-666-
AN ANALYSIS OF CONGRESSIONAL FROCESS
BASED ON THE WORK OF KARL W. DEUTSCH:
€& SYSTEM DYNAMICS MODEL
Sonja C. Fowell
Washington, D. C.
Paper presented at
The 1985 International Conference of the System Dynamics Society
Keystone, Colorado
duly 2-5, 1985
-667-
ABSTRACT
In The Nerves of Government (1963), Karl W. Deutsch postulated
the crucial problems of "steering" and of the "creative intelligence
function” -— the ability to invent and carry out fundamentally
new policies to meet new conditions, the ability to combine
items of information into new patterns so as to find and recognize
relevant new solutions — that increasingly confront government
institutions and that constitute an essential aspect of the
decision- and policy-making processes on which the political
system may depend.
This paper is an initial effort to conceive a System Dynamics
{S/D) model of the U.S. Congressional system from this perspective,
to promote further investigation of Deutsch’s work in this area,
and thus to effect appropriate change in Congressional institutional
structure and function in this respect.
INTRODUCTION
This paper is concerned with design of a model of the dynamics
of legislative decision-making.
Based fundamentaily on the work of Karl W. Deutsch, the proposed
model is intended, in fact, to portray the dynamic relationship
among the components of the Congressional system precisely
as analyzed by Deutsch in his The Nerves of Government (principally
Chapters 10 and 11).
BACKGROUND
According to a report prepared at the request of the Subcommittee
on Fisheries and Wildlife Conservation and the Environment of
the House Committee on Merchant Marine and Fisheries (July 71,
1975) entitled “Computer Simulation Methods to Aid National
Growth Policy":
It is not clear that the existing decision-making
capabilities of the Congress and the Executive
Branch permit full comprehension of the
multi-dimensional nature af societal problems
or proposed solutions...There is an increasingly
recognized need within the Congress for techniques
and resources which will better enable Members
to understand the complexity of the problems
they confront and the relationship among
issues.....They have found it increasingly
hard to rest content with partial models
of isolated traits or situations.
The acknowledged need is for "the development of multi-dimensional
analytical and predictive capabilities within the
-668-
Congress..... Clearly the problems of information gathering and
analysis confronted by the Congress are more challenging taday
than in the past. Certainly the Members have needed an increased
capability to obtain and evaluate information, due both to the
growing complexity of societal problems and the increasing masses
of existing relevant data. They also have needed an improved
ability to identify trends in relationships and to esimate the
short- and long-range effects of such trends. These developments
could be expected to enhance greatly Congress’s capacity to
assess the adequacy of past or proposed legisiative and executive
responses to societal problems.” (U. S. Library of Congress,
i?7S, pages 2-3)
Senator Moynihan expressed the view that:
Tracing the complex and involute interconnections
by which inputs produce outputs in a large
sacial system is not the work of amateurs..... iNdo
one really knows how to do it..... I¥ the
analysis and discussion of public issues
is to continue to move in this new direction,
it becomes necessary to lay down certain
principles which ought to guide all of those
involved if the end result is to be a more
creative democracy, and not simply a more
effective gavernment. (U. S. Library of
Congress, i975, pages 7-8)
The foregoing remarks and others contained in the Subcommittee
Report cited were specifically focussed on the matter of application
of computer simulation models as a “tool for conducting strategic
policy analysis and a potential tool for aiding in the construction
af the decision-making and implementing mechanisms needed toa
define and carry out..-policy on a continuing basis." (U. S. Library
of Congress, 1975, page 1) Deutsch, while also attuned to this
matter, was devoted nonetheless to the broader scope of the
political function:
‘atta @ (We are near the end of a long period
of human history and in transition to a very
different one. Accelerating processes of
change, we feel, have been carrying us ever
closer to the edge of the area in which our
traditional intellectual equipment has been
adequate. Such an intellectual climate of
partial disengagement from tradition and
uncertain but persistent expectations of
change is not new in history, but the breadth
and strength of this sentiment is unmistakable
among such very different minds as Fope John
XXIII, Teilhard de Chardin, A. J. Toynbee,
Ernst Bloch, Ernst Junger, Karl Jaspers,
Norbert Wiener, Leo Szilard, John Platt,
Richard Meier, Bertrand de Jouvenel, and
-669-
many others..... Coming up toward this edge
ef cur past understanding, we must either
accept intellectual impotence and probable
defeat, ar, we must trust in irrationality,
blind luck, and ‘muddling through,’ or else
we must increase substantially our powers
of thought and perception. We must increase
the capacities of our intellectual equipment
~..by increasing and improving our personal
efforts together with our man-made mind extensions
and mind-like artifacts, and with our patterns
af effective teamwork among men, and among
men and machines..... (Deutsch, 19466, page
xvid
WHY AN S/D MODEL?
David Easton (1964, 1965) conceived the notion of political
life as “a system of behavior,...a camplex set of processes
through which certain kinds of inputs are converted into the
type of outputs we may call authoritative policies, decisions
and implementing actions,...a boundary-maintaining set of
interactions imbedded in and surrounded by other social systems
to the influence of which it is constantly exposed.”
Models are attempts to imitate systems.
They try to capture the major components
and interactions of a system. By referring
to models one can obtain valuable insights
into the behavior of a system. (U. S. Library
of Congress, 1975, page 24)
A model is an abstraction of reality and
Can be conceptually regarded as a substitute
for the real system. It is used to capture
the functional essence...of a system.....
To the extent that a particular model is
an appropriate representation of the system,
it can be a valuable aid to policy analysis
and policymaking.....
@& model is anything that illuminates and
clarifies the interrelations of component
parts, of action and reaction and of cause
and effect. (U. S. Library of Congress, 1975,
page 293; Harton, 1972)
System dynamics modeling is a methodology that deals with
deterministic, dynamic, nonlinear, closed boundary systems.
The systems approach is a way of thinking which strives to be
rational, logical, consistent, objective and quantitative in
analyzing complex systems and solving complex problems.
It is on the basis of all these considerations that a system
dynamics model is deemed a most appropriate means for study
of the behavior of the Congressional decision process. As noted
-670-
by Forrester, computer-modeled and —generated representation
of this process makes possible the explicit formulation of
assumptions about the behavior of the system and determination
of the dynamic consequences of the interactions of these
assumptions. As compared with the mental (that is, intuitive)
model, the computerized (that is, mathematical) model used for
describing behavior af the system is clearly advantageous:
it is unambiguous, clearer, simpler, more precise.
A structure (or theory) is essential if
we are to effectively interrelate and interpret
our observations in any field of knowledge.
wees. (Mow the concepts of "feedback" systems
seem to be emerging as the Llong-sought basis
for structuring our observations of social
systems. Over the last century the theory
cof systems has slowly been developed to apply
to mechanical and electrical systems..... (It
is only in the last decade that the principles
of dynamic interactions in systems have been
developed far enough to become practical
and useful in dealing with systems of people.
Around the system dynamics principles ... it
should be possible to structure our confusing
observations about political and business
systems. (Forrester, 1968)
@ great asset of the (S/D) method is that
it forces comprehensive consideration of
the system rather than singling out a particular
facet and trying to understand it alone.
Forrester has made us aware that interrelations
in complex systems often tend to hide ultimate
causes far from the point where results are
seen and has shown that simulation technique
gives us a feasible approach to understanding
such systems. (Henize, 1975)
Major criticisms have abounded concerning the sparing use of
measured supportive empirical data in application af the approach
to a wide range of disparate areas. Deutsch was particularly
keen about the need to check structural data against actual,
observed data in his work on governmental process. He consistently
elucidateds and substantiateds his points by numerous concrete
examples that indicate clear paths to quantification of relevant
data.
It is primarily the concept of feedback, along with other related
engineering principles so clearly delineated in Deutsch’s analysis
of government process, that has drawn this author to attempt
design af the S/D model. Deutsch cited the important works
of many other investigators who have applied these concepts
te the field of political science--William J. Foltz, Ithiel
de Sala Pool, Richard L. Meier, David Easton, John W. Burton,
-671-
Dieter Senghaas, Richard C. Snyder, Ernst B. Haas, Amitai Etzioni,
Harold D. Lasswell, Anatol Rapoport, Hayward B. Alker, dr...
to name but a few. In setting forth his adaptation of the concepts
of feedback and cybernetics to the process of government, Deutsch
draws attention to the etymology of the word “government”.
Let us recall that our word "government”
comes from a Greek root that refers to the
art of the steersman. The same underlying
concept is reflected in the double meaning
of the modern word "governor" as a person
charged with the administrative control of
a political unit, and as a mechanical device
controlling the performance of a steam engine
or an automobile. On closer investigation
we found that there is indeed a certain underlying
similarity between the governing or self-governing
of ships or machines and the governing of
human organizations. Steering a ship implies
guiding the future behavior of the ship on
the basis of information concerning the past
performance and present position of this
ship itself in relation to some external
course, goal, or target. In such cases,
the next step in the behavior of the system
must be guided in part by information concerning
its own performance in the past. (Deutsch,
1966, page 182)
Deutsch expanded on the obvious connection here with engineering
concepts in his description of the coincidence of development
of communication theory, control theory and cybernetics, Deutsch
aS applied to the social, hence, political sciences:
The first is the most obvious and well-known.
It is the vast rise of electronic communications
technology, large-scale computers, and automatic
control systems.
The second is less obvious but readily
discerned. It is the confrontation of the
social sciences with many problems for which
the traditional mechanistic, organismic,
historical, or literary forms of thinking
had proved inadequate. In order ta reformulate
these problems in ways that made them. more |
amenable ta empirical analysis and research
methods, and to mathematical treatment, it
was necessary to develop concepts and models
embodying notions of feedback, information,
memory, self-steering, automatic pattern
recognition, and the like. Such notions
proved increasingly helpful, regardless of
whether they were called "cybernetic" or
went under some other label. . . . . (Deutsch,
-672-
1966, pages xv-xvi)
The third source of the increase in interest may be social and
historical -- deriving from a sense of the current state of
transition of human history and of disengagement of human society
from tradition, accompanied by the stark realization of the
need for substantial increase in our intellectual capacities
and for new tools for thinking.
Forrester has expounded fairly identical linkages and confluence
in describing the background threads out of which system dynamics
has arisen:
(1) the classical, liberal arts approach
to social behavior
(2) the formal theories and principles of
cybernetics (or, servomechanisms, or feedback
systems)
(3) the development of the high-speed electronic
computer
(Oltmans, 1974)
THE MODEL
As a first step in designing the dynamic system model,
Figure 1 presents a simple causal loop diagram -- basic structural
Glement -- describing the system behavior in terms of its boundary
and feedback loops.
PMC > ~y
p—> 16 ——>MG DIFF —
D, I, c
Exec
COLL,
EB
)
cosR
p—----J
1 /
acc —~-—-4 yok
accsoc----4
j +
‘
PRIOR — —-— \ —cFSR
pac -e| eee
+ MEM
i! wm
VIMVEM |
vnver—d
-673-
Table 1 lists and defines parameters of the model conceptualized
in Figure 1, as derived fram Deutsch’s analysis.
Acc
ACCSOC
Cc
CFSR
COLL
cosR
D
DIFF
pon
*#DRIFT
EB
EXEC
VIMVEM
VMVCI
VPVF
Acceptance and Support of Solutions to New Problems
Acceptance of Solutions by Individuals & Groups
Challenge
Capacity for Structural Rearrangement
Apparatus for Collecting Information from Outside World
Commitment of Structures & Resources
Data from Past
Diffusion af Fundamental Items of Knowledge & Information
Death of Memory
Habits, Preferences, Beliefs, Personality Structures
Changes in External Behavior
Execution of Solutions
Feedback
Facilities
Information
Intermediate Goals
Information Turned Into New Form
Innovating Capacity
Inventing Capacity
Intake of Information from Outside World
(Creative) Learning
Learning Capacity
Manpower
Memory
Memory Facilities Available for Recall & Application
to Action
Major or Strategic Goal Preference or Value
Set of New Kinds of Behavior for Learning
Power Fursuit
Present Decisions
Probability of Meeting Challenge
Operating Freferences or Priorities
Propensity to Innovate
Fropensity tao Invent
Range of Intake of Information from Outside World
Range of Inner Recombinations
Responses Needed to Solve Problem to Meet Challenge
Range of Internally Available Recombinations of Knowledge
Steering Capacity or Coordination
Steering & Decision-Making
System Ferformance in Present
Self—Steering
Uncommitted Resources/Committed Resources
Uncommitted Resources/Responses Needed
Valuation of Current Ranges of Intake/Valuation of New
Data
Valuation of Internal Messages/Valuation of External
Messages
Valuation of Memories/Valuation of Current Ranges
Valuation of Fresent/Valuation of Future
~674-
Overall polarity of the loop is indicated as negative. This
is a goal-seeking system. The goal or purpose of the political
decision system is “to invent and carry out fundamentally new
policies to meet new conditions;..... to invent and execute an
effective new ‘response’ to some new ‘challenge’ presented ta
the state or the society by its environment". (Deutsch, 1966,
Page 143)
Deutsch explains the negative-feedback, goal-seeking process
thus:
(The) system first of all is given a major
internal imbalance or disequilibrium that
functions as its drive, in the sense that
the system tends to mave taward a state in
which this internal disequilibrium will be
reduced, or more loosely expressed, in which
its internal "tension" will be lowered.
