Harold A Linstone Multiple Perspective Theory

TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE 20, 275-325 (1981)
The Multiple Perspective Concept

With Applications to Technology Assessment
and Other Decision Areas
HAROLD A. LINSTONE ET AL. 1
ABSTRACT
In the area of sociotechnological systems the science/technology based paradigms of analysis haver serious
limitations. Building on the work of Graham Allison, we explore the use of multiple perspectives to develop
greater insights; specifically, an organizational/societal perspective and a personal/individual perspective aug-
ment the conventional “technical” perspective. Application of the three perspectives to several technology
assessments and other complex decision areas indicates that the concept can effectively narrow the gap between
model and reality for decision making.
Section I describes the background leading to the study and Section II focuses on the limitations of the
traditional perspective. Section III develops the multiple perspectives and Section IV summarizes illustrations
and applications to date. The concluding section (V) offers guidelines to assist assessors, forecasters, policy
analysts, and other users.
Section I. Introduction
The origin of this research can be traced to the principal investigator’s “Confessions
of a Forecaster” [31:3-121. We recall that the fifties had seen impressive successes in
operations research and systems analysis: linear programming and inventory optimization
as management aids, nuclear war strategy (deterrence, second strike) in military planning.
Researchers developed game theory and decision analysis; by the sixties cost-benefit
calculations and the Planning-Programming-Budgeting System (PPBS) became de rigueur
among Washington cognoscenti. The computer was the ideal tool and its rapid spread had a
‘DR. LINSTONE was the Principal Investigator. He is Director of the Futures Research Institute and
University Professor, Portland State University. The Core Team of the project included:
DR. ARNOLD J. MELTSNER, Deputy Principal Investigator, Professor at the Graduate School of Public
Policy, University of California, Berkeley;
DR. MARVIN ADELSON, Professor at the Graduate School of Architecture and Urban Planning, Univer-
sity of California, Los Angeles;
DR. ARNOLD MYSIOR of Mysior-Runion and Associates, Washington, D.C.;
DR. LINDA UMBDENSTOCK, Planner, Los Angeles County Hospital;
DR. BRUCE CLARY, Assistant Professor of Public Administration, Portland State University;
DR. DONNA WAGNER, Assistant Professor, University of Bridgeport;
DR. JACK SHUMAN, National Oceanic and Atmospheric Administration, Rockville, MD.
The full report is indicated in this paper as Ref. 34 and is based in part upon research supported by the National
Science Foundation under Grant No. PRA 7910020. Any opinions, findings, and conclusions or recom-
mendations expressed in this publication are those of the authors and do not necessarily reflect the views of the
National Science Foundation.
@ H.A. Linstone, 1981

276 HAROLD A. LINSTONE ET AL
seductive effect: computer modeling became a major occupation. The great expectations
placed on systems analysis reached their pinnacle in 1967 when Max Ways wrote in
Fortune’s survey “The Road to 1977”:
The further advance of this new style [systems analysis] is the most significant prediction that can be made
about the next ten years. By 1977 this new way of dealing with the future will be recognized at home and
abroad as a salient American characteristic. 1571
It was about the same time that the idea of technology assessment (TA) took root.
Congressman Emilio Q. Daddario introduced H.R. 6698 in the Congress on March 7,
1967. The bill was designed to stimulate discussions to formalize the concept of a priori
assessment of the impact of technology, thus strengthening the legislative process in the
area of technology policy.
Ten years later bleak reality had displaced the euphoria about systems analysis in
general and technology assessment in particular. Ida Hoos writes:
In our technological era, the dominant paradigm is so technically oriented that most of our problems are
defined as technical in nature and assigned the same treatment4octoring by systems analysts. The
“experts” are methodological Merlins. Technology assessments are conducted by latter-day intellec-
tual condortieri, the brains-for-hire.
Most of the technology assessments I have reviewed must be taken with a large measure of
skepticism lest they lead us to regrettable, if not disastrous, conclusions. [23]
Linstone’s twenty years of systems analysis activity-in industry (corporate plan-

ning), institute (RAND), and academe (developing a systems science doctoral pro-
gram)-mirrored this shift from confidence to concern. There was a deepening frustration
with the chasm between systems analysis and decision making. And the conviction grew
that better systems analysis would not bridge this chasm. The 1976-1978 project on “The
Use of Structural Modeling for Technology Assessment” made the dilemma very vivid. In
the Conclusion section of the Final Report, we wrote:
There is still only lip service paid to any but “rational” or traditional “systems analysis” approaches and
models in addressing the TA process. [32: I321
Decision making inherently involves organizations and individuals. Their perspec-

tives are very different from those of the rational systems analyst or technology assessor.
The signpost that showed a way out of the dilemma was Allison’s book, Essence cf
Decision: Explaining the Cuban Missile Crisis [ 11, which uses three perspectives to see a
single decision process. The present work is the direct result.
The three perspectives differ in their underlying paradigms and thereby unavoidably
create a kind of “Catch 22” situation. In consonance with modes of inquiry preferred by
scientists and technologists, we could regard technology assessment (TA) as a scientific
endeavor and deal with the formulation of theorems, the design of models, and the
validation of hypotheses. But it is not a science, and our fundamental concept, the use of
multiple perspectives, inexorably moves us beyond the paradigms associated with science
and technology. Experimental design and validation of hypotheses are intraparadigmatic:
they operate only within the framework of a perspective. They cannot prove that a model
gives the most useful or correct representation of reality; they cannot give assurance that
the variables chosen are sufficiently inclusive or appropriate. They tell us nothing about
other perspectives.
THE MULTIPLE PERSPECTIVE CONCEPT 211
It should be emphasized that the concept of multiple perspectives need not be con-
fined to applications in the area of technology assessment. As the literature review [34:
Table 21 shows, the value of similar multiple perspectives has been demonstrated in other
contexts (e.g., understanding the Cuban missile crisis, management of policy modeling in
public sector bureaucracies). The illustrations in this study also range beyond TA.2 The
factors which applications to date have in common are:
l Ill-structured nature of the problem (typically sociotechnological systems);

9 Significant policy andlor decision analysis content.
Thus social impact assessment (SIA), environmental impact statements (EIS), corpo-
rate planning, venture analysis, and the design or management of complex systems which
are not purely technological, offer significant opportunities for the introduction of
multiple perspectives.
Section II. Our Usual Perspective and Its Limitations

Science and technology represent the most successful “religion” of modem times.
From Galileo to the Apollo lunar landing, from Darwin to recombinant DNA, the
paradigms of science and technology have yielded dazzling triumphs.
They include the following characteristics:
a. The definition of problems abstracted from the world around us and the implicit
assumption that problems can be solved;
b. Optimization or the search for a best solution;
C. Reductionism, i.e., study of a system in terms of a very limited number of
elements (or variables) and the interactions among them;
d. Reliance on data and models, and combinations thereof, as modes of inquiry;
e. Quantification of information;
f. Objectivity, the assumption that the scientist is an unbiased observer outside of
the system he or she is studying, i.e., truth is observer-invariant;
g. A tendency to ignore the individual in the search for generality, supported by a
trend of reductionism and quantification (e.g., use of averages) as well as objec-
tivity;
h. Time movement seen as linear, i.e., at a universally accepted pace reckoned by
precise physical measurement.
In a technology assessment, as in other technical studies, systems analysis tools are

considered appropriate and the traditional guidelines for analysis apply. Technical impacts
are carefully described and, where possible, quantified. Frequently, cause-and-effect
modeling is carried out to study the static and dynamic behavior of the variables which
describe the system and its environment. Structural and system dynamics models are
illustrative of such tools [32,331. At times the models even drive the analysis, i.e., the
analyst’s modeling background and experience may be instrumental in determining what
is analyzed and how. Strong reliance is placed on technical experts as well as technical
reports containing empirical data or theoretical models and data. Rationality is assumed to
‘See Section IV
determine decisions, e.g., the alternative with the highest calculated effectiveness is
selected. Figure la selectively summarizes the general approach.
The success of this mode of thought and its paradigms has led very naturally to
increasing pressure to extend its use beyond science and technology, i.e., society and all
its systems. The attitude is typified by the planning, programming, and budgeting drive in
the 1960s and the popularity of econometric models in the 1970s. As a Dean of the School
of Electrical Engineering at MIT put it: “I doubt if there is any such thing as an urban
crisis, but if there were MIT could lick it in the same way we handled the Second World
War.” [.53a:48] Organizations become cybernetic systems, utility theory determines pref-
erences, social benefits are synthesized, decision analysis provides the key to decision
making, policy analysis selects strategies. There is a mathematical theory of war and, of
course, “management science.” Hugh Miser, in his recent state-of-the-art review of
systems analysis and operations research for the centennial issue of Science, recalls
Herbert Simon’s description of systems analysis as a “celebration of human rationality”
and concludes:
The challenge is to enlarge this celebration to include the rational management of all of society’s systems
and their problems. [40: 1461
Figure lb portrays the situation. Without question the technical perspective is ideal for
well-structured problems in physical science and technology. Why, then, is there deep
trouble in relying on it in ill-structured areas such as technology assessment?
Richard M. Cyert, co-author of A Rrhu~~iuml Theor! cfrhr Firm, and President of

Carnegie-Mellon University:
As a professor of organization theory and management I used to wonder about the practical value of
these academic fields. For the last eight years, I’ve had some first-hand experience finding out And
I’ve concluded that the study of management makes a useful, but only a limited, contribution to the
practicing manager.
Orgamzatmn theory hasn’t provided me any framework to judge possible appointees. The theory
hasn’t even been very useful in developing new organizational structures Finally, theory doesn’t shed
much light on how a manager should get information about how his organization operates.
Wall Street Journal, April 7. 1980
To answer this question we examine the eight paradigms listed above in more detail.
THE PROBLEM-SOLUTION VIