Balance of Power and System Stability: Simulating Complex Anarchical Environments Over the Internet
| Author | Douglas A. Van Belle |
| Published date | 01 March 1998 |
| Date | 01 March 1998 |
| DOI | http://doi.org/10.1177/106591299805100112 |
| Subject Matter | Articles |
265
Balance
of
Power
and
System
Stability:
Simulating
Complex
Anarchical
Environments
Over
the
Internet
DOUGLAS
A.
VAN
BELLE,
UNIVERSITY
OF
NEW
ORLEANS
Game
theoretic
examinations
of
balance
of
power
hypotheses
in
the
context
of N-player
games
have
identified
certain
distributions
of
resources
which
should
lead
to
system
stability,
defined
as
the
continued
survival
of
all
participants
(Wagner
1986,
Niou,
Ordeshook,
and
Rose
1989;
Niou
and
Ordeshook
1990).
This
article
uses
data
from
a
seven-player
simula-
tion
of
realist
international
politics
to
analyze
hypotheses
on
power
dis-
tributions,
the
number
of
actors,
and
system
stability
By
conducting
the
simulation
over
the
Internet,
a
tremendous
number
of
runs
through
the
simulation
can
be
completed
in
a
relatively
short
time,
thereby
generat-
ing
sufficient
data
for
the
statistical
analysis
of
the
distributions
of
power
that
occur.
These
data
can
then
be
used
to
examine
how
those
distribu-
tions
relate
to
the
number
of
participants
found
in
the
eventual
end-state
of
the
system.
The
primary
contribution
of
this
article
is
empirical.
Data
generated
with
a
simulation
of
realist
international
politics
are
used to
examine
the
applica-
bility
of
game
theoretic
propositions
to
humans
interacting
in
environments
that
are
similar,
but
still
much
more
complex
than
the
precise
and
certain
contexts
necessary
for
formal
models.
In
this
way
the
analysis
provides
an
intermediate
step
between
formal
theory
and
real-world
complexity.
Within
the
context
of
a
brief
outline,
four
existing
propositions
concerning
the
distri-
bution
of
power,
the
number
of
actors,
and
stability
within
an
anarchic
NOTE:
The
author
would
like
to
thank
Pat
Kenney,
James
Van
Belle,
the
editors
of
Political
Research
Quarterly,
the
anonymous
reviewers,
and
all
of
the
participants
in
the
research
simulation.
266
environment
are
presented.
After
presenting
some
of
the
details
of
the
simula-
tion,
these
propositions
are
then
operationalized
into
five
hypotheses
that
can
be
tested
using
the
simulation
data.
The
bivariate
analyses
produce
some
clear
statistical
results
while
the
multivariate
analysis
produces
a
modest
surprise.
In
exploring
the
unexpected
multivariate
results,
some
suggestions
are
made
concerning
future
theoretical
and
empirical
work.
One
thing
that
this
article
does
not
address
is
the
contested
nature
of
the
overall
concept
of
the
balance
of
power.
No
concept
causes
more
difficulty
or
is
a
greater
source
of
confusion
in
discussions
of
international
relations
than
the
concept
of
balance
of
power.
Scholars
are
uncertain
about
whether
this
term
refers
to
a
theory
of
con-
flict
and
coalitions,
to
a
description
of
international
systems,
to
the
goals
of
key
decision
makers,
or
to
a
normative
prescription
about
how
inter-
national
systems
ought
to
achieve
peace.
(Niou,
Ordeshook,
and
Rose
1989:
28.1
Wagner
(1986)
as
well
as
Niou
and
Ordeshook
(1987)
and
Niou,
Ordeshook,
and
Rose
(1989)
address
the
contested
concept
of
balance
of
power
at
great
length
and
in
great
detail.
They
all
focus
on
balance
of
power
as
a
relationship
between
the
distribution
of
resources
and
what
they
define
as
system
stability.
According
to
their
definition,
an
anarchical
system
is
stable
when
each
and
every
state
in
the
system
can
guarantee
its
own
continued
existence.
In
their
models,
this
is
considered
separate
from
what
they
refer
to
as
resource
stability,
where
all
states
in
the
system
can
guarantee
both
their
continued
existence
and
maintain
their
current
level
of
resources.
An
anarchic
system
that
is
resource
stable
is
frozen:
no
alterations
in
the
distribution
of
power
can
occur.
If
no
changes
in
the
distribution
of
power
occur,
then
no
players
can
be
reduced
to
zero-power
and
eliminated.
Therefore,
the
system
must
also
be
system
stable.
The
models
they
construct
demonstrate
that
in
game
theoretic
contexts
of
rational
interaction,
system
stability
can
be
achieved
with
certain
distributions
of
power
among
various
numbers
of
self-interested
actors.
Some
of
these
are
dynamic
situations
that
are
not
frozen
and
are
not
resource
stable
even
though
they
are
system
stable.
Balances
of
power
are
defined
as
distributions
of
resources
that
lead
to
one
or
both
of
these
types
of
stability
This
limited
depiction
of
balance
of
power
does
not
capture
the
entire
range
of
concepts
that
could
be
included
under
the
rubric
of
a
balance
of
power,
but
it
does
provide
a
carefully
defined
and
theoreti-
cally
grounded
foundation
for
the
empirical
analyses
presented
below
1
Niou
and
Ordeshook
make
a
similar
statement
in
Niou
and
Ordeshook
(1987:
685).
To continue reading
Request your trial