3 n=2
5 4 3 2 1 0
Use of Adaptive Del?nses
absen 0 to 5 [style dominant).
Use of adaptive defenses was rated on a scale from 0 (unimportant or
Third, the number of stressful life events in the adult
lives of the College men from ages 20 to 60 was studied
prospectively (Cui & Vaillant, 1996). The number and sever-
ity of such life events both predicted-and resulted from--the
occurrence of major depressive disorder. Figure 2 illustrates
that major depressive disorder occurred only among men with
high life stress scores. However, the men who deployed the
most adaptive defenses could still experience multiple stress-
ful life events without risk of major depression.
Finally, adaptive defenses also mitigated the strong
association between severe combat and later symptoms of
post-traumatic stress disorder (PTSD) among the College
sample. (In our study, symptoms of PTSD could be almost
entirely explained by severity of World War II combat
carefully quantified in 1946; Lee, Vaillant, Torrey, & El-
der, 1995; Wells & Woods, 1946.) Of the 33 College men
who experienced the most severe combat, the 16 men who
deployed most adaptive defenses reported an average of
0.19 PTSD symptoms. The 17 men with less adaptive
defenses who had had similarly high combat exposure
reported an average of 1.70 PTSD symptoms, t(31) = 2.75,
p = .0t, two-tailed. It is significant that prior to the war, the
two groups of men did not differ in physical symptoms
with stress, and in late middle life they did not differ in
How Do Defenses Work?
How do mature defenses work to promote a positive psychol-
ogy (enhanced ability to work, love, and play) and at the same
time to reduce conflict and cognitive dissonance? Table 1
presented a range of defenses rank ordered as in the DSM-IV.
The DSM-1V suggests that the mechanisms at "the high adap-
tive level" not only maximize gratification but also "promote
an optimum balance among conflicting motives" (American
Psychiatric Association, 1994, p. 752). Again, whether one
views such a response as making the best of a bad situation or
as transformative depends on the vantage point of the ob-
server. Thus, unlike less adaptive mechanisms, mature de-
fenses synthesize and attenuate rather than deny and distort
conflicting sources of human behavior like conscience, reality,
interpersonal relations, and emotions. The best-of-a-bad-situ-
ation point of view would note that predominant use of
adaptive defenses simply means that such individuals did not
cope by using less balanced mechanisms like schizoid fantasy
and projection, which are strongly predictive of poor out-
comes. Unlike acting out. which denies conscience, or reac-
tion formation, which denies emotion, or schizoid fantasy
which denies real people, or projection, which denies the
subject, or psychotic defenses, which deny objective reality,
mature defenses elegantly balance and attenuate these multi-
ple sources of conflict. Ballet dancing, Albert Rothenberg's
"Janusian creativity," Beethoven producing a symphony
fueled by despair and rage, people with physical disabilities
deriving hope and self-esteem from helping others with dis-
abilities all reflect the transformative nature of achieving psy-
Beyond the above suggestions, psychology really does
not know how defenses work. Do adaptive defenses reflect
inborn traits te.g., perfect pitch or a capacity for higher math-
ematics)? Or do adaptive defenses reflect traits that are ac-
quired through education and maturation (e.g., good diction or
a graceful backhand)? Should psychology view adaptive de-
fenses as virtues (like empathy and creativity)? Or should
psychology view such defenses as adaptive self-deceptions to
resolve conflict as did Anna Freud when she quipped that
altruism came not from the goodness "but from the badness of
his heart" (Sandler & Freud, 1985, p. 176)? I believe that the
correct answer to all four questions is yes, but more research
As Table 2 shows, the etiology of adaptive defenses is
as obscure as the etiology of creativity or athletic prowess.
Although genes, social environment, and the absence of
brain disease undoubtedly each play a role, the association
of adaptive defenses with positive psychology is most
pronounced among individuals from dysfunctional families
(Vaillant, 1993). The best definition of creativity--or of an
adaptive defense--is putting something of value in the
world that was not there before. It is the transformative,
creative quality that makes the adaptive defenses more than
just healthy wound healing.
This article raises questions that must be solved if psychol-
ogists are to develop a science of positive psychology.
First, how should psychology quantify positive mental
health? At present, psychology has no metric except per-
haps scores of greater than 85 on the DSM-1V's Axis V
(Global Assessment of Functioning). If more reliable meth-
January 2000 ° American Psychologist 97
ods for assessing the relative maturity of defenses can be
developed, psychology may gain a means of quantifying
the theoretical formula for positive mental health that
Marie Jahoda (1959) offered to psychology 40 years ago.
She suggested the same synthesis between affective life
and practical reality that is reflected in the conceptualiza-
tion of adaptive-level defenses. Jahoda suggested that men-
tally healthy individuals should be oriented toward the
future and efficient in problem solving. They should be
resistant to stress and perceive reality without distortion.
They should possess empathy and be able to love and to
play as well as to work. They should remain in touch with
their own feelings. In short, they should manifest anticipa-
tion, suppression, altruism, humor, and sublimation.
In addition, psychology needs to understand how best
to facilitate the transmutation of less adaptive defenses into
more adaptive defenses. My own suggestions (Vaillant,
1995) have been first to increase social supports and inter-
personal safety and second to facilitate the intactness of the
central nervous system (e.g., rest, nutrition, and sobriety).
However, the newer forms of integrative psychotherapies
also can catalyze such change, and throughout this journal
issue there are further clues.
American Psychiatric Association. (1994). Diagnostic and statistical
manual of mental disorders (4th ed.). Washington, DC: Author.
Block, J. H., & Block, J. (1980). The role of ego-control and ego-
resiliency. In W. A. Collins (Ed.), The Minnesota Symposium on Child
Psychology: Vol. 13. Development of cognitive, ~'ect and social rela-
tions (pp. 39-101). Hillsdale, NJ: Erlbaum.
Costa, P. T., & McCrae, R. R. (1985). The NEO Personality Inventor'
manual. Odessa, FL: Psychological Assessment Resources.
Cramer, P. (1991). The development of defense mechanisms. New York:
Cui, X., & Vaillant, G. E. (1996). The antecedents and consequences of
negative life events in adulthood: A longitudinal study. American
Journal qf Psychiatrv, 152, 21-26.
Ezekiel, R. S. (1968). The personal future and Peace Corps competence
[Monograph]. Journal of Personal and Social Psychology, 8, (2, Pt. 2).
Forbes, E. (1969). Thayer's liJe of Beethoven. Princeton, NJ: Princeton
Freud, S. (1960). Jokes and their relation to the unconscious. In The
Standard Edition of the Complete Works of Sigmund Freud (Vol. 8, pp.
225 233). London: Hogarth Press. (Original work published 1905)
Friedman, S. B., Chodoff, P., Mason, J. W., & Hamburg, D. (1963).
Behavioral observations on parents anticipating the death of a child.
Pediatrics, 32, 610 625.
Glueck, S., & Glueck, E. (I 950). Unraveling juvenile delinquency. New
York: Conunonwealth Fund.
Glueck, S., & Glueck, E. (1968). Delinquents and non-delinquents in
perspective. Cambridge, MA: Harvard University Press.
Haan, N. (1963). Proposed model of ego functioning: Coping and defense
mechanisms in relationship to IQ change. Psychological Monographs,
ttaan, N. (1977). Coping and defimding. New York: Academic Press.
Heath, C. (1945). What people are. Cambridge, MA: Harvard University Press.
Holahan, C. K., & Sears, R. R. (1995). The gifted group in later maturity.
Stanford, CA: Stanford University Press.
Hollingshead, A., & Redlich, F. C. (1958). Social class and mental illness.
New York: Wiley.
Horowitz, M. J. (1988). Introduction to psychodynamics. New York:
Jahoda, M. (1959). Current concepts of positive mental health. New York:
Janis, I. (1958). Psychological stress. New York: Wiley.
