HAPPINESS IS LOVE: FULL STOP

George E. Vaillant, M.D.

From Harvard Medical School

And
Brigham and Women's Hospital
Boston, MA

Acknowledgments: This work was supported by research grants KO5-MHOO364 and

MH42248 from the National Institute of Mental Health

Address correspondence: George E. Vaillant, M.D.

Brigham and Women's Hospital
1249 BOYLSTON ST
Boston, MA 02215
(617) 525 6140

[email protected]

1
 HAPPINESS IS LOVE: FULL STOP

George E. Vaillant M.D.

Modern ethology and neuroscience make clear that all mammals are hard-wired
for love. Of all the fauna on earth, however, Homo sapiens is the most radically
dependent on love. Thus, ethologist Konrad Lorenz called love ―the most wonderful
product of ten million years of evolution‖; psychoanalyst Erich Fromm wrote, ―without
love humanity could not exist even for a day‖; and evangelist St. Paul concluded, ―And
now abideth faith, hope, love, these three; but the greatest of these is love.‖

Mammalian love involves attachment that is selective, enduring and often

remarkably unselfish. If the loved one leaves, the grief that ensues is also selective and
enduring. Not all mammals manifest such attachment, but many do. The Greek
philosophers did not understand love. Their agape (universal unselfish love) is not
selective, and the Greeks’ Eros (instinctual desire and lust) is not enduring. Satisfy his
lust and the romantic troubadour loses interest.

For 40 years I have directed a 70-year Harvard project, The Study of Adult
Development. I was recently quoted as asserting that the most salient finding to be drawn
from the study was: ―Happiness equals love—full stop.‖ A newspaper editor challenged
me. Did my sweeping, sentimental generalization really reflect the most important finding
of a multi-million dollar, seven-decade study designed to identify the key ingredients that
lead to a rewarding life? I spent a month reviewing the data, and my conclusions
remained those of Virgil: Omnia vincit amor. Only Virgil was arguing from sentiment and I
was arguing from real data. Let me explain.

Begun in 1938 the Study of Adult Development had intended to harness medical,
anthropological, and psychological sciences in order to understand what determines
health rather than illness. The study had been funded in its early years by a department
store magnate, William T. Grant. He had hoped that a longitudinal study of healthy men
would help him identify effective managers. The student health service director, Dr.

2
Arlie Bock, the designer of the study, hoped among other things that the research would
help the United States, about to be embroiled in World War II, select better officer
candidates. They were interested in identifying super men, not best friends.

Originally, the study had been planned to last twenty to twenty-five years—in those
days an incredibly ambitious goal. The originators never dreamed that it would continue
for seventy years. I have been the director of the study for the last 40 years; thus, I am
privy to who actually became the leaders and the best friends.

Nevertheless, when I began to respond to the editor, I wondered: Was love really
that important? Perhaps I was speaking from my heart and not from science. In order to
find out, I went back to the drawing board. I reviewed the findings on 268 Harvard
sophomores selected in 1938-42 and followed prospectively until 2010.

In order to demonstrate that our intimate attachments to other people—and them to

us—matter, and matter more than anything else in the world. I shall examine the power of
the childhood and young adult variables to predict ―rewarding‖ lives from age 50-80.
Criteria for a ―rewarding‖ late life included a decathlon of 10 ―events‖ in Table 1. Two
―events‖ reflected economic success: high earned income and high occupational prestige.
Four ―events‖, reflected subjective and objective biological success. Three ―events‖
reflected good relationships: a happy marriage (40-70), close father-child relationships,
and social supports at age 70. The final ―event‖ was never in 70 years receiving
professional help for mental distress. (This variable was an objective means of defining
the vague term ―mental health‖ and correlated more highly with longevity than almost
any variable in the Study.)

What young adult variables could predict success in these 10 events in later life?
There were 8 potential predictors of a ―rewarding‖ late-life adjustment that were favorites
among the original investigators. (Table 2). These predictors were relatively independent
of close relationships. Two were obvious: IQ and parental social class. (Remember, these
Harvard graduates came from very different backgrounds. Half of the students’ mothers
had not been to college, and fully half of the students needed scholarships and/or worked
during the academic year to pay their tuition. Other study fathers had made over a million
dollars a year -- during the Great Depression.)