Moreover, this inner disequilibrium must
be of a particular kind, such that it can
be reduced by bringing the whole system into
some particular situation or relation vis-a-vis
the outside world. This situation of the
system to the outside world we may call a
goal _ situation, or briefly a goal: once the
system has reached such a goal its inner
disequilibrium wili be lower.
Second, in order for the system to approach
the goal effectively, the feedback condition
must be given. The system must receive
information concerning the position of the
goal and concerning its own distance from
it; and it must receive information concerning
the changes in its distance from the goal
brought about by its own performance. The
messages are aften negative in that they
oppose the previosus actions of the system,
so as to oppose overshooting of the target.
In the third place, the system must be
abie to respond to this information by further
changes in its own position or behavior.
With these facilities, and given sufficient
freedom, the system will therefore tend to
approach its goal.
Finaliy, if these changes are effective
and the system reaches the goal, same of
its drive or inner tension usually will be
lowered.....-
Governments may seek goals in domestic
oar foreign policies. In arder to approach
these goals they must guide their behavior
by a stream of information concerning their
own position in relation to these goals;
their remaining distance from them; and the
-675-
actual, as distinct from the intended, results
of their own most recent steps or attempts
to approach them. (Deutsch, 1966, pages
183-185)
(T)he goal may be a changing goal. It may
change both its position,...and even its
speed and direction..... (Deutsch, 1966,
Page 187)
It is also quite possible for the goal itself to be changed.
This may occur gradually, through a drift
in the characteristics or behavior of some
parts of the system. Gradual changes in
the culture patterns or personality structures
of a population, or in the personnel of a
political elite, may thus change the goals
sought by a political decision system. Studies
of the political effects of changes in the
“national character" or in prevailing personality
patterns, such as the change toward
“other—directedness" suggested by David Riesman,
might be developed in this direction.
In some organizations, goal-changing is
a part of the pattern of feedback processes
itself......
Isolated instances of goal-changing are
well known in politics..... How, when, and
how quickly goals are changed by individuais,
groups, and organizations might be a fruitful
subject for political research.
A more specifically political problem arises
in situations where a major strategic goal
must be achieved through a sequence of
intermediate or tactical goals. (Deutsch,
1946, page 196)
Aan organization, having pursued one kind
of goal, might come ta pursue a very different
kind of goal..... This may involve more than
the change of just one or several values.
Rather, where such changes in major goals
occur, we may find at work a process of lang-range
reconstruction; and where they occur in a
relatively short time we may face the phenomena
of renovation, reformation, revolution, or
conversion. All these involve a major change
in over-all function and behavior, as well
as major structural rearrangements of the
political decision system, and usually of
the rest of the society. If we ask, “Haw
likely are such major changes to occur in
a particular political or social system?
-676-
And how likely are they to occur without
self-destruction on the part af the system?" -—
then we have gone beyond the problem of simple
goal-changing feedback. We are asking about
the learning capacity or innovating capacity
of that society. (Deutsch, 1966, page 199)
As a final note on this subject of goal-changing, Senator Moynihan
has stated in the Report of the National Goals Research Staff:
For well over a century observers of American
saciety have been turning out elaborations
of de Tocqueville's original perception that
as conditions for a group improve, the gap
that remains grows steadily less tolerable,
with the rough result that the better things
are the worse they are said to be. More
recently sacial scientists have formulated
this in terms of "goal gradients, with the
hope that the phenomenon cannot only be described,
but can be measured. But it remains part
of the reality; part of the price a society
pays when it consciously seeks to change
things for the better. (Moynihan, 1970, page
472)
The reference mode —- which identifies the essential problem
of interest being modeled over time -- is the measure of the
creativity of political decisions, of “learning capacity,” which
Deutsch defines as the praportion of uncommitted to committed
inner resources within the system compared to the set of responses
needed to solve a particular problem, or to meet a particular
situation or chalienge. It is an idea which he adapted from
the concept of the learning curve.