Kobasa, S. C., Maddi, S. R., & Kahn, S. (1982). Hardiness and health: A
prospective study. Journal of Personality and Social Psychology, 42,
Lazarus, R., & Folkrnan, S. (1984). Stress, appraisal and coping. New
Lee. K. A., Vaillant, G. E., Torrey, W. C., & Elder, G. H. (i995). A
50-year prospective study of the psychological sequelae of World War
II combat. American Journal of Psychiatry, 152, 516-522.
Long, J. V. F.. & Vaillant, G. E. (1984). Natural history of male psycho-
logical health: XI. Escape from the underclass. American Journal of
Psychiatry', 141, 341-346.
McCrae, R. R., & Costa, P. T. (1985). Comparison of EPI, Psychoticism
scales with measures of the five-factor model of personality. Person-
alitv and hldividual Differences, 6, 587-597.
Moos, R. H., & Billings, A. G. (1982). Conceptualizing and measuring
coping resources and processes. In L. Goldberger & S. Breznitz (Eds.),
Handbook of stress: Theoretical and clinical aspects (pp. 212-230).
New York: Free Press.
Sandier, J.. & Freud, A. (1985). The analysis of defense: The ego and the
meclumisms of defense revisited. New York: International Universities Press.
Semrad, E. (1967). The organization of ego defenses and object loss. In
D. M. Moriarity (Ed.), The loss of loved ones (pp. 126-134). Spring-
field, IL: Charles C Thomas.
Taylor, S. (1989). Positive illusions: Creative self deception and the
healthy mind. New York: Basic Books.
Tennan, L. M. (1925). Genetic studies of genius: Vol. 1. Mental and
physical traits of a thousand gifted children. Stanford, CA: Stanford
Tennan, L. M., & Oden, M. H. (1959). Genetic studies of genius: Vol. 5.
The gifted group at midlife. Stanford, CA: Stanford University Press.
Vai/lant, G. E. ( 1971). Theoretical hierarchy of adaptive ego mechanisms.
Archives ~f General Psychiatry. 24, 107-118.
Vai[lant, G. E. (1977). Adaptation to life. Boston: Little Brown.
Vaillant, G. E. (1979). Natural history of male psychological health:
Effects of mental health on physical health. New England Journal of
Medicine. 301, 1249-1254.
Vai[lant, G. E. (1992). Ego mechanisms of defense. Washington, DC:
American Psychiatric Press.
Vaillant, G. E. (1993). Wisdom of the ego. Cambridge, MA: Harvard
Vaillant, G. E. (1995). Natural history of alcoholism revisited. Cam-
bridge, MA: Harvard University Press.
Vaillant, G. E. (1999). Lessons learned from living. Scientific American,
Vaillant, G. E., Meyer, S. E., Mukamal, K., & Soldz, S. (1998). Are social
supports in late midlife a cause or a result of successful physical aging?
Psychological Medicine, 28, 1159-1168.
Vaillant, G. E., Orav, J., Meyer, S. E., McCullough, L., & Roston, D.
(1996). Late life consequences of affective spectrum disorder. Interna-
tional Psychogeriatrics, 8. 1 20.
Vaillant, G. E., & Schnurr, P. (1988). What is a case? A forty-five year
tbllow-up of a college sample selected for mental health. Archives q/
General Psychiat~_, 45, 313-319.
Vaillant, G. E., & Vaillant, C. O. (1990a). Determinants and consequences
of creativity in a cohort of gifted women. Psychology of Women
Quarterlv, 14, 607-616.
Vaillant, G. E., & Vaillant, C. O. (1990b). Natural history of male
psychological health: XII. A forty-five year study of successful aging at
age 65. American Journal of Psychiat~', 147, 31-37.
Weinberger, D. A., Schwartz, G. E., & Davidson, R. J. (1979). Low-
anxious, high-anxious, and repressive coping styles: Psychometric pat-
terns and behavioral and physiological responses to stress. Journal oJ
Abnormal Psychology, 88, 369-380.
Wells, F. L., & Woods, W. L. (1946). Outstanding traits: In a selected
college group with some reference to career interests and war records.
Genetic Psychology Monographs. 33, 127-249.
98 January 2000 ° American Psychologist