3
 TABLE 1
A “DECATHLON’ FOR DEFINING A
FLOURSHING LIFE FROM AGE 50 TO AGE 80

1. Maximum earned income (usually between age 55 and 60)

2. In ―Who’s Who in America‖

3. Relative success at work, love and play from age 50-65

4. Good objective physical health at age 70

5. Physically active at age 75

6. Subjective and objective good physical and mental health at age 80

7. Social supports other than wife and kids from age 50-75

8. Close to kids age 50-80

9. Good marriage age 50-80

10. Low lifetime distress (no use of therapists or psych meds)

4
 TABLE 2

PREDICTOR VARIABLES

A. Early Study Variables Independent of Attachment

(15% of 80 Predictions Significant)

1. Parental Social Class
2. IQ
3. Somatotype
4. Masculine Body Build
5. ―Vital Affect‖
6. Athletic Prowess
7. ―Friendly‖
8. Treadmill Endurance

B. Other Common Early Risk Variables Independent of Attachment

(16 % of 50 Predictions Significant)

1. Ancestral Longevity

2. Alcoholic Relatives

3. Depressed Relatives

4. Years of Education

5. Childhood developmental problems

C. Early Variables Reflecting Attachment

(85% of 40 Predictions Significant)

1. Warm Childhood
2. Overall College Soundness
3. Empathic Coping Style (age 20-35)

5
4. Warm Adult Relationships 30-45
Six other items were theorized by the study to predict good officers and store
managers: Mesomorphy, Masculine body build, ―Vital Affect‖, Athletic Prowess,
Friendly and Treadmill Endurance. In only 8 out of a possible 80 matches (8 predictors
times 10 outcomes) did any of these 8 variables significantly predict ―rewarding‖
adjustment to life—none of them strongly. (By strongly I mean a correlation that could
have occurred by chance less than one time in 1,000.)

Five additional commonly cited childhood risk variables to aging: early ancestral
death, alcoholic relatives, depressed relatives, years of education and developmental
problems in childhood were significant in only 8 out a possible 50 matches, none of them
strongly. However, if the three types of genetic vulnerability (to depression, to short life
and to alcoholism) were summed, heredity did become a significant risk variable—
affecting 7 of the 10 outcome variables and two strongly. In short genes are important –
just not as important as love.

Thus, if, as the most important prologue to a good life, I substitute Love (variables
reflecting attachment) for genes and physical vigor, prediction became far more
successful. Since warm relationships are hard enough to measure in the 21st century let
alone in 1940, I used the four indirect measures illustrated in Table 2, part C.

During college both the men and their parents were interviewed in depth. To assess
the first attachment indicator, the assessment of a cohesive home-life was combined with
a warm relationship with mother, with father and with siblings to reflect a 20-point scale.
However, no one had thought to make such a rating until the men were age 50, when it
was made by young raters kept blind to all data after age 18.

The second predictor was the study staff’s A, B, and C original consensus rating of
the men’s overall soundness at age 21 but the facets of the scale that I have underscored
below were not those emphasized originally.

A = Would have no ―serious problem in handling problems that might

confront them.‖
B = ―If a boy was lacking in warmth in his touch with people‖ or too
―sensitive.‖

6
 C = Men who showed ―marked mood variations‖ or were ―markedly
asocial.‖

The third predictor was the ―maturity‖ or ―immaturity‖ of the men’s involuntary
coping style from 20 to 35. Mature coping mechanisms included ―suppression‖ (patience
and stoicism), ―altruism‖ (doing for others what you wished had been done for yourself),
and ―anticipation‖ (allowing painful emotions to come consciously to mind before the
event). Immature coping mechanisms included ―fantasy‖ (imaginary friends), projection
(externalizing blame), ―hypochondriasis‖ (help rejecting complaining) and ―acting out‖
(tantrums). While often soothing the subject, these latter behaviors do not win friends. In
effect, the extremes of coping style reflected narcissism and empathy during young
adulthood.

The final predictor ―Object Relations (age 30-47)" subtracted points: for not being
married for more than ten years, not having children, being distant from own children,
few friends, no enjoyable contact with family of origin, no clubs, and no games with
others. Although not assessed until mid adulthood, this variable was used because,
astonishing as it seems in the 21st century, there was no earlier measure of capacity for
attachment.

These four measures of warm relationships all strongly correlated with each other.
More important, these four variables were highly predictive of not only Love in the future
but also of future income and occupational prestige. Out of the 40 possible matches (four
predictors times ten outcomes), 34 were significant and 12 strongly (p< .001). For
example, the 84 men with poor childhood relationships reported an average maximum
earned income of $150,000 a year (in 2009 dollars). The 41 men with the warmest
childhoods earned a maximum of $343,000 a year. The 12 men with the most mature
(empathic) coping style reported a maximum income of $369,000 a year; the 16 men with
the most immature (narcissistic) coping style reported a maximum income of $159,000 a
year. To put these differences in perspective, the difference in average maximum earned
income between men whose parents had been in the upper-upper and in the lower-middle
social class was only $12,000—a statistically insignificant difference. (A Harvard

7
diploma is a great social equalizer.) Still more important, of course, these four early life
variables predicted intimate relations in late life even more powerfully. Indeed, to my
surprise maintaining even physical health was a reflection of having had many warm
relationships in the past and close companions to play with in time present.

