The Need to Belong: Desire for Interpersonal Attachments as a Fundamental Human Motivation

Abstract

A hypothesized need to form and maintain strong, stable interpersonal relationships is evaluated in light of the empirical literature. The need is for frequent, nonaversive interactions within an ongoing relational bond. Consistent with the belongingness hypothesis, people form social attachments readily under most conditions and resist the dissolution of existing bonds. Belongingness appears to have multiple and strong effects on emotional patterns and on cognitive processes. Lack of attachments is linked to a variety of ill effects on health, adjustment, and well-being. Other evidence, such as that concerning satiation, substitution, and behavioral consequences, is likewise consistent with the hypothesized motivation. Several seeming counterexamples turned out not to disconfirm the hypothesis. Existing evidence supports the hypothesis that the need to belong is a powerful, fundamental, and extremely pervasive motivation.

Full text

Psychological
Bulletin
1995,
Vol.
117,
No. 3,
497-529
Copyright
1995
by the
American
Psychological
Association,
Inc.
0033-2909/95/S3.00
The
Need
to
Belong:
Desire
for
Interpersonal Attachments
as a
Fundamental Human Motivation
Roy
F.
Baumeister
Case Western Reserve University
Mark
R.
Leary
Wake
Forest
University
A
hypothesized need
to
form
and
maintain strong, stable interpersonal relationships
is
evaluated
in
light
of the
empirical literature.
The
need
is for
frequent,
nonaversive interactions within
an
ongoing
relational bond. Consistent
with
the
belongingness hypothesis, people
form
social attachments
readily
under most conditions
and
resist
the
dissolution
of
existing
bonds.
Belongingness
appears
to
have
multiple
and
strong
effects
on
emotional patterns
and on
cognitive processes. Lack
of
attach-
ments
is
linked
to a
variety
of ill
effects
on
health, adjustment,
and
well-being. Other evidence, such
as
that concerning satiation, substitution,
and
behavioral consequences,
is
likewise consistent
with
the
hypothesized motivation. Several seeming counterexamples turned
out not to
disconfirm
the
hypothesis.
Existing evidence supports
the
hypothesis that
the
need
to
belong
is a
powerful,
funda-
mental,
and
extremely pervasive motivation.
The
purpose
of
this review
is to
develop
and
evaluate
the hy-
pothesis that
a
need
to
belong
is a
fundamental human motiva-
tion
and to
propose that
the
need
to
belong
can
provide
a
point
of
departure
for
understanding
and
integrating
a
great deal
of
the
existing literature regarding human interpersonal behavior.
More precisely,
the
belongingness hypothesis
is
that human
be-
ings
have
a
pervasive drive
to
form
and
maintain
at
least
a
min-
imum
quantity
of
lasting,
positive,
and
significant
interpersonal
relationships.
Satisfying
this drive involves
two
criteria: First,
there
is a
need
for
frequent,
affectively
pleasant interactions
with
a few
other people, and, second, these interactions must
take
place
in the
context
of a
temporally stable
and
enduring
framework
of
affective
concern
for
each
other's
welfare.
Interac-
tions with
a
constantly changing sequence
of
partners will
be
less
satisfactory than repeated interactions with
the
same
person
(s),
and
relatedness without
frequent
contact
will
also
be
unsatisfactory.
A
lack
of
belongingness should constitute severe
deprivation
and
cause
a
variety
of ill
effects.
Furthermore,
a
great
deal
of
human behavior, emotion,
and
thought
is
caused
by
this
fundamental
interpersonal motive.
The
hypothesis
that
people
are
motivated
to
form
and
maintain
interpersonal
bonds
is not
new,
of
course. John Donne
(1975)
has
been
widely
quoted
for the
line
"No
[person]
is an
island."
In
psy-
chology,
the
need
for
interpersonal contact
was
asserted
in
several
ways
by
Freud (e.g., 1930), although
he
tended
to see the
motive
as
derived
from
the sex
drive
and
from
the filial
bond. Maslow
Roy
F.
Baumeister, Department
of
Psychology,
Case Western Reserve
University;
Mark
R.
Leary, Department
of
Psychology,
Wake
Forest
University.
We
thank
Bob
Hogan,
Ned
Jones, Richard Moreland, Dave Myers,
Len
Newman,
Paula Pietromonaco, Harry Reis,
Dan
Wegner,
and Di-
anne
Tice
for
comments
on
preliminary
drafts.
Correspondence concerning this article should
be
addressed
to Roy
F.
Baumeister,
Department
of
Psychology,
Case
Western Reserve Uni-
versity,
10900 Euclid
Avenue,
Cleveland, Ohio 44106-7123. Electronic
mail
may be
sent
via
Internet
to
rfb2@po.cwru.edu.
(1968)
ranked "love
and
belongingness
needs"
in the
middle
of his
motivational hierarchy; that
is,
belongingness needs
do not
emerge
until
food, hunger,
safety,
and
other basic needs
are
satisfied,
but
they
take precedence over esteem
and
self-actualization.
Bowlby's
(e.g.,
1969,1973)
attachment
theory
also
posited
the
need
to
form
and
maintain relationships.
His
early thinking
followed
the
Freud-
ian
pattern
of
deriving attachment needs
from
the
relationship
to
one's mother;
he
regarded
the
adult's need
for
attachment
as an
effort
to
recapture
the
intimate contact that
the
individual had,
as
an
infant,
with
his or her
mother.1
Horney
(1945),
Sullivan
(1953),
Fromm
(1955,
1956),
de
Rivera(1984),
Hogan
(1983),
Epstein
(1992),
Ryan
(1991),
Guisinger
and
Blatt
(1994),
and
others
have
made similar suggestions.
The
existence
of a
need
to
belong
is
thus
a
familiar point
of
theory
and
speculation, although
not all
theorists
have
anticipated
our
particular formulation
of
this
need
as the
combination
of
frequent
interaction plus persistent
caring. Moreover, most theorists
have
neglected
to
provide system-
atic empirical evaluation
of
this hypothesis.
For
example,
Mas-
low's
(1968)
influential
assertion
of a
belongingness need
was ac-
companied
by
neither original data
nor
review
of
previous
find-
ings.
Thus,
despite
frequent, speculative
assertions
that
people
need
to
belong,
the
belongingness hypothesis needs
to be
critically
evaluated
in
light
of
empirical evidence.
A
main goal
of
the
present
article
is to
assemble
a
large body
of
empirical
findings
pertinent
to the
belongingness hypothesis
to
evaluate
how
well
the
hypothe-
sis
fits the
data.
Another goal
of
this article
is to
demonstrate
the
broad
appli-
cability
of the
need
to
belong
for
understanding human motiva-
tion
and
behavior. Even though many psychological theorists
have
noted human
affiliative
tendencies
in one
form
or
another,
the
field as a
whole
has
neglected
the
broad applicability
of
this
1
His
later thinking may,
however,
have moved beyond this
view
to
regard attachment needs
as
having
a
separate, even innate basis rather
than being derived
from
the
contact
with
one's
mother;
in
this
later
view,
he
treated
the
relationship
to
one's
mother
as
simply
an
influential
prototype
of
attachment.
497

498
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
need
to a
wide range
of
behaviors. Thus,
for
example,
the mo-
tive
literature
has
been dominated
by
research
on the
respective
needs
for
power, achievement, intimacy, approval, and,
to a
lesser extent,
affiliation.
But the
need
for
power
may
well
be
driven
by the
need
to
belong,
as we
suggest later. Likewise, peo-
ple
prefer achievements that
are
validated, recognized,
and
val-
ued
by
other people over solitary achievements,
so
there
may
be a
substantial interpersonal component behind
the
need
for
achievement.
And the
needs
for
approval
and
intimacy
are un-
doubtedly
linked
to the
fact
that approval
is a
prerequisite
for
forming
and
maintaining social bonds,
and
intimacy
is a
denn-
ing
characteristic
of
close
relationships.
The
need
to
belong
could
thus
be
linked
to all of
them.
Furthermore,
even
a
quick glance
at
research
on
social
be-
havior
from
the
perspective
of the
belongingness
hypothesis
raises
the
possibility that much
of
what human beings
do is
done
in
the
service
of
belongingness. Thus,
the
belongingness hypoth-
esis
might have considerable value
for
personality
and
social
psychology
and
even
for
psychology
as a
whole.
As a
broad
in-
tegrative hypothesis,
it
might help
rectify
what some observers
have
criticized
as
fragmentation
and
atomization
in the
concep-
tual
underpinnings
of the field
(see Vallacher
&
Nowak,
1994;
West,
Newsom,
&
Fenaughty,
1992).
At
the
interdisciplinary level,
the
belongingness hypothesis
might help psychology
recover
from
the
challenge
posed
by
cul-
tural
materialism. Cultural materialism (e.g., Harris, 1974,
1978,
1979)
is
based
on the
assumption that human culture
is
shaped primarily
by
economic needs
and
opportunities,
and so
historical, anthropological, sociological,
and
other cultural pat-
terns should
mainly
be
analyzed
with
reference
to
economic
causes.
In
that framework, psychology
is
reduced
to a
vastly
subordinate
role; psychological phenomena
are
regarded
merely
as
symptoms
or
coping mechanisms that
follow
from
economic
realities.
In
contrast,
the
belongingness
hypothesis
would
suggest that human culture
is at
least partly
adapted
to
enable people
to
satisfy
the
psychological need
to
live
together
(along
with economic needs,
to be
sure), thereby assigning
some
fundamental
causal power
to
psychological forces.
We
suggest
that belongingness
can be
almost
as
compelling
a
need
as
food
and
that human culture
is
significantly
conditioned
by
the
pressure
to
provide belongingness.
Modern personality
and
social psychologists have shown
a
pervasive
reluctance
to
entertain sweeping generalizations
and
broad hypotheses. This reluctance
may
well
be a
response
to
speculative excesses
of
earlier generations
of
theorists,
who
sup-
posedly rushed
to
formulate
broad
theories
from
intuition
and
impression. Today there
may be a
sense that
it is
more appro-
priate
to
await
the
passing
of a
substantial interval, until con-
siderable
empirical work
has
been done.
We
propose that such
an
interval
has
passed, however, making
it
possible
to
begin con-
sidering
broad hypotheses
in
light
of the
evidence accumulated
through
the
last three
decades.
That
is
what
we
undertake
here.
Conceptual Background
Fundamental Motivations:
Metatheory
Before
proceeding with
our
examination
of the
need
to be-
long,
we
must consider
briefly
the
metatheoretical
requirements
of
our
hypothesis. That
is,
what criteria must
be
satisfied
to
con-
clude that
the
need
to
belong,
or any
other drive,
is a
fundamen-
tal
human motivation?
We
suggest
the
following.
A
fundamen-
tal
motivation should
(a)
produce
effects
readily under
all but
adverse conditions,
(b)
have
affective
consequences,
(c)
direct
cognitive processing,
(d)
lead
to ill
effects
(such
as on
health
or
adjustment)
when
thwarted,
(e)
elicit goal-oriented behavior
designed
to
satisfy
it
(subject
to
motivational patterns such
as
object
substitutability
and
satiation),
(f) be
universal
in the
sense
of
applying
to all
people,
(g) not be
derivative
of
other
motives,
(h)
affect
a
broad
variety
of
behaviors,
and
(i)
have
implications that
go
beyond immediate psychological
function-
ing.
We
consider each
of
these criteria
in
turn.
The first
criterion
is
that
a
fundamental motivation should
operate
in a
wide variety
of
settings:
any
motive that requires
highly
specific
or
supportive circumstances
to
produce
effects
cannot properly
be
called fundamental. Certain circumstances
may
retard
or
prevent
its
operation,
but in
general
the
more
widely
it can
produce
effects,
the
stronger
its
claim
to
being
a
fundamental
motivation.
The
second
and
third criteria
refer
to
emotional
and
cognitive
patterns. Cognitive
and
emotional responses
reflect
subjective
importance
and
concern,
and a
motivation that
fails
to
guide
emotion
and
cognition
(at
least sometimes)
can
hardly
be
con-
sidered
an
important
one.
In
addition,
most
motivational
and
drive
systems involve hedonic consequences that alert
the
indi-
vidual
to
undesired state changes that motivate behavior
to re-
store
the
desired state
and
whose removal serves
as
negative
re-
inforcement
for
goal attainment.
The
fourth
criterion
is
that
failure
to
satisfy
a
fundamental
motivation should produce
ill
effects
that
go
beyond temporary
affective
distress.
A
motivation
can be
considered
to be
funda-
mental only
if
health, adjustment,
or
well-being requires that
it
be
satisfied. Also,
motivations
can be
sorted
into
wants
and
needs,
the
difference
being
in the
scope
of ill
effects
that
follow
from
nonsatisfaction: Unsatisfied needs should lead
to
pathol-
ogy
(medical,
psychological,
or
behavioral),
unlike
unsatisfied
wants. Thus,
if
belongingness
is a
need rather than simply
a
want,
then people
who
lack belongingness should exhibit patho-
logical
consequences beyond mere temporary distress.
Substitution
and
satiation
are two
familiar hallmarks
of mo-
tivation.
If the
need
to
belong
is a
fundamental need, then
be-
longing
to one
group should satisfy
it and
hence obviate
or re-
duce
the
need
to
belong
to
another group. People
may be
driven
to
form
social
bonds
until they have
a
certain number, whereaf-
ter the
drive
to
form
attachments would presumably subside.
Furthermore, attachment partners should
be to
some degree
in-
terchangeable.
Of
course, this does
not
mean that
a
20-year
spouse
or
friend
can be
simply replaced with
a new
acquain-
tance.
In the
long run, however,
a new
spouse
or
friend
should
do as
well
as the
previous one.
The
sixth
and
seventh criteria involve universality
and
non-
derivativeness.
Any
motivation
that
is
limited
to
certain human
beings
or
certain circumstances,
or any
motivation that
is de-
rived
from
another motive, cannot
be
regarded
as
fundamental.
Universality
can be
indicated
by
transcending cultural bound-
aries. Establishing that
a
motive
is not
derivative
is not
easy,
although path-analytic models
can
suggest derivative
patterns.
Satisfying
the first
criterion
may
also help
satisfy
the
seventh,

THE
NEED
TO
BELONG
499
because
if the
motivation operates
in a
broad variety
of
situa-
tions without requiring particular, favorable circumstances,
then
it may be
presumed
to be
fundamental. Meanwhile,
if the
evidence contradicts evolutionary patterns
or
fails
to
indicate
physiological
mechanisms, then
the
hypothesis
of
universality
or
innateness would lose credibility.
The
eighth criterion
is the
ability
to
affect
a
wide
and
diverse
assortment
of
behaviors.
The
more behaviors that appear
to be
influenced
by a
particular motive,
the
stronger
its
case
for
being
one of the
fundamental motives. Lastly,
we
suggest
that
a
fun-
damental motive should have implications
that
go
beyond psy-
chological
functioning.
If a
motivation
is
truly fundamental,
it
should
influence
a
broad range
of
human activity,
and
hence
it
should
be
capable
of
offering
viable
and
consistent inter-
pretations
of
patterns observed
in
historical, economic,
or so-
ciological
studies.
Falsification
is
only
one
relevant approach
to
evaluating
a
broad hypothesis about
belongingness
being
a
fundamental
mo-
tivation.
The
belongingness hypothesis could indeed
be
falsified
if
it
were shown,
for
example, that many people
can
live
happy,
healthy
lives
in
social isolation
or
that many people show
no
cognitive
or
emotional responses
to
looming
significant
changes
in
their belongingness
status.
In
addition
to
such criteria, how-
ever,
hypotheses about
fundamental
motivations must
be
evalu-
ated
in
terms
of
their capacity
to
interpret
and
explain
a
wide
range
of
phenomena. Part
of the
value
of
such
a
theory
is its
capacity
to
provide
an
integrative
framework,
and
this value
is
a
direct
function
of the
quantity
and
importance
of the
behavior
patterns that
it can
explain
in a
consistent, intelligible
fashion.
We
therefore
pay
close attention
to the
potential range
of im-
plications
of the
belongingness hypothesis,
in
addition
to
exam-
ining
how
many
falsification
tests
the
hypothesis
has
managed
to
survive.
The
Need
to
Belong:
Theory
In
view
of the
metatheoretical requirements listed
in the
pre-
vious
section,
we
propose
that
a
need
to
belong, that
is, a
need
to
form
and
maintain
at
least
a
minimum quantity
of
interper-
sonal
relationships,
is
innately prepared (and hence nearly
universal)
among human beings. Thus, unlike
the
Freudian
(1930)
view
that regarded sexuality
and
aggression
as the
major
driving
psychological forces,
and
unlike
the
most ambitious
be-
haviorist
views
that considered each newborn
a
tabula
rasa,
our
view
depicts
the
human being
as
naturally driven toward
estab-
lishing
and
sustaining
belongingness.
The
need
to
belong should
therefore
be
found
to
some degree
in all
humans
in all
cultures,
although
naturally
one
would expect there
to be
individual
differences
in
strength
and
intensity,
as
well
as
cultural
and in-
dividual
variations
in how
people express
and
satisfy
the
need.
But
it
should prove
difficult
or
impossible
for
culture
to
eradi-
cate
the
need
to
belong
(except
perhaps
for an
occasional, seri-
ously
warped
individual).
The
innate quality presumably
has an
evolutionary basis.
It
seems clear that
a
desire
to
form
and
maintain social bonds
would
have both survival
and
reproductive
benefits
(see
Ains-
worth,
1989; Axelrod
&
Hamilton, 1981; Barash, 1977;
Bowlby,
1969;
D. M.
Buss, 1990,
1991;
Hogan, Jones,
&
Cheek,
1985; Moreland,
1987).
Groups
can
share food, pro-
vide
mates,
and
help care
for
offspring
(including
orphans).
Some survival tasks, such
as
hunting large animals
or
main-
taining
defensive vigilance against predatory enemies,
are
best
accomplished
by
group cooperation. Children
who de-
sired
to
stay together with adults
(and
who
would resist being
left
alone)
would
be
more
likely
to
survive until their
repro-
ductive years than other children because they would
be
more
likely
to
receive care
and
food
as
well
as
protection. Cues
that
connote possible harm, such
as
illness, danger, nightfall,
and
disaster, seem
to
increase
the
need
to be
with others
(see
also
Rofe,
1984),
which again underscores
the
protective value
of
group membership. Adults
who
formed attachments would
be
more likely
to
reproduce than those
who
failed
to
form
them,
and
long-term relationships would increase
the
chances that
the
offspring
would reach maturity
and
repro-
duce
in
turn (see also Shaver, Hazan,
&
Bradshaw,
1988).2
Competition
for
limited resources could also provide
a
pow-
erful
stimulus
to
forming
interpersonal connections. There
are
several
potential, although
debatable,
advantages
to
forming
a
group under conditions
of
scarcity.
For
example, groups
may
share resources
and
thus prevent
any
individual
from
starving
(although
sharing deprives other group members
of
some
of
their
resources),
and
groups
may
appropriate resources
from
nonmembers
(although
there
is the
problem
of how to
distrib-
ute
them
in the
group).
What appears less debatable
is the se-
vere
competitive disadvantage
of the
lone individual confront-
ing
a
group when both want
the
same resource. When other
people
are in
groups,
it is
vital
to
belong
to a
group
oneself,
particularly
a
group
of
familiar,
cooperative people
who
care
about
one's
welfare.
Thus,
an
inclination
to
form
and
sustain
social bonds would have important benefits
of
defending
oneself
and
protecting
one's
resources against external threats.
The
likely
result
of
this evolutionary selection would
be a set
of
internal mechanisms that guide individual human beings
into social groups
and
lasting relationships. These mechanisms
would
presumably include
a
tendency
to
orient toward other
members
of the
species,
a
tendency
to
experience
affective
dis-
tress when deprived
of
social contact
or
relationships,
and a
ten-
dency
to
feel
pleasure
or
positive
affect
from
social contact
and
relatedness. These
affective
mechanisms would stimulate learn-
ing
by
making positive social contact
reinforcing
and
social
de-
privation punishing.
Our
version
of the
belongingness hypothesis does
not
regard
the
need
as
derived
from
a
particular relationship
or
focused
on
a
particular individual.
In
this,
it
differs
from
the
early, Freudian
version
of
Bowlby's work,
in
which
the
relationship
to the
mother
was
regarded
as the
cause
of the
desire
for
attachment.
Thus,
Bowlby
suggested that adult attachments
to
work organi-
zations, religious groups,
or
others
are
derived
from
the
child's
tie
to
mother
and
revolve around personal attachment
to the
group leader
or
supervisor
(Bowlby,
1969,
p.
207).
In
contrast,
2
A
possible
sex
difference
could
be
suggested
in the
mode
of
express-
ing
this need, however,
in
that
men may be
more oriented toward
form-
ing
relationships, whereas women
may be
more oriented toward main-
taining
them.
Men can
reproduce
many
times
by
forming
many brief
relationships,
whereas
women
can
reproduce
only
about once
per
year,
and so
their most
effective
reproductive strategy would
be to
enable each
child
to
receive maximal care
and
protection
(D.
M.
Buss,
1991).

500
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
we
propose that
the
need
to
belong can,
in
principle,
be
directed
toward
any
other human being,
and the
loss
of
relationship with
one
person
can to
some extent
be
replaced
by any
other.
The
main
obstacle
to
such substitution
is
that formation
of new re-
lationships
takes time, such
as in the
gradual accumulation
of
intimacy
and
shared experience
(see
Sternberg, 1986,
on the
time
course
of
intimacy).
Social contact
with
a
long-term inti-
mate
would therefore provide
some
satisfactions, including
a
sense
of
belonging, that would
not be
available
in
interactions
with
strangers
or new
acquaintances.
The
belongingness
hypothesis
can be
distinguished
from
a hy-
pothesized need
for
mere social contact
in
terms
of
whether
in-
teractions
with
strangers
or
with
people
one
dislikes
or
hates
would
satisfy
the
need.
It can be
distinguished
from
a
hypothe-
sized
need
for
positive, pleasant social contact
in
terms
of
whether
nonhostile interactions with strangers would
satisfy
it.
The
need
to
belong entails that relationships
are
desired,
so in-
teractions
with
strangers would mainly
be
appealing
as
possible
first
steps
toward
long-term
contacts
(including practicing
so-
cial
skills
or
learning about one's capacity
to
attract
partners),
and
interactions with disliked people would
not
satisfy
it.
Additional
differences
between
the
belongingness hypothesis
and
attachment theory could
be
suggested, although
it may be a
matter
of
interpretation whether these
are
merely
differences
of
emphasis
or
fundamental theoretical
differences.
In our
under-
standing,
the
(very
real)
strengths
of
attachment theory
are
two-
fold.
First, attachment theory
has
emphasized
the
task
of
elab-
orating
individual
differences
in
attachment style (e.g., Hazan
&Shaver,
1994a, 1994b;
Shaver
etal.,
1988),
whereas
we
focus
on
the
commonality
of the
overarching
need
to
belong.
Second,
attachment
theory
has
emphasized certain emotional needs
and
satisfactions
implicit
in
certain
kinds
of
relationships, whereas
we
regard
it as at
least plausible that
the
need
to
belong could
be
satisfied
in
other ways.
For
example,
one
might imagine
a
young
fellow
without
any
family
or
intimate relationships
who
is
nonetheless
satisfied
by
being heavily involved
in an
ideologi-
cally
radical political movement. There
are
undoubtedly strong
emotional
mechanisms associated with belongingness,
as we
show
later,
but
these could
be
understood
as
mediating mecha-
nisms
rather than
as
essential properties.
Asa
fundamental
motivation,
the
need
to
belong
should
stim-
ulate
goal-directed activity designed
to
satisfy
it.
People should
show
tendencies
to
seek
out
interpersonal contacts
and
cultivate
possible
relationships,
at
least until they have reached
a
mini-
mum
level
of
social contact
and
relatedness. Meanwhile, social
bonds should
form
easily, readily,
and
without
requiring
highly
particular
or
conducive settings.
(Indeed,
if
social attachments
form
through shared unpleasant experiences, contrary
to
what
simple
association models might predict, this would
be
espe-
cially
compelling support
for the
belongingness hypothesis.)
Cognitive
activity should
reflect
a
pervasive concern with form-
ing
and
maintaining
relationships.
Emotional
reactions
should
follow
directly
from
outcomes that pertain
to the
need
to be-
long.
More precisely, positive
affect
should
follow
from forming
and
solidifying
social bonds,
and
negative
affect
should ensue
when
relationships
are
broken, threatened,
or
refused.
If
belongingness
is
indeed
a
fundamental need, then aver-
sive
reactions
to a
loss
of
belongingnsss
should
go
beyond
negative
affect
to
include some types
of
pathology.
People
who
are
socially deprived should exhibit
a
variety
of ill
effects,
such
as
signs
of
maladjustment
or
stress, behavioral
or
psychological pathology,
and
possibly health problems. They
should
also show
an
increase
in
goal-directed activity aimed
at
forming
relationships.
In
addition,
the
belongingness hypothesis entails that people
should
strive
to
achieve
a
certain minimum quantity
and
qual-
ity
of
social
contacts
but
that
once
this
level
is
surpassed,
the
motivation should diminish.
The
need
is
presumably
for a
cer-
tain minimum number
of
bonds
and
quantity
of
interaction.
The
formation
of
further
social attachments beyond that mini-
mal
level
should
be
subject
to
diminishing returns; that
is,
peo-
ple
should experience less satisfaction
on
formation
of
such
ex-
tra
relationships,
as
well
as
less distress
on
terminating them.
Satiation patterns should
be
evident, such that people
who are
well
enmeshed
in
social relationships would
be
less inclined
to
seek
and
form
additional bonds than would people
who are so-
cially
deprived. Relationships should substitute
for
each other,
to
some extent,
as
would
be
indicated
by
effective
replacement
of
lost relationship partners
and by a
capacity
for
social related-
ness
in one
sphere
to
overcome potential
ill
effects
of
social
de-
privation
in
another sphere (e.g.,
if
strong
family
ties compen-
sate
for
aloneness
at
work).
We
propose that
the
need
to
belong
has two
main features.
First, people need frequent personal contacts
or
interactions
with
the
other person.
Ideally,
these interactions would
be
affec-
tively
positive
or
pleasant,
but it is
mainly important that
the
majority
be
free
from
conflict
and
negative
affect.
Second, people need
to
perceive that there
is an
interpersonal
bond
or
relationship
marked
by
stability,
affective
concern,
and
continuation into
the
foreseeable future. This
aspect
provides
a
relational context
to
one's
interactions with
the
other person,
and so the
perception
of the
bond
is
essential
for
satisfying
the
need
to
belong. When compared with essentially identical
in-
teractions with other people with whom
one is not
connected,
a
strictly
behavioral record might reveal nothing special
or re-
warding
about these interactions.
Yet an
interaction with
a
per-
son
in the
context
of an
ongoing relationship
is
subjectively
different
from
and
often
more rewarding than
an
interaction
with
a
stranger
or
casual acquaintance.
To
satisfy
the
need
to
belong,
the
person
must believe
that
the
other
cares
about
his or
her
welfare
and
likes
(or
loves)
him or
her.
Ideally
this concern would
be
mutual,
so
that
the
person
has
reciprocal
feelings
about
the
other.
M. S.
Clark
and her
col-
leagues (e.g., Clark,
1984;
Clark
&
Mills,
1979;
Clark, Mills,
&
Corcoran,
1989;
Clark, Mills,
&
Powell,
1986)
have shown that
a
framework
of
mutual concern
produces
a
relationship quali-
tatively
different
from
one
based
on
self-interested social
ex-
change.
Still,
it is
plausible that mutuality
is
merely desirable
rather than essential.
The
decisive
aspect
may be the
perception
that
one is the
recipient
of the
other's
lasting
concern.
Viewed
in
this
way,
the
need
to
belong
is
something
other
than
a
need
for
mere
affiliation.
Frequent
contacts
with nonsup-
portive,
indifferent
others
can go
only
so far in
promoting
one's
general well-being
and
would
do
little
to
satisfy
the
need
to be-
long.
Conversely, relationships characterized
by
strong feelings
of
attachment, intimacy,
or
commitment
but
lacking regular
contact will
also
fail
to
satisfy
the
need. Simply knowing
that
a
bond exists
may be
emotionally reassuring,
yet it
would
not

