A Vietnam Simulation: a Report On the Canadian/English Joint Project
| DOI | 10.1177/002234336700400101 |
| Published date | 01 March 1967 |
| Date | 01 March 1967 |
| Author | Paul Smoker,John Macrae |
| Subject Matter | Articles |
A
VIETNAM
SIMULATION:
A
REPORT
ON
THE
CANADIAN/ENGLISH JOINT
PROJECT*
By
JOHN
MACRAE
and
PAUL
SMOKER
Peace
Research
Centre,
Lancaster
Peace
Research
Centre,
Lancaster,
and
Northwestern
University,
Evanston,
Illinois
1.
Introduction
This
paper
is
a
description
of
the
Viet-
nam
simulation
undertaken
on
a
joint
basis
by
the
Canadian
Peace
Research
Institute
and
the
Lancaster
Peace
Re-
search
Centre.
No
empirical
findings
are
included,
but
a
theoretical
framework
is
put
forward
and
its
relevance
to
the
Vietnam
simulation
is
elaborated.
Inter-
spersed
with
the
main
argument
of
the
article
are
descriptive
sections
which
illus-
trate
events
within
the
Vietnam
simu-
lation.
Technical
details
concerning
the
simulation
are
given
in
the
Appendix,
together
with
the
major
assumptions
in-
volved
in
the
gaming
and
mathematical
sections
of
the
simulation.
2.
Models
Richard
Brody
defines
a
model
as
a
collection
of
assertions
about
some
reality-
past,
present
or
predicted.’
Further,
the
statements
purport
to
describe
patterns
of
relationships
holding
between
compo-
nents -
units
and
variables -
of
that
reality.
At
the
same
time,
Brody
suggests
a
model
is
simplified
as
it
is
less
than
a
total
representation
of
the
details
of
the
phenomena
of
interest.
Russell
Ackoff
argues
that
the
mean-
ing
of
the
word
’model’
as
a
noun
differs
from
the
meaning
as
an
adjective
or
a
verb.2
As
a
noun,
a
’model’
is
a
represen-
tation
in
the
sense
in
which
an
architect
constructs
a
small
scale
model
of
a
building
or
a
physicist
a
large
scale
model
of
an
atom.
As
an
adjective,
’model’
implies
perfection
or
idealization,
and
as
a
verb
’to
model’
means
to
demonstrate
or
reveal
or
show
what
a
thing
is
like.
Ackoff
suggests
that
in
science
models
have
all
these
connotations,
as
(1)
they
are
rep-
resentations
of
states,
objects,
and
events,
(2)
they
are
idealized
as
they
are
simplified
and
include
only
relevant
properties
of
reality,
and
(3)
they
model
in
the
sense
that
they
hopefully
demonstrate
or
reveal
or
show
what
reality
is
like.
Brody
classifies
models
into
four
distinct
types -
pictorial,
verbal,
mathematical,
and
simulational.
He
does
not
suggest
that
these
four
types
exhaust
the
possible
universe
of
models,
rather
that
they
are
those
most
commonly
found
in
social
science.
Pictorial
(or
iconic)
models
are
used
primarily
for
description
and
are
essentially
static;
verbal
models
are
able
to
describe
relationships
and
can
possess
a
certain
amount
of
logical
consistency;
while
mathematical
models
are
powerful
scientific
tools
that
can
facilitate
the
re-
cognition
of
similarities
and
congruences
between
models
and
the
realities
they
represent.
Further,
mathematical
models
can
be
used
to
explore
alternative
states
of
a
system.3
3
.
Ackoff
suggests
just
three
types
of
models,
namely
iconic,
analogue
and
symbolic.
Iconic
models
are
defined
as
large
or
small
scale
representations
of
states,
ob-
jects,
or
events.
Iconic
models,
such
as
road
maps
and
aerial
photographs,
look
like
what
they
represent,
as
they
show
relative
distances
between
places
and
relative
positions
of
roads.
There
is,
in
other
words,
a
similarity
in
appearance
between
the
model
and
the
relevant
aspects
of
the
thing
being
modeled.
An
2
analogue,
however,
uses
one
property
to
represent
another
(for
example,
repre-
senting
an
electrical
system
by
a
hydraulic
system
through
making
the
flow
of
water
in
pipes
an
analogue
of
current);
while
a
symbolic
model
expresses
the
properties
of
the
thing
represented,
as
in
the
case
of
the
Lanchester
equations.4
Ackoff
argues
that
(1)
iconic
models
are
the
most
specific
and
concrete
but
are
usually
the
most
difficult
to
manipulate
for
purposes
of
determining
the
effect
of
changes
on
the
real
thing;
(2)
analogue
models
substitute
easier
to
manipulate
properties
and
as
a
consequence
are
more
abstract
and
gen-
eral ;
and
(3)
symbolic
models,
which
are
the
most
abstract
and
general,
are
the
easiest
to
manipulate.5
While
there
is
considerable
similarity
between
the
three
types
of
models
sug-
gested
by
Ackoff
and
the
first
three
types
suggested
by
Brody,
Ackoff
does
not
view
simulation
as
a
model.
Brody
sees
simu-
lations
as
physical
and/or
biological
rep-
resentations
of
systems
which
attempt
to
replicate
sociopolitical
processes,
and
ar-
gues
that
they
are
models
which
yield
infor-
mation
about
entity
and
variable
changes
over
time.
Ackoff
argues
that
simulation
is
a
way
of
using
a
model,
that
it
is
ex-
perimentation
on
a
model
rather
than
on
the
phenomenon
itself.
While
Brody
also
makes
the
point
that
simulation
involves
experimentation,
the
distinction
Ackoff
makes
between
simulations
and
models
may
be
of
some
value.6
3.
Simulations
Brody
suggests
three
subtypes
of
simu-
lation,
namely
machine
or
computer
simu-
lation,
mixed
or
man-machine
simulations,
and
all-man
simulations
or
games.
Ma-
chine
simulations
he
sees
as
operational
models
(which
have
been
programmed
for
high
speed
computing
equipment)
where
the
humans
are
involved
primarily
as
experimenters.7
All-man
simulations
are
nearly
opposite
in
purpose
to
the
all-
machine
or
computer
simulation.
Here
the
focus
is
on
testing
general
subjective
statements
about
the
international
be-
havior
of
states
and,
to
a
lesser
extent,
checking
certain
hypotheses
about
strate-
gic
behavior
suggested
by
the
theory
of
games
against
the
actual
behavior
of
participants
in
the
simulation.8
The
Vietnam
simulation
can
be
classed
as
an
example
of
Brody’s
third
subtype,
namely
man-machine
or
mixed
simula-
tion.
Mixed
simulations
try
to
reap
the
benefits
of
all-computer
and
all-man
simulations
by
using
both
men
and
machines.9
Ackoff
points
out
that
simulation
has
come
to
mean
different
things
to
different
people,
but
in
line
with
his
background
(operations
research)
he
concentrates
on
its
application
for
problem
solving.
He
suggests
four
principal
uses
of
simu-
lations
in
this
context.
They
are
( 1 )
to
determine
the
optimizing
values
of
con-
trolled
variables;
(2)
to
study
transitional
processes;
(3)
to
estimate
values
of
model
parameters
or
the
model’s
functional
form;
(4)
to
treat
courses
of
action
which
cannot
be
formulated
in
the
model.
Ackoff
suggests
three
types
of
simulation,
namely
iconic
simulation,
analogue
simu-
lation,
and
symbolic
simulation.
Given
Ackoff’s
three
types
of
models
and
the
view
that
simulation
is
experimentation
on
a
model,
the
above
types
of
simulation
are
self-explanatory. 10
Superimposed
on
this
classification,
Ackoff
points
out
that
a
simulation
in
which
human
decision-
makers
are
involved
is
often
called
gam-
ing.ll
He
suggests
three
uses
of
gaming
in
problem
solving
research:
( 1 )
to
help
develop
a
decision
model,
(2)
to
help
find
the
solution
to
such
a
model,
and
(3)
to
help
evaluate
proposed
solutions
to
problems
modeled
by
the
game.
The
principal
methodological
problem
in
the
use
of
gaming,
Ackoff
correctly
observes,
lies
in
the
process
of
inferring
from
the
experimental
to
the
’real’
situ-
ation.
This
validity
problem
is
currently
receiving
increasing
attention
in
gaming
To continue reading
Request your trial