The dynamic hypothesis being examined states that the ability
of the Congressional system to invent and carry out policies
appropriate to canditions of the environment of the system is
related ta its ability to combine information into new patterns
for new solutions as may be required and appropriate. In addition
to being invented and recagnized, new solutions and sources
must be acted on to be effective. This can be done only to
the extent that uncommitted resources are and can be available
within the system; and then what counts is the ease or probability
with which they are available for unexpected (re)cammitment.
Learning consists in this case in internal structural changes
followed by changes in external behavior, causing the system
to give a different and possibly more effective response to
repeated external stimulus. Consequently, learning capacity
of the system is related to the amount and kinds of its uncommitted
resources.
-677-
In contrast to the simpler causal loop diagram presented in
Figure 1, a more detailed diagram of the dynamic multiple-loop,
nonlinear feedback system would elaborate the interrelations
of the major levels of the system -- Range of Possible Intake
of Information from the Outside World, Range of Inner Recombinations
or Range af Internally Available Recombinations of Knowledge,
Manpower & Facilities, Responses Needed, Commitment of Structures
& Resources and principal decisions/actions cancerning Creative
Learning, Changes in External Behavior, Internal Structural
Changes, as well as key intermediate concepts and factors —-
Load, Lag, Gain, Lead, Information, Goals/Purposes.
By way of explaining the mutual effects of the system components
and the ensuing fluctuations in the state of the system, with
respect to the resourcefulness or "creativity" of political
decisions, Deutsch notes Toynbee’s analysis of the failure of
rulers ta invent and execute an effective "response" to some
new "challenge" presented to the state or the society by its
environment.
This ability to produce novelty, and to recagnize
relevant new solutions once they have been
found (this creative intelligence function),
seems related to the combinatorial richness
of the system by which information is stored,
processed, and evaluated. (Deutsch, 1966,
Page 144)
The combinatorial effects of multiple level variables introduce
time delays (lag) -- critical factors in the Congressional decision
process —- which result in oscillation in the mode of behavior
of the system. Deutsch describes the mutual relationships among
the intermediate concepts of load, lag, gain and lead - among
others — and the contribution that the introduction of such
concepts makes to increasing understanding about the performance
of governments:
Feedback analysis permits us to identify
and in principle to measure a number of elements
in either goal-seeking or homeostatic processes.
We can evaluate the efficiency of a feedback
process in terms of the number and the size
of its mistakes, that is, the under—- or
over-correction it makes in reaching the
goal. (Whether or not the goal will be approached
successfully...depends on the mutual relationship
between four quantitative factors.
1. The load in terms of information, that
is, the extent and speed of changes in the
Position of the target relative to the
goal-seeking system.....
2. The lag in the response of the system,
that is, the amount of time between the reception
-678-
of information concerning the position of
the target and the execution of the corresponding
step in the goal-seeking behavior of the
system.....Clearly, this lag may be influenced
by a number of factors, such as slowness
in the reception of target information, ar
in its interpretation or transmission; or
by delays in the response of parts of the
system in executing the new course; by the
inertia af the system; and so on.
3. The gain in each corrective step taken
by the system, that is, the amount of actual
change in behavior that results.....
4. The lead, that is, the distance of the
accurately predicted position of the moving
target from the actual position from which
the most recent signals were received......The
amount of lead, in turn, depends on the efficiency
of predictive processes available to the
gGal-seeking system, and aon the amount af
inaccuracy that can be tolerated.....
The chances of success in goal~seeking
are thus always inversely related to the
amounts of load and lag. Up to a point,
they may be positively related to the amount
of gain, although, at high rates of gain,
this relationship may be reversed; and they
are always positively related ta the amount
of lead. (Deutsch, 19646, pages 187-190)
The influence exerted on the system by other equally important
structural and dynamic elements has also been explored by Deutsch.
CONCLUSION
Focussing on the four key concepts of load, lag, gain and lead
in Deutsch’s analysis, this paper concludes:
& feedback model of this kind permits us
to ask a number of significant questions
about the performance of governments that
are apt to receive less attention in terms
of traditional analysis:
i. What are the amount and rate of change
in the international or domestic situation
with which the government must cope? In
other words, what is the load upon the political
decision system of the state?.....