Certainly my rash conclusions about the importance of relationships would not have
occurred to any of the originators of the study. Nor did the capacity for warm, intimate
relationships commonly cross the minds of social scientists anywhere else. It is hard to
remember that the pre-World War II Anglo-American social sciences had helped fuel
Nazi eugenics. Many psychiatrists believed that personality was determined by body
build and there was a popular psychiatric diagnosis called ―constitutional inferiority‖. In
1938, when the Study began, constitution and eugenics were commonly seen as more
potent than environment. Physical anthropology was still dominant. Franz Boaz’s cultural
anthropology that would capture the hearts and minds of college students in the 1960s
was still relatively esoteric. The Grant Study investigators were far more interested in
understanding the physiology and somatotypes of the Grant Study men than their intimate
relationships. The Nazi’s had not yet given the study of ―race‖ a bad name.

The initial data collection years of the Study were the early years of World War II.
Thus, it was understandable that besides seeking the antecedents of ―normality‖ and
mental health, efficient officer selection was on the investigators’ minds. The study’s
predictive criteria for ―good officer material‖ included: "vital affect," athletic prowess, a
mesomorphic body build, masculine (as contrasted to feminine) body build, high IQ,
perseverance on a treadmill, and ―friendliness‖ (a variable that further follow-up revealed
that was more correlated with extraversion than capacity for intimacy).

The first book on the Grant Study by Earnest Hooton (a distinguished Harvard
physical anthropologist), Young Man You are Normal, was based in a world of
constitutional medicine. Hooton (1945) wrote, ―When physique, studied from different
stand points, turns out to be so intimately related to various personality traits, it is clear
that body build must also furnish clues to the social capacities of the individual.‖ He
quotes a study of ROTC recruits in which 41% of men with a strong masculine
component were considered ―excellent officer material‖; and not a single man with
feminine body build (narrow shoulders and broad hips) was considered of equal value.

8
Of course, there was no follow-up to document these claims. Instead of testing the claims
of social environmentalists experimentally, to discover if their predictions could also be
correct, Hooton dismissed his opposition as ―crass environmentalists.‖ Thus, Study
physicians and physiologists studied their subjects in minute biological detail. But
capacities for love, intimacy, ―emotional intelligence,‖ and close friendships were largely
ignored. Of interest was that these childhood variables were the only ones in Table 2 that
actually predicted final World War II military rank- in other words, good officer material.
(The range was from private to major.)

Let me offer two further pieces of evidence for my assertion that social science has
only considered the capacity for loving attachment as important over the last 50 years.
First, the not uncommon malady of Infantile Autism was not discovered until 1943; its
close ―relative,‖ Asbergers syndrome, was not identified until 1944. It took fifty years
more until these two disorders were included in standard diagnostic nomenclature. Until
1943 physicians lived in a world where many understood arcane phenomena like quantum
mechanics and the transmission of photographs via television, but where none could
conceptualize a disorder due to a congenital absence of empathy. Yet this genetic
incapacity affects roughly 300,000 Americans with devastating consequences.

My second piece of evidence is that until recently Love from Aristotle to Freud—
was conceptualized as Eros, not as attachment. Love was thought to be due to
hypothalamic instinct, not limbic pair bonding. Not until 1950 did psychoanalyst-
ethologist John Bowlby popularize the concept of attachment—and that the fact that
babies become attached to their mothers because the mothers cuddled them, sang to them
and gazed into their eyes. Later, psychologist-ethologist Harry Harlow had to teach
psychiatrists and psychologists that mother-child attachment did not, as the classical
Freudians maintained, occur just because mothers fed their babies; but pair bonding in
humans evolves and depends on a loving environment in childhood. In his 1958
presidential address to the America Psychological Association Harry Harlow was driven
to exclaim, ―Psychologists not only show no interest in the origin and development of
love and affection, but they seem to be unaware of its very existence.‖

Thus, it is understandable that in 1940, during ten hours of psychiatric interviews

the Study men were queried about their religious upbringing, masturbation, and their

9
views upon Freudian theory, and premarital sex but not about best friends or girlfriends.
In the twenty-first century it is difficult to remember that our understanding of attachment
and love is more recent than our understanding of how to make atomic bombs.

To summarize, love begets love; while adolescent social class, intelligence, body
build and a hail-fellow-well-met approach to life contributed much less to successful
aging. Happiness IS love - full stop.

10