THE
NEED
TO
BELONG
501
provide
full
belongingness
if one
does
not
interact with
the
other
person. Thus,
we
view
the
need
to
belong
as
something more
than either
a
need
for
affiliation
or a
need
for
intimate
attachment.
The
notion that people need relationships characterized
by
both regular contact
and an
ongoing bond
has
been anti-
cipated
to
some degree
by
Weiss
(1973;
see
also Shaver
&
Buhrmester,
1983),
who
suggested that
feelings
of
loneliness
can
be
precipitated either
by an
insufficient
amount
of
social
contact (social loneliness)
or by a
lack
of
meaningful,
inti-
mate relatedness
(emotional
loneliness).
Weiss's
distinction
has
been criticized
on
conceptual
and
empirical grounds
(e.g.,
Paloutzian
&
Janigian, 1987; Perlman, 1987),
and
efforts
to
operationalize
and
test
the
distinction have
met
with
mixed
results (DiTommaso
&
Spinner, 1993; Saklofske
&
Yackulic,
1989;
Vaux,
1988).
In our
view,
the
difficulty
with
this
distinction arises
from
the
assumption that people have
a
need
for
mere social contact
and a
separate need
for
intimate
relationships.
Rather,
the
need
is for
regular social
contact
with
those
to
whom
one
feels
connected. From
an
evolution-
ary
perspective, relationships characterized
by
both
of
these
features
would have greater survival
and
reproductive value
than
would
relationships characterized
by
only one. Accord-
ingly,
the
need
to
belong should
be
marked
by
both
aspects.
Review
of
Empirical
Findings
We
searched
the
empirical literature
of
social
and
personality
psychology
for
findings
relevant
to the
belongingness hypothe-
sis.
The
following
sections summarize
the
evidence
we
found
pertaining
to the
series
of
predictions
about
belongingness.
Forming
Social
Bonds
A
first
prediction
of the
belongingness hypothesis
is
that
so-
cial
bonds should
form
relatively easily, without requiring spe-
cially
conducive circumstances. Such evidence
not
only would
attest
to the
presence
and
power
of the
need
to
belong
but
would
suggest
that
the
need
is not a
derivative
of
other needs
(insofar
as it is not
limited
to
circumstances that meet other require-
ments
or
follow
from
other
events).
There
is
abundant evidence that social bonds
form
easily.
In-
deed, people
in
every society
on
earth belong
to
small primary
groups
that
involve
face-to-face,
personal interactions
(Mann,
1980).
The
anthropologist Coon
(1946)
asserted
that natural
groups
are
characteristic
of all
human beings. Societies
differ
in
the
type, number,
and
permanence
of the
groups that people
join,
but
people
of all
cultures quite naturally
form
groups.
The
classic Robbers Cave study conducted
by
Sherif,
Harvey,
White,
Hood,
and
Sherif
(1961/1988)
showed
that
when pre-
viously
unacquainted boys were randomly assigned
to
newly
created groups, strong loyalty
and
group identification ties
en-
sued
rapidly.
In
fact,
later
in
that study,
the two
strongly opposed
groups were recombined into
a
single group with cooperative
goals,
and
emotional
and
behavioral patterns quickly accom-
modated
to the new
group (although
the
prior antagonistic
identifications
did
hamper
the
process).
The
tendency
for
laboratory
or
experimentally created
groups
to
quickly
become cohesive
has
also been noted
in the
minimal
intergroup
situation (Brewer,
1979).
Tajfel
and his
colleagues
(Billig&Tajfel,
1973;
Tajfel,
1970;
Tajfel
&Billig,
1974;
Tajfel,
Flament,
Billig,
&
Bundy,
1971)
showed
that
assigning
participants
to
categories
on a
seemingly arbitrary
basis
was
sufficient
to
cause them
to
allocate
greater rewards
to
in-group members than
to
out-group members. Indeed,
the
original goal
of
Tajfel
et
al.
(1971)
was not to
study group
formation
but to
understand
the
causes
of
in-group favorit-
ism.
To do
this, they sought
to set up an
experimental group
that would
be so
trivial that
no
favoritism would
be
found,
intending then
to add
other
variables progressively
so as to
determine
at
what point favoritism would
start.
To
their sur-
prise, however, in-group favoritism
appeared
at
once, even
in
the
minimal
and
supposedly trivial situation
(see
also Turner,
1985).
This preferential treatment
of
in-group members
does
not
appear
to be due to
inferred
self-interest
or to
issues
of
novelty
and
uncertainty about
the
task
(Brewer
&
Silver, 1978;
Tajfel,
1970;
Tajfel
&
Billig,
1974).
Inferred
similarity
of
self
to in-
group members
was a
viable explanation
for
many
of the
early
findings, but
Locksley,
Ortiz,
and
Hepburn
(1980)
ruled this
out by
showing that people show in-group
favoritism
even when
they
have
been assigned
to
groups
by a
random lottery. Thus,
patterns
of
in-group
favoritism,
such
as
sharing rewards
and
categorizing others relative
to the
group, appeared quite readily,
even
in the
absence
of
experiences designed
to
bond people
to
the
group emotionally
or
materially.
Several
other studies suggest
how
little
it
takes
(other
than
frequent
contact)
to
create social attachments.
Bowlby
(1969)
noted
that
infants
form
attachments
to
caregivers
very
early
in
life,
long before
babies
are
able
to
calculate
benefits
or
even
speak. Festinger, Schachter,
and
Back
(1950)
found
that mere
proximity
was a
potent factor
in
relationship formation; people
seemed
to
develop social bonds with each other simply because
they
lived near each other. Nahemow
and
Lawton
(1975)
repli-
cated
those
findings and
also showed that pairs
of
best
friends
who
differed
by age or
race were particularly
likely
to
have lived
very
close together, suggesting that extreme proximity
may
over-
come tendencies
to
bond with similar others. Wilder
and
Thompson
(1980)
showed that people seem
to
form
favorable
views
toward whomever they spend time
with,
even
if
these oth-
ers are
members
of a
previously disliked
or
stereotyped out-
group.
In
their study, intergroup biases decreased
as
contact
with
members
of the
out-groups increased (and
as
in-group
contact
decreased).
We
noted that
the
formation
of
social attachments under
ad-
verse
circumstances would
be
especially compelling evidence
because
it
avoids
the
alternative explanations based
on
classical
conditioning
(i.e.,
that
positive associations breed attraction).
Latane,
Eckman,
and Joy
(1966)
found
that
participants
who
experienced electric shock together tended
to
like each other
more than control participants
who did not
experience shock,
although
the
effect
was
significant
only among
firstborns.
Ken-
rick
and
Johnson
(1979)
found
that participants rated each
other more positively
in the
presence
of
aversive
than nonaver-
sive
noise. Elder
and
Clipp
(1988)
compared
the
persistence
of
attachments among military veterans
and
found
that
the
great-
est
persistence
occurred among groups that
had
undergone
heavy
combat resulting
in the
deaths
of
some
friends
and
com-

502
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
rades. Although
it
would
be
rash
to
suggest that
all
shared neg-
ative
experiences increase
attraction,
it
does
appear
that posi-
tive
bonding
will
occur even under adverse circumstances.
The
development
of
interpersonal attraction under
fearful
circumstances
has
been explained
in
terms
of
both
misattribution
(i.e., people
may
misinterpret their anxious
arousal
as
attraction
to
another person)
and
reinforcement
theory
(i.e.,
when
the
presence
of
some other person reduces
one's distress,
a
positive emotional response becomes associ-
ated with that person; Kenrick
&
Cialdini,
1977).
The
mis-
attribution
explanation
is
largely irrelevant
to the
belong-
ingness
hypothesis,
but the
reinforcement explanation
is
ger-
mane.
Specifically,
although others
may
reduce
one's
distress
through
various routes (such
as
distraction, humor,
or
reassurance), evidence suggests strongly that
the
mere pres-
ence
of
other people
can be
comforting (Schachter,
1959).
Such
effects
may
well
be
conditioned through years
of
experi-
ence with supportive others,
but
they also
may
indicate that
threatening
events stimulate
the
need
to
belong.
The
fact
that people sometimes
form
attachments with for-
mer
rivals
or
opponents
is
itself
a
meaningful
indicator
of a
gen-
eral
inclination
to
form
bonds. Cognitive consistency pressures
and
affective
memories would militate against forming positive
social bonds
with
people
who
have been rivals
or
opponents.
Yet,
as we
have already noted,
the
Robbers Cave study (Sherif
et
al.,
1961
/1988)
showed that people could join
and
work
to-
gether
with
others
who had
been bitterly opposed
very
recently,
and
Wilder
and
Thompson
(1980)
showed that social contact
could
overcome established intergroup prejudices
and
stereo-
types. Orbell,
van de
Kragt,
and
Dawes
(1988)
likewise showed
that impulses toward forming positive attachments
could
over-
come oppositional patterns.
In
their study using
the
prisoner's
dilemma
game,
having
a
discussion period
led to
decreased
competition
and
increased cooperation,
as a
result
of
either
the
formation
of a
group identity that joined
the
potential rivals
together
or
explicit agreements
to
cooperate. Thus, belong-
ingness
motivations appear
to be
able
to
overcome some antag-
onistic,
competitive,
or
divisive tendencies.
Similar
shifts
have been suggested
by M. S.
Clark
(1984,
1986;
Clark,
Mills,
&
Powell,
1986;
Clark, Ouellette, Powell,
&
Milberg,
1987),
who
showed that people move toward
a
com-
munal
orientation when there
is a
chance
to
form
a
relation-
ship.
When
participants
were confronted with
a
person
who
seemingly
would
not be
amenable
to
relationship formation
(i.e.,
because
she was
already married), they interacted
with
her
on the
basis
of
norms
of
equitable exchange
and
individual-
ity;
when they believed
she
would
be a
possible relationship
partner,
however, they interacted
with
her on a
communal basis
(i.e.,
mutuality
and
sharing, without respect
to
individual
eq-
uity
concerns).
Critical
assessment.
The
remarkable ease with which social
bonds
form
has
been shown with experimental methods
and
confirmed
by
other methods.
The
main limitation would
be
that
people
do not
always form
relationships
with
all
available
or
proximal others, which could mean that satiation
processes
limit
the
number
of
relationships people seek
and
which
also
indicates that other factors
and
processes
affect
the
formation
of
relationships. Some
patterns
(e.g.,
in-group
favoritism
in
minimal
groups)
have been well replicated
with
careful
efforts
to
rule
out
alternative explanations.
Conclusion.
In
brief, people seem
widely
and
strongly
in-
clined
to
form
social relationships quite easily
in the
absence
of
any
special
set of
eliciting circumstances
or
ulterior motives.
Friendships
and
group allegiance seem
to
arise spontaneously
and
readily, without needing evidence
of
material advantage
or
inferred
similarity.
Not
only
do
relationships emerge quite nat-
urally,
but
people invest
a
great deal
of
time
and
effort
in
foster-
ing
supportive relationships with others. External threat seems
to
increase
the
tendency
to
form strong
bonds.
Not
Breaking Bonds
The
belongingness
hypothesis predicts that people should
generally
be at
least
as
reluctant
to
break social bonds
as
they
are
eager
to
form
them
in the first
place.
A
variety
of
patterns
supports
the
view that people
try to
preserve relationships
and
avoid
ending them.
In
fact,
Hazan
and
Shaver
(1994a,
p.
14)
recently
concluded that
the
tendency
for
human beings
to re-
spond
with
distress
and
protest
to the end of a
relationship
is
nearly universal, even
across
different cultures
and
across
the
age
span.
Some relationships
are
limited
in
time
by
external factors,
and so
these
are
logically
the first
place
to
look
for
evidence that
people show distress
and
resistance
to
breaking bonds.
En-
counter groups
and
training groups,
for
example,
are
often
con-
vened
with
the
explicit understanding that
the
meetings
will
stop
at
a
certain point
in the
future.
Even
so, it is a
familiar observa-
tion
in the
empirical literature (e.g., Egan, 1970;
Lacoursiere,
1980;
Lieberman,
Yalom,
&
Miles,
1973)
that
the
members
of
such
groups resist
the
notion that
the
group
will
dissolve.
Even
though
the
group's
purpose
may
have been
fulfilled,
the
partici-
pants want
to
hold
on to the
social
bonds
and
relationships they
have
formed
with
each other. They promise individually
and
sometimes collectively
to
stay
in
touch with each other, they
plan
for
reunions,
and
they take other steps
to
ensure
a
continu-
ity
of
future
contacts.
In
actuality, only
a
small minority
of
these
envisioned reunions
or
contacts take place,
and so the
wide-
spread exercise
of
making them
can be
regarded
as a
symptom
of
resistance
to the
threatened dissolution
(Lacoursiere,
1980,
p.216).
Other relationships
are
limited
in
time
by
external transitions
such
as
graduating
from
college, moving
to a
different city,
or
getting
a new
job.
As
such transitions approach, people com-
monly
get
together
formally
and
informally
and
promise
to re-
main
in
contact,
to
share meals
or
other social occasions
to-
gether,
to
write
and
call each other,
and to
continue
the
relation-
ship
in
other ways. They also
cry or
show other signs
of
distress
over
the
impending separation
(Bridges,
1980).
These
patterns
seem
to
occur even
if the
dissolving relationship
(e.g.,
with
neighbors)
had no
important
practical
or
instrumental
func-
tion
and
there
is no
realistic likelihood
of
further
contact.
More generally, many social institutions
and
behavior pat-
terns
seem
to
serve
a
need
to
preserve
at
least
the
appearance
of
social
attachment
in the
absence
of
actual,
continued interac-
tion. Reunions constitute
an
occasion
for
people
to see
former
acquaintances.
The
massive exchange
of
greeting
cards
during
the
Christmas holiday season includes many
cases
in
which
the

THE
NEED
TO
BELONG
503
card
is the
sole
contact
that
two
people
have
had
during
the
entire
year,
but
people still resist dropping each
other's
name
from the
mailing list because
to do so
signifies
a final
dissolution
of
the
social bond.
In
fact,
most people will send Christmas
cards
to
perfect strangers
from
whom they receive cards
(Kunz
&
Woolcott,
1976).
People seem
not to
want
to
risk damaging
a
relationship even
if
they
do not
know
the
identity
of the
other
person!
Likewise,
social rituals
involving
greetings
and
farewells
serve
to
assure others
of the
continuation
of
one's
relationships with
them. Many greetings, particularly those directed
at
family
members
and
close
friends,
seem designed
to
indicate that
one's
relationship
has
remained intact since
the
last contact,
and
fare-
wells
often
include some hint that
the
relationship
will
be
main-
tained until
the
people
see one
another again
(Goffman,
1971).
The
importance
of
such rituals
in the
maintenance
of
belong-
ingness
is
reflected
in the
distress people sometimes experience
when
they
feel
that
another's
greeting
is
inadequately warm
or
that
the
other's
farewell
expresses
insufficient
concern about
the
impending
separation.
In
many cases, people seem reluctant
to
dissolve even
bad or
destructive
relationships.
The
apparent unwillingness
of
many
women
to
leave abusive, battering spouses
or
boyfriends
(Roy,
1977;
Strube,
1988)
has
prompted several generations
of
spec-
ulative
explanations,
ranging
from
masochistic
or
self-destruc-
tive
liking
for
abuse
to
calculations
of
economic self-interest
that supposedly override considerations
of
physical harm.
The
belongingness
hypothesis
offers
yet one
more potential perspec-
tive:
The
unwillingness
to
leave
an
abusive intimate partner
is
another manifestation
of the
strength
of the
need
to
belong
and
of
the
resulting reluctance
to
break social bonds.
The
fact
that
people resist breaking
off
an
attachment that causes pain attests
to how
deeply
rooted
and
powerful
the
need
to
belong
is.
Moreover,
when people
do
decide
to
break
off an
intimate
relationship,
they typically experience considerable distress
over
the
dissolution
(which
we
cover
in
more detail
in the
later
section
on
emotion).
This
is
ironic: Although goal attainment
is
usually
marked
by
positive
affect
such
as
satisfaction
and
joy,
attaining
the
goal
of
getting
a
divorce
is
generally accompanied
by
negative
affect.
To be
sure,
in
some cases
the
distress over
divorce
is
accompanied
by an
admixture
of
positive
affect,
but
the
negative
affect
nonetheless indicates
the
resistance
to
break-
ing
the
bond.
It
is
also relevant
and
noteworthy that
the
social bond
often
continues
despite
the
divorce.
In her
study
on
divorce,
Vaughan
(1986)
concluded that
"in
most cases [marital] relationships
don't end. They change,
but
they
don't
end"
(p.
282). Weiss
(1979)
also
found
that some
form
of
(often
ambivalent)
attach-
ment
persists
after
divorce.
The
persistence
of
intimate relation-
ships
past
the
occasion
of
mutually
agreed
and
formally
institu-
tionalized
dissolution
may be yet
another indication
of
people's
reluctance
to
break social bonds.
Critical
assessment. Because ethical
and
practical con-
straints prevent laboratory experimentation
on the
ending
of
significant
relationships,
the
evidence
in
this section
was
drawn
from
observational studies
and
other methods,
and so the hy-
pothesis
of
resistance
to
relationship dissolution
is not as
con-
clusively
supported
as
might
be
desired. Alternative explana-
tions
exist
for
some
of the findings. For
example,
the
persistence
of
relatedness
after
divorce
is
partly
due to
ongoing
practical
concerns, such
as
joint responsibility
for
child care; although
Vaughan
(1986)
was
emphatic
in
asserting
that
such pragmatic
concerns
fall
far
short
of
explaining
the
extent
of
continuing
attachments,
she was
vague about
the
evidence
to
back
up her
assertion. Also,
as we
noted,
the
tendency
for
battered
women
to
return
to
their abusive partners
has
been explained
in
many
ways,
and the
hypothesized reluctance
to
break
off
a
relation-
ship
is
only
one of
them.
On the
positive side,
however,
the
persistence
of
such bonds
has
been observed
by a
variety
of
researchers.
The
fact
that
these researchers
are
from
different
disciplines suggests that
these
conclusions
do not
stem
from
a
single methodological
or
theoretical bias. More systematic research
on
possible bound-
ary
and
limiting conditions
of the
resistance
to
dissolve bonds
would
be
desirable.
Conclusion.
Despite some methodological weaknesses
and
ambiguities,
the
weight
of the
evidence does
favor
the
conclu-
sion
that people strongly
and
generally resist
the
dissolution
of
relationships
and
social bonds. Moreover, this resistance
ap-
pears
to go
well
beyond rational considerations
of
practical
or
material advantage.
Cognition
Intelligent
thought
is
generally recognized
as the
most impor-
tant adaptive trait among human beings,
and so it
seems rea-
sonable
to
assume that issues
of
fundamental
concern
and im-
portance
are
likely
to be the
focus
of
cognitive activity.
The be-
longingness
hypothesis therefore would predict
that
people
will
devote considerable cognitive processing
to
interpersonal
in-
teractions
and
relationships.
Basic
patterns
of
thought appear
to
reflect
a
fundamental
concern
with
social
relationships.
Sedikides,
Olsen,
and
Reis
(1993)
showed that relationships
are
natural categories;
that
is,
people spontaneously
classify
incoming information
in
terms
of
social relationships. Participants stored information about
relationship partners together,
and
they
did
this more
for
strong,
close relationships (marriage) than
for
weak
or
distant ones
(e.g.,
acquaintanceship).
Pryor
and
Ostrom
(1981)
showed
that people
use the
individual person
as a
cognitive unit
of
anal-
ysis
for
familiar
people more than
for
unfamiliar
people. These
researchers began
by
questioning
the
basic assumption that
the
person
is the
fundamental unit
of
social perception. That
is,
information
is not
necessarily
or
inherently processed
and
stored
in
memory
on a
person-by-person
basis,
but it is, in
fact,
processed
and
stored
on
such
a
basis when
it
pertains
to
signifi-
cant others. Ostrom, Carpenter, Sedikides,
and Li
(1993)
pro-
vided
evidence that information about out-group members
tends
to be
stored
and
organized
on the
basis
of
attribute cate-
gories
(such
as
traits, preferences,
and
duties),
whereas
in-
group
information
is
processed
on the
basis
of
person catego-
ries. Thus, social bonds create
a
pattern
in
cognitive processing
that gives priority
to
organizing information
on the
basis
of the
person with whom
one has
some
sort
of
connection.
Several
studies have pursued
the
notion that people process
information
about close relationship partners
differently
from
the way
they
process
information about strangers
or
dis-
tant acquaintances.
For
example, research
has
shown that,

504
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
when
a
group
of
people take turns reading words aloud, they
each have high
recall
for the
words they personally speak
but
have
poor recall
for the
words preceding
and
following
their
performance. Brenner
(1976)
found that this next-in-line
effect
occurs
not
only
for
one's
own
performance
but
also
for
words spoken
by
one's
dating partner (and
the
words imme-
diately
preceding
and
following).
In
a
series
of
studies, Aron, Aron, Tudor,
and
Nelson
(1991)
showed
that close relationship partners, unlike strangers, have
cognitive
effects
similar
to
those
of the
self.
Thus, when people
form
an
image
of
themselves
or
their mothers interacting with
some object, they have more
difficulty
recalling that object than
if
they imagined
a
famous
but
personally unacquainted person
interacting
with that same object.
In
another study, participants
had
more
difficulty
in
making
me-not
me
judgments about
traits
on
which they
differed
from
their spouse than
in
making
judgments about traits
on
which they resembled
the
spouse.
These results suggest that cognitive processes tend
to
blur
the
boundaries between relationship partners
and the
self,
in the
form
of
"including [the] other
in the
self"
(p.
241).
In
short,
these studies
confirm
that information about relationship
part-
ners
is
singled
out for
special processing,
and
they raise
the
pos-
sibility
that
the
need
to
belong leads
to a
cognitive merging
of
self
with
particular other people. Such patterns
of
subsuming
the
individual
in the
interpersonal unit indicate
the
importance
of
these relationships.
Many
of the
special biases that people exhibit
for
processing
information
in
ways that
favor
and flatter
themselves
are ex-
tended
to
partners
in
close relationships. Fincham, Beach,
and
Baucom
(1987)
showed that self-serving biases that take credit
for
success
and
refuse
blame
for
failure
operate
just
as
strongly—or
even more
strongly—when
people interpret their
spouses' outcomes
as
when they interpret their
own
outcomes.
That
is,
events
are
interpreted
in a way
that
is
maximally
flat-
tering
to the
spouse, just
as
they
are
interpreted
in
ways that
enhance
and
protect
the
self.
(These
patterns
are
extended only
to
partners
in
good, strong, happy relationships, however; high
marital
distress
is
correlated with
a
breakdown
in
these partner-
serving
attributions.)
Likewise,
the
"illusion
of
unique invulnerability" (Perloff
&
Fetzer,
1986)
turns
out not to be as
unique
as first
thought.
Although
people
are
more extremely
and
unrealistically
optimistic
about themselves than about some vague target
such
as the
average person, they
are
equally optimistic about
their
closest
friends
and
family
members. That
is,
they think
that
bad
things
are not as
likely
to
happen either
to
them-
selves
or to
their close friends
as to
strangers
or to a
hypothet-
ical
average
person.3
Along
the
same lines, Brown
(1986)
showed
that people
(particularly
those with high self-esteem)
tend
to
extend self-serving biases
to
their friends.
Specifically,
people rate both self
and a
same-sex
friend
more favorably
than
they rate people
in
general.
Group memberships also appear
to
exert important
influ-
ences
on
cognitive patterns. People expect more favorable
and
fewer
objectionable actions
by
their
in-group
than
by
out-group
members,
and
these expectations bias information processing
and
memory, leading people
to
forget
the bad
things (relative
to
good
things)
that their
fellow
in-group members
do
(Howard
&
Rothbart,
1980).
People also make group-serving
or
"sociocen-
tric"
attributions
for the
performance
of the
groups
to
which
they
belong. Members
of a
successful group
may
make group-
serving
attributions that
put the
entire group
in a
good light,
whereas,
after
failure,
group members
may
join together
in ab-
solving
one
another
of
responsibility
(Forsyth
&
Schlenker,
1977;
Leary
&
Forsyth,
1987;
Zander,
1971).
Linville
and
Jones
(1980)
showed that people tend
to
process
information
about out-group members
in
extreme, black-and-
white,
simplistic, polarized ways, whereas similar information
about members
of
their
own
group
is
processed
in a
more com-
plex
fashion. Thus,
the
mere existence
of a
social bond leads
to
more complex (and sometimes more biased) information
processing.
Of
broader interest
is
evidence that
belongingness
can
affect
how
people process information about nearly
all
categories
of
stimuli
in the
social world. Wegner
(1986)
noted
the
irony that
traditional theories
of the
"group mind" tended
to
assume that
all
members would essentially think
the
same thing, because
much more
far-reaching
advantages could
be
realized through
a
group mind
if
each member
was
responsible
for
different
in-
formation,
thereby enabling
the
group
to
process considerably
more information than
any one
person could. Wegner went
on
to
propose that transactive memory processes operate
in
close
relationships
and
groups
by
assigning each person
a
significant
category
of
expertise, with
the
result that each person becomes
expert
in one or a few
areas
and
simply consults others when
alternative
areas come
up. An
empirical study conducted
by
Wegner,
Erber,
and
Raymond
(1991)
supported
the
transactive
memory hypothesis
by
showing that partners
in
close relation-
ships
apparently have established procedures
for
determining
which
person
should
remember
which information. Partici-
pants were people
in
dating couples
who
were paired either
with
their partner
or
with
a
stranger.
The
preexisting couples showed
better memory
for
experimental stimuli than
the
impromptu
assigned
couples, except when
the
researchers assigned people
at
random
to be the
expert responsible
for
various categories
of
stimuli.
In
this latter condition, apparently,
the
assignment
disrupted
the
couples' preexisting system
and
hence impaired
the
processing
of
information.
Another
broad
and
very basic issue
is how
often
interpersonal
belongingness
is
used
as an
interpretive category.
C. A.
Ander-
son
(e.g.,
1991)
sought
to
establish
the
fundamental
dimensions
people
use for
making attributions about
the
causes
of
events.
His
study coded participants' attributional activity along
13
di-
mensions, including
all of the
ones featured
in the
major attri-
3
Perloff
and
Fetzer
(1986)
favored
an
interpretation
for
their results
in
terms
of the
vagueness
of the
comparison
target over
the
motivational
explanation
that
people
want
to
regard
their
closest
relationship
part-
ners
as
equally invulnerable (equal
to
themselves). Their discrimina-
tion between
the two
hypotheses
rested
on the
"one
of
your friends"
condition
in
their
second
study: They found
that
the
"closest
friend"
was
seen
as
being highly invulnerable, whereas when
participants
chose
one
of
their
other
friends, this
person
was
seen
as
more
vulnerable. Their
findings
suggested that participants
in
that
condition
selected
a
friend
who
seemed
most
likely
to
have
the
problem asked
about,
so it is
diffi-
cult
to
evaluate
the
motivational hypothesis. Thus,
the
interpretation
emphasized
here
is
consistent
with
all of
Perloff
and
Fetzer's
findings, as
they acknowledged, even though their
own
interpretations
tended
to
favor
explanation
in
terms
of
vague versus specific targets.

THE
NEED
TO
BELONG
505
butional theories (e.g., locus, stability,
globality,
and
controllability).
To his
surprise, however,
the
strongest single
dimension
was
what
he
called
interpersonalness,
which
was de-
nned
as the
degree
to
which
the
causes
of the
focal
event
re-
flected
on
the
relationship
between
the
individual
attributor
and
other people (e.g., doing something because
one is
married).
Thus, although interpersonalness
was not a
central concern
of
his
investigation (because
the
major attribution theories
had
largely
ignored
it),
it
emerged
as a
major dimension
in the way
people
normally think about
and
interpret
the
causes
of
events.
The
unexpected emergence
of
interpersonalness
as a
powerful
fundamental
dimension
of
causal attribution
is
consistent with
the
view
that
belongingness
is one of
humanity's basic concerns.
Thus
far we
have provided
evidence
that
interpersonal
rela-
tionships
are
centrally important
in the way
people think.
Ad-
ditional
predictions about cognitive activity
can be
derived
from
the
belongingness hypothesis. Although
the
evidence
is
consis-
tent with these predictions,
it
tends
to be
subject
to
alternative
explanations
based
on
short-term, pragmatic concerns,
so it is
less
compelling
for
present purposes.
We
include brief coverage
for
the
sake
of
thoroughness.
Clearly,
one
would predict,
on the
basis
of a
need
to
belong,
that people should tend
to
think
particularly
about
actual
and
potential relationship partners more than about other people.
This would
be
reflected
in
increased cognitive processing
caused
by the
expectation
of
future
or
further
interactions,
be-
cause these conditions hold
the
possibility
of
forming
a
relation-
ship.
Devine, Sedikides,
and
Fuhrman
(1989)
confronted par-
ticipants
with
advance information about various stimulus per-
sons
and
found
that this information received more thorough
and
detailed processing when
it
pertained
to a
future
interaction
partner.
Monson,
Keel, Stephens,
and
Genung
(1982)
found
that
people made more
extreme—and
more
valid—trait
attri-
butions
from
identical information when
it
pertained
to a fu-
ture
interaction partner than when
it
pertained
to
someone with
whom
no
interaction
was
anticipated.
Erber
and
Fiske
(1984)
showed
that interpersonal dependency
(outcome
dependency)
overcame
the
usual tendency
to
ignore information that runs
counter
to
expectations. When participants were outcome
de-
pendent
on the
confederate, they paid extra attention
to
incon-
sistent
information about
the
confederate
and
seemed
to
think
more
in
terms
of
dispositional attributions about
the
partner.
Thus,
belonging
to
another person changes
the way one
pro-
cesses information about that person.
Some
of
these
interaction
effects
could
be
interpreted
as
guided
by
short-term concerns. Still,
the
prospect
of
forming
a
relationship
with
a
recently
met
person appears
to be
sufficient
to
alter
the way
people process
the
interaction.
M. S.
Clark
(1984)
showed that people keep track
of
information
differently
when
the
interaction partner
is a
potential relationship partner.
Furthermore,
recent work
by
Tice,
Butler,
Muraven,
and
Still-
well
(1994)
showed that when people were interacting with
friends
as
opposed
to
strangers, they changed
the way
they pre-
sented information about themselves (i.e., they became more
modest).
Moreover,
the way
they encoded
and
recalled
the in-
teraction depended
on the
relationship: Memory
was
best
if one
had
been modest with
friends
or
self-enhancing
with
strangers,
and
otherwise
it was
impaired.
Critical
assessment.
The
evidence that interpersonal con-
cerns
affect
cognitive
processing
is
methodologically
strong
and
extensive.
A
broad
variety
of
experimental procedures
has
been
involved
in
demonstrating such
effects.
For
present purposes,
the
main critique would
be
that some
of the
studies have
not
been directly concerned with close relationships. Some have
shown
that
the
expectation
of
interaction with
a
stranger
or new
acquaintance
is
sufficient
to
alter
cognitive processing.
Al-
though
it is
reasonable
to
infer
that people regard meeting
new
people
as the first
step
in
possible relationship formation
(perhaps especially among
the
young adult populations
who
constitute most
of the
experimental
samples),
this inference
re-
quires
further
validation before
one can
have
full
confidence
in
interpreting those
findings as
evidence
for the
need
to
belong,
because
short-term
concerns
of
practical
or
material
advantage
may
also play
a
role
in
some such situations.
Nonetheless, many
of the findings
reviewed
in
this section
do
pertain
to
close relationship partners,
and
there
is
evidence that
information
pertaining
to
interaction partners
is
processed
differently
depending
on its
relevance
to
lasting relationships.
It
is
thus quite clear that relatedness
affects
cognitive processing;
only
the
extent
of
that
influence
and
some
of its
processes
are
still
open
to
debate.
Conclusion.
Concern
with
belongingness
appears
to be a
powerful
factor shaping human thought. People interpret situa-
tions
and
events with regard
to
their implications
for
relation-
ships,
and
they think more thoroughly about relationship
(and
interaction)
partners
than about other people. Moreover,
the
special patterns
of
processing information about
the
self
are
sometimes used
for
information about relationship partners
as
well.
Thus, both actual
and
potential bonds exert substantial
effects
on how
people think.
Emotion
The
main emotional implication
of the
belongingness
hy-
pothesis
is
that
real,
potential,
or
imagined changes
in
one's
be-
longingness
status
will
produce emotional responses, with posi-
tive
affect
linked
to
increases
in
belongingness
and
negative
affect
linked
to
decreases
in it.
Also, stable
or
chronic conditions
of
high belongingness should produce
a
general abundance
of
positive
affect,
whereas chronic deprivation should produce
a
tendency
toward abundant negative
affect.
Positive
affect.
In
general,
the
formation
of
social bonds
is
associated with positive emotions. Perhaps
the
prototype
of re-
lationship
formation
is the
experience
of
falling
in
love, which
is
typically marked
by
periods
of
intense bliss
and
joy,
at
least
if
the
love
is
mutual (e.g., Steinberg,
1986).
When
love
arises
without
belongingness,
as in
unrequited love,
the
result
is
typi-
cally
distress
and
disappointment
(Baumeister
&
Wotman,
1992).
Belongingness
is
thus crucial
if
love
is to
produce bliss.
Likewise,
occasions such
as new
employment, childbirth, fra-
ternity
or
sorority pledging,
and
religious conversion,
all of
which
are
based
on the
entry into
new
relationships
and the
formation
of
new
social bonds,
are
typically marked
by
positive
emotions
and
celebrated
as
joyous. Childbirth
is
especially sig-
nificant
in
this regard because
the
data show that parenthood
reduces happiness
and
increases stress, strain,
and
marital dis-
satisfaction
(e.g.,
S. A.
Anderson, Russell,
&
Schumm, 1983;
Campbell, Converse,
&
Rodgers,
1976; Glenn
&
McLanahan,

506
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
1982;
for
reviews,
see
Baumeister,
1991;
Bernard,
1982;
Camp-
bell,
1981;
Spanier
&
Lewis,
1980),
yet
people nonetheless
re-
tain
a
positive image
of it,
celebrate
it, and
feel
positive about
it,
both
in
advance
and in
retrospect.
It is
plausible that
the
formation
of the new
social bond
is
directly responsible
for the
joy and
positive
feelings,
whereas
the
negative aspects
and
feel-
ings
associated with parenthood arise indirectly
from
the
has-
sles, conflicts,
and
stresses that accompany
the
social bond.
If
the
formation
of
bonds
is one
occasion
for
joy,
a
second
occasion comes when
the
bond
is
formalized into
a
more recog-
nizably
permanent status.
A
wedding,
for
example, does
not
create
a new
relationship,
at
least
in
modern Western cultures,
because
the
bride
and
groom typically have known each other
intimately
for
some time.
The
wedding does, however,
signify
an
increase
in
commitment
to
maintaining
the
relationship per-
manently,
and the
joyful
celebration
of the
wedding
can be re-
garded
as an
affective
consequence
of
solidifying
the
social
bond.
It is
noteworthy that many traditional wedding vows
in-
clude
an
actuarially implausible pledge that
the
marriage
will
never
end
("till
death
do us
part").
In
essence, such
vows
are
an
institutionalized mechanism
for
committing people
to
meet
their
spouse's
belongingness
needs.
Although
we
have emphasized
the
view
of
affect
as a
result
of
attachment, positive
affect
may in
turn help
solidify
social
attachment. Probably
the
most
influential view
of
this
sort
was
developed
by
Shaver
et
al.
(1988),
who
portrayed romantic
love
as a
kind
of
glue designed
by
nature
to
solidify
the
attachment
between
two
adults whose interaction
is
likely
to
lead
to
parent-
ing.
In
their
view,
love elaborates
on
sexual attraction
in a way
that
will
hold
the
couple together when their sexual intercourse
leads
to
reproduction. Along
the
same lines, various studies
have
found
that positive
affective
experiences increase attrac-
tion
and
solidify
social bonds
(L. A.
Clark
&
Watson, 1988;
Gouaux, 1971;
May &
Hamilton,
1980; Veitch
&
Griffitt,
1976).
Moreland
(1987)
concluded that
the
development
of
shared emotions
is one of the
principal causes
of the
formation
of
small groups.
More generally, happiness
in
life
is
strongly correlated with
having
some close personal relationships. Research suggests
that
it
does
not
seem
to
make
a
great deal
of
difference
what
sort
of
relationship
one
has,
but the
absence
of
close social bonds
is
strongly
linked
to
unhappiness, depression,
and
other woes
(e.g.,Argyle,
1987;
Freedman,
1978;
Myers,
1992).
People with
high
levels
of
intimacy motivation tend
to
enjoy higher levels
of
happiness
and
subjective well-being
(McAdams
&
Bryant,
1987),
which
is
likely
a
result
of
their tendency
to
form
and
maintain
a
rich network
of
friendships
and
other social bonds
(McAdams,
1985).
Having some intimate bond appears
to be
important
and
perhaps even necessary
for
happiness. Social iso-
lation
is
practically incompatible with high levels
of
happiness.
Negative
affect.
Threats
to
social attachments, especially
the
dissolution
of
social bonds,
are a
primary source
of
negative
affect.
People
feel
anxious
at the
prospect
of
losing
important
relationships,
feel
depressed
or
grief stricken when their con-
nections
with certain other people
are
severed,
and
feel
lonely
when
they lack important relationships
(Leary,
1990;
Leary
&
Downs,
in
press; Tambor
&
Leary,
1993).
Anxiety
is
often
regarded
as the
extreme
or
prototype
of
neg-
ative
affect,
and it is
clearly linked
to
damaged, lost,
or
threat-
ened social bonds.
In
fact,
social exclusion
may
well
be the
most
common
and
important cause
of
anxiety (Baumeister
&
Tice,
1990).
Horney
(1945)
identified
the
source
of
"basic
anxiety"
as the
feeling
of
"being isolated
and
helpless
in a
potentially
hostile world"
(p.
41);
of
course, that
formula
mixes
two
different
sources, insofar
as
isolation
is a
function
of the
belong-
ingness
need, whereas helplessness
is a
frustration
of
control
(which
is
probably another fundamental
motivation).
Anxiety
and
general distress seem
to be a
natural consequence
of
being
separated
from
important others. Children
as
young
as 1
year
old
show extreme
distress—separation
anxiety—on
being sep-
arated
from
their mothers
(Bowlby,
1973),
and
adults show
similar reactions when they must leave loved ones
for an ex-
tended period
of
time. Furthermore, people's memories
of
past
rejections
are
tainted
with
anxiety (Tambor
&
Leary, 1993),
and
even just imagining social rejection increases physiological
arousal
(Craighead,
Kimball,
&
Rehak,
1979).
Consistent with
the
social exclusion theory
of
anxiety,
Barden,
Garber,
Leiman, Ford,
and
Masters
(1985)
found
that
anxiety
ensues
if
people
are
excluded
from
social groups,
but
experiences
of
social inclusion appear
to
counteract
the
effects
of
exclusion
and
remove
the
anxiety. Mathes, Adams,
and Da-
vies
(1985)
predicted that
a
threat
to
self-esteem would mediate
the
link
between jealousy
and
anxiety,
but
their results
did not
support
their
hypothesis.
Instead,
they found
that
the
loss
of
relationship
led
directly
to
anxiety.
Like
anxiety, depression
may be
precipitated
by a
variety
of
events,
but
failing
to
feel
accepted
or
included
is
certainly
one
of
them. Both general depression
and
social depression (i.e.,
dysphoria about
the
nature
of
one's social
relationships)
are
inversely
related
to the
degree
to
which
one
feels
included
and
accepted
by
others
(Tambor
&
Leary,
1993).
Hoyle
and
Craw-
ford
(in
press)
found
that both depression
and
anxiety were sig-
nificantly
correlated
(negatively)
with
students'
sense
of
belong-
ing
to
their university.
Jealousy
is
another negative
affective
state that
is a
com-
mon
response
to
threats
to
one's
relationships. Pines
and Ar-
onson
(1983)
reported that,
in a
series
of
surveys, some
ex-
perience
of
jealousy
was
essentially universal,
in the
sense
that everyone experiences
it
sooner
or
later. Moreover, more
than half
of
their respondents
described
themselves
as
being
"a
jealous
person"
and
correctly
estimated
that slightly more
than half
of the
other
participants
would
respond
in
that
same way; however, they also said that
the
true incidence
of
jealous people
was
even higher, because some jealous people
deny
their jealousy. Pines
and
Aronson emphasized that
"feeling excluded"
is a
major cause
of
jealousy.
Regarding jealousy, perhaps
the
most relevant
finding for our
purposes
was
that
of
Reiss(
1986),
who
concluded that jealousy
is
cross-culturally universal. Reiss
carefully
investigated
the ex-
travagant claims made
by
some observers
and
anthropologists
that,
in
certain cultures, people
are
able
to
exchange sexual
partners
and
intimate partners without
any
possessiveness
or
jealousy,
and in
every case
the
claim turned
out to be
unwar-
ranted.
Cultures
may
indeed vary
as to
which particular actions
or
signs
of
affection
are
regarded
as
justifying
jealous
reactions,
and
they
may
differ
in how
people express their jealousy,
but
sexual
jealousy
is
found
in all
cultures.
Loneliness reflects
"an
individual's subjective perception
of

THE
NEED
TO
BELONG
507
deficiencies
in his or her
social
relationships"
(Russell,
Cu-
trona, Rose,
&
Yurko, 1984,
p.
1313).
In
other words, people
feel
lonely when their
belongingness
needs
are
being
insuffi-
ciently
met. Moreover,
it
appears
that belongingness, rather
than
mere social contact,
is the
crucial factor. Mere social con-
tact does not,
by
itself,
buffer
people against loneliness. Lonely
and
nonlonely people
do not
differ
markedly
in the
amount
of
time
they spend
with
other people. However, lonely people
spend less time
with
friends
and
family—those
who are
most
likely
to
fulfill
their
needs
to
belong—than
nonlonely
people
(Jones,
1981).
Furthermore, loneliness
is
much more strongly
related
to
one's sense
of
social isolation than
to
objective
in-
dexes
of
one's
social network, such
as
one's sheer number
of
friends
(Williams
&
Solano,
1983).
In one
study,
the
correla-
tion
between self-reported loneliness
and the
degree
to
which
people
felt
included
and
accepted
by
others
was
found
to be
-.71
(Spivey,
1990).
Generally, loneliness seems
to be a
matter
more
of a
lack
of
intimate connections than
of a
lack
of
social
contact (Reis, 1990; Wheeler, Reis,
&
Nezlek,
1983).
Yet
another
highly
aversive
emotional state
is
guilt. Despite
a
long
tradition
of
analyzing
guilt
in
terms
of
self-evaluation
accord-
ing
to
abstract moral standards, recent work
has
increasingly
em-
phasized
the
interpersonal structure
of
guilt (Baumeister,
Stillwell,
&
Heatherton,
1994;
Cunningham, Steinberg,
&
Grev, 1980;
Jones
&
Kugler,
in
press; Jones,
Kugler,
&
Adams, 1995; Miceli,
1992;
Tangney,
1992). Empirical studies
of how
people induce
guilt
in
others
have
found
that such inductions
are
almost entirely
confined
to
close interpersonal relationships
and
that
a
major rea-
son
for
inducing
guilt
is to
cause one's partner
to
exert himself
or
herself
more
to
maintain
the
interpersonal relationship (e.g.,
by
spending
more time with
or
paying
more attention
to
oneself;
Baumeister,
Stillwell,
&
Heatherton,
in
press;
Vangelisti,
Daly,
&
Rudnick,
1991).
Many
episodes
of
guilt
can
thus
be
understood
as
responses
to
disturbances
or
threats
to
interpersonal attachments.
Two
specific
events that thwart people's need
to
belong
are
divorce
and
death. Divorce
is not
generally recognized
as an
occasion
for
joyful
celebration, even
if the
divorce
was
desired
more
fervently
than
the
wedding
had
been. Divorce produces
varied
forms
of
distress,
including
anger, depression, desolation,
and
loneliness,
in
nearly everyone. Weiss
(1979)
concluded that
some
"emotional
upset.
. .
appears
to be a
nearly inevitable
accompaniment
to
marital separation"
(p.
210)
and is
found
"even though [the] marriage
had
become unhappy"
(p.
202).
In
contrast, Spanier
and
Casto
(1979)
and
Goode
(1956)
did
find a
minority
of
participants
who
reported relatively little dis-
tress
in
response
to
divorce. Spanier
and
Casto
(1979)
thought
that
one
possible explanation
for the
discrepancy
was
that
their
single
interview
technique
(in
contrast
to
Weiss's multiple
sessions)
was
less sensitive
to
some deep
or
occasional
feelings.
Consistent
with
this, they concluded that certain forms
of
dis-
tress, such
as
regret, yearning,
and
bitterness, "actually
may in-
crease over time"
(p.
226). Price
and
McKenry
(1988)
sug-
gested another
reason
that
one-time
measures
may
fail
to find
universal
distress
after
divorce: Many couples
may
have passed
through
the
most distressing phase before
the
researchers col-
lect
their data.
Spanier
and
Casto
(1979)
listed
the
emotional turmoil
after
divorce
as
mixed
from
among
feelings
about
the
(former)
spouse, such
as
love, hate, bitterness,
guilt,
anger, envy, concern,
and
attachment;
feelings
about
the
mar-
riage,
such
as
regret, disappointment, bitterness, sadness,
and
fail-
ure;
and
more general feelings, such
as
failure,
depression,
eupho-
ria,
relief,
guilt, lowered self-esteem,
and
lowered
self-confidence.
(p.213)
Price
and
McKenry
(1988)
summarized
the
common emo-
tional
reactions
to
divorce
as
including "extreme stress, includ-
ing
feelings
of
rejection, depression, hostility, bitterness, loneli-
ness,
ambivalence,
guilt,
failure, confusion,
disorganization,
and
sometimes relief"
(p.
42).
It is
clear that plenty
of
negative
affect
accompanies divorce.
Perhaps
the
strongest emotional reactions human beings
ex-
perience involve death, both
the
death
of
oneself
and the
death
of
other people.
The
death
of a
spouse, child,
or
close
friend
ranks
among
the
most stressful events that people experience
(T.
H.
Holmes
&
Rahe, 1967; Weiss, 1979). Grief
often
takes
the
form
of an
especially severe depression. Some conceptual-
izations
of
grief portray
it not as a
reaction
to the
loss
of the
person
but as a
reaction
to the
loss
of a
linkage with another
person
(Lofland,
1982).
It is
interesting that people even grieve
deeply
over
the
death
of
spouses with whom they
had had
trou-
bled
marriages.
As
Weiss
(1979)
observed,
"Apart
from
minor
variations,.
. .
nearly disabling grief
was
the
rule,
even
among
individuals
who
could
say
about
the
preceding marriage,
as one
widow
did,
'Ours
wasn't
the
best marriage
in the
world'"
(p.
202).
Anxiety
about death, whether
of
oneself
or
others,
can be re-
garded
as
stemming
(at
least
in
part)
from
a
threat
to
belong-
ingness
(Baumeister
&
Tice,
1990).
As
Lofland
(1982)
pointed
out, when people die, relationships end. Along these lines,
Conte,
Weiner,
and
Plutchik
(1982)
linked death anxiety
to
fear
of
loneliness (see also Mijuskovic, 1980).
In a
study
of
death
anxieties,
Bednarski
and
Leary
(1994)
found
that
a
primary
basis
of
people's
fears
about
death
involved
concerns
with being
separated
from
friends
and
family.
These interpersonal con-
cerns appeared
to be a
more important source
of
death anxiety
than
fears
about
no
longer existing
or
uncertainty about what
happens
after
death. This
link
between death anxiety
and
sepa-
ration anxiety
may
explain
why
most positive depictions
of
life
after
death have emphasized togetherness with
family
and
loved
ones,
with
a
broad community
of
like-minded believers, with
a
loving
deity,
or
with
all
of
the
above (e.g., Baumeister, 1991).
If
death anxiety
is
rooted
in
threats
to
belongingness
and
social
inclusion,
then
fears
of
death
can
best
be
soothed
by
emphasiz-
ing
that death
will
involve
a
continuation
or
even
an
improve-
ment
in
one's
belongingness status.
Indirect
effects.
Although
we
have emphasized emotional
consequences
of
changes
in
belongingness, there
may
also
be
indirect ways
in
which belongingness
affects
emotion.
As
shown
earlier with cognitive processes, emotional processes
may
change
when
the
situation involves
a
close
friend
or
intimate
partner. Tesser
(1991)
has
reviewed
a
number
of
such
effects.
The
main implication
is
that
emotional
responses
to the
relative
outcomes
of
self
and
other depend heavily
on
whether
the
other
person
is a
close relationship partner such
as a
good
friend.
When
the
performance involves
a
domain that
is
important
to
the
self,
it is
upsetting
to be
outperformed
by
another person,
and the
emotional distress
is
magnified
if the
other person
is a

508
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
close
friend
(see also Tesser, Millar,
&
Moore, 1988).
In
con-
trast,
if the
performance involves some ability that
is not im-
portant
to
self-definition,
then superior performances
by
friends
(but
not
strangers) produce positive
affect.
One key
difference
is
what
Tesser
(1991)
called
the
reflection
process:
The
positive achievements
of
one's relationship partners
reflect
favorably
on the
self
(as
long
as
they
do not
make
the
self look
bad by
comparison
in
some important
way).
Similar achieve-
ments
by
strangers
do not
reflect
on the
self,
of
course,
and so
they
do not
produce positive
affect.
Meanwhile,
it
appears that
the
positive accomplishments
of
close others
in
domains rele-
vant
to
one's
own
identity have
a
special capacity
to
generate
distress
by
threatening one's cherished
views
of
one's
own im-
portant abilities. Thus,
the
existence
of a
close relationship with
another
person changes
the way one
responds emotionally
to
that person's performance outcomes
in
complex
but
predict-
able
ways.
Critical
assessment.
The
evidence reviewed
in
this section
was
drawn
from
sociology, anthropology,
and
several
subfields
of
psychology,
and it is
based
on a
variety
of
methods including
surveys,
observational studies, cross-cultural comparisons,
au-
tobiographical narratives,
and
experiments. Although several
of
these methods
are
generally regarded
as
less conclusive than
experimentation,
the
consistency
of the
conclusion across
multiple
methodologies
is
itself
a
source
of
confidence.
Thus,
for
example,
one
could dispute Pines
and
Aronson's
(1983)
de-
termination
about
the
pervasiveness
of
jealousy
by
noting that
their
sample
was
possibly skewed
to
include
a
high proportion
of
people
who
were interested
in
jealousy,
but the
very
high
(indeed,
universal)
incidence
of
jealousy across
different
cul-
tures,
as
attested
by
Reiss's
(1986)
review,
makes
it
seem
un-
likely
that Pines
and
Aronson were wrong
in
concluding that
jealousy
is
very
common.
Probably
the
greatest ambiguity
in
this section's evidence
at-
taches
to the
discussion
of
death.
To be
sure,
it is
implausible
to
dispute
that emotional distress very typically attends
the
death
of
a
loved
one or
relationship partner. Still, there
are
alternate
explanations
for
this distress that could possibly dispense with
the
need
to
belong.
A
partner's death
may
have
effects
on
mate-
rial
and
pragmatic concerns (e.g., loss
of
income),
may
create
distressing
uncertainty about
one's
own
future,
may
affect
the
self-concept,
and may
activate worries about
one's
own
death.
Conclusion.
Many
of the
strongest emotions people experi-
ence, both positive
and
negative,
are
linked
to
belongingness.
Evidence
suggests
a
general conclusion that being accepted,
in-
cluded,
or
welcomed leads
to a
variety
of
positive emotions
(e.g.,
happiness, elation, contentment,
and
calm), whereas
be-
ing
rejected, excluded,
or
ignored leads
to
potent negative
feel-
ings
(e.g., anxiety, depression,
grief,
jealousy,
and
loneliness).
The
near universality
of
distress associated with divorce
and be-
reavement
is
consistent with
the
belongingness hypothesis;
in-
deed, there
is no firm
evidence
in
those literatures that
signifi-
cant social bonds
can
ever
be
broken without
suffering
or
dis-
tress, even though
(as
noted)
not
every recently divorced
or
bereaved person
will
necessarily
be
suffering
acutely when
the
interviewer
happens
to
call.
Although
the
evidence
was not
equally abundant
or
equally
strong
for all
emotions,
the
consistency
across
multiple emo-
tions
was
impressive.
It
seems quite
safe
to
conclude that both
positive
and
negative emotional reactions
are
pervasively linked
to
relationship status.
The
existence
of an
interpersonal bond
changes
the way one
responds emotionally
to the
performances
and
actions
of a
relationship partner
and
indeed
intensifies
many
emotional reactions. Moreover, actual
or
possible changes
in
belongingness status constitute
an
important cause
of
emo-
tions.
The
evidence
is
sufficiently
broad
and
consistent
to
sug-
gest
that
one of the
basic functions
of
emotion
is to
regulate
behavior
so as to
form
and
maintain social bonds.
Consequences
of
Deprivation
The
general argument
is
that deprivation
of
belongingness
should lead
to a
variety
of
affiliative
behaviors
and
cause various
undesirable
effects,
including decrements
in
health, happiness,
and
adjustment.
We
have already documented
(in
the
preceding
section)
that loss
of
social bonds causes emotional
distress,
which
is
sufficient
to
show that belongingness
is
something peo-
ple
want.
To
regard
it as a
need, however,
it is
necessary
to
show
effects
that
go
beyond mere frustration
and
emotional
distress.
Considerable research shows
that
people
who do not
have
ad-
equate supportive relationships experience greater
stress
than
those
who do. In
part, this
is
because having other people avail-
able
for
support
and
assistance
can
enhance coping
and
provide
a
buffer
against stress. However, evidence suggests that simply
being part
of a
supportive social network reduces
stress,
even
if
other people
do not
provide explicit emotional
or
practical
assistance (Cohen
&
Wills, 1985). Although this
finding has
been interpreted
in
terms
of the
stress-reducing
effects
of
social
support,
an
equally plausible explanation
is
that
the
deprivation
of
the
need
to
belong
is
inherently stressful.
Direct evidence that deprivation
of
belongingness
is
maladap-
tive
was
provided
by
DeLongis,
Folkman,
and
Lazarus
(1988).
They
found
that happily married couples were less likely
to ex-
perience psychological
and
somatic health problems, both
on
and
after
stressful
days, than other participants. Medical
re-
search
has
suggested that these
beneficial
effects
extend beyond
mere health complaints. Lynch
(1979)
summarized
the
evi-
dence
from
many studies
by
stating
that
"U.S. mortality rates
for
all
causes
of
death
. . . are
consistently higher
for
divorced,
single,
and
widowed individuals" than
for
married individuals
(p.
38).
Lynch's
own
data
showed
the
greater incidence
of
fatal
heart attacks among unattached individuals than among mar-
ried people,
but he
noted that similar
effects
can be
found
for
tuberculosis, cancer,
and
many other illnesses,
as
well
as
overall
patterns.
Of
course, there
are
multiple possible explanations
for
such
an
effect
that might have nothing
to do
with belongingness,
but
efforts
to
control
for
these variables have
often
found
a
per-
sistent, independent, robust
effect
of
social relations. Goodwin,
Hunt, Key,
and
Samet
(1987)
found that married participants
survived
cancer better than single ones even
after
the
timing
of
diagnosis, likelihood
of
receiving
treatment,
and
cigarette
smoking
had
been controlled,
and
they
cited
other evidence
that
the
effect
remains
after
family
income
has
been controlled.
Indeed, being deprived
of
belongingness
may
have direct
effects
on the
immune system.
Kiecolt-Glaser,
Garner,
et
al.
(1984)
found
that loneliness
was
associated
with
a
decrease
in
immunocompetence,
specifically
in
natural killer cell activity,
and
this
effect
was
independent
of
changes
in
perceived distress.

THE
NEED
TO
BELONG
509
Kiecolt-Glaser,
Ricker,
et
al.
(1984)
replicated this
effect
and
also
found
elevated urinary
cortisol
levels among lonely partic-
ipants.
Kiecolt-Glaser
et al.
(1987)
found
poorer
immune
func-
tion
on
several measures among women
suffering
from
marital
disruption, including divorce,
separation,
and
unhappy
marriage.
The
effects
of
belongingness
on
mental illness parallel those
on
physical illness. Rejected children have
a
higher incidence
of
psychopathology
than other children (Bhatti, Derezotes, Kim,
&
Specht, 1989; Hamachek, 1992). Children
who
grow
up
without
receiving adequate attention
from
caregivers
show
emotional
and
behavioral pathologies,
as
demonstrated experi-
mentally
by
Harlow, Harlow,
and
Suomi
(1971)
with animals
and
as
corroborated
by
observations
of
human children
by
Bowlby(1969,1973;seealsoRutter,
1979).4
Marital
status
also
has
strong
correlations
with mental ill-
ness. Bloom, White,
and
Asher
(1979)
reviewed
the
literature
and
concluded that,
in all
studies, mental hospital admission
rates
are
highest among divorced
and
separated
people, inter-
mediate among never-married people,
and
lowest among mar-
ried
people.
In
fact,
as
measured
by
admissions
to
mental
hospitals,5
mental illness
is at
least
3 and
possibly
up to 22
times
higher
among divorced people than among married people.
Even
problems that might
at first
seem unrelated
to
social
interaction
and
relationships
are
sometimes
found
to
have
so-
cial
deprivation
or
failed
belongingness
as an
underlying cause.
Problems
with
attachment
have
been identified
as a
major fac-
tor in
eating disorders. Sours
(1974),
for
example, noted that
patients with eating disorders tended
to
have
been
(as
children)
overly
sensitive
to
separation
from
their mothers. Armstrong
and
Roth
(1989)
found
that women
with
eating disorders
had
significantly
more intense
and
severe separation
and
attach-
ment
difficulties
than
a
normal comparison group.
Combat-related stress
is
also moderated
by
belongingness.
Veterans
who
perceive that they have
a
high degree
of
social
support
are
significantly
less
likely
to
experience post-traumatic
stress disorder than those
who
have
lower perceived support
(Hobfall
&
London, 1986; Solomon, Waysman,
&
Mikulincer,
1990).
In
fact,
the
authors
of one
study concluded that loneli-
ness
"is the
most direct antecedent
of
psychopathology
and so-
cial
dysfunction"
in
combat stress reactions (Solomon
et
al.,
1990,
p.
468).
Crime
may
also
be
affected
by
belongingness. Sampson
and
Laub
(1993)
showed that having
a
good marriage
and a
stable
job
each
had a
strong negative
effect
on
adult crime, consistent
with
other evidence. Other evidence suggests that
social
bonds
to
other criminals
or to
criminal groups
may
foster crime.
Re-
cent
news
coverage
of
gangs
has
repeatedly suggested that
a
need
to
belong attracts unattached young people
to
join violent
gangs,
which
tend
to
serve
as a
surrogate "family"
(Olmos,
1994;
cf.
Jankowski,
1991).
Sampson
and
Laub likewise
found
that having social relationships with delinquent
peers
was one
of
the
strongest independent predictors
of
juvenile
delinquency,
consistent
with
plenty
of
previous evidence. They did, however,
caution
that
this well-established
link
is
based
on
largely
corre-
lational
data
and
that ambiguities about
the
direction
of
causal-
ity
remain
to be
addressed.
Still,
for
present purposes,
the
link
is
important evidence that belongingness needs
are
important
among
deviants, regardless
of
whether
the
link arises because
having
delinquent
peers
causes delinquent activity
or
because
delinquent activity leads
to
bonding with delinquent
peers.
Meanwhile,
in
laboratory experimentation,
Geis
and
Moon
(1981)
sought
to
involve
participants
in
lying,
cheating,
and
stealing
at the
behest
of an
assigned group
partner
(a
confederate).
They
found
that
67% of a
sample
of
college stu-
dents acquiesced
in an act of
cheating
and in a
monetary
theft
by
their partner
and
that they actively lied
to
conceal
the
theft.
Thus,
it
appears
that even recently formed group bonds
may be
strong enough
to
overcome some salient prohibitions
of
tradi-
tional morality.
(It
is
noteworthy
that
the
group loyalty
in
that
study
may
have been
intensified
by the
presence
of a
hostile rival
group.) More extreme versions
of the
phenomenon
of
going
along with objectionable actions
by
fellow
group members
be-
cause
of
loyalty
have
been commonly observed
as
central fac-
tors
in
group violence, such
as
spontaneous
atrocities
commit-
ted by the Ku
Klux
Klan
(Wade,
1987),
Nazi police guards
(Browning,
1992),
and
others
(Staub,
1989;
see
also
Groth,
1979,
on
gang
rape).
The
relevance
of
belongingness
to
suicide
was
suggested
nearly
a
century
ago by
Durkheim
(1897/1963).
His
seminal
work
proposed
that
suicide could
be
explained
as a
result
of a
failure
of
social integration. People
who are
well
integrated into
society
by
multiple
and
strong relationships
are
unlikely
to
commit suicide, whereas
unintegrated
people
are
much more
likely
to
kill
themselves. Durkheim's hypothesis
has
held
up far
better than most social science hypotheses over
the
decades,
and
the
evidence continues
to
show that
a
lack
of
social integration
increases
the
likelihood
of
suicide (Trout, 1980).
For
example,
single, divorced,
and
widowed people
are
more
likely
to
commit
suicide than married people (e.g.,
Rothberg
&
Jones, 1987).
Those
who are
unemployed
have
a
higher suicide
rate
than
those
who are
employed. People
who
belong
to
subcultural
groups
that
are
shrinking have increased suicide rates. People
who
work
in
occupations
that
are
shrinking
are
also more
likely
than
others
to
commit suicide. Indeed,
the
main criticism that
can be
leveled against Durkheim's hypothesis
is
that
it is
incom-
plete
in the
sense
that
it
does
not
explain everything about sui-
cide
(e.g.,
Baumeister,
1990;
Douglas,
1967),
but it is
correct
as
far
as it
goes.
For
present purposes,
the
important point
is
that
strong social ties
are
associated
with
a
lower risk
of
suicide,
probably because such ties help restrain people
from
killing
themselves.
Social support research
is
relevant
to the
belongingness
hy-
pothesis because social support
is
based
on
relationships
and
4
Several
studies
have
shown
that
physically
unattractive
people
have
a
higher
incidence
of
psychopathology
than
attractive
people
(e.g.,
Ba-
rocas
&
Vance,
1974;
Cash,
1985;
Farina,
Burns,
Austad,
Bugglin,
&
Fischer,
1986;
O'Grady,
1989).
One
reason
may be
that
they
tack
be-
longingness,
because
society
tends
to
reject
unattractive
individuals
(Berscheid
&
Walster,
1974).
5
Admittedly,
hospital
admissions
is an
imprecise
measure.
One
might
object
that
married
people
can
stay
out of
institutions
because
they
have
someone
at
home
to
take
care
of
them.
On the
other
hand,
many
people
are
admitted
to
such
institutions
at the
behest
of
family
members,
and so one
could
argue
that
the
true
difference
is
even
larger.
Given
the
size
and
consistency
of the
effect,
it
seems
reasonable
to
con-
clude
that
marital
status
is
related
to
mental
illness,
although
further
and
methodologically
better
evidence
is
needed.

510
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
positive
interactions with others,
and so any
benefits
of
such
support would constitute
further
confirmation
of the
belong-
ingness
hypothesis.
The
benefits
of
social support appear
to be
well
established. Thus,
for
example, Cohen, Sherrod,
and
Clark
(1986)
showed that
the
availability
of
social
support—which
can
be
restated
as the
existence
of
social
bonds—buffers
people
against
the ill
effects
of
stress. Cutrona
(1989)
showed that
so-
cial
support reduced depression during pregnancy
and
postpar-
tum
depression among adolescent girls. Responding
to
method-
ological
criticisms that
had
attacked social support research
as
merely
self-report bias, Cutrona's study included ratings
of
each
girl's support network
by an
adult informant
who
knew
the
girls
well,
and
these external informants' ratings predicted health
outcomes
(in
some cases, even better than
the
girls'
own
ratings
of
their support). Thus,
the
benefits
of
belongingness
in
coping
with
major
life
stress appear
to go
beyond mere self-report bias.
Older adults
who
have
a
close, intimate
friend
(i.e.,
a
"confidant")
maintain higher morale
in the
face
of
life
stresses
such
as
retirement
and
spousal death than individuals
who
lack
such
a
relationship.
For
example, Lowenthal
and
Haven
(1968)
found
that widows
who
have
a
confidant have been
found
to
be
only
slightly
more depressed than married women, whereas
those without
a
confidant have been
found
to be
much more
dysphoric.
These researchers also
found
that
the
majority
of
older adults
who
recently lost
a
confidant
were depressed,
but
the
majority
who
currently
had a
confidant
were satisfied.
Rook
(1987b)
distinguished between social support
and
companionship.
Social support
was in
this case rather narrowly
interpreted
in
terms
of
direct help, whereas companionship
meant
the
expressive aspects
of
social interaction. Both were
found
to be
important
and
beneficial,
but
companionship
may
be the
more important
of the
two, especially
for
psychological
well-being,
social satisfaction,
and
coping with minor stress.
These data
are
particularly important
for the
relevance
of
social
support
research
to the
belongingness hypothesis because
one
could
conceivably argue that belongingness
per se is
irrelevant
and
that
the
practical, material help that people derive
from
their
social networks
is
solely responsible
for the
benefits
of so-
cial
support. Rook's data suggested,
on the
contrary, that
the
practical
help
is
secondary (except
in
extreme circumstances
in
which
major assistance
is
needed),
whereas belongingness
is
highly
beneficial
by
itself.
Perhaps most generally, general well-being
and
happiness
in
life
depend
on
having some close social ties. Social isolation
is
strongly
related
to
various patterns
of
unhappiness
(for
reviews,
see
Argyle,
1987; Baumeister,
1991;
Freedman, 1978; Myers,
1992).
Indeed, Baumeister
(1991)
noted that
it is
about
the
only
objective factor that shows
a
substantial correlation with
subjective
well-being. Happiness also appears
to be
fairly
stable
across time
and
circumstance
(e.g.,
Costa,
McCrae,
&
Zonder-
man,
1987),
leading many
to
conclude that
it is
linked
to
per-
sonality
factors.
The
broad
trait
of
extraversion
appears
to be
strongly
related
to
happiness
and
positive
affectivity
(see
Costa
&
McCrae, 1980, 1984),
and
extraversion
encompasses
several
factors,
such
as
sociability,
gregariousness,
warmth,
and
social
involvement,
that seem
likely
to
enhance
the
tendency
to
form
and
maintain social ties. Moreover, belongingness
appears
to
be
sufficient
to
overcome
the
relative
deficit
in
happiness that
introverts
suffer.
Hotard, McFatter, McWhirter,
and
Stegall
(1989)
found
that introverts
who
have
a
good network
of
social
relationships
are
just
as
happy
as
extraverts. Thus, introverts'
deficit
in
happiness
may be a
result
of
their experiencing less
belongingness.
Further support
for the
importance
of
belongingness
to
psy-
chological well-being
is
provided
by the
fact
that
the
psychother-
apeutic process
is
facilitated
by
close personal bonds. Numer-
ous
therapeutic orientations stress
the
importance
of the
rela-
tionship between
the
therapist
and the
client. Rogers
(1959),
for
example, urged psychotherapists
to
display
a
willingness
to
accept
and
support
the
client regardless
of his or her
behavior
or
contribution
to the
relationship. Such "unconditional positive
regard"
is
perhaps
the
ultimate
way to
fulfill
another
person's
belongingness
needs. From
the
standpoint
of the
belongingness
hypothesis, however,
the
essential ingredient
in
client-centered
therapy
is not
unconditional positive regard (i.e.,
appraisal)
but
unconditional social acceptance (i.e.,
belongingness).6
Furthermore, some have suggested that
one
goal
of
psycho-
therapy
should
be to
enhance clients' ability
to
elicit social
sup-
port
in
their everyday lives
(Brehm,
1987).
To the
extent that
people
who
have strong connections
with
others
are
happier,
healthier,
and
better able
to
cope with
the
stresses
of
everyday
life,
most clients would presumably benefit
from
enhancing
their
belongingness.
The
psychotherapeutic usefulness
of
belonging
can
also
be
seen
in the
effectiveness
of
group therapy.
As
Lewin
(1951)
flatly
stated,
"It is
easier
to
change individuals formed into
a
group than
to
change them
separately"
(p.
228).
In
part,
the
effectiveness
of
group therapy seems
to
depend
on
engendering
a
sense
of
belongingness,
as
some authors have asserted
(Larkin,
1972;
Yalom,
1985).
Forsyth
(1991),
in his
review
of
research
on
group therapy, observed that therapeutic groups provide
the
member "with
a
sense
of
belonging, protection
from
harm,
and
acceptance"
(p.
675).
People
differ,
of
course,
in the
degree
to
which they believe
that their belongingness needs
are
being
met
irrespective
of the
extensiveness
of
their social networks
or the
strength
of
social
support they receive. Lakey
and
Cassady
(1990)
provided data
suggesting
that perceived social support
operates
much
like
a
cognitive
schema. People have relatively stable, organized
be-
liefs
about
the
extent
and
quality
of
their interpersonal relation-
ships. These belief systems lead
to
biased interpretation
of so-
cial
interactions,
as
well
as to a
biased recall
of
past interper-
sonal events.
As a
result, some people have
a
predisposition
to
perceive
others
as
unsupportive, leading them
to
experience
be-
longingness
deprivation even when others
are in
fact
being
supportive.
Critical
assessment.
The
diversity
of
methodologies
and the
multiplicity
of
disciplines that have furnished
the
evidence
re-
viewed
in
this section make
it
highly implausible
to
suggest that
all
such evidence
can be
explained
away
as the
result
of
con-
founds
or
artifacts.
At
worst, some
of the findings
have
alternate
explanations.
Not all
studies have maintained
careful
distinc-
tions
between
the
pragmatic benefits
of
certain
relationships
and the
direct
benefits
of
belongingness.
The
fact
that
happily
6
The two
overlap
in
many ways,
of
course.
Cutrona
(1986)
has
noted
that
esteem
support
is an
important
element
of
social
support,
particu-
larly
for
helping
people
avoid
depressive
reactions
to
stressful events.

THE
NEED
TO
BELONG
511
married people commit
fewer
crimes than other adults,
for ex-
ample, might
be
partly
(or
even
wholly)
due to the
material
benefits
of
being married. Even
so,
researchers
who
have main-
tained such
distinctions
(such
as
several
of the
social
support
researchers)
have
found
pragmatic benefits
to be a
secondary
factor.
Belongingness
thus
has
important
and
direct benefits.
A
more serious limitation
is
that several
of the findings are
correlational.
The
higher
rates
of
mental
and
physical illness
among loners could
reflect
a
tendency
for
people
to
reject
deviants
as
potential relationship partners.
By the
same
to-
ken,
the
higher levels
of
life
satisfaction
found
among happily
married people could
be
partly
due to a
tendency
for
chroni-
cally
unhappy people
to be
rejected
as
marriage partners.
Still,
those studies that have provided evidence about
the di-
rection
of
causality have consistently identified belongingness
as the
causal factor.
Conclusion.
Deprivation
of
stable, good relationships
has
been
linked
to a
large array
of
aversive
and
pathological conse-
quences.
People
who
lack belongingness
suffer
higher levels
of
mental
and
physical illness
and are
relatively highly prone
to a
broad range
of
behavioral problems, ranging
from
traffic
acci-
dents
to
criminality
to
suicide. Some
of
these
findings may be
subject
to
alternative explanations,
and for
some
the
direction
of
causality
has not
been established;
however,
the
weight
of ev-
idence
suggests that lack
of
belongingness
is a
primary cause
of
multiple
and
diverse problems.
It
therefore seems appropriate
to
regard belongingness
as a
need rather than simply
a
want.
Partial
Deprivation:
Relatedness
Without
Interaction
We
have
proposed that
the
need
to
belong
has two
aspects:
People
require
frequent
interactions with
the
same person,
and
people
want
a
stable, enduring context
of
concern
and
caring.
This section examines some instances
in
which people
have
the
first
of
these
while
being deprived
of the
second,
and the
next
section
examines
the
opposite case. These
cases
are
important
for
establishing whether
the
need
to
belong does indeed involve
both aspects.
To
confirm
this version
of the
belongingness
hy-
pothesis,
a
rather precise pattern
of
comparisons
is
needed,
one
based
on the
assumption that
satisfying
only
one of the
compo-
nents
should bring only partial satisfaction. People with only
one
of the two
components should presumably
be
slightly better
off
than people
who
have neither,
but
they should
be
worse
off
than
people
who
have
both.
One
example
of
relatedness without interaction involves peo-
ple
in
prison. Many prisoners have
families
or
loved ones
on
the
outside,
but
interactions with them
are
severely restricted.
Although
systematic, quantitative data
are
scarce, works
on
prison
life
appear
to be in
emphatic agreement that prisoners
treasure
and
cling
to
these ties
yet
suffer
greatly over
the
lack
of
interaction
(Baunach, 1985;
Isenberg,
1991; Toch, 1977).
At
least some
efforts
at
prison
reform
and the
cultivation
of al-
ternatives
to
standard imprisonment emphasize that increasing
contact
with
family
members
is
beneficial
to the
prisoner
(Scudder,
1952). Toch
(1977)
documented
the
extensive
an-
guish
suffered
by
prisoners
over
lack
of
contact with
family
and
romantic partners, although
he
noted that
the
perceived threat
of
losing
the
bond
was
often
a
source
of
suffering,
which suggests
that
the
deprivation
of
interaction
is not
fully
responsible
for
the
distress.
One
group
that
might
be
relatively immune
to
this
fear
would
be
imprisoned mothers, insofar
as
mother-child bonds
cannot
be
broken through divorce
or
other mechanisms (unlike
ro-
mantic ties
and
friendships). Baunach
(1985)
and
Giallom-
bardo
(1966)
both
reported
that imprisoned mothers lamented
the
loss
of
interaction with their children
and
used every avail-
able means
to
maximize contact. They noted
that
these
efforts
were
especially impressive
in
that
both
the
prison institution
and
the
collective wisdom
of the
prison subculture stress
the
need
to
suspend
all
emotional interest
in
events outside
the
prison, because such concerns produce frustration
and
helpless-
ness. Thus, these women's
efforts
to
maintain such ties
are
op-
posed
by
pervasive situational pressures,
but
they remain strong
anyway,
suggesting
that
the
bonds continue
to
offer
satisfaction
of
some
powerful
need despite
the
effort
and
frustration
in-
volved
in
maintaining them.
Noncustodial divorced parents represent another group
re-
strained
from
interacting with their children. Wilbur
and
Wil-
bur
(1988)
observed that most such
parents
refused
to
accept
that
the
bond
to
their children
was
severed
or
even damaged,
even
when their lawyers advised them
to
abandon
efforts
to
con-
tinue
the
relationship. Thus,
the
bond
is
apparently
very
impor-
tant
to
these parents even
if it is
mainly associated with frustra-
tion,
aggravation,
and
disappointment. Meanwhile,
the
lack
of
interaction
and
contact with
the
children
was
often
very upset-
ting
to
these noncustodial parents,
and
indeed most
of the di-
lemmas
that
Wilbur
and
Wilbur
associated
with noncustodial
parenthood revolved around
a
lack
of
contact
and
interaction.
Children
of
divorce
are
often
in a
similar position
of
losing
interaction access
to the
noncustodial parent.
R.
Rosen
(1979)
found
that children
who had
free,
unlimited access
to
interact
with
the
noncustodial parent were least
likely
to
perceive
the
divorce
as
traumatic, although that
finding was
based
on
corre-
lational
data
and
both variables
(access
and
trauma)
could
have
been confounded with
how
well
the
parents
got
along with each
other
after
the
divorce. Rosen also
found
that most children
ex-
pressed
a
clear preference
for
such
free
access
to the
noncusto-
dial parent,
and a
large minority indicated
that
they
had
less
contact than they wanted
to
have with that parent. Thus, even
if
the
relational bond continues
to
exist, many children seem
to
suffer
from
the
reduction
in
interaction.
A
later study
by
Drill
(1987)
concluded strongly that most children want
to
maintain
the
bond
despite
the
reduced interaction. Drill observed
that
children
of
divorce were most prone
to
depression
if
they per-
ceived
the
noncustodial parent
as
being lost,
in the
sense
of
hav-
ing
the
bond severed. Fortunately, most children apparently
perceive
the
bond
to
remain
in
existence, which presumably
accounted
for
Drill's
finding
that children
of
divorced parents
were
no
more likely overall than children
of
non-divorced par-
ents
to be
depressed.
Weiss
(1973)
reported that housewives
who had
recently
moved
to the
Boston
area
often
reported loneliness despite hav-
ing
a
strong marital bond. They were lonely because they were
deprived
of
interactions most
of the
time: They
had no
local
friends,
and
their husbands were
away
all day and
preoccupied
with
their
new
jobs.
Hoyle
and
Crawford
(in
press)
found
that
students'
sense
of
belonging
to
their university involved more

512
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
than mere
identification
with
the
institution;
it
also
had a
strong
component
of
behavioral involvement. This sense
of
belonging
was
heavily correlated
(.65)
with their involvement
in
univer-
sity
activities, suggesting that daily interactions
are an
impor-
tant
part
of
belongingness.
Although alternative explanations
for
these
findings
cannot
be
ruled out,
the findings are
consis-
tent
with
the
general pattern that
a
bond alone
is not
enough
to
satisfy
the
need
to
belong.
Long-distance
relationships
and
commuter marriages
offer
another
set of
circumstances
in
which people have
an
interper-
sonal
bond
but are
relatively deprived
of
interaction. Gerstel
and
Gross
(1982)
observed that people cling
to
these relation-
ships,
which suggests that they
are
positively valued
and
provide
some
rewards,
but
also
find
them
stressful,
consistent with
the
view
that relationship without interaction
is
less than
fully
sat-
isfactory.
Similarly intermediate results were reported
by Go-
vaerts
and
Dixon
(1988):
Commuters
did not
show
any
sig-
nificant
drop
in
overall marital satisfaction,
but
they
did
express
dissatisfaction
with
time spent together
and
affectional
commu-
nication.
Gerstel
and
Gross
found
that
the
stressful
aspect
of
commuter
marriage
was
significantly
reduced
by
regular week-
end
visits; thus,
the
opportunity
for
regular
and
fairly
frequent
interactions
was
very
beneficial
(see also Holt
&
Stone,
1988).
They
also
found
that couples
who had
been married longer
and
therefore
had a
greater sense
of
stability
suffered
less
from
the
stress
of
separation than other commuter couples, presumably
because they could remain more secure
and
confident that
the
attachment
to the
spouse would survive. Thus, these people
have
a
solid bond
but
still
express
a
strong need
for
interactions.
In
a
later work, Gerstel
and
Gross
(1984)
reported that com-
muter
couples valued
the
bond
but
suffered
over
the
loss
of in-
teraction.
Couples seemed
to find it
ironic that small talk over
trivial
matters would turn
out to be
something they missed,
but
as
Gerstel
and
Gross
noted, these seemingly insubstantial
in-
teractions
are
believed
to be an
important aspect without which
the
marital bond
is not
fully
satisfactory
or
fulfilling.
Frequent
(long-distance)
telephone conversations were
a
common
but
not
fully
satisfactory solution
to the
deprivation
of
interaction.
Respondents
in
that study noted that telephone conversations
seemed
adequate
for
sharing information
and
discussing prac-
tical
affairs
but
were
frequently
deficient
for
producing pleasant
social
interactions
or
enjoying
one
another's company. This sug-
gests
that regular interactions
do
have something
to
offer
that
is
not
contained
in
merely knowing that
the
social bond exists
and
exchanging
information.
In
addition, loss
of
shared leisure
ac-
tivities
was a
common
complaint.7
Winfield's
(1985)
study
of
commuter marriages confirmed
many
of
Gerstel
and
Gross's
(1982,
1984) conclusions.
In ad-
dition,
Winfield
found
a
surprisingly
low
rate
of
sexual
infidelity
(see
also Gerstel
&
Gross, 1984)
and
concluded that married
people
who
live
apart are, ironically, only about half
as
likely
to
be
unfaithful
as
married people
who
live
together (despite
the
presumably
much greater opportunity
and
temptation).
She
cited commitment
to the
relationship
as an
important
reason
for
this increased
fidelity, and so it
reflects
on how
these
people
value
the
social bond. Still,
it was
clear that many
couples
suffer
and
feel
deprived because
of the
lack
of
interaction,
and
Win-
field
observed
that loneliness
was a
frequent problem.
A
similar
point
was
made
by
Bunker, Zubek,
Vanderslice,
and
Rice
(1992),
who
found
that commuting spouses were less satisfied
with
their marital relationship,
family
life,
and
overall quality
of
life
than were spouses
who
lived together.
For
present
purposes,
the
implication
is
that
the
bond
to an
absent spouse
appears
to
furnish
some positive
benefits
and
satisfactions,
but
people
still
suffer
over
the
lack
of
contact.
The
evidence
from
commuter marriages thus appears
to
confirm
the
importance
of
two
separate components
of
belongingness, namely
the
secure
confidence
in an
enduring bond
of
mutual caring
and the
regu-
lar
experience
of
pleasant,
affectively
positive interactions.
Similar
findings
have emerged
from
studies
of the
spouses
of
military
personnel. Several articles
on the
wives
of
submariners
have
shown that these women
suffer
anxiety, depression,
and
physical
illness during
the
long absences
of
their husbands
(K.
Beckman,
Marsella,
&
Finney, 1979;
A. I.
Snyder, 1978;
see
Harrison
&
Connors,
1984,
for a
review).
Pearlman
(1970)
ob-
served
that each departure typically involved
a
crisis.
Critical
assessment. Evidence
from
multiple
fields and
seemingly
quite
different
populations points
to the
same con-
clusion
about
the
need
for
interactions.
All of the
studies
can
be
criticized
on
methodological grounds, however.
The
prison
samples
may be
atypical
and
pathological. Commuters
may be
atypical
because
of
having chosen
to
live
apart (although
the
fact
that they still
suffer
from
the
deprivation despite this choice
would
seemingly strengthen rather than weaken
the
argument
that frequent interactions
are
needed).
The
observations about
children
of
divorce seem less tainted
by
such concerns,
but it
may
be
difficult
to
disentangle
the
multiple causes
of
distress.
The
spouses
of
military personnel
may be
most representative
of
the
population
at
large.
A
further
problem
is
that most
of
these studies have used samples
of
convenience rather than sys-
tematically created ones.
For
prisoners
in
particular,
and in
some
of the
studies
of
other groups,
the
data
are
largely obser-
vational
and
impressionistic,
and it
would
be
much better
to
have
quantified
comparisons with well-chosen control groups.
Alternative
explanations also plague
the
prison studies. Pris-
oners derive practical benefits
from
maintaining contact with
people outside
the
prison
who can
bring them material goods
and do
them
favors
(Isenberg,
1991).
As
already noted, some
of
the
concern about lack
of
contact
with
loved ones
may
reflect
a
fear
of
losing
the
bond,
so it is not
safe
to
regard prisoners
as a
pure example
of
people
who
have
a
stable bond
but
lack
interac-
tions.
To
some extent, this problem
can be
rectified
by
consid-
ering
mothers,
who
should
be
less worried about being aban-
doned
by
their children;
in
some cases, however, they
too
fear
that
the
child
will
bond with someone else
and
become
es-
tranged
from
them
(Baunach,
1985),
so it may be
appropriate
to
regard this
fear
as
merely reduced,
not
eliminated, among
them. Baunach
(1985)
also noted that
it is
impossible
to
rule
7
One
might
wonder
whether sexual
deprivation
was
responsible
for
the
problems
reported
in
commuter
marriages.
However,
studies
of
these
couples
indicate
that
most
see
each
other
a
couple
of
days
each
week, which
in
principle
would
be
sufficient
for the
approximately
weekly sexual
intercourse
that
is the
norm
among
married
couples.
Ger-
stel
and
Gross
(1984)
found
that
most
of
these
couples
had had sex
only
on
weekends
even when they lived together,
so
there
was
little
decline
in
sexual frequency
as a
result
of
commuting;
most
couples
reported
that
their
sex
lives were
basically
the
same
after they
started
commuting.

THE
NEED
TO
BELONG
513
out
the
alternative explanation that some imprisoned mothers'
displays
of
concern
for
their children
are
feigned
ploys
to im-
press
the
parole
board.
The
growing literatures
on
commuter marriages
and filial at-
tachments
to
divorced, noncustodial parents
are
less subject
to
alternative explanations than
the
prison studies,
but
they
too
are far
from
controlled, prospective studies
that
conclusively
demonstrate causal
effects.
There have been
attempts
to
study
direct
effects
of
frequency
of
interaction independent
of
prag-
matic
and
other benefits. Most
of the findings are
still correla-
tional,
but on a
priori grounds
it
seems implausible
to
suggest
the
reverse causal hypothesis
(e.g.,
that unhappiness over lack
of
interaction causes people
to
spend less time
together).
Despite these concerns,
the
convergence across
different
groups
and
methods
is
encouraging.
At
present,
it
seems
appro-
priate
to
accept
the
converging
conclusions
from
these studies,
at
least
until
contrary evidence
is
found.
Conclusion.
Broad
and
consistent
but
methodologically
weak
evidence supports
the
conclusion that having
a
relation-
ship
without
frequent
interactions
offers
only partial, incom-
plete
satisfaction
of the
need
to
belong.
Researchers
have
stud-
ied
several
different
circumstances
in
which people
find
them-
selves
having
relationships without interactions,
and in
each
case
the
same conclusion
has
emerged: People with such bonds
do
seem
to
treat them
as
desirable
and
valuable (consistent with
the
view
that
they
do
offer
some
rewards)
but
suffer
over
the
lack
of
direct contact
with
the
other person.
Partial
Deprivation:
Interaction
Without
a
Bond
of
Caring
Interaction without
an
ongoing bond
of
caring should also
be
only partly satisfactory.
Two
predictions
can be
made.
First, insofar
as the
need
to
belong requires
that
some interac-
tions
reflect
a
relationship context,
it can be
predicted that
interactions
with changing
series
of
partners should
be
less
than
satisfying.
Second,
if the
interactions
are
supposed
to
reflect
the
context
of
positive emotional concern, then people
should
not be
satisfied
by
interactions within
the
context
of
an
ongoing relationship
or
social bond that
is not
marked
by
positive caring.
We
look
for
evidence
for the
specifically
mu-
tual
nature
of the
bond.
Need
for
r-elatedness.
Can
people
be
satisfied
by
frequent
interactions
without stable relationships? Weiss
(1973)
ob-
served
that "loneliness
is not
simply
a
desire
for
company,
any
company;
rather
it
yields only
to
very
specific
forms
of
relation-
ship"
(p.
13).
Wheeler
et
al.
(1983)
showed that loneliness
is
largely
independent
of
one's
amount
of
social contact, thereby
confirming
Weiss's observation.
In the
next section
of
this arti-
cle,
we
cover several studies showing that people seem
to
prefer
a
few
close
friendships
over
a
high number
of
transient
or
super-
ficial
encounters
and
that
evidence could
be
taken
to
indicate
that
the
relationship bond
is
essential
to
full
satisfaction.
One
possible population
of
people
who
have many interac-
tions without
the
bond
of
mutual caring would consist
of
pros-
titutes,
who may
have
a
high frequency
of
physically intimate
interactions
with
partners with whom there
is no
ongoing bond.
Sure
enough, prostitutes
often
describe their occupation
as
hav-
ing
the
benefits
of
meeting interesting people
and not
being
as
boring
as
other
jobs
(e.g., McLeod, 1982,
p.
31).
If
intimate
interactions were
sufficient
to
satisfy social needs without
any
lasting bond,
prostitutes
might
be
very happy
and
well
adjusted.
On the
contrary, however,
it
appears
that prostitutes
are far
from
satisfied
by
these interactions
and
instead seek
and
culti-
vate lasting bonds with others.
The
desire
for
bonds
of
mutual
caring
is
apparently
often
responsible
for
irrational, even
self-
destructive attachments
to
procurers
and
other
men
(Adler,
1980; McLeod,
1982;
Symanski,
1980).
Also, many prostitutes
are
single mothers,
and the
bond with
the
child
is
very
impor-
tant
(McLeod,
1982).
Several signs indicate
that
prostitutes
do
like
to
cultivate long-term relationships with clients,
as
evi-
denced
by
some brothel rules designed
to
prevent
the
formation
of
such attachments (Symanski, 1980). Indeed, Symanski
(1980)
calculated that prostitutes would maximize their
finan-
cial
earnings
by
working
in
brothels
and
serving
the
most cus-
tomers,
yet
many
specifically
objected
to the
procedures involv-
ing
many brief contacts
and
sought
to
work
in
other settings
where
they could have more time with each client
and
cultivate
repeat customers. These observations must
be
regarded
as
ten-
tative, however,
because
the
studies
lack
methodological
rigor.
Bond
of
caring.
The
next issue
is
whether
all
relationship
bonds
can
satisfy
the
need
to
belong.
It
appears that only bonds
marked
by
positive concern
and
caring
offer
satisfaction. Even
if
a
person
has
both
an
enduring bond
and
frequent
interactions,
he
or she may
feel
that
the
need
to
belong
is not
fully
satisfied.
We
turn
now to
relevant evidence involving cases
in
which
the
person
is firmly
linked
to
others
but has
unpleasant
or
unsatis-
fying
interactions with them.
Earlier,
we
listed
a
series
of
apparent benefits
of
social bonds
for
health, adjustment,
happiness,
and
general welfare.
There
is
an
important qualification, however.
In
many cases,
it is not
the
mere
fact
of
having
an
interpersonal attachment,
but
rather
having
an
attachment that brings positive interactions, that
is
decisive.
Relationships marked
by
conflictual
interactions
are
much
less beneficial
and
sometimes harmful.
DeLongis
et al.
(1988)
found
that happily married people were much healthier
than were people
in
unsupportive social relationships. Thus,
it
is
not the
mere
fact
of
marriage,
but
rather having
a
supportive
marital
relationship, that provides health benefits,
and
people
who
are
deprived
of
such
a
satisfying relationship
are
more vul-
nerable. Coyne
and
DeLongis
(1986)
reviewed evidence
and
concluded that
bad
marriages
may be
worse than being alone
in
terms
of
effects
on
happiness
and
health.
Kiecolt-Glaser
et al.
(1987)
found
decrements
in
immune function among unhap-
pily
married women
and
among women
who
were separated
from
their husbands while remaining emotionally attached
to
them. Myers's
(1992)
review
of the
literature
on
happiness con-
cluded
that whereas good marriages provide
a
powerful
boost
to
happiness,
bad
marriages lead
to
extreme unhappiness. Like-
wise,
research
on
social participation
by
Reis, Wheeler,
Kernis,
Spiegel,
and
Nezlek
(1985)
found
that
the
quality rather than
the
quantity
of
social interactions predicted health.
Specifically,
participants
(particularly
women)
who had
better quality
in-
teractions
(defined
in
terms
of
intimacy, pleasantness, satisfac-
tion, mutual disclosure, initiation,
and
influence)
fared
better
on
a
variety
of
measures
of
physical
and
mental health.
Although
the
lack
of a
good marital relationship
appears
to
be
detrimental
to
mental health,
the
existence
of a bad
marital

514
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
relationship
is
arguably worse. Having
a
spouse
or
close partner
may
preclude
the
person
from seeking other, more satisfying
and
beneficial
relationships,
and the
pervasive
and
salient
conflic-
tual
interactions
may
intensify
the
person's
feeling
of not be-
longing.
Thus,
to
complement
the
standard
finding
that good
social support
is
beneficial
for
mental health, Vinokur
and van
Ryn
(1993)
showedthat
social undermining (i.e., conflict, crit-
icism, making
life
difficult,
and
inducing
feelings
of
being
unwanted)
in
close relationships
has a
strongly negative
effect
on
mental health. Indeed,
in
their sample
of
unemployed peo-
ple,
the
effect
of
social undermining
was
stronger than
the
effect
of
social support. Carnelley, Pietromonaco,
and
Jaffe
(1994)
confirmed
the
link between problematic relationships
to
par-
ents
and
subsequent depression,
and
they also
found
that
the
current
romantic involvements
of
depressed adults tended
to be
characterized
by
fearful
avoidance
and
anxious ambivalence.
Although
their results
are
correlational, they
are
quite consis-
tent with
the
view
that problems
and
deficiencies
in
close rela-
tionships
contribute
to
depression
(with
attachment style
as a
mediating
variable).
We
also mentioned Sampson
and
Laub's
(1993)
finding
that
linked
marriage
and job
involvement
to
reduction
in
criminal
activity.
These reductions
in
crime were limited
to
people
who
had
good, stable, happy marriages
and who
were employed
in
steady
jobs.
(The
marital
and job
effects
were independent.)
In
contrast,
the
mere
fact
of
being married,
or the
level
of
one's
income,
had no
relation
to
crime. Thus, being
well
integrated
into
good relationships, rather than merely having
a
social
at-
tachment,
reduces
criminality.
Also
relevant
are
studies
on how a
good marital relationship
affects
offspring;
indeed,
for
evolutionary analyses,
these
inves-
tigations
may be
especially important. Several reviews have con-
cluded that
conflict
between parents leads
to
aggressive, antiso-
cial
behavior (such
as
juvenile delinquency)
and
perhaps other
behavior
problems
in
children (Belsky,
1981;
Emery, 1982;
Rutter
&
Garmezy,
1983).
Indeed, Emery
(1982)
concluded
that parental conflict, rather than separation,
is the
main factor
responsible
for the bad
effects
of
divorce
on
children, because
the
problems covary much more closely with
conflict
(in
either
intact
or
separated parents) than with separation. Recent work
indicates that
a
good marital relationship tends
to
cause greater
warmth toward
the
children, which
in
turn reduces angry
and
defiant
misbehavior
on the
part
of the
children
(N.
B.
Miller,
Cowan, Cowan,
Hetherington,
&
Clingempeel, 1993;
see
also
Belsky,
1979).
Mutuality.
The
last issue concerns
how
important
it is
that
caring,
concern,
and
affection
be
mutual
and
reciprocal.
One
can
well
understand
why
people
are
better
off
to
interact with
partners
who
care about them, because
the
partners might pro-
vide
more material rewards
and
other benefits.
But is
there
any
value
to
caring about
the
other person,
as
opposed
to
being
merely
cared about?
In
the first
place,
it
does
appear
that mutuality
is the
norm.
M.
S.
Clark
et
al.
(1987)
showed that
the
desire
to
receive help
from
others
was
correlated with
the
desire
to
give help
and re-
spond
to
others' needs. This suggests that
the
desire
for
commu-
nal
relationships
is
based partly
on the
appeal
of a
framework
in
which people have mutual concern
for
each
other's
welfare.
The
alternative explanation
for
Clark
et
al.'s
findings
would
be
based
on
social exchange theory, which would propose that peo-
ple
might prefer
to be
involved
in
one-way relationships,
so
that
they
would receive
the
benefits
of the
other
person's
care
but not
incur
the
costs
of
having
to
care
for the
other person. Perhaps
mutuality
is the
norm only because people cannot
find
others
who
will
care
for
them without getting anything
in
return.
The
evidence
runs contrary
to
this
view,
however, despite
its
eco-
nomic
and
utilitarian logic. Hays
(1985)
examined relationship
satisfaction
as a
function
of the
costs
and
benefits
to the
individ-
ual. From
a
behavioristic standpoint,
he
predicted that satisfac-
tion would
be
predicted
by an
index
of the
rewards minus
the
costs, which
is
precisely what economic rationality would
favor.
Contrary
to
that prediction, however, Hays
found
that satisfac-
tion
was
much better predicted
by an
index
of
rewards plus
costs.
In
other words, people preferred relationships
in
which
both parties gave
and
received care.
Mutuality
seems
to
improve
and
strengthen
the
relation-
ship. Rusbult, Verette,
and
Drigotus
(1994)
found
that
mu-
tuality
of
commitment predicted good marital adjustment.
This
effect
was
independent
of the
actual level
of
commit-
ment,
which shows that mutuality
per se is
indeed beneficial.
The
other side
of
this
was
demonstrated
by
Hill, Rubin,
and
Peplau
(1976),
who
showed that unequal involvement
was a
strong predictor
of
romantic breakup. Moreover,
it was not
simply
the
case that
the
less involved partner
was
more likely
to
break
off
the
attachment, because
in
many cases
the
more
involved
person initiated
the
breakup. Only when both
part-
ners reported that both were equally involved
was the
couple
likely
to
still
be
together
2
years later.
If
mutuality
is
good
for
relationships,
it is
also good
for
indi-
viduals,
as
indicated
by
recent
findings
from
studies
of
unre-
quited love (Baumeister
&
Wotman, 1992; Baumeister, Wot-
man,
&
Stillwell, 1993). These studies compared people
who
received
love without giving
it and
people
who
gave love without
receiving
it. To the
researchers' surprise, both groups tended
to
describe
the
experience
as
aversive.
Apparently, love
is
highly
satisfying
and
desirable only
if it is
mutual.
The
parent-child
bond
is
inevitably asymmetrical, insofar
as
the
child cannot provide
the
parent with
the
nurturant care
and
concern that
the
parent must provide
the
child.
If
there
are any
exceptions
to the
principle that mutuality
is
optimal, they
would
presumably
be
found among parents.
The
difficulty,
of
course,
is
determining what
is the
appropriate comparison.
One
strategy would
be to
compare mutual
and
nonmutual two-per-
son
families, that
is,
compare
families
consisting only
of two
adults
(i.e.,
childless marriages)
and
families consisting only
of
a
parent
and
child (i.e., single
parents).
Research
has
abun-
dantly shown that those
two
types
of
families
differ
dramatically
in
terms
of
happiness
(of
the
adult):
The
childless spouses
are
happier than average,
and the
single parents
are
less happy than
average.
In
other words,
if an
adult woman
is to
have only
one
other person
in her
family,
she
will
be
happier
if
this person
is a
husband rather than
a
child (e.g.,
S. A.
Anderson
et
al.,
1983;
Baumeister,
1991;
Bernard, 1982; Campbell,
1981;
Campbell
etal.,
1976).
One
reason
for the
importance
of
mutuality
may be
trust.
J. G.
Holmes
and
Rempel
(1989)
reviewed evidence that trust
is
often
a
crucial
and
influential
feature
of
good, beneficial,
and
satisfying
relationships
and
concluded that trust depends

THE
NEED
TO
BELONG
515
heavily
on
mutuality, especially
the
mutual recognition
of re-
ciprocal concern
and
attachment. Dissimilar
feelings
and un-
equal
involvement prevent
the
growth
of
trust
and
thereby
thwart
or
weaken relationships.
Critical
assessment.
The
evidence
in
this section
was un-
even
in
quality
and
quantity.
We
have found
no
methodolog-
ically
rigorous evidence indicating that frequent interactions
without
an
ongoing
relationship
offer
partial
or
intermediate
satisfaction
of the
need
to
belong.
The
prostitution studies
were
merely correlational
and
impressionistic, and, even
if
they
had
been based
on
systematic samples
of
prostitutes
(which
they were
not),
one
would
be
reluctant
to
generalize
from
prostitutes
to the
rest
of the
population.
In
contrast,
the
evidence
is
stronger with regard
to the
inadequacy
of
negative
or
conflictual
interactions
to
provide
satisfaction.
Although much
of
this evidence
is
correlational,
there
is
some time-sequence evidence suggesting that
un-
happy
marriages
and
other problematic relationships lead
to
distress
and
illness.
The
evidence
for
mutuality
is
scattered
and
fragmentary,
al-
though
it is
consistent. Most
of it is
somewhat indirect. Further
research
is
needed
to
provide direct evidence about
the
impor-
tance
of
mutuality,
particularly whether
one's
own
caring
for
the
partner
is
important
for
satisfying
one's
own
need
to
belong.
Conclusion.
First, there
is
some evidence that interactions
with
a
changing series
of
partners, without
any
ongoing rela-
tionship
bond,
fail
to
satisfy
people,
but
this evidence
is
sug-
gestive
rather than conclusive. Second, several studies have
indicated
that problematic
or
unhappy marriages
fail
to
pro-
duce
the
benefits normally linked
to
belongingness
and,
in
fact,
may
make things worse. Thus,
the
mere
fact
of a
social
bond
is not
enough
to
protect people
from
these problems
and
pathologies. Rather,
it
appears that people require their
primary
social bonds
to be
characterized
by
affectively
posi-
tive
interactions that
signify
the
other's
affectionate
concern.
Third,
there
are
several indications
that
people prefer rela-
tionships
marked
by
mutual, reciprocal concern,
but
stronger
and
more direct evidence
is
needed.
It is
also plausible that
mutuality
is
merely
a
preference rather than
a
need.
Satiation
and
Substitution
The
belongingness hypothesis holds that individuals need
a
certain
amount
of
social relatedness. Social relationships
and
partners should therefore
be to
some extent interchangeable.
Moreover,
people
who
have
sufficient
social bonds
to
satisfy
the
need
to
belong should
be
less interested
in
forming
additional
relationships
than people
who do not
already have
sufficient
bonds. These corollaries
of the
belongingness hypothesis
can be
expressed
in
terms
of
satiation
and
substitution. Satiation
refers
to the
diminished motivation that ensues when
the
need
to be-
long
is
already well satisfied,
and
substitution refers
to the re-
placeability
of one
social bond with another. Satiation
and
sub-
stitution
are not
unrelated,
of
course, because both invoke
the
basic assumption that people need
a
certain quantity
of
belong-
ingness,
and
attachments
or
interactions beyond that minimum
should
be
subject
to a
pattern
of
diminishing
returns.
Satiation implies
a
diminishing returns principle
in the
pur-
suit
of new
relationships
and
partners.
Even
in
people-rich
en-
vironments
such
as
colleges, people
appear
to
restrict their
so-
cial
lives
to
some extent. Studies show that
the
vast majority
of
the
average
student's
meaningful interactions
are
with
the
same
six
people (Wheeler
&
Nezlek, 1977). Reis (1990) surveyed
students about their interpersonal goals,
and
although "having
lots
of
friends" received
one of the
lowest ratings, most
of the
top-rated
items referred
to
intimate sharing with
a few
close,
caring friends.
Caldwell
and
Peplau
(1982)
found
that
a
strong
majority
of
both
men and
women expressed
a
clear preference
for
a few
close friendships over
a
large number
of
good
but
less
intimate friendships. Thus, people appear
to
devote their time
and
efforts
toward deepening
a
limited number
of
relationships
rather than toward meeting ever
new
people
or
cultivating
a
wider
range
of
acquaintanceships. Consistent with
the
satiation
hypothesis, people seem
to
believe that,
in
terms
of
friendships,
quality
(closeness)
is far
more important than quantity.
Audy
(1980)
suggested that this satiation
is
more
or
less
es-
sential
if a
species
is to
survive. Organisms evolve
a
"limited
requirement
for the
frequency
of
social
transactions
and a
cor-
responding optimum group
size"
that permit
a
maximum
of
social
gratification balanced
by
socially induced frustration
(pp.
123-124).
As he
noted,
there
is
evidence that people have
evolved
"a
physiological structure
and
basic mental require-
ments
suited
to a
particular group size that corresponds
to
[their]
need
for a
certain level
of
social
transactions"
(p.
124).
Satiation patterns,
in the
form
of
diminishing
effects
of
social
approval
as
reinforcement, have also been investigated
in the
context
of
learning theory. Gewirtz
and
Baer
(1958)
replicated
the
standard pattern that children's task performance would
improve
in
response
to
verbal approval reinforcers such
as
praise
and
other approving remarks; moreover,
the
reinforce-
ment
effect
was
intensified
if
the
children
had first
been deprived
of
social approval
by
being kept
in
isolation
for a
brief period.
In
another condition, however,
the
children were
first
given
an
interview
in
which they received praise
and
admiration
for
whatever
they said about themselves.
After
this interview,
the
standard praise
and
approval remarks
failed
to
elicit
improve-
ments
in
task performance, which suggests that these partici-
pants
had
been satiated with approval
and
were
unaffected
by
further
doses.
Eisenberger
(1970)
reviewed
the
subsequent stud-
ies
on the
same topic
and
found
that
the
initial
results
were well
replicated. Moreover, these
effects
were
not a
result
of
sensory
deprivation,
and
they also
failed
to
alter
the
subsequent respon-
siveness
to
nonsocial rewards. Eisenberger concluded that social
deprivation
and
social
(approval)
satiation
effects
operated
by
altering
short-term motivation
for
obtaining approval.
Al-
though
these studies were generally conducted with children
and did not
involve lasting relationships, they
do
indicate that
the
motive
to
gain social approval
is
susceptible
to
satiation.
Social interaction patterns that accompany
the
formation
of
an
intimate romantic relationship
are
especially relevant,
be-
cause
both
satiation
and
substitution
are
implicit.
Milardo,
Johnson,
and
Huston
(1983)
found
that
as an
intimate relation-
ship
develops, people reduce
the
amount
of
time they spend
in-
teracting with other people, including
old
friends.
Thus,
the ro-
mantic
relationship
appears
to
supplant
the
others
and
satisfy
the
belongingness need previously satisfied
by the
other
friendships.
The
belongingness hypothesis
is, of
course,
not
limited
to the

516
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
mere
existence
of
some
formal
attachment
but
also depends
on
the
quality
of the
relationship. Consistent with this, Berman's
(1988)
research
on
attachment
to
ex-spouses
found
that
the
positive
qualities
of the
relationship were important determi-
nants
of the
distress over
the
loss,
as
indicated
by the finding
that people
who had
more favorable memories
of the
marriage
also
had
more distress
after
it
ended.
For
present purposes,
the
important
implication
is
that
if bad
marriages
fail
to
satisfy
the
need
to
belong, then,
as a
result, they should stimulate
a
search
for
new
attachments.
Vaughan
(1986)
observed that when mar-
riages begin
to
develop
significant
problems that
will
eventually
lead
to
their dissolution,
the
individuals
often
begin
to
seek
out
new
friendships
and
relationships. Along
the
same lines, Law-
son's
(1988)
research
on
adultery
found
evidence that substitu-
tion
can be an
important
factor,
particularly
for
women.
She
found
that
the
reason most commonly cited
by
women
for en-
gaging
in
extramarital
sex was the
husband's
failure
to
satisfy
the
wife's
intimacy needs.
(For
husbands, other factors such
as
sexual
novelty
and
variety were
influential,
but
these factors
are
irrelevant
to the
belongingness
hypothesis.)
The
implication
is
that
when
the
marital relationship satisfies
the
need
to
belong,
women
are
unlikely
to
seek extramarital relationships,
but
when
the
marriage
is not
satisfactory, extramarital substitutes
may
be
sought. This conclusion supports both
the
satiation
and
substitution
hypotheses.
Spanier
and
Casto
(1979)
found
that most people relied
heavily
on
(and
benefited
from)
social support
from
friends
and
family
during divorce. When friends
and
family
were
not
sup-
portive,
however,
"this
lack
of
support seemed
to
increase
the
overall
difficulties
in
adjusting
to the
separation, especially
the
emotional
adjustments"
(p.
217).
Spanier
and
Casto also noted
that
a
failure
to
make
new
friends
made
the
adjustment worse.
In
a
direct test
of the
hypothesis that more social interaction
will
lead
to
less adjustment problems
after
divorce, they found
a
strong relationship between social activity
and
adjustment
problems. They also
found
that
forming
new
heterosexual
or
romantic
relationships
eased
the
transition
of
divorce
and led to
far
fewer
difficulties
of
adjustment. When
new
relationships
fail
to
form,
the
emotional distress associated with
the
divorce
and
the
ex-spouse
may
actually increase
rather
than decrease over
time
(Spanier
&
Casto,
1979,
p.
226),
which again implies that
substitution
is an
effective
way of
recovering
from
relationship
dissolution.
A
very
different
source
of
evidence
for the
same
conclusion
is
Bowlby's(
1969,1973)
observation that children's
anxiety
and
distress over separation
from
the
mother seemed
to
be
greatly reduced
if
the
children were accompanied
by a
famil-
iar
other person
at
that time.
Populations
of old
people
offer
a
useful
way to
examine
progressive
social deprivation, because
in
many cases
old
people have retired
from
work,
are
losing spouses
to
death,
and
cease
to
make
new
intimate friendships
(e.g.,
Kaufman,
1986).
Like Kaufman,
L. J.
Beckman
(1981)
found
that
old
women's relationships
to
adult children
had
become
increas-
ingly
important
to
their lives.
The
happiness
of old
women
with
children
was
unrelated
to the
amount
of
social
interac-
tions
with other people; among childless
old
widows, however,
happiness
in
life
was
significantly
correlated with
the
quality
and
quantity
of
social interaction with other people. Thus,
the
rewards
of
social interaction with children
appear
to be
"exchangeable
and
interchangeable"
(L. J.
Beckman,
1981,
p.
1085) with
the
rewards
of
interacting with other people.
Similarly,
older adults
who
have
a
close
friend
are no
more
likely
to
become depressed
if, for one
reason
or
another,
the
amount
of
social interaction they have with other people
de-
creases.
In
contrast, older adults without
a
confidant
who de-
crease their interactions with others
are at a
very high risk
for
depression (Lowenthal
&
Haven,
1968).
These results sup-
port
the
view
that people need some social attachments
to be
happy
and
that these attachments
are to
some extent inter-
changeable.
In
particular, close relations with nonrelatives
can
apparently substitute
for
relationships
with
offspring,
at
least
in
terms
of
preventing
any
significant loss
of
happiness.
L.
J.
Beckman
(1981)
also obtained
findings
relevant
to
the
satiation hypothesis.
She
found that
the
total amount
of
social interaction with others
was a
significant predictor
of
happiness among childless women
but not
among
old
women
who
did
have children,
and she
suggested that restriction
of
range
may
account
for
this
differential
predictability. Spe-
cifically,
according
to
Beckman, most
old
women with chil-
dren
do
have
at
least
a
certain minimal level
of
social interac-
tion, provided
by the
children,
and so
these women hardly
ever
fall
into
the
category
of
extreme loneliness
and
social
deprivation. Although Beckman repeatedly found that, iron-
ically,
interactions
with
nonoffspring
had a
bigger impact
on
happiness than interactions with children, having children
visit
occasionally seemed
to be
enough
to
satisfy
the
need
to
belong
sufficiently
to
prevent
the
most severe problems
of de-
privation.
Above that minimum,
further
quantity
of
social
interaction
did not
appear
to
have
an
effect.
Substitutability
patterns were suggested
in a
very
different
way
by
Rusbult, Zembrodt,
and
Gunn
(1982;
see
also Rusbult,
1980). They suggested that people remain
in
their close rela-
tionships
for
several reasons,
and one
important factor
is the
availability
of
desirable alternative potential partners.
In
other
words, people
are
more
likely
to
leave
an
intimate relationship
if
they have some
prospect
of
forming
another intimate rela-
tionship
with
someone else soon.
Divorced people
are at
risk
for a
multitude
of bad
outcomes,
including
illness, homicide, suicide, criminality,
and
accidental
injury
or
death (Bloom
et
al.,
1979).
One
explanation
is
that
the
divorced population represents
a
self-selected group
of
pathologically
inclined individuals,
but
this
dispositional
argu-
ment
is
weakened
by the finding
that remarriage tends
to
reduce
or
eliminate
the
elevation
in
risk.
The
trauma
of
divorce itself
may
be
partly responsible
for the
increase
in
vulnerability,
but
although
the
risks
are
highest immediately
after
divorce, they
do not
fully
subside until remarriage.
The
fact
that remarriage
appears
to
eliminate many negative consequences
of
divorce
can
indeed
be
explained
in
several ways,
but it is at
least strongly
consistent with
the
hypothesis that
the new
marriage substitutes
for
the old
one.
Divorce
may
often
be
voluntary,
but
imprisonment
is
not,
and
prisoners
suffer
deprivation
of
contact
with relationship
partners
outside
the
prison.
Men's
prisons
are
physically dan-
gerous,
and
both
the
gang bonding
and the
cultivation
of
social
isolation
commonly found among male prisoners
may
reflect
more
a
concern with physical
safety
than anything
else.
In
con-
trast,
female
prisons
are far
less dangerous
to
inmates,
and so it

THE
NEED
TO
BELONG
517
is
relevant
to
examine
how
women prisoners adapt
to the
depri-
vation
of
contact with outside partners. Several studies have
found
that female prisoners commonly
form
substitute
families
based
on
imaginary kinship ties with other prisoners
(Burkhart,
1973; Giallombardo, 1966;
Toch,
1975). Some
pseudomarital
bonds appear
to
involve voluntary homosexuality, which
is
typ-
ically
renounced
on
leaving
the
prison (Burkhart,
1973).
In
some cases, these pseudofamilies become quite extensive
and
complex,
with many women playing
parts
of
parent, child, hus-
band,
wife,
grandparent,
and
other roles
for
each other. These
patterns appear
to be
simply
an
adaptation
to
prison
life
based
on
the
desire
to
experience romantic
and
family-style
belong-
ingness
during
the
time
one is
deprived
of
contact with
the
usual
relationship
partners.
As
we
noted earlier, when groups break apart
or
people move
away
from
relationship partners, there
is
often
initially strong
resistance
to
dissolving
the
relationship,
but
this resistance
tends
to
diminish over time
(e.g.,
Lacoursiere,
1980;
Lieberman
et
al.,
1973).
These
efforts
to
maintain
the
bond
may be
driven
by
the
absence
of
social
ties
in the new
environment,
and as
people gradually
form
new
attachments they lose
the
need
to
sustain
the old
ones.
If
this
is
correct, then
future
research
should
find
that people's
efforts
to
sustain friendships across
long
distances
are
inversely proportional
to
their opportunities
to
develop
new
friendships.
For
example, when people move
overseas, where cultural
differences
may
hinder
the
develop-
ment
of new
intimate friendships, they should
be
more
likely
to
stay
in
touch
with
old
friends
than when moving
to
another
place
in the
same country.
Not all
relationships
are
interchangeable,
of
course. Close
re-
lationships
based
on
romantic love
may
offer
a
variety
of
satis-
factions
that
are not
easily obtained through nonromantic, non-
sexual
friendships. Ruehlman
and
Wolchik
(1988)
found
that
there were indeed particular benefits connected with
the
rela-
tionship
to the
most significant other
in a
person's
life.
More
precisely,
they
found
that once
the
social support
and
hindrance
provided
by the
most
significant
other person
in
someone's
life
were
taken into account, there were
no
additional
significant
effects
of the
support
and
hindrance provided
by
other people.
This
pattern
of
results suggests that people need
at
least
one
particularly
strong, close attachment
and
that once they have
that,
further
attachments
are
subject
to
some principle
of di-
minishing
returns.
A
similar point
was
made
by
Coyne
and
DeLongis
(1986),
who
concluded,
from
a
review
of the
social
support literature, that
the
harmful
effects
of a bad
marriage
are not
offset
by
having other good relationships; thus, again,
the
special importance
of the
marital bond
was
confirmed.
Likewise,
although women prisoners adapt
to
prison
by
sus-
pending
their emotional attachments
to
most outsiders
and
forming
substitute
family
relationships with other prisoners,
they
do
exert themselves extensively
to
maintain
the
bond with
their
real children
who
remain outside
the
prison
(Baunach,
1985).
And of
course,
as
Kazan
and
Shaver
(1994a)
have
pointed out, although children could conceivably
affiliate
with
any
available person, they nearly
always
focus
on one
particular
person,
and
their need
to
interact
(as
evidenced,
in
part,
by
dis-
tress over
separation)
becomes mainly focused
on
that
person.
Critical
assessment. There
is an
assortment
of
evidence
consistent
with
the
hypotheses
of
satiation
and
substitution,
but
the
evidence
is
neither systematic enough
nor
unambiguous
enough
to
regard those hypotheses
as
strongly supported. Thus,
the
fact
that
forming
a
close romantic attachment leads
to
with-
drawal
from
other friendships could
be
partly
due to
having
a
limited amount
of
time
to
spend socializing rather than
to any
reduction
in
need
for the
other friendships. Likewise,
the
culti-
vation
of
external friendships when
a
marriage goes
bad
could
be
due to a
need
to
discuss
the
marital problems with
a
sympa-
thetic outsider rather than
a
quest
to find a new
social
bond that
could
furnish
what
the
marriage
no
longer provides. Although
the
diversity
of
spheres yielding consistent
findings
encourages
one to
expect that
further
evidence
will
continue
to fit the
satia-
tion
and
substitution hypotheses, more systematic work
is
needed
to
rule
out
alternative explanations.
Also,
there appear
to be
limits
on the
extent
to
which rela-
tionships
can be
substituted.
A
close romantic attachment
to a
partner, with sexual attraction,
appears
to
have special benefits
that cannot
be
compensated
by
other relationships.
Still,
when
such
a
relationship ends, forming
a new one
appears
to be
sufficient
to
bring
the
person
back
to an
equally high level
of
adaptation
and
happiness, which suggests that,
in the final
anal-
ysis,
a new
spouse
may be an
effective
substitute
for a
previous
one.
Conclusion.
People's interaction patterns
and
surveys
of
preferences
suggest that people seek
a
limited number
of
rela-
tionships,
consistent with
the
view
that
the
need
to
belong
is
subject
to
satiation
and
diminishing returns.
The first
few
close
social bonds appear
to be the
most important, beyond which
additional ones
furnish
ever lesser benefits. When people lose
such
bonds
or find
their particular partners inadequate, they
can
often
derive similar benefits
from
others, suggesting that
partners
can be
substituted
to
some extent. There
are
certain
kinds
of
relationships that cannot
effectively
be
replaced with
other
kinds
of
relationships, although
finding a new
relationship
of
the
same type appears,
in
many cases,
to be
viable
and
effective.
These conclusions
are
tentative, however,
and
further,
more systematic work
is
desirable.
Innateness,
Universality,
and
Evolutionary
Perspectives
We
proposed that
a
fundamental need would presumably
be
innate, which would
entail
that
it is
found
in all
human beings
and
is not
derivative
of
other motives. This
will,
of
course,
be
quite
difficult
to
verify,
because empirical criteria
for
testing
such
a
hypothesis
are not
widely recognized.
One
approach,
however,
would
be to
examine
how
well
the
empirical evidence
conforms
to
evolutionary arguments.
If
evolution
has
instilled
the
motivation, then
it is
presumably universal among human
beings
and
will
be
present
in
each
person
without needing
to be
derived
from
other motives.
Barchas
(1986)
has
asserted that "over
the
course
of
evolu-
tion,
the
small group became
the
basic survival strategy devel-
oped
by the
human
species"
(p.
212).
He
went
on to
suggest
that
the
brain
and
small groups evolved
and
adapted together,
with
multiple interrelationships.
The
evidence reviewed
by
Bar-
chas remains preliminary,
but it
does seem that
any
link
be-
tween
brain structures
and
small-group formation would
strengthen
the
case
for an
innate motivation.
Although
the
psychobiological
systems involved
in
social
at-

518
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
tachment
are not yet
well
understood, early evidence implicates
the
brain opioid system. According
to
Panksepp,
Siviy,
and
Nor-
mansell
(1985),
both
the
tendency
to
form social bonds
and
the
emotional
effects
of
social loss (e.g., sadness
or
grief)
are
mediated
by
opioids.
The
formation
and
validation
of
relation-
ships
apparently stimulate opioid production, whereas
the
dis-
solution
of
relationships impedes
it. As
Panksepp
et
al.
put it,
"social
affect
and
social bonding
are in
some fundamental neu-
rochemical sense opioid
addictions"
(p.
25).
Thus,
in
their
view,
the
tendency
to
seek social connections with others
is
based
not
only
on the
secondary reinforcements that other peo-
ple
provide
but on
psychophysiological mechanisms
as
well.
Multiple
evolutionary reasons could
be
suggested
for the
readiness
to
form
groups easily. Groups
can
share labor,
re-
sources,
and
information;
diffuse
risk;
and
cooperate
to
over-
come stress
or
threat
(Hogan
et
al.,
1985).
Defense against rival
groups would also
be a
significant
factor:
If
other people
form
into groups, lone individuals would
be at a
competitive disad-
vantage
in
many situations,
and so
evolution
may
have selected
for
people
who
would
form
groups
defensively.
Hence,
the
evo-
lutionary
argument would
fit any
evidence that group forma-
tion
or
cohesion patterns
are
increased
by
external threat.
It
has
long been noted that external threats increase group
cohesion,
and
some writers have treated this
as
axiomatic. Stein
(1976)
reviewed these
views
in
light
of the
evidence
and
found
that
a
broad variety
of
methods have yielded generally consis-
tent
findings;
that
is,
external threats
do
increase cohesion most
of
the
time. There
are
some circumstances
in
which groups dis-
integrate under threat, especially
if the
threat pertains only
to
some members
of the
group
or
if
group members must compete
against each other
to
survive
the
threat (e.g.,
if
there
are too
few
lifeboats). Staw, Sandelands,
and
Dutton
(1981)
also
found
evidence
that group cohesion
is
sometimes weakened
in the af-
termath
of a
threat, especially
if the
group
has
failed
to
defeat
the
threat
and the
group members blame each other. Apart
from
these circumscribed exceptions, however,
it is
safe
to
conclude
that external threats
do
generally increase group cohesion.
A
remarkable demonstration
of the
power
of
external threat
to
forge
lasting bonds
was
provided
by
Elder
and
Clipp's
(1988)
study
of
World
War II
veterans' groups.
In
Elder
and
Clipp's
results,
the
effects
of
maximum
threat
were discernible
40
years
later.
That
is,
four
decades
after
the
war,
the
most enduring
and
strongest
ties
were
found
among veterans
who had
experienced
heavy
combat together
and had
suffered
the
deaths
of
some close
comrades. Units that
had
experienced combat without fatalities
were
less close
40
years later,
but
they retained stronger ties than
the
units that
had not
been
in
combat together.
In
other words,
the
sharing
of
military experience provided some lasting bonds,
these bonds were
intensified
by
shared experience
of
combat,
and
they were especially strong
if it had
been heavy combat that
had
killed some members
of the
group.
It
seems clear
that
there
would
be
survival benefits
to a
pattern
in
which
the
death
of a
group member strengthened
the
ties among
the
survivors, espe-
cially
in the
face
of
external danger.
The
group formation
effects
in the
Robbers Cave study
(described
earlier; Sherif
et
al.,
1961
/1988)
accelerated
rapidly
after
the
mutual discovery
of the
existence
of the two
rival
groups; that
is, the
implicit threat posed
by the
opposing
group
seemed
to
motivate each
boy to
cling
to his own
group
more
strongly.
Similar processes have been observed
in
terrorist
groups, which mainly become cohesive
in the
face
of
external
threat
and
danger. During periods when
the
conflict
with out-
siders lapses, terrorist groups experience internal dissent
and
conflict
and may
fall
apart
(see
McCauley
&
Segal,
1987).
Compelling
evidence
in
favor
of
emphasizing
the
competitive
disadvantage
motive
for
affiliating
was
provided
by
Hoyle,
Pinkley,
and
Insko
(1989).
These researchers noted
the
irony
that encounters between individuals
are
generally pleasant
and
supportive, whereas encounters between groups
are
frequently
unpleasant
and
confrontational,
and
their
first
study
confirmed
these general expectations
and
stereotypes empirically.
In
their
second study, they sought
to
determine
the
decisive factor
by
comparing interactions between persons, between groups,
and
between
one
person
and one
group.
To
their surprise, they
found
that participants' expectations about
the
interaction were deter-
mined
mainly
by the
other party rather than
by
participants'
own
belongingness status. When participants expected
to in-
teract with
a
group, they expected
an
abrasive interaction; when
they
expected
to
interact
with
an
individual, they anticipated
a
pleasant, agreeable interaction. Identical
effects
were
found
regardless
of
whether
the
participant expected
to be
alone
or to
be
part
of a
group. Thus, apparently,
the
presence
of an
out-
group causes people
to
anticipate
conflict
and
problematic
in-
teractions. Such
an
expectation could well elicit
a
motivation
to
form
a
group
to
protect oneself.
A
similar
conclusion
was
suggested
by
Lauderdale,
Smith-
Cunnien,
Parker,
and
Inverarity
(1984).
Following Schachter's
(1951)
studies
on
group rejection
of
deviants, they
found
that
increasing
an
external threat
led to
increased rejection.
The im-
plication
was
that groups become increasingly oriented toward
solidarity
when confronted with
an
external threat.
Apart
from
threat,
the
possibility
of
gaining resources also
seems
to
trigger group cohesion, even when
it is
functionally
irrelevant. Rabbie
and
Horwitz
(1969)
assigned participants
randomly
to two
groups.
The
random assignment alone yielded
no
effects
of
group cohesion
on
their measures
of
in-group pref-
erence,
but
they
did find
significant
effects
after
a
manipulation
in
which
one
group
was
given
a
prize (transistor radio) based
on
a
coin
flip. The
rewarded group
and the
deprived group both
showed increased in-group preference.
The
prize
was
logically
irrelevant
to
subsequent group activities
and
preferences.
The
implication
is
apparently that
the
combination
of
limited
re-
sources
and
multiple groups triggers
an
in-group preference
re-
sponse that
has no
apparent practical
or
rational basis, which
is
consistent with
the
view
that
it is a
deeply rooted
and
possibly
innate
tendency rather than
a
strategic
or
rational choice.
Critical
assessment.
The
evidence linking external threat
to
increased group cohesion
is
convincing
but
does
not
prove
an
evolutionary hypothesis
of
innateness
or
universality.
The
evi-
dence
for
brain mechanisms
is
likewise supportive
but
inade-
quate
to
prove innateness.
The
evidence
in
this section
is
per-
haps
best
described
by
stating that
the
evolutionary hypothesis
nicely
survived several
tests
that could have
contradicted
it.
Conclusion.
Several patterns seem consistent with evolu-
tionary reasoning.
It
remains plausible (but
unproven)
that
the
need
to
belong
is
part
of the
human biological inheritance.
If so,
the
case
for
universality
and
nonderivativeness would
be
strong.

THE
NEED
TO
BELONG
519
At
present,
it
seems
fair
to
accept these hypotheses
as
tentative
working
assumptions while waiting
for
further evidence.
Apparent
Counterexamples
Although
the
evidence presented thus
far has
been largely
supportive
of the
belongingness hypothesis,
one
might object
that
our
literature search
has
been structured
in
ways that pre-
disposed
it
toward just such favorable indications.
It is
therefore
desirable
to
examine behavioral patterns that would seemingly
constitute boundary conditions
or
counterexamples
to the
need
to
belong. This section
briefly
considers several.
Refusal
to
help
or
cooperate.
People generally show
a
sig-
nificant
willingness
to
help others,
but
often
there
may be
self-
interested
motives
lurking
behind
the
apparent altruism (e.g.,
Cialdini,
Darby,
&
Vincent,
1973;
Manucia,
Baumann,
&
Cial-
dini,
1984).
To be
sure,
in
many cases people appear
to put
self-
interest
ahead
of the
welfare
of
others, leading them
to
disdain
opportunities
for
helping others
or
cooperating. Entering into
the
long-running debate about
the
possibility
and
reality
of
truly
altruistic behavior
is
beyond
the
scope
of
this article;
our
goal
is
merely
to ask
whether such cases
do
indeed contradict
the
belongingness hypothesis.
In
particular,
it is
necessary
to ask
whether
these
nonhelpful,
noncooperative behavior patterns
are
reduced
or
eliminated
by
belongingness.
One of
social psychology's best-known
findings
concerns
the
unhelpfulness
of
multiple bystanders
at an
emergency site.
As
Darley
and
Latane's
(1968)
study
and
many subsequent inves-
tigations
(see Latane
&
Nida,
1981)
showed, people tend
not to
come
to the aid of an
emergency victim
when
many other people
are
also
present. Among
the
reasons
for the
bystander
effect
are
the
sense that
it is not
one's
own
responsibility
to
help
and the
fear
that
helping
may
bring negative consequences
to the
self.
Various
findings
suggest, however, that belongingness
can
over-
come
the
nonresponsiveness
of
bystanders.
The
bystander
effect
is
apparently robust among strangers (e.g., Darley
&
Latane,
1968),
but in
cohesive groups,
the
opposite pattern
is
found,
namely
that larger groups produce more helping
(Rutkowski,
Cruder,
&
Romer,
1983).
Even
the
mere anticipation
of
future
interaction
among group members
is
enough
to
eliminate
the
bystander
effect,
making
group members quite
willing
and
likely
to
come
to
each other's
aid
(Gottlieb
&
Carver,
1980).
Social
loafing
is
another pattern
in
which people
put
self-in-
terest ahead
of
cooperative concern
for
others
(e.g.,
Latane, Wil-
liams,
&
Harkins, 1979).
In
social
loafing,
people reduce their
efforts
when submerged
in the
group, thereby gaining benefits
of
the
group success without having
to
exert themselves maxi-
mally.
Group membership
can
foster
a
sense
of
duty
or
obliga-
tion
that
can
effectively
override tendencies
to
engage
in
social
loafing,
however. Harkins
and
Petty
(1982)
showed that
if
peo-
ple
believe that they
can
make
a
unique contribution
to the
group, they
do not
engage
in
social
loafing,
even
if
individual
contributions
to the
group
will
not be
identified
(and
thus even
if
they
will
not
receive credit
for
their contribution;
see
also
Hardy
&
Grace,
1991).
The
prisoner's dilemma game
has
been widely used
to
exam-
ine
how
people choose between
a
self-interested, individualistic
(competitive)
response
and a
cooperative response that
can po-
tentially
maximize everyone's collective outcomes
at the
cost
or
risk
of
individual
vulnerability
to
loss.
Once again,
the
presence
or
apparent
possibility
of
social
attachments
seems
to
shift peo-
ple
away
from
the
self-oriented mode toward
a
more coopera-
tive,
collectively beneficial mode
of
response.
The
expectation
of
future
interaction increases
helpful
cooperation
in the
pris-
oner's dilemma game, although this
effect
appears
to
obtain
mainly
among high self-monitors (Danheiser
&
Graziano,
1982).
The
opportunity
to
meet
and
talk with strangers
appears
to be
sufficient
to
alter responses
to a
subsequent
prisoner's
di-
lemma game
in
favor
of
increased cooperation
and
decreased
exploitation-defensiveness
(Orbell
et
al.,
1988).
Lastly,
the
commons dilemma
(in
which people deplete
renewable resources
for
short-term individual gain)
is an-
other pattern
in
which people typically seek personal advan-
tage
at the
expense
of
long-range collective welfare.
The
com-
mons dilemma also
can be
reduced
or
overcome
by
belong-
ingness, however. Kramer
and
Brewer
(1984)
showed that
when
belongingness
is
stimulated
by
making
the
group iden-
tity salient, people
are
more likely
to
restrain their self-inter-
ested tendencies
and
instead cooperate with others
for the
greater good
of the
group.
More generally, helping appears
to be
increased
by the
exis-
tence
of
social bonds. Schoenrade, Batson, Brandt,
and
Loud
(1986)
found
that
the
existence
of a
social relationship
in-
creases
the
motivation
for
helping.
In the
absence
of a
relation-
ship, people help
only
for
egoistic reasons (i.e.,
self-interest);
when
a
relationship exists, however, people
will
help
for
rela-
tively
selfless,
altruistic reasons
(see
also
Toi &
Batson,
1982).
Even
among strangers,
familiarity
leads
to
increased helping,
as
does
a
sense
of
interpersonal dependency (Pearce, 1980).
The
fact
that
belongingness
can
overcome self-interested
patterns
is
shown
by
evidence that people
prefer
reciprocity
in
social
ex-
change
to the
extent that even
overbenefited
individuals some-
times
feel
uncomfortable
and
distressed even though material
self-interest
is
maximally served under conditions
of
being
overbenefited
(Rook, 1987a).
The
concern with equity
and
with
aiding others
is
further
indicated
by the
occasionally nega-
tive
responses
of
would-be helpers
to
having their
helpful
efforts
spurned
by the
intended recipients
(e.g.,
S.
Rosen, Mickler,
&
Collins,
1987).
Throughout this article,
we
have suggested
mat the
need
to
belong
may be
biologically
prepared.
Evidence with animal spe-
cies
is
therefore relevant here. Masserman, Wechkin,
and
Terris
(1964)
taught rhesus monkeys
to
pull
a
chain
for
food
and
then,
in
one
condition, added
the
contingency that pulling
the
chain
would
cause
a
shock
to be
delivered
to
another monkey. Most
monkeys
refrained
from
pulling
the
chain under those condi-
tions, even
to the
extent
of
starving themselves
for
several days
rather than cause another monkey
to be
shocked. These pat-
terns were particularly strong when
the 2
animals
had
pre-
viously
been cage mates
and
thus
may be
presumed
to
have
formed
some sort
of
bond; when
the
animals were strangers
to
each other, less than
a
third showed this
form
of
altruistic,
self-
sacrificing
behavior.
Nonreciprocation
of
love.
Although mutual love provides
strong satisfactions
and
hedonic benefits, there
are
many
cases
in
which people
do not
reciprocate
another's
affection
and ro-
mantic interest. Such
refusals
to
form
a
social bond might
be
taken
as
evidence against
the
belongingness hypothesis.

520
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
On
closer examination, however, inspection
of
patterns
of un-
requited
love
does
not
provide
a
serious challenge
to the
belong-
ingness
hypothesis,
for
several reasons. First, most people
do
want
to
form
a
close romantic relationship,
and
their
refusals
are
typically
based
on
either already having such
a
relationship
with
another partner
(consistent
with
the
satiation
hypothesis)
or
perceiving
the
aspiring partner
as
unsuitable
for
some reason,
such
as
unattractiveness
or
incompatibility. Moreover,
in
many
cases, rejectors experience considerable distress such
as
guilt
and
empathic pain when rejecting
another's
offer
of
love. This
distress
is
consistent with
the
view
that rejecting social attach-
ment
goes against some deeply
rooted
aspect
of
human nature,
even
when
the
person
is
quite certain that
he or she
does
not
want
this particular attachment
(Baumeister
&
Wotman,
1992;
Baumeisteretal,
1993).
Shyness.
Shy
behavior patterns
may
seem antisocial insofar
as the shy
person sometimes avoids social encounters, with-
draws
from
ongoing interactions,
and
acts
in
other ways that
reduce
the
chances
of
forming
relationships (Leary,
1983).
In
fact,
however,
shy
people
are
strongly motivated
to
form
rela-
tionships,
and
shyness
may
reflect
an
interpersonal strategy that
partially
protects
the
individual against rejection.
When
people
do not
believe that they
will
be
regarded
in
ways
that
will
result
in
social acceptance, they
may
avoid absolute
rejection
by
disaffiliating.
Although reticence
and
withdrawal
are
unlikely
to
make particularly good impressions
or to
bring
hearty
acceptance
from
others, they reduce
the
risk
of
saying
or
doing something that
others
might regard negatively. When
one
fears
rejection,
the
best tactic
may
seem
to be to
participate
as
little
as
possible, thereby
giving
others
few
reasons
to
reject
one
(Shepperd&Arkin,
1990).
At
the
same time that they pull back, however,
shy
people
engage
in
behaviors that have been characterized
as
"innocu-
ously
sociable" (Leary, 1983). They smile more (even though
they
feel
anxious rather than
happy),
nod
their heads more
in
agreement,
ask
more questions,
and use
more verbal reinforcers
when
others
are
speaking. These behaviors
may
reflect
last-re-
sort tactics
to
maintain
a
minimum degree
of
interpersonal con-
nection
in
otherwise
difficult
or
threatening encounters
(Leary,
Knight,
&
Johnson, 1987).
General Discussion
We
have considered
a
broad assortment
of
evidence pertain-
ing
to the
hypothesis that
the
desire
for
interpersonal attach-
ments—the
need
to
belong—is
a
fundamental human motiva-
tion. Most
of the
metatheoretical requirements
we
outlined
for
evaluating
such
a
hypothesis appear
to be
satisfied, although
some issues remain.
We
begin
by
reviewing
the
major
conclusions.
Again
and
again,
we
found evidence
of a
basic
desire
to
form
social attachments. People
form
social
bonds
readily, even
un-
der
seemingly adverse conditions. People
who
have anything
in
common,
who
share common (even unpleasant)
experiences,
or who
simply
are
exposed
to
each other
frequently
tend
to
form
friendships
or
other attachments. Moreover, people
resist
losing
attachments
and
breaking social bonds, even
if
there
is no ma-
terial
or
pragmatic reason
to
maintain
the
bond
and
even
if
maintaining
it
would
be
difficult.
Abundant evidence also
attests
that
the
need
to
belong shapes
emotion
and
cognition. Forming
or
solidifying
social attach-
ments generally produces positive emotion, whereas real, imag-
ined,
or
even potential threats
to
social bonds generate
a
variety
of
unpleasant emotional states.
In
short, change
in
belong-
ingness
is a
strong
and
pervasive cause
of
emotion
in
ways that
support
the
hypothesis
of a
need
to
belong.
It is
also evident that
people think
a
great deal about
belongingness.
They devote
a
disproportionate amount
of
cognitive processing
to
actual
or
possible relationship partners
and
interaction partners,
and
they reserve particular, more extensive,
and
more favorable pat-
terns
of
information processing
for
people
with
whom they
share social bonds.
Deficits
in
belongingness apparently lead
to a
variety
of ill
effects,
consistent with
the
view
that belongingness
is a
need
(as
opposed
to
merely
a
want).
Both psychological
and
physical
health problems
are
more common among people
who
lack
so-
cial
attachments. Behavioral pathologies, ranging
from
eating
disorders
to
suicide,
are
more common among people
who are
unattached. Although most
of
these
findings are
correlational
and
many alternative explanations
can be
suggested, recent
efforts
have begun controlling
for
these other factors,
and the
pure, primary
effects
of
belongingness appear
to
remain strong.
It
appears,
then, that belongingness
is not
only pleasant
but
also
apparently
very
beneficial
to the
individual
in
multiple ways.
We
proposed
two
aspects
of the
need
to
belong,
and
both
ap-
pear
to be
important. That
is,
people seem
to
need
frequent,
affectively
pleasant
or
positive interactions with
the
same indi-
viduals,
and
they need these interactions
to
occur
in a
frame-
work
of
long-term, stable caring
and
concern. People
who can
satisfy
one
component
but not the
other tend
to be
less
satisfied
and
less
well
off
then people
who can
satisfy
both,
but
they
do
seem
to
derive some benefits
from
satisfying
the one
component
(as
opposed
to
satisfying
neither). More
and
better evidence
is
needed
on
this point,
however;
most evidence pertains
to
people
who
have
the
bond
and
lack interactions, rather than
the re-
verse. Also,
it is
unclear whether
the
interactions must
be
pleas-
ant or can be
satisfactory
if
they
are
merely neutral.
The
evi-
dence suggests merely that aversive
or
conflictual
interactions
fail
to
satisfy
the
need. Some evidence suggests that
a
framework
of
mutual, reciprocal concern
is
best,
but the
effects
and
impor-
tance
of
mutuality need
further
investigation.
The
need
to
belong also
appears
to
conform
to
motivational
patterns
of
satiation
and
substitution. People need
a few
close
relationships,
and
forming
additional bonds beyond those
few
has
less
and
less impact. Having
two as
opposed
to no
close
relationships
may
make
a
world
of
difference
to the
person's
health
and
happiness;
having
eight
as
opposed
to six may
have
very
little consequence. When
a
social bond
is
broken, people
appear
to
recover best
if
they form
a new
one, although each
individual
life
tends
to
involve some particularly
special
rela-
tionships (such
as filial or
marital bonds) that
are not
easily
replaced.
People without intimate partners engage
in a
variety
of
activities
to find
partners,
but
people
who
have partners
al-
ready
are
much less active
at
seeking additional relationships,
consistent with
the
satiation hypothesis.
We
reviewed evidence that
the
need
to
belong
affects
a
broad
variety
of
behaviors; indeed,
the
range
is
sufficiently
broad
as to
render less plausible
any
notion that
the
need
to
belong
is a

THE
NEED
TO
BELONG
521
product
of
certain other factors
or
motives.
We
also noted that
evidence
about
belongingness
seems
to
implicate some brain
mechanisms
and to
conform
to
patterns that evolutionary the-
ory
would suggest, both
of
which seem consistent with
the ar-
gument
that
the
need
is
innate
in
humans. Still,
the
nonderiva-
tive
hypothesis
is
probably
the
least
well
supported
aspect
of our
theory,
not
because
of any
clear evidence deriving
the
need
to
belong
from
other motives
but
simply perhaps because
it is
rel-
atively
difficult
to
collect compelling data
to
show that
a
motive
is not
derivative.
The
issue
of
which motives derive
from
which
others appears
to be an
important challenge
for
future
motiva-
tion research.
We
also considered several counterexamples that
at
least
su-
perficially
suggested tendencies
to
reject social attachment.
On
close inspection,
these
patterns
did not
stand
up as
counterex-
amples,
and
indeed there
was
generally strong evidence
of a
pos-
itive
need
to
belong that increased
the
subjective
difficulty
of
rejecting
or
avoiding attachment.
We
conclude, then, that
the
present state
of the
empirical
ev-
idence
is
sufficient
to
confirm
the
belongingness hypothesis.
The
need
to
belong
can be
considered
a
fundamental human
motivation.
Implications
for
Psychological Theory
If
the
belongingness hypothesis
is
indeed correct, then
it
seems
plausible that many psychological phenomena
may be
affected
by
this motivation. Clearly, patterns
of
group behavior
and
close relationships
can be
understood
as
serving
the
need
to
belong.
It is
thus
not
necessary
to
derive
all
group
and
intimate
affiliation
patterns
from
other motives, such
as the
fact
that
groups
may
confer
pragmatic
benefits
or
bolster self-esteem
(see
also
Turner,
1985). People
may
simply
desire
to
belong
to
groups.
Patterns
of
self-presentation (Baumeister, 1982; Leary,
1994;Leary&Kowalski,
1990;
Schlenker,
1980),
interpersonal
redress
and
excuse making
(R. S.
Miller
&
Leary, 1992;
Schlenker,
1980;
C. R.
Snyder,
Higgins,
&
Stucky,
1983),
and
group
conformity
(Moreland
&
Levine,
1989)
may all be
seen
in
the
context
of
enhancing
one's chances
of
inclusion
in
groups
and
relationships. Also,
it may be no
accident
that
people seem
most
likely
to be
prejudiced against members
of
groups
to
which
they
have
little
or no
opportunity
to
belong. Thus,
the
most common
and
widespread
bases
of
prejudice
are
race,
gen-
der,
and
national origin. People bolster their
own
in-group
at
the
expense
of
out-groups
from
which they
are
excluded (e.g.,
Meindl
&
Lerner, 1984).
Although
antisocial behavior might,
at first
glance,
be re-
garded
as
another potential counterexample
for the
belong-
ingness
hypothesis (because antisocial behavior makes enemies
and
alienates other
people),
it is
readily apparent that belong-
ingness
has
close ties
to it.
Members
of
some groups
are
pres-
sured
to
commit violent acts, ranging
from
vandalism
to
mur-
der,
to be
accepted
by and to
demonstrate commitment
to the
group
(e.g., Breitman, 1991; Freud,
1913/1956;
Hogan
&
Jones,
1983;
Rosenberg,
1991;
Sarbin,
1982;Staub,
1989;Toch,
1992).
It
seems likely that aggression
as
well
may
have some
belongingness
as a
prerequisite, because aggression risks
alienating
other people
and so
only people with
firm
attach-
ments
can
safely
engage
in
aggressive behavior.
The
centrality
of
belongingness
to
human psychological
func-
tioning also
has
implications
for
the
treatment
of
emotional
and
behavioral problems. From
our
standpoint,
a
great
deal
of
peo-
ple's
psychological
difficulties
reflects emotional
and
behavioral
reactions
to
perceived
threats
to
social bonds.
As has
been
shown,
many
of the
emotional problems
for
which people seek
professional
help (anxiety,
depression,
grief,
loneliness, rela-
tionship problems,
and the
like)
result
from
people's
failure
to
meet their belongingness needs. Furthermore,
a
great deal
of
neurotic, maladaptive,
and
destructive behavior seems
to
reflect
either
desperate attempts
to
establish
or
maintain relationships
with
other people
or
sheer frustration
and
purposelessness
when
one's
need
to
belong goes unmet.
Implications
for
Other Fields
We
suggested
at the
outset that
the
belongingness hypothesis
ought
to
have
implications
that
go
beyond
immediate
psycho-
logical
functioning
and, indeed, that
it
might prove
useful
as an
explanatory
construct
for the
phenomena studied
by
research-
ers in
other
fields. We now
consider
briefly some nonpsycholog-
ical
applications
of the
need
to
belong.
Contrary
to
cultural materialism,
we
have proposed that
many
aspects
of
human culture
are
directly
and
functionally
linked
to
enabling people
to
satisfy
the
psychological need
to
belong.
If
this
is
correct, then some historical
and
sociological
changes
in the
structures
of
society should
be
linked
to
changes
in
the
bases
for
belongingness.
For
example, membership
in
many
organizations
(including
corporate employment)
has
largely
ceased
to
depend
on
family
connections
the way it
once
did, with corresponding changes
in the
definition
and
power bal-
ance
in
families,
the
educational system (which provides
credentials),
and
other placement systems (e.g., Burgess
&
Locke,
1945;
Pass,
1977).
By
the
same token,
the
decline
of
arranged marriages
and the
increasing availability
of
divorce have made romantic attach-
ment
more
dependent
on
individual attractiveness
and
other
traits. Concern over
the
self
as an
instrument
for
attracting oth-
ers and
maintaining attachments should therefore increase.
Thus,
becoming
old or fat
would
be
less threatening
if
divorce
were
impossible
or if
marriages were arranged. With
the in-
creasing threat, social structures should emerge,
for
example,
to
help people look young
or
lose
weight. Also, sexuality
has a
frequently
changing relationship
to
social inclusion,
and
vari-
ous
eras have included
or
excluded people
on the
basis
of
sexual
chastity, skill,
appeal,
and
perceived
healthiness.
A
general pattern
may
well
be
that cultures
use
social inclu-
sion
to
reward,
and
exclusion
to
punish, their members
as a
way
of
enforcing
their values.
As is
well
known, many early civiliza-
tions
equated
exile
with death, which seems
to
suggest that
life
is
desirable
only
within
the
network
of
close relationships
to
which
the
person belongs. Modern civilizations tend
to use
prison
to
punish people, which again invokes
the
principle that
depriving
people
of
contact
with
relationship partners
is
highly
aversive;
solitary confinement
is
generally recognized
as the
most severe
and
aversive
form
of
imprisonment.
On the
positive
side,
the
evolution
of
modern society
has
seen
an
increasingly
broad
and
fundamental
quest
for
fame.
Braudy's
(1986)
history
of
fame
characterizes
the
desire
for
fame
as
based
on a
"dream

522
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
of
acceptance"
that holds
the
(often
illusory)
promise that once
a
person achieves
fame,
he or she
will
be
embraced
and
sought
by
others
for the
rest
of his or her
life.
Fame
may
well
be
thus
another
instance
of the use of
social inclusion
as a
reward.
A. H.
Buss
(1983)
has
pointed
out
that both
the
presence
of
others
and
the
attention
of
others
are
important social rewards,
and
the
deprivation
of
such contact
has
often
been used
as
powerful
social punishment.
Turning
to
political science,
a
well-known article
by
Morgenthau
(1962)
argued that
the
pursuit
of
power
can be
understood
as the
counterpart
to the
pursuit
of
love
in
that
both involve
an
attempt
to
escape
from
loneliness.
In
Mor-
genthau's
analysis,
the
human condition
suffers
from
the
threat
of
isolation,
and by
breaking down
the
barriers
be-
tween
one
another people hope
to
achieve
a
sense
of
together-
ness.
The
main
difference
between love
and
power
is
that
love
aspires
to a
mutual dissolving
of
personal boundaries, leading
to an
egalitarian merging into
a new
whole, whereas power
seeks
a
unilateral overcoming
of
boundaries,
by
which
the
will
of the
more
powerful
person becomes
the
will
of
both.
Morgenthau
noted that
the
pursuit
of
power
often
fails
to
overcome loneliness,
so
that,
ironically,
the
most
powerful
in-
dividuals
end up
feeling
still isolated
and
lonely
(hence,
the
tendency
for
rulers
to
demand that their subjects love them
too).
For
present purposes,
the
main point
is
that
the
need
to
belong
may be
regarded
as a
major source
of the
desire
for
power.
The
role
of
belongingness
is
also apparent
in
religion.
Al-
though
ideological belief
and
acceptance
of
metaphysical doc-
trines
are
often
regarded
as the
essence
of
religious participa-
tion,
Stark
and
Bainbridge
(1985)
reviewed considerable evi-
dence suggesting that
the
need
to
belong
may be a
more
compelling factor than
the
need
to
believe. They
noted
that
movement
into
and out of
religious groups (including cults,
sects,
and
mainstream denominations) depends much more
heavily
on
social ties than
on
ideological belief. Indeed, many
people
do not
fully
grasp
or
understand
the
theological belief
structure
of
their
own
religion (e.g.,
the
subtle
differences
be-
tween
the
many Protestant
denominations),
but
they
are
well
aware
of
what sort
of
people
in
their community belong
to
which
religion. Cults mainly attract people
who are
socially iso-
lated
or
lonely,
and
these individuals
are
often
attracted partic-
ularly
by the
promise
of
becoming part
of a
community
or
gain-
ing
a
sense
of
belonging.
Those
who
form
social
attachments
to
other
members
of the
cult tend
to
remain, whereas those
who
do not
form
social bonds tend
to
leave
soon.
By the
same token,
Kirkpatrick
and
Shaver
(1992)
have shown multiple links
be-
tween
religious
beliefs
and
adult attachment styles
or
relation-
ship
patterns.
Thus
far we
have
focused
on the
broad need
itself,
but
some
specific
patterns regarding human sociality
may
also have
im-
plications
for
other
fields. For
example,
we
have
noted
that
an
interesting
psychological issue involves
the
factors that deter-
mine
whether previously opposed groups
do or do not
integrate
into larger
wholes
(cf.
Sherifetal.,
1961/1988),
such
that
indi-
viduals
redefine
their allegiance
so as to
belong
to the new
group.
Such
reidentifications
have been
important
throughout
history.
The
Wars
of the
Roses were
finally
decided
by the
battle
of
Bosworth,
in
which Richard
III was
killed, thereby enabling
Henry
Tudor
to
establish
the
dynasty that ruled Elizabethan
England;
that battle
(like
others
in the
conflict)
turned
on the
dubious
loyalty
and
betrayal
of
several major groups that were
incompletely
merged into their respective sides (Ross, 1976).
The
Zulu empire
in
South
Africa
was
formed
by
incorporating
many
other groups that
had
been rivals
of and
neighbors
to the
original
Zulus,
and
those
new
identifications persist even today,
long
past
the
fall
of
that empire (Morris,
1965).
Meanwhile,
however,
Angola, Rwanda,
and
Uganda have
suffered
repeated
bouts
of
cruel violence
and
civil
war
between
former
rival
groups
that
failed
to
integrate
and
identify with
the
national
unity.
Nor is
this problem unique
to
Africa;
the
former
Yugo-
slavia
provided
a
vivid
example
of
bitter factional violence
re-
emerging
after
decades
of
seemingly
peaceful
coexistence,
and
the
same goes
for Sri
Lanka.
In the
United States, melting
pot
ideology
has
recently gone
out of
fashion
as the
nation
has be-
gun
to
accept
the
problematic reality
of
multiple, separate
en-
claves
defined
by
racial
and
ethnic backgrounds.
In
short,
it ap-
pears that asking people
to
redefine
their belongingness
to ac-
commodate
new
realities
is
difficult
and
only
sometimes
successful.
These
applications
are not
intended
as
exhaustive,
nor
even
as
the
most compelling
or
important. They
are
merely intended
as an
indication that
the
need
to
belong could
be
used
as an
interpretive
construct outside
of
psychological laboratories.
Concluding
Remarks
At
present,
it
seems
fair
to
conclude
that
human beings
are
fundamentally
and
pervasively motivated
by a
need
to
belong,
that
is, by a
strong desire
to
form
and
maintain enduring inter-
personal attachments. People seek frequent,
affectively
positive
interactions
within
the
context
of
long-term, caring relation-
ships.
As a
speculative point
of
theory
or
impressionistic obser-
vation,
the
need
to
belong
is not a new
idea; indeed,
we
noted
a
variety
of
previous psychological theorists
who
have
proposed
it
in one
form
or
another. What
is
new, however,
is the
existence
of
a
large body
of
empirical evidence with which
to
evaluate that
hypothesis.
If
psychology
has
erred with regard
to the
need
to
belong,
in
our
view,
the
error
has not
been
to
deny
the
existence
of
such
a
motive
so
much
as to
underappreciate
it.
This review
has
shown
multiple
links between
the
need
to
belong
and
cognitive pro-
cesses, emotional patterns, behavioral
responses,
and
health
and
well-being.
The
desire
for
interpersonal attachment
may
well
be
one
of the
most far-reaching
and
integrative
constructs cur-
rently
available
to
understand human nature.
References
Adler,
P.
(1980).
On
becoming
a
prostitute.
In E.
Muga
(Ed.),
Studies
in
prostitution (pp.
22-26).
Nairobi,
Kenya:'
Kenya Literature
Bureau.
Ainsworth,
M. D.
(1989).
Attachments beyond
infancy.
American
Psy-
chologist,
44,
709-716.
Anderson,
C. A.
(1991).
How
people think about causes: Examination
of
the
typical phenomenal organization
of
attributions
for
success
and
failure.
Social Cognition,
9,
295-329.
Anderson,
S.
A.,
Russell,
C. S., &
Schumm,
W. R.
(1983).
Perceived

THE
NEED
TO
BELONG
523
marital
quality
and
family
life-cycle
categories:
A
further analysis.
Journal
of
Marriage
and the
Family,
45,
127-139.
Argyle,
M.
(1987).
The
psychology
of
happiness.
London: Methuen.
Armstrong,
J.
G.,
&
Roth,
D. M.
(1989).
Attachment
and
separation
in
eating disorders:
A
preliminary investigation. International Jour-
nal
of
Eating Disorders,
8,
141-155.
Aron,
A.,
Aron,
E. N.,
Tudor,
M.,
&
Nelson,
G.
(1991).
Close relation-
ships
as
including other
in the
self. Journal
of
Personality
and
Social
Psychology.
60,
241-253.
Audy,
J. R.
(1980).
Man the
lonely animal: Biological roots
of
loneli-
ness.
In J.
Hartog,
J. R.
Audy,
& Y. A.
Cohen
(Eds.),
The
anatomy
of
loneliness
(pp.
111-128).
New
York: International Universities
Press.
Axelrod,
R.,
&
Hamilton,
W. D.
(1981).
The
evolution
of
cooperation.
Science.
211,
1390-1396.
Barash.
D. P.
(1977).
Sociobiology
and
behavior.
New
York:
Elsevier.
Barchas,
P.
(1986).
A
sociophysiological
orientation
to
small groups.
In
E.
Lawler
(Ed.),
Advances
in
group
processes (Vol.
3, pp.
209-246).
Greenwich,
CT:
JAI
Press.
Barden.
R.
C,
Garber,
J.,
Leiman,
B.,
Ford,
M.
E.,
&
Masters,
J. C.
(1985).
Factors governing
the
effective
remediation
of
negative
affect
and its
cognitive
and
behavioral consequences. Journal
of
Personality
and
Social
Psychology,
49,
1040-1053.
Barocas,
R.,
&
Vance,
F. L.
(1974).
Physical appearance
and
personal
adjustment
counseling. Journal
of
Counseling
Psychology,
21, 96-
100.
Baumeister,
R. F.
(1982).
A
self-presentational view
of
social phenom-
ena.
Ps,vchological
Bulletin,
91,
3-26.
Baumeister,
R. F.
(1990).
Suicide
as
escape
from
self.
PsychologicalRe-
view,
97,
90-113.
Baumeister,
R. F.
(1991).
Meanings
of
life.
New
York:
Guilford Press.
Baumeister,
R. F,
Stillwell,
A.
M.,
&
Heatherton,
T. F.
(1994).
Guilt:
An
interpersonal approach.
Psychological
Bulletin,
115,
243-267.
Baumeister,
R.
F.,
Stillwell,
A.
M.,
&
Heatherton,
T. F. (in
press).
Per-
sonal
narratives about guilt: Role
in
action control
and
interpersonal
relationships.
Basic
and
Applied Social
Psychology.
Baumeister,
R. F, &
Tice,
D. M.
(1990).
Anxiety
and
social exclusion.
Journal
oj'Social
and
Clinical
Psychology,
9,
165-195.
Baumeister,
R. F, &
Wotman,
S. R.
(1992).
Breaking hearts:
The two
sides
of
unrequited
lave.
New
York:
Guilford
Press.
Baumeister,
R. F,
Wotman,
S.
R.,
&
Stillwell,
A. M.
(1993).
Unre-
quited
love:
On
heartbreak, anger, guilt,
scriptlessness,
and
humilia-
tion.
Journal
of
Personality
and
Social
Psychology,
64,
377-394.
Baunach,
P. J.
(1985).
Mothers
in
prison.
New
Brunswick,
NJ:
Transac-
tion
Books.
Beckman,
K..,
Marsella,
A.
J.,
&
Finney,
R.
(1979).
Depression
in the
wives
of
nuclear submarine personnel. American Journal
of
Psychia-
try,
136,
524-526.
Beckman,
L. J.
(1981).
Effects
of
social interaction
and
children's rela-
tive
inputs
on
older women's psychological well-being. Journal
of
Per-
sonality
and
Social
Psychology,
41,
1075-1086.
Bednarski,
R.,
&
Leary,
M. R.
(1994).
Self-esteem
and
fear
of
death.
Unpublished
manuscript, Wake Forest University,
Winston-Salem,
NC.
Belsky,
J.
(1979).
The
interrelation
of
parental
and
spousal behavior
during
infancy
in
traditional nuclear families:
An
exploratory analy-
sis.
Journal
of
Marriage
and the
Family,
41,
749-755.
Belsky,
J.
(1981).
Early
human experience:
A
family
perspective.
De-
velopmental
Psychology,
17,
3-23.
Berman,
W. H.
(1988).
The
role
of
attachment
in the
post-divorce
ex-
perience. Journal
of
Personality
and
Social
Psychology,
54,
496-503.
Bernard,
J.
(1982).
The
future
of
marriage.
New
Haven,
CT:
Yale
Uni-
versity
Press.
Berscheid,
E.,
&
Walster,
E.
(1974).
Physical attractiveness.
In L.
Ber-
kowitz
(Ed.),
Advances
in
experimental social
psychology
(Vol.
7, pp.
158-215).
New
York:
Academic
Press.
Bhatti,
B.,
Derezotes,
D.,
Kim,
S.,
&
Specht,
H.
(1989).
The
association
between child
maltreatment
and
self-esteem.
In A. M.
Mecca,
N. J.
Smelser,
& J.
Vasconcellos
(Eds.),
The
social importance
of
self-es-
teem
(pp.
24-71).
Berkeley: University
of
California
Press.
Billig,
M.,
&
Tajfel,
H.
(1973).
Social categorization
and
similarity
in
intergroup
behavior. European Journal
of
Social
Psychology,
3, 27-
51.
Bloom,
B.
L.,
White,
S. W, &
Asher,
S. J.
(1979).
Marital disruption
as
a
stressful
life
event.
In G.
Levinger
& O. C.
Moles
(Eds.),
Divorce
and
separation: Context, causes,
and
consequences
(pp.
184-200).
New
York:
Basic Books.
Bowlby,
J.
(1969).
Attachment
and
loss: Vol.
1.
Attachment.
New
York:
Basic Books.
Bowlby,
J.
(1973).
Attachment
and
loss:
Vol.
2.
Separation anxiety
and
anger.
New
York:
Basic Books.
Braudy,
L.
(1986).
The
frenzy
of
renown: Fame
and its
history.
New
York:
Oxford University
Press.
Brehm,
S. S.
(1987).
Social support
and
clinical practice.
In J. E.
Mad-
dux,
C. D.
Stoltenberg,
& R.
Rosenwein
(Eds.),
Social processes
in
clinical
and
counseling
psychology
(pp.
26-38).
New
York:
Springer-
Verlag.
Breitman,
R.
(1991).
The
architect
of
genocide:
Himmler
and
the final
solution. Hanover,
NH:
Brandeis University
Press.
Brenner,
M. W.
(1976).
Memory
and
interpersonal
relations.
Unpub-
lished
doctoral
dissertation, University
of
Michigan,
Ann
Arbor.
Brewer,
M. B.
(1979).
Ingroup
bias
in the
minimal intergroup situation:
A
cognitive-motivational analysis.
Psychological
Bulletin,
86,
307-
324.
Brewer,
M.
B.,
&
Silver,
M.
(1978).
Ingroup bias
as a
function
of
task
characteristics. European Journal
of
Social
Psychology,
8,
393-400.
Bridges,
W.
(1980).
Transitions: Making sense
of
life's
changes. Read-
ing,
MA:
Addison-Wesley.
Brown,
J. D.
(1986).
Evaluations
of
self
and
others:
Self-enhancement
biases
in
social judgments. Social Cognition,
4,
353-376.
Browning,
C. R.
(1992).
Ordinary men: Reserve Police Battalion
101
and
the
final
solution
in
Poland.
New
York:
Harper Collins.
Bunker,
B.
B.,
Zubek,
J.
M.,
Vanderslice,
V.
J.,
&
Rice,
R. W.
(1992).
Quality
of
life
in
dual-career families: Commuting versus single-resi-
dence couples. Journal
of
Marriage
and the
Family,
54,
399-407.
Burgess,
E.
W.,
&
Locke,
H. J.
(1945).
The
family:
From institution
to
companionship.
New
York:
American Book.
Burkhart,
K.
(1973).
Women
in
prison.
Garden
City,
NY:
Doubleday.
Buss,
A. H.
(1983).
Social
rewards
and
personality.
Journal
of
Person-
ality
and
Social
Psychology,
44,
553-563.
Buss,
D. M.
(1990).
The
evolution
of
anxiety
and
social exclusion.
Journal
of
Social
and
Clinical
Psychology,
9,
196-210.
Buss,
D. M.
(1991).
Evolutionary personality psychology. Annual
Re-
view
of
Psychology,
42,
459-491.
Caldwell,
M.
A.,
&
Peplau,
L. A.
(1982).
Sex
differences
in
same-sex
friendship.
Sex
Roles,
8,
721-732.
Campbell,
A.
(1981).
The
sense
of
well-being
in
America.
New
York:
McGraw-Hill.
Campbell,
A.,
Converse,
P.
E.,
&
Rodgers,
W. L.
(1976).
The
quality
of
American
life:
Perceptions, evaluations,
and
satisfactions.
New
York:
Russell
Sage Foundation.
Carnelley,
K.
B.,
Pietromonaco,
P.
R.,
&
Jaffe,
K.
(1994).
Depression,
working
models
of
others,
and
relationship functioning. Journal
of
Personality
and
Social
Psychology,
66,
127-140.
Cash,
T. F.
(1985).
Physical appearance
and
mental health.
In J. A.
Graham&A.
KJigman(Eds.),
The
psychology
of
cosmetic treatments
(pp.
196-216).
New
York:
Praeger.
Cialdini,
R.
B.,
Darby,
B.,
&
Vincent,
J.
(1973).
Transgression
and
al-

524
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
truism:
A
case
for
hedonism. Journal
of
Experimental
Social Psy-
chology,
9,
502-516.
Clark,
L.
A.,
&
Watson,
D.
(1988).
Mood
and the
mundane: Relations
between daily
life
events
and
self-reported
mood.
Journal
of
Person-
ality
and
Social
Psychology,
54,
296-308.
Clark,
M. S.
(1984).
Record keeping
in two
types
of
relationships.
Journal
of
Personality
and
Social
Psychology,
47,
549-557.
Clark,
M. S.
(1986).
Evidence
for the
effectiveness
of
manipulations
of
communal
and
exchange relationships. Personality
and
Social Psy-
chology
Bulletin,
12,
414-425.
Clark,
M.
S.,
&
Mills,
J.
(1979).
Interpersonal attraction
in
exchange
and
communal relationships. Journal
of
Personality
and
Social
Psy-
chology^?,
12-24.
Clark,
M.
S.,
Mills,
J.,
&
Corcoran,
D. M.
(1989).
Keeping track
of
needs
and
inputs
of
friends
and
strangers.
Personality
and
Social
Psy-
chology
Bulletin,
15,
533-542.
Clark,
M.
S.,
Mills,
J.,
&
Powell,
M. C.
(1986).
Keeping track
of
needs
in
communal
and
exchange
relationships.
Journal
of
Personality
and
Social
Psychology,
51,
333-338.
Clark,
M.
S.,
Ouellette,
R.,
Powell,
M.
C.,
&
Milberg,
S.
(1987).
Recip-
ient's mood, relationship type,
and
helping. Journal
of
Personality
and
Social
Psychology,
53,
94-103.
Cohen,
S.,
Sherrod,
D.
R.,
&
Clark,
M. S.
(1986).
Social skills
and
the
stress-protective role
of
social
support.
Journal
of
Personality
and
Social
Psychology,
50,
963-973.
Cohen,
S.,
&
Wills,
T. A.
(1985).
Stress, social support,
and the
buff-
ering
hypothesis.
Psychological
Bulletin,
98,
310-357.
Conte,
H.
R.,
Weiner,
M.
B.,
&
Plutchik,
R.
(1982).
Measuring death
anxiety:
Conceptual, psychometric,
and
factor-analytic
aspects.
Journal
of
Personality
and
Social
Psychology,
43,
775-785.
Coon,
C. S.
(1946).
The
universality
of
natural groupings
in
human
societies. Journal
of
Educational
Sociology,
20,
163-168.
Costa,
P.
T.,
&
McCrae,
R. R.
(1980).
Influence
of
extraversion
and
neuroticism
on
subjective well-being: Happy
and
unhappy
people.
Journal
of
Personality
and
Social
Psychology,
38,
668-678.
Costa,
P.
T.,
&
McCrae,
R. R.
(1984).
Personality
as a
lifelong
determi-
nant
of
wellbeing.
In C. Z.
Malatesta
& C. E.
Izard
(Eds.),
Emotion
in
adult
development
(pp.
141-157).
Beverly Hills,
CA:
Sage.
Costa,
P.
T.,
McCrae,
R.
R.,
&
Zonderman,
A. B.
(1987).
Environmen-
tal
and
dispositional
influences
on
well-being: Longitudinal
follow-
up of an
American national sample. British Journal
of
Psychology,
78,
299-306.
Coyne,
J.
C.,
&
DeLongis,
A.
(1986).
Going beyond social
support:
The
role
of
social relationships
in
adaptation. Journal
of
Consulting
and
Clinical
Psychology,
54,454-460.
Craighead,
W.
E.,
Kimball,
W.
H.,
&
Rehak,
R. S.
(1979).
Mood
changes, physiological responses,
and
self-statements during social
rejection
imagery. Journal
of
Consulting
and
Clinical
Psychology,
47,
385-396.
Cunningham,
M.
R.,
Steinberg,
J.,
&
Grev,
R.
(1980).
Wanting
to and
having
to
help: Separate motivations
for
positive mood
and
guilt-in-
duced
helping. Journal
of
Personality
and
Social
Psychology,
38,181-
192.
Cutrona,
C. E.
(1986).
Behavioral manifestations
of
social
support:
A
microanalytic
investigation. Journal
of
Personality
and
Social
Psy-
chology,
51,
201-208.
Cutrona,
C. E.
(1989).
Ratings
of
social
support
by
adolescents
and
adult
informants: Degree
of
correspondence
and
prediction
of de-
pressive symptoms. Journal
of
Personality
and
Social
Psychology,
57,
723-730.
Danheiser,
P.
R.,
&
Graziano,
W. G.
(1982).
Self-monitoring
and
coop-
eration
as a
self-presentational strategy. Journal
of
Personality
and
Social
Psychology,
42,497-505.
Darley,
J.
M.,
&
Latane,
B.
(1968).
Bystander intervention
in
emergen-
cies:
Diffusion
of
responsibility. Journal
of
Personality
and
Social
Psychology,
8,
377-383.
DeLongis,
A.,
Folkman,
S.,
&
Lazarus,
R. S.
(1988).
The
impact
of
daily
stress
on
health
and
mood:
Psychological
and
social
resources
as
mediators.
Journal
of
Personality
and
Social
Psychology,
54,
486-
495.
de
Rivera,
J.
(1984).
The
structure
of
emotional relationships.
In
P.
Shaver
(Ed.),
Review
of
personality
and
social
psychology,
Vol.
5:
Emotions, relationships,
and
health (pp.
116-145).
Beverly Hills,
CA:
Sage.
Devine,
P.
G.,
Sedikides,
C.,
&
Fuhrman,
R. W.
(1989).
Goals
in
social
information
processing:
The
case
of
anticipated
interaction. Journal
of
Personality
and
Social
Psychology,
56,
680-690.
DiTommaso,
E.,
&
Spinner,
B.
(1993).
The
development
and
initial
validation
of the
social
and
emotional loneliness scale
for
adults. Per-
sonality
and
Individual
Differences,
14,
127-134.
Donne,
J.
(1975).
Devotions
upon
emergent occasions. Montreal, Can-
ada:
McGill
Queens
University
Press.
Douglas,
J. D.
(1967).
The
social meanings
of
suicide.
Princeton,
NJ:
Princeton University
Press.
Drill,
R. L.
(1987).
Young
adult children
of
divorced parents: Depres-
sion
and the
perception
of
loss.
Journal
of
Divorce,
10,
169-187.
Durkheim,
E.
(1963).
Suicide.
New
York: Free
Press.
(Original
work
published
1897)
Egan,
G.
(1970).
Encounter: Group
processes
for
interpersonal
growth.
Monterey,
CA:
Brooks/Cole.
Eisenberger,
R.
(1970).
Is
there
a
deprivation-satiation
function
for so-
cial
approval?
Psychological
Bulletin,
74,
255-275.
Elder,
G.
H.,
&
Clipp,
E. C.
(1988).
Wartime
losses
and
social bonding:
Influence
across
40
years
in
men's
lives.
Psychiatry,
51,
177-198.
Emery,
R. E.
(1982).
Interparental
conflict
and the
children
of
divorce.
Psychological
Bulletin,
92,
310-330.
Epstein,
S.
(1992).
The
cognitive
self,
the
psychoanalytic
self,
and the
forgotten selves.
Psychological
Inquiry,
3,
34-37.
Erber,
F.,
&
Fiske,
S. T.
(1984).
Outcome dependency
and
attention
to
inconsistent information. Journal
of
Personality
and
Social
Psychol-
ogy,
47,
709-726.
Farina,
A.,
Burns,
G.
L.,
Austad,
C.,
Bugglin,
C.
S.,
&
Fischer,
E. H.
(1986).
The
role
of
physical attractiveness
in the
readjustment
of
discharged psychiatric patients. Journal
of
Abnormal
Psychology,
86,
510-517.
Fass,
P.
(1977).
The
damned
and the
beautiful:
American youth
in the
1920s.
Oxford, England: Oxford University
Press.
Festinger,
L.,
Schachter,
S.,
&
Back,
K.
(1950).
Social pressures
in in-
formal
groups:
A
study
of
a
housing community. Palo Alto,
CA:
Stan-
ford
University
Press.
Fincham,
F.
D.,
Beach,
S.
R.,
&
Baucom,
D. H.
(1987).
Attribution
processes
in
distressed
and
nondistressed
couples:
4.
Self-partner
at-
tribution differences. Journal
of
Personality
and
Social
Psychology,
52,
739-748.
Forsyth,
D. R.
(1991).
Change
in
therapeutic
groups.
In C. R.
Snyder
& D. R.
Forsyth
(Eds.),
Handbook
of
social
and
clinical
psychology
(pp.
664-680).
New
York:
Pergamon
Press.
Forsyth,
D.
R.,
&
Schlenker,
B. R.
(1977).
Attributing
the
causes
of
group performance:
Effects
of
performance quality, task
importance,
and
future
testing.
Journal
of
Personality,
45,
220-236.
Freedman,
J.
(191
S).
Happy
people:
What
happiness
is, who has it, and
why.
New
Ybrk:
Harcourt Brace
Jovanovich.
Freud,
S.
(1930).
Civilization
and its
discontents
(J.
Riviere,
Trans.).
London:
Hogarth
Press.
Freud,
S.
(1956).
Totem
und
Tabu
[
Totem
and
taboo
].
Frankfurt, Ger-
many:
Fischer
Biicherei.
(Original
work published
1913)
Fromm,
E.
(1955).
The
sane
society.
New
York:
Holt,
Rinehart
&
Winston.

THE
NEED
TO
BELONG
525
Fromm,
E.
(1956).
The art
of
loving.
New
York:
Harper
&
Brothers.
Geis,
F. L., &
Moon,
T. H.
(1981)..Machiavellianism
and
deception.
Journal
of
Personality
and
Social
Psychology,
41,
766-775.
Gerstel,
N., &
Gross,
H.
(1982).
Commuter
marriages:
A
review. Mar-
riage
and
Family Review,
5,
71-93.
Gerstel,
N.,
&
Gross,
H.
(1984).
Commuter marriage:
A
study
of
work
and
family.
New
York:
Guilford
Press.
Gewirtz,
J.
L.,
&
Baer,
D. M.
(1958).
Deprivation
and
satiation
of
social
reinforcers
as
drive conditions. Journal
of
Abnormal
and
Social
Psy-
chology,
57,
165-172.
Giallombardo,
R.
(1966).
Society
of
women:
A
study
of a
women's
prison.
New
\brk:
Wiley.
Glenn,
N.
D.,
&
McLanahan,
S.
(1982).
Children
and
marital happi-
ness:
A
further
specification
of the
relationship. Journal
of
Marriage
and
the
Family,
44,
63-72.
Goffman,
E.
(1971).
Relations
in
public.
New
York: Harper
Colophon.
Goode,
W. J.
(1956).
After divorce.
New
York: Free
Press.
Goodwin,
J.
S.,
Hunt,
W.
C,
Key,
C.
R.,
&
Samet,
J. M.
(1987).
The
effect
of
marital status
on
stage,
treatment,
and
survival
of
cancer
patients. Journal
of
the
American Medical Association, 258,
3125-
3130.
Gottlieb,
J.,
&
Carver,
C.
(1980).
Anticipation
of
future
interaction
and
the
bystander
effect.
Journal
of
Experimental
Social
Psychology,
16,
253-260.
Gouaux,
C.
(1971).
Induced
affective
states
and
interpersonal attrac-
tion.
Journal
of
Personality
and
Social
Psychology,
20,
37-43.
Govaerts,
K.,
&
Dixon,
D. N.
(1988).
. . .
Until careers
do us
part:
Vocational
and
marital satisfaction
in the
dual-career commuter mar-
riage. International Journal
for the
Advancement
of
Counselling,
11,
265-281.
Groth,
A. N.
(1979).
Men who
rape.
New
York:
Plenum Press.
Guisinger,
S.,
&
Blatt,
S. J.
(1994).
Individuality
and
relatedness: Evo-
lution
of a
fundamental dialectic. American Psychologist,
49,
104-
111.
Hamachek,
D.
(1992).
Encounters with
the
self(4lh
ed.).
San
Diego,
CA:
Harcourt Brace Jovanovich.
Hardy,
C.
J.,
&
Grace,
R. K.
(1991).
The
effects
of
task structure
and
teammate competence
on
social loafing. Journal
of
Sport
and
Exer-
cise
Psychology,
13,
372-381.
Harkins,
S. G., &
Petty,
R. E.
(1982).
Effects
of
task
difficulty
and
task
uniqueness
on
social
loafing.
Journal
of
Personality
and
Social Psy-
chology,
43,
1214-1229.
Harlow,
H.
F.,
Harlow,
M.
K..,
&
Suomi,
S. J.
(1971).
From thought
to
therapy:
Lessons
from
a
primate laboratory. American Scientist,
59,
538-549.
Harris,
M.
(1974).
Cows, pigs,
wars,
and
witches:
The
riddles
of
culture.
New
York:
Random House.
Harris,
M.
(1978).
Cannibals
and
kings:
The
origins
of
cultures.
New
York:
Random House.
Harris,
M.
(1979).
Cultural materialism:
The
struggle
for
a
science
of
culture.
New
York: Random House.
Harrison,
A.
A.,
&
Connors,
M. M.
(1984).
Groups
in
exotic environ-
ments.
In L.
Berkowitz
(Ed.),
Advances
in
experimental social psy-
chology
(Vol.
18,
pp.
49-87).
New
York:
Academic Press.
Hays,
R. B.
(1985).
A
longitudinal study
of
friendship development.
Journal
of
Personality
and
Social
Psychology,
48,
909-924.
Hazan,
C.,
&
Shaver,
P. R.
(1994a).
Attachment
as an
organizational
framework
for
research
on
close relationships.
Psychological
Inquiry,
5,
1-22.
Hazan,
C.,
&
Shaver,
P. R.
(1994b).
Deeper
into.attachment
theory.
Psychological
Inquiry,
5,
68-79.
Hill,
C. T,
Rubin,
Z., &
Peplau,
L. A.
(1976),
Breakups before mar-
riage:
The
end
of 103
affairs.
Journal
of
Social Issues,
32,
147-168.
Hobfall,
S.
E.,
&
London,
P.
(1986).
The
relationship
of
self-concept
and
social
support
to
emotional distress among women during
the
war.
Journal
of
Social
and
Clinical
Psychology,
4,
189-203.
Hogan,
R.
(1983).
A
socioanalytic
theory
of
personality.
In M.
Page
&
R.
Dienstbier
(Eds.),
Nebraska Symposium
on
Motivation,
1982
(pp.
55-89).
Lincoln: University
of
Nebraska
Press.
Hogan,
R.,
&
Jones,
W. H.
(1983).
A
role theoretical model
of
criminal
conduct.
In W. S.
Laufer
& J. M.
Days
(Eds.),
Personality
theory,
moral development,
and
criminal behavior (pp.
3-21).
Lexington,
MA:
Lexington Books.
Hogan,
R.,
Jones,
W.
H.,
&
Cheek,
J. M.
(1985).
Socioanalytic theory:
An
alternative
to
armadillo psychology.
In B. R.
Schlenker
(Ed.),
The
self
and
social
life
(pp. 175-198).
New
York:
McGraw-Hill.
Holmes,
J.
G.,
&
Rempel,
J. K.
(1989).
Trust
in
close relationships.
In
M.
Clark
(Ed.),
Close relationships: Review
of
personality
and
social
psychology
(Vol.
10,
pp.
187-220).
Newbury
Park,
CA:
Sage.
Holmes,
T.
H.,
&
Rahe,
R. H.
(1967).
The
social readjustment rating
scale.
Journal
of
Psychosomatic
Research,
11,
213-218.
Holt,
P.
A.,
&
Stone,
G. L.
(1988).
Needs, coping strategies,
and
coping
outcomes
associated
with
long-distance
relationships. Journal
of
Col-
lege
Student Development,
29,
136-141.
Horney,
K.
(1945).
Our
inner
conflicts:
A
constructive
theory
of
neuro-
sis.
New
York:
Norton.
Hotard,
S. R.,
McFatter,
R.
M.,
McWhirter,
R.
M.,
&
Stegall,
M. E.
(1989).
Interactive
effects
of
extraversion,
neuroticism,
and
social
relationships
on
subjective well-being. Journal
of
Personality
and So-
cial
Psychology,
5
7,
321
-331.
Howard,
J.
W.,
&
Rothbart,
M.
(1980).
Social categorization
and
mem-
ory for
in-group
and
out-group behavior. Journal
of
Personality
and
Social
Psychology,
38,
301-310.
Hoyle,
R.
H.,
&
Crawford,
A. M.
(in
press).
Use of
individual-level
data
to
investigate group phenomena: Issues
and
strategies. Small Group
Research.
Hoyle,
R.
H.,
Pinkley,
R.
L.,
&
Insko,
C. A.
(1989).
Perceptions
of
social behavior: Evidence
of
differing
expectations
for
interpersonal
and
intergroup
interaction. Personality
and
Social
Psychology
Bulle-
tin,
15,
365-376.
Isenberg,
S.
(1991).
Women
who
love
men who
kill.
New
York:
Simon
&
Schuster.
Jankowski,
M. S.
(1991).
Islands
in the
street: Gangs
and
American
urban
society.
Berkeley: University
of
California
Press.
Jones,
W. H.
(1981).
Loneliness
and
social
contact.
Journal
of
Social
Psychology,
113,
295-296.
Jones,
W.
H.,
&
Kugler,
K. (in
press). Interpersonal correlates
of the
guilt
inventory. Journal
of
Personality
Assessment.
Jones,
W.
H.,
Kugler,
K.,
&
Adams,
P.
(1995).
You
always hurt
the one
you
love: Guilt
and
transgressions against relationship
partners.
In K.
Fischer
& J.
Tangner
(Eds.),
Self-conscious
emotion (pp.
301-321).
New
York:
Guilford
Press.
Kaufman,S.
R.(
1986).
The
ageless
self:
Sources
of
meaning
in
late
life.
New
York:
Meridian.
Kenrick,
D. T, &
Cialdini,
R. B.
(1977).
Romantic
attraction:
Misat-
tribution
versus reinforcement explanations. Journal
of
Personality
and
Social
Psychology,
35,
381-391.
Kenrick,
D. T, &
Johnson,
G. A.
(1979).
Interpersonal
attraction
in
aversive
environments:
A
problem
for the
classical conditioning par-
adigm? Journal
of
Personality
and
Social
Psychology,
37,
572-579.
Kiecolt-Glaser,
J.
K.,
Fisher,
L.
D.,
Ogrocki,
P.,
Stout,
J.
C.,
Speicher,
C.
E.,
&
Glaser,
R.
(1987).
Marital quality, marital disruption,
and
immune
function. Psychosomatic Medicine,
49,
13-34.
Kiecolt-Glaser,
J.
K.,
Garner,
W.,
Speicher,
C.,
Penn,
G.
M.,
Holliday,
J.,
&
Glaser,
R.
(1984).
Psychosocial
modifiers
of
immunocompe-
tence
in
medical students. Psychosomatic Medicine,
46,
7-14.
Kiecolt-Glaser,
J.
K.,
Ricker,
D.,
George,
J.,
Messick,
G.,
Speicher,
C.
E.,
Garner,
W.,
&
Glaser,
R.
(1984).
Urinary
cortisol
levels,
cellu-

526
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
lar
immunocompetency,
and
loneliness
in
psychiatric inpatients. Psy-
chosomatic Medicine,
46,
15-23.
Kirkpatrick,
L.
A.,
&
Shaver,
P. R.
(1992).
An
attachment-theoretical
approach
to
romantic love
and
religious belief. Personality
and
Social
Psychology
Bulletin,
18,
266-275.
Kramer,
R. M., &
Brewer,
M.
B.
(1984).
Effects
of
group identity
on
resource
use in a
simulated commons dilemma. Journal
of
Personal-
ity
and
Social
Psychology,
46,
1044-1057.
Kunz,
P.
R.,
&
Woolcott,
M.
(1976).
Season's greetings: From
my
sta-
tus
to
yours. Social Science Research,
5,
269-278.
Lacoursiere,
R. B.
(1980).
The
life
cycles
of
groups:
Group
developmen-
tal
stage
theory.
New
York:
Human Sciences
Press.
Lakey,
B.,
&
Cassady,
P. B.
(1990).
Cognitive processes
in
perceived
social
support. Journal
of
Personality
and
Social
Psychology,
59,
337-
343.
Larkin,
M.
(1972).
Experiential groups:
The
uses
of
interpersonal
en-
counter,
psychotherapy
groups,
and
sensitivity
training. Morristown,
NJ:
General
Learning
Press.
Latane,
B.,
Eckman,
J., &
Joy,
V.
(1966).
Shared stress
and
interper-
sonal attraction. Journal
of
Experimental Social
Psychology,
7(Suppl.),
80-94.
Latane,
B.,
&
Nida,
S.
(1981).
Ten
years
of
research
on
group size
and
helping.
Psychological
Bulletin,
89,
308-324.
Latane,
B.,
Williams,
K., &
Harkins,
S. G.
(1979).
Many hands make
light
the
work:
The
causes
and
consequences
of
social loafing. Journal
of
Personality
and
Social
Psychology,
37,
822-832.
Lauderdale,
P.,
Smith-Cunnien,
P.,
Parker,
J.,
&
Inverarity,
J.
(1984).
External
threat
and the
definition
of
deviance. Journal
of
Personality
and
Social
Psychology,
46,
1058-1068.
Lawson,
A.
(1988).
Adultery:
An
analysis
of
love
and
betrayal.
New
York:
Basic Books.
Leary,
M. R.
(1983).
Understanding
social anxiety: Social, personality,
and
clinical
perspectives.
Beverly Hills,
CA:
Sage.
Leary,
M.
R.
(1990).
Responses
to
social exclusion: Social
anxiety,
jeal-
ousy,
loneliness, depression,
and low
self-esteem. Journal
of
Social
and
Clinical
Psychology,
9,
221-229.
Leary,
M. R.
(1994).
Self-presentation:
Impression management
and
interpersonal
behavior.
Dubuque,
IA:
Brown
&
Benchmark.
Leary,
M.
R.,
&
Downs,
D. L. (in
press).
Interpersonal functions
of the
self-esteem
motive:
The
self-esteem system
as a
sociometer.
In
M.
Kernis
(Ed.),
Efficacy,
agency,
and
self-esteem.
New
York:
Plenum
Press.
Leary,
M.
R.,
&
Forsyth,
D. R.
(1987).
Attributions
of
responsibility
for
collective endeavors.
In C.
Hendrick
(Ed.),
Review
of
personality
and
social
psychology.
Vol.
8:
Group processes (pp. 167-188). New-
bury
Park,
CA:
Sage.
Leary,
M.
R.,
Knight,
P.
D.,
&
Johnson,
K. A.
(1987).
Social anxiety
and
dyadic conversation:
A
verbal response analysis. Journal
of
So-
cial
and
Clinical
Psychology,
5,
34-50.
Leary,
M.
R.,
&
Kowalski,
R. M.
(1990).
Impression
management:
A
literature
review
and
two-component model.
Psychological
Bulletin,
107,
34-47.
Lewin,
K.
(1951).
Field theory
in
social sciences.
New
York: Harper
&
Row.
Lieberman,
M.
A.,
Yalom,
I.
D.,
&
Miles,
M. B.
(1973).
Encounter
groups:
First
facts.
New
York:
Basic Books.
Linville,
P. W., &
Jones,
E. E.
(1980).
Polarized appraisals
of
out-group
members.
Journal
of
Personality
and
Social
Psychology,
38,689-703.
Locksley,
A.,
Ortiz,
V.,
&
Hepburn,
C.
(1980).
Social
categorization
and
discriminatory behavior: Extinguishing
the
minimal
intergroup
discrimination
effect.
Journal
of
Personality
and
Social
Psychology,
39.
773-783.
Lofland,
L. H.
(1982).
Loss
and
human connection:
An
exploration
into
the
nature
of the
social
bond.
In W.
Ickes
& E. S.
Knowles
(Eds.),
Personality,
roles,
and
social behavior (pp.
219-242).
New
York:
Springer-Verlag.
Lowenthal,
M. F, &
Haven,
C.
(1968).
Interaction
and
adaptation:
In-
timacy
as a
critical variable. American Sociological Review,
33, 20-
30.
Lynch,
J. J.
(1979).
The
broken heart:
The
medical consequences
of
loneliness.
New
York:
Basic Books.
Mann,
L.
(1980).
Cross-cultural studies
of
small groups.
In H.
Triandis
& R.
Brislin
(Eds.),
Handbook
of
cross-cultural
psychology:
Social
psychology
(Vol.
5, pp.
155-209).
Boston:
Allyn
&
Bacon.
Manucia,
G.
K.,
Baumann,
D.
J.,
&
Cialdini,
R. B.
(1984).
Mood
in-
fluences on
helping: Direct
effects
or
side
effects?
Journal
of
Person-
ality
and
Social
Psychology,
46,
357-364.
Maslow,
A. H.
(1968).
Toward
a
psychology
of
being.
New
York:
Van
Nostrand.
Masserman,
J.
H.,
Wechkin,
S.,
&
Terris,
W.
(1964).
"Altruistic"
be-
havior
in
rhesus monkeys. American Journal
of
Psychiatry,
121, 584-
585.
Mathes,
E.
W.,
Adams,
H.
E.,
&
Davies,
R. M.
(1985).
Jealousy: Loss
of
relationship rewards, loss
of
self-esteem,
depression,
anxiety,
and
anger.
Journal
of
Personality
and
Social
Psychology,
48,
1552-1561.
May,
J.
L.,
&
Hamilton,
P. A.
(1980).
Effects
of
musically evoked
affect
on
women's interpersonal attraction
and
perceptual judgments
of
physical
attractiveness
of
men. Motivation
and
Emotion,
4,117-228.
McAdams,
D. P.
(1985).
Motivation
and
friendship.
In S.
Duck
&
D.
Perlman
(Eds.),
Understanding
personal relationships:
An
inter-
disciplinary
approach
(pp.
85-105).
Beverly Hills,
CA:
Sage.
McAdams,
D. P., &
Bryant,
F. B.
(1987).
Intimacy motivation
and
sub-
jective mental health
in a
nationwide sample. Journal
of
Personality,
55,395-413.
McCauley,
C.
R.,
&
Segal,
M. E.
(1987).
Social psychology
of
terrorist
groups.
In C.
Hendrick
(Ed.),
Group processes
and
intergroup
re-
lations:
Review
of
personality
and
social
psychology
(Vol.
9, pp.
231
-
256).
Newbury Park,
CA:
Sage.
McLeod,
E.
(1982).
Women working: Prostitution
today.
London:
Croom
Helm.
Meindl,
J.
R.,
&
Lerner,
M. J.
(1984).
Exacerbation
of
extreme
re-
sponses
to an
out-group. Journal
of
Personality
and
Social
Psychol-
ogy,
47,
71-84.
Miceli,
M.
(1992).
How to
make
someone
feel
guilty: Strategies
of
guilt
inducement
and
their
goals.
Journal
for
the
Theory
of
Social Behav-
iour,
22,
81-104.
Mijuskovic,
B.
(1980).
Loneliness:
An
interdisciplinary
approach.
In
J.
Hartog,
J. R.
Audy,
& Y. A.
Cohen
(Eds.),
The
anatomy
of
loneli-
ness
(pp.
65-94).
New
York: International Universities
Press.
Milardo, R.
M.,
Johnson, M. P., & Huston, T. L.
(1983).
Developing
close relationships: Changing
patterns
of
interaction between pair
members
and
social networks. Journal
of
Personality
and
Social Psy-
chology,
44,
964-976.
Miller,
N.
B.,
Cowan,
P.
A.,
Cowan,
C. P.,
Hetherington,
E.
M.,
&
Clingempeel,
W. G.
(1993).
Externalizing
in
preschoolers
and
early
adolescents:
A
cross-study replication
of a
family
model.
Develop-
mental
Psychology,
29,
3-18.
Miller,
R.
S.,
&
Leary,
M. R.
(1992).
Social sources
and
interactive
functions
of
emotion:
The
case
of
embarrassment.
In M.
Clark
(Ed.),
Review
of
personality
and
social psychology: Vol.
14.
Emotion
and
social
behavior (pp.
202-221).
Newbury Park,
CA:
Sage.
Monson,
T.
C.,
Keel,
R.,
Stephens,
D.,
&
Genung,
V.
(1982).
Trait
at-
tributions: Relative validity, covariation with behavior,
and
prospect
of
future
interaction.
Journal
of
Personality
and
Social
Psychology,
42,
1014-1024.
Moreland,
R. L.
(1987).
The
formation
of
small groups.
In C.
Hendrick
(Ed.),
Group processes: Review
of
personality
and
social
psychology
(Vol.
8, pp.
80-110).
Newbury Park,
CA:
Sage.

THE
NEED
TO
BELONG
527
Moreland,
R. L., &
Levine,
J. M.
(1989).
Newcomers
and
oldtimers
in
small
groups.
In P.
Paulus
(Ed.),
Psychology
of
group
influence
(Vol.
2,
pp.
143-186).
Hillsdale,
NJ:
Erlbaum.
Morgenthau,
H.
(1962).
Love
and
power. Commentary,
33,
247-251.
Morris,
D. R.
(1965).
The
washing
of
the
spears:
The
rise
and
fall
of
the
Zulu
nation.
New
York:
Simon
&
Schuster.
Myers,
D.
(1992).
The
pursuit
of
happiness.
New
York: Morrow.
Nahemow,
L.,
&
Lawton,
M. P.
(1975).
Similarity
and
propinquity
in
friendship
formation. Journal
of
Personality
and
Social
Psychology,
32,205-213.
O'Grady,
K.
E.
(1989).
Physical attractiveness, need
for
approval,
so-
cial
self-esteem,
and
maladjustment. Journal
of
Social
and
Clinical
Psychology,
8,
62-69.
Olmos,
E.
(Producer).
(1994,
April
8).
Lives
in
hazard.
New
York:
National Broadcasting Company.
Orbell,
J.
M.,
van de
Kragt,
A.,
&
Dawes,
R. M.
(1988).
Explaining
discussion-induced cooperation. Journal
of
Personality
and
Social
Psychology,
54,
811-819.
Ostrom,
T.
M.,
Carpenter,
S.
L.,
Sedikides,
C., & Li, F.
(1993).
Differ-
ential
processing
of
in-group
and
out-group information. Journal
of
Personality
and
Social
Psychology,
64,
21-34.
Paloutzian,
R. E, &
Janigian,
A. S.
(1987).
Models
and
methods
in
loneliness
research: Their status
and
direction.
In M.
Hoja
& R.
Cran-
dall
(Eds.),
Loneliness:
Theory,
research,
and
applications
(pp.
31-
36).
San
Rafael,
CA:
Select Press.
Panksepp,
J.,
Siviy,
S.
M.,
&
Normansell,
L. A.
(1985).
Brain opioids
and
social emotions.
In M.
Reite
& T.
Field
(Eds.),
Thepsychobiology
of
attachment
and
separation
(pp.
3-49).
New
York:
Academic
Press.
Pearce,
P. L.
(1980).
Strangers, travelers,
and
Greyhound terminals:
A
study
of
small-scale helping behaviors. Journal
of
Personality
and
Social
Psychology,
38,
935-940.
Pearlman,
C. A.
(1970).
Separation reactions
of
married women.
American Journal
of
Psychiatry,
126,
946-950.
Perlman,
D.
(1987).
Further reflections
on the
present state
of
loneli-
ness research.
In M.
Hoja
& R.
Crandall
(Eds.),
Loneliness:
Theory,
research,
and
applications
(pp.
17-26).
San
Rafael,
CA:
Select
Press.
Perloff,
L.
S.,
&
Fetzer,
B. K.
(1986).
Self-other judgments
and
per-
ceived
vulnerability
to
victimization. Journal
of
Personality
and
So-
cial
Psychology,
50,
502-510.
Pines,
M.,
&
Aronson,
E.
(1983).
Antecedents,
correlates,
and
conse-
quences
of
sexual jealousy. Journal
of
Personality,
51,
108-135.
Price,
S.
J.,
&
McKenry,
P. C.
(1988).
Divorce. Beverly Hills,
CA:
Sage.
Pryor,
J.
M.,
&
Ostrom,
T. M.
(1981).
The
cognitive organization
of
social
life:
A
converging-operations approach. Journal
of
Personality
and
Social
Psychology,
41,
628-641.
Rabbie,
J.
M.,
&
Horwitz,
M.
(1969).
Arousal
of
ingroup-outgroup
bias
by a
chance
win or
loss. Journal
of
Personality
and
Social Psy-
chology,
13,
269-277.
Reis,
H. T.
(1990).
The
role
of
intimacy
in
interpersonal relations.
Journal
of
Social
and
Clinical
Psychology,
9,
15-30.
Reis,
H.
T.,
Wheeler,
L.,
Kernis,
M.
H.,
Spiegel,
N., &
Nezlek,
J.
(1985).
On
specificity
in the
impact
of
social participation
on
physical
and
psychological
health. Journal
of
Personality
and
Social
Psychology,
48,456-471.
Reiss,
I. L.
(1986).
A
sociological journey into sexuality. Journal
of
Marriage
and the
Family,
48,
233-242.
Rofe,
Y.
(1984).
Stress
and
affiliation:
A
utility theory. Psychological
Review,
91,
235-250.
Rogers,
C. R.
(1959).
A
theory
of
therapy,
personality,
and
interper-
sonal relationships,
as
developed
in the
client-centered framework.
In
S.
Koch
(Ed.),
Psychology:
A
study
of
a
science
(Vol.
3, pp.
184-256).
New
York:
McGraw-Hill.
Rook,
K. S.
(1987a).
Reciprocity
of
social exchange
and
social
satisfac-
tion among older women. Journal
of
Personality
and
Social
Psychol-
ogy,
52,
145-154.
Rook,
K. S.
(1987b).
Social
support
versus companionship:
Effects
on
life
stress,
loneliness,
and
evaluations
by
others.
Journal
of
Personal-
ity
and
Social
Psychology,
52,
1132-1147.
Rosen,
R.
(1979).
Some crucial issues concerning children
of
divorce.
Journal
of
Divorce,
3,
19-25.
Rosen,
S.,
Mickler,
S.
E.,
&
Collins,
J. E.
(1987).
Reactions
of
would-be
helpers whose
offer
of
help
is
spurned. Journal
of
Personality
and
Social
Psychology,
53,
288-297.
Rosenberg,
T.
(1991).
Children
of
Cain:
Violence
and the
violent
in
Latin
America.
New
\brk:
Penguin Books.
Ross,
C.
(1976).
The
Wars
of
the
Roses:
A
concise history.
New
York:
Thames
&
Hudson.
Rothberg,
J.
M.,
&
Jones,
F. D.
(1987).
Suicide
in the
U.S. Army: Epi-
demiological
and
periodic
aspects.
Suicide
and
Life-Threatening
Be-
havior,
17,
119-132.
Roy,
M.
(Ed.).
(1977).
Battered
women.
New
York:
Van
Nostrand.
Ruehlman,
L.
S.,
&
Wolchik,
S. A.
(1988).
Personal goals
and
interper-
sonal support
and
hindrance
as
factors
in
psychological
distress
and
well-being. Journal
of
Personality
and
Social
Psychology,
55,
293-
301.
Rusbult,
C. E.
(1980).
Commitment
and
satisfaction
in
romantic
asso-
ciations:
A
test
of the
investment model. Journal
of
Experimental
Social
Psychology,
16,
172-186.
Rusbult,
C.
E.,
Verette,
J.,
&
Drigotus,
S. M.
(1994).
Absolute commit-
ment
level,
mutuality
of
commitment,
and
couple
adjustment
in
mar-
ital relationships.
Unpublished
manuscript,
University
of
North
Car-
olina
at
Chapel Hill.
Rusbult,
C.
E.,
Zembrodt,
I.
M.,
&
Gunn,
L. K.
(1982).
Exit, voice,
loyalty,
and
neglect: Responses
to
dissatisfaction in'romantic involve-
ments.
Journal
oj
Personality
and
Social
Psychology,
43,
1230-1242.
Russell,
D.,
Cutrona,
C.,
Rose,
J.,
&
Yurko,
K..
(1984).
Social
and
emo-
tional loneliness:
An
examination
of
Weiss's typology
of
loneliness.
Journal
of
Personality
and
Social
Psychology,
46,
1313-1321.
Rutkowski,
G.
K.,
Cruder,
C.
L.,
&
Romer,
D.
(1983).
Group cohesive-
ness, social norms,
and
bystander intervention. Journal
of
Personal-
ity
and
Social
Psychology,
44,
545-552.
Rutter,
M.
(1979).
Maternal deprivation,
1972-1978:
New findings,
new
concepts,
new
approaches.
Child
Development,
50,
283-305.
Rutter,
M.,
&Garmezy,
N.
(1983).
Developmental
psychopathology.
In
E.
M.
Hetherington
(Ed.),
Handbook
ofchild
psychology
(Vol.
4, pp.
775-911).
New
York:
Wiley.
Ryan,
R. M.
(1991).
The
nature
of the
self
in
autonomy
and
related-
ness.
In J.
Strauss
& G. R.
Goethals
(Eds.),
The
self:
Interdisciplinary
approaches
(pp.
208-238).
New
York:
Springer-Verlag.
Saklofske,
D.
H.,
&
Yackulic,
R. A.
(1989).
Personality predictors
of
loneliness.
Personality
and
Individual
Differences,
10,
467-472.
Sampson,
R.
J.,
&
Laub,
J. H.
(1993).
Crime
in the
making: Pathways
and
turning points
through
life.
Cambridge,
MA:
Harvard University
Press.
Sarbin,
T.
(1982).
The
dangerous individual:
An
outcome
of
social
identity transformations.
In V. L.
Allen
& K. E.
Scheibe
(Eds.),
The
social
context
of
conduct
(pp.
113-118).
New
York:
Pergamon
Press.
Schachter,
S.
(1951).
Deviance, rejection,
and
communication. Journal
of
Abnormal
and
Social
Psychology,
46,
190-207.
Schachter,
S.
(1959).
The
psychology
of
affiliation.
Stanford,
CA:
Stan-
ford
University Press.
Schlenker,
B. R.
(1980).
Impression management:
The
self-concept,
so-
cial
identity,
and
interpersonal
relations. Monterey,
CA:
Brooks/
Cole.
Schoenrade,
P.
A.,
Batson,
C.
D.,
Brandt,
J.
R.,
&
Loud,
R. E.
(1986).
Attachment, accountability,
and
motivation
to
benefit
another
not in
distress. Journal
of
'Personality
and
Social
Psychology,
51,
557-563.

528
ROY F.
BAUMEISTER
AND
MARK
R.
LEARY
Scudder,
K.
J.
(1952).
Prisoners
are
people. Garden City,
NY:
Doubleday.
Sedikides,
C,
Olsen,
N., &
Reis,
H. T.
(1993).
Relationships
as
natural
categories. Journal
of
Personality
and
Social
Psychology,
64,
71-82.
Shaver.
P., &
Buhrmester,
D.
(1983).
Loneliness, sex-role orientation,
and
group
life:
A
social
needs
perspective.
In P.
Paulus
(Ed.),
Basic
group
processes (pp.
259-288).
New
York:
Springer-Verlag.
Shaver,
P.,
Hazan,
C.,
&
Bradshaw,
D.
(1988).
Love
as
attachment:
The
integration
of
three behavioral systems.
In R. J.
Sternberg
&
M.
L.
Barnes
(Eds.),
The
psychology
of
love
(pp.
68-99).
New
Haven,
CT:
Yale
University Press.
Shepperd,
J.
A.,
&
Arkin,
R. M.
(1990).
Shyness
and
self-presentation.
In
W. R.
Crozier
(Ed.),
Shyness
and
embarrassment: Perspectives
from social
psychology
(pp.
286-314).
Cambridge, England: Cam-
bridge University Press.
Sherif,
M.,
Harvey,
O.
H.,
White,
B.
J.,
Hood,
W.
R.,
&
Sherif,
C. W.
(1988).
The
Robbers Cave
experiment:
Intergroup
conflict
and
coop-
eration.
Middletown,
CT:
Wesleyan University
Press.
(Original
work
published
1961)
Snyder,
A. 1.
(1978).
Periodic marital separation
and
physical illness.
American Journal
of
Orthopsychiatry,
48,
637-643.
Snyder,
C.
R.,
Higgins,
R.
L.,
&
Stucky,
R. J.
(1983).
Excuses: Mas-
querades
in
search
of
grace.
New
York:
Wiley.
Solomon,
Z.,
Waysman,
M.,
&
Mikulincer,
M.
(1990).
Family
func-
tioning,
perceived social support,
and
combat-related
psychopathol-
ogy:
The
moderating role
of
loneliness. Journal
of
Social
and
Clinical
Psychology.
9,
456-472.
Sours,
J. A.
(1974).
The
anorexia nervosa syndrome. International
Journal
of
Psychoanalysis,
55,
567-576.
Spanier,
G.
B.,
&
Casto,
R. F.
(1979).
Adjustment
to
separation
and
divorce:
A
qualitative analysis.
In G.
Levinger
& O. C.
Moles
(Eds.),
Divorce
and
separation:
Context, causes,
and
consequences (pp.
211-
227).
New
York:
Basic Books.
Spanier,
G.
B.,
&
Lewis,
R. A.
(1980).
Marital quality:
A
review
of the
seventies. Journal
of
Marriage
and the
Family,
42,
825-839.
Spivey,
E.
(1990).
Social exclusion
as a
common
factor
in
social anxiety,
loneliness,
jealousy,
and
social depression. Unpublished
master's
the-
sis,
Wake Forest
University,
Winston-Salem,
NC.
Stark,
R.,
&
Bainbridge,
W. S.
(1985).
The
future
of
religion:
Secular-
ization,
revival,
and
cult
formation.
Berkeley: University
of
California
Press.
Staub,
E.
(1989).
The
roots
of
evil:
The
origins
of
genocide
and
other
group
violence.
Cambridge, England: Cambridge University
Press.
Staw,
B.
M.,
Sandelands,
D.
E.,
&
Dutton,
J. E.
(1981).
Threat-rigidity
effects
in
organizational behavior:
A
multi-level
analysis.
Administra-
tive
Science Quarterly,
26,
501-524.
Stein,
A. A.
(1976).
Conflict
and
cohesion:
A
review
of the
literature.
Journal
of
Conflict
Resolution,
20,
143-172.
Sternberg,
R. J.
(1986).
A
triangular theory
of
love. Psychological
Re-
view,
93,
119-135.
Strube,
M. J.
(1988).
The
decision
to
leave
an
abusive relationship:
Empirical
evidence
and
theoretical
issues.
Psychological Bulletin,
104,
236-250.
Sullivan,
H. S.
(1953).
The
interpersonal
theory
of
psychiatry.
New
York:
Norton.
Symanski,
R.
(1980).
Prostitution
in
Nevada.
In E.
Muga
(Ed.),
Stud-
ies
in
prostitution
(pp.
246-279).
Nairobi, Kenya: Kenya
Literature
Bureau.
Tajfel,
H.
(1970).
Experiments
in
intergroup discrimination. Scientific
American, 223,
96-102.
Tajfel,
H.,
&
Billig,
M.
(1974).
Familiarity
and
categorization
in in-
tergroup behavior. Journal
of
Experimental Social
Psychology,
10,
159-170.
Tajfel,
H.,
Flament,
C.,
Billig,
M.
G.,
&
Bundy,
R. F.
(1971).
Social
categorization
and
intergroup behaviour. European Journal
of
Social
Psychology,
I,
149-177.
Tambor,
E.
S.,
&
Leary,
M. R.
(1993).
Perceived
exclusion
as a
common
factor
in
social anxiety, loneliness, jealousy, depression,
and
low
self-
esteem.
Manuscript submitted
for
publication.
Tangney,
J. P.
(1992).
Situational determinants
of
shame
and
guilt
in
young
adulthood.
Personality
and
Social
Psychology
Bulletin,
18,
199-206.
Tesser,
A.
(1991).
Emotion
in
social comparison
and
reflection pro-
cesses.
In J.
Suls
& T. A.
Wills
(Eds.),
Social comparison: Contempo-
rary
theory
and
research (pp.
117-148).
Hillsdale,
NJ:
Erlbaum.
Tesser,
A.,
Millar,
M.,
&
Moore,
J.
(1988).
Some
affective
consequences
of
social comparison
and
reflection
processes:
The
pain
and
pleasure
of
being close. Journal
of
Personality
and
Social
Psychology,
54, 49-
61.
Tice,
D.
M.,
Butler,
J.
L.,
Muraven,
M.
B.,
&
Stillwell,
A. M.
(1994).
When
modesty prevails:
Differential
favorability
of
self-presentation
to
friends
and
strangers.
Manuscript submitted
for
publication.
Toch,
H.
(1975).
Men in
crisis. Chicago: Aldine.
Toch,
H.
(1977).
Living
in
prison:
The
ecology
of
survival.
New
York:
Free Press.
Toch,
H.
(1992).
Violent
men. Washington,
DC:
American Psychologi-
cal
Association.
Toi,
M.,
&
Batson,
C. D.
(1982).
More evidence that empathy
is a
source
of
altruistic motivation. Journal
of
Personality
and
Social
Psy-
chology,
43,
281-292.
Trout,
D. L.
(1980).
The
role
of
social isolation
in
suicide. Suicide
and
Life-Threatening
Behavior,
W,
10-23.
Turner,
J. C.
(1985).
Social categorization
and the
self-concept:
A
social
cognitive
theory
of
group behavior.
In E. J.
Lawler
(Ed.),
Advances
in
group
processes: Theory
and
research (Vol.
2, pp.
77-121).
Green-
wich,
CT:
JAI
Press.
Vallacher,
R.
R.,
&
Nowak,
A.
(1994).
The
chaos
in
social
psychology.
In
R.
Vallacher
& A.
Nowak
(Eds.),
Dynamical systems
in
social psy-
chology
(pp.
1-16).
San
Diego,
CA:
Academic Press.
Vangelisti,
A.
L.,
Daly,
J.
A.,
&
Rudnick,
J. R.
(1991).
Making people
feel
guilty
in
conversations: Techniques
and
correlates. Human Com-
munication
Research,
18,
3-39.
Vaughan,
D.
(1986).
Uncoupling.
New
York:
Oxford University
Press.
Vaux,
A.
(1988).
Social
and
emotional loneliness:
The
role
of
social
and
personal characteristics. Personality
and
Social
Psychology
Bulletin,
14,
722-734.
Veitch,
R.,
&
Griffin,
W.
(1976).
Good
news,
bad
news:
Affective
and
interpersonal
effects.
Journal
of
Applied
Social
Psychology,
6,
69-75.
Vinokur,
A.
D.,
& van
Ryn,
M.
(1993).
Social support
and
undermin-
ing
in
close relationships: Their independent
effects
on
mental health
of
unemployed
persons.
Journal
of
Personality
and
Social Psychol-
ogy,
65,
350-359.
Wade,
W. C.
(1987).
The fiery
cross:
The
Kit
Klux
Klan
in
America.
New
"York:
Touchstone/Simon
&
Schuster.
Wegner,
D. M.
(1986).
Transactive memory:
A
contemporary analysis
of
the
group mind.
In B.
Mullen
& G. R.
Goethals
(Eds.),
Theory
of
group
behavior
(pp.
185-208).
New
York:
Springer-Verlag.
Wegner,
D.
M.,
Erber,
R.,
&
Raymond,
P.
(1991).
Transactive memory
in
close relationships. Journal
of
Personality
and
Social
Psychology,
61,
923-929.
Weiss,
R. S.
(1973).
Loneliness:
The
experience
of
emotional
and
social
isolation. Cambridge,
MA: MIT
Press.
Weiss,
R. S.
(1979).
The
emotional impact
of
marital separation.
In
G.
Levinger
& O. C.
Moles
(Eds.),
Divorce
and
separation: Context,
causes,
and
consequences (pp.
201-210).
New
York:
Basic Books.
West,
S.,
Newsom,
J. T, &
Fenaughty,
A. M.
(1992).
Publication trends
in
Journal
of
Personality
and
Social
Psychology:
Stability
and
change

THE
NEED
TO
BELONG
529
in
topics, methods,
and
theories
across
two
decades.
Personality
and
Social
Psychology
Bulletin,
18,
473-484.
Wheeler,
L., &
Nezlek,
J.
(1977).
Sex
differences
in
social participation.
Journal
of
Personality
and
Social
Psychology,
35,
742-754.
Wheeler,
L.,
Reis,
H.
T,
&
Nezlek,
J.
(1983).
Loneliness, social interac-
tion,
and sex
roles. Journal
of
Personality
and
Social
Psychology,
45,
943-953.
Wilbur,
J. R., &
Wilbur,
M.
(1988).
The
noncustodial parent: Dilem-
mas and
interventions. Journal
of
Counseling
and
Development,
66,
434-437.
Wilder,
D.
A.,
&
Thompson,
J. E.
(1980).
Intergroup
contact with
in-
dependent manipulations
of
in-group
and
out-group interaction.
Journal
of
Personality
and
Social
Psychology,
38,
589-603.
Williams,
J. G., &
Solano,
C. H.
(1983).
The
social
reality
of
feeling
lonely:
Friendship
and
reciprocation.
Personality
and
Social
Psychol-
ogy
Bulletin,
9,
237-242.
Winfield,
F. E.
(1985).
Commuter marriage.
New
York:
Columbia Uni-
versity
Press.
Yalom,
I.
(1985).
The
theory
and
practice
of
group
psychotherapy.
New
York:
Basic Books.
Zander,
A.
(1971).
Motives
and
goals
in
groups.
New
York:
Academic
Press.
Received
August
27,1993
Revision
received
November
4,
1994
Accepted
November
7,
1994
•
AMERICAN
PSYCHOLOGICAL ASSOCIATION
SUBSCRIPTION
CLAIMS
INFORMATION
Today's
Date:.
We
provide
this
form
to
assist
members,
institutions,
and
nonmember
individuals
with
any
subscription
problems.
With
the
appropriate
information
we can
begin
a
resolution.
If
you
use the
services
of
an
agent,
please
do
NOT
duplicate
claims
through
them
and
directly
to us.
PLEASE
PRINT
CLEARLY
AND IN INK IF
POSSIBLE.
PRINT
FULL
NAME
OR
KEY
NAME
OF
INSTITUTION
ADDRESS
MEMBER
OR
CUSTOMER
NUMBER
(MAY
BE
FOUND
ON
ANY
PAST ISSUE LABEL)
DATE YOUR ORDER
WAS
MAILED
(OR
PHONED)
PREPAID CHECK
CHARGE
CHECK/CARD
CLEARED
DATE:_
CITY
STATECOUNTRY
YOUR
NAME
AND
PHONE NUMBER
TITLE
(If
possible, send
a
copy,
front
and
back,
of
your cancelled check
to
help
us in our research
of
your claim.)
ISSUES:
MISSING DAMAGED
VOLUME
OR
YEAR
NUMBER
OR
MONTH
Thank
you.
Once
a
claim
is
received
and
resolved,
delivery
of
replacement
issues
routinely
takes
4-6
weeks.
^———
(TO
BE
FILLED
OUT
BY
APA
STAFF)
——1—^__———._
DATE
RECEIVED:.
ACTION
TAKEN:
_
STAFF
NAME:
DATE
OF
ACTION:
_
INV.
NO. &
DATE:
LABEL
NO.
&
DATE:
Send this
form
to APA
Subscription
Claims,
750
First Street,
NE,
Washington,
DC
20002-4242
PLEASE
DO
NOT
REMOVE.
A
PHOTOCOPY
MAY BE
USED.