2. What is the lag in the response of a
government...ta a new emergency or challenge?
How much time do policymakers require to
become aware of a new situation, and haw
much additional time do they need to arrive
at a decision? How much delay is imposed
-679-
by broader consuitation or participation?
How much time is required to transmit a series
of changing orders to the officials...and
citizens who are to execute them, and how
much time do these persons require to readjust
their previous behavior patterns, habits,
and values, so as to be able to comply
effectively?.....What is the lag in the response
to new information that is brought into the
political decision system through one channel
rather than another, for example, the lag
in the reaction to information that is reported
more or less "straight from the top"...,
in contrast to the information that is first
accepted among some particular social or
eccupational groups?.....
3. What is the gain of the response, that
is, the speed and size of the reaction of
a political system to new data it has accepted?
How quickly do bureaucracies, interest groups,
Political organizations, and citizens respond
with major recommitments of their resources?
4. What is the amount of lead, that is,
of the capability of a government to predict
and to anticipate new problems effectively?
To what extent do governments attempt to
improve their rate of lead by setting up
specific intelligence organizations, strategy
and planning boards, and other devices? (Deutsch,
1966, pages 187-1970)
In a 1983 Conference paper, George FP. Richardson noted John
Platt‘s arguments in The Steps to Man (1966) concerning the
implicitness of the feedback loop concept in political thinking
and devices of the eighteenth century: "that the ‘checks and
balances’ assiduously built into the U.S. Constitution were
a conscious effort to design a system of ‘stabilization
feedbacks,’ and that...there are clear indications (in The Federalist
Papers) of designing (the new Constitution) for stability, for
speed of response without instability, for different time constants
of response for different purposes, and for achieving a desirably
self-regulating structure out of the natural self-interests
of the participants."
An effort has been made here to illustrate the vital S/D role
in appraising and sustaining this dynamic governmental system
and process, as originally conceived.
HERKKIKE KETENE EEE
-680-
REFERENCES
i. U.S. Library of Congress. Futures Research Group, Congressional
Research Service, Computer Simulation Methods to Aid National
Growth Policy. Washington, D.C.: July 30, 1975.
2. Alphonse Chapanis, "Men, machines and models", American
Psychologist, v. 46, March 1961, p. 115, quoted in U.S. Library
of Congress, p. 29.
3. Karl W. Deutsch, The Nerves of Government, New York: The
Macmillan Co., 1966.
4. David Easton, A_ Framework for Political Analysis, Englewood
Cliffs: Prentice-Hall, 1965, p. 25.
S. . A Systems Analysis of Political Life, New Yark:
dohn Wiley & Sons, 1964, pp. 17-18.
6& Jay W. Forrester, Principles of Systems, Cambridge, Mass.:
Wright-Allen Fress, Inc., 1968, quoted in U.S. Library of Congress,
pp. 1.2-1.4.
7. Geoffrey Gordon, System Simulation, Englewood Cliffs, N.J.:
Prentice-Hall, Inc., 1969, quoted in U.S. Library of Congress,
p. 30.
8. John W. Henize, Statement by Wesley H. Long, in "A Framework
for the Evaluation of Large-Scale Social System Models," presented
at workshop on modeling large scale systems at regional and
national levels, Washington, D.C.: Brookings Institution, February
10-12, 1975, quoted in U.S. Library of Congress, p. 29.
9. Forest W. Horton, "Models," in Eorest W. Horton Reference
Guide to Advanced Management Methods, New York: American Management
Association, Inc., 1972, reprinted in U. S. Library of Congress,
p. 29.
10. Daniel P. Moynihan, “Counsellor‘’s Statement: Toward Balanced
Growth: Quantity with Quality," Report of the National Goals
Research Staff, 1970, quoted in U.S. Library of Congress, pp. 442,
472.
ii. Willem L. Oltmans, "Jay W. Forrester," in his Gn Growth,
New York: G. P. Putnam's Sons, 1974, quoted in U.S. Library
ef Congress, pp. 247-349,
12. George FP. Richardson, "The Feedback Concept in American
Social Science, with Implications for System Dynamics," paper
presented at 1983 International System Dynamics Conference,
Chestnut Hill, Mass.: Fine Manor College, July 27-30, 1983,
pp. 1-39,
SEES EERE EEE EHR A EH BH BA
CC BY-NC-SA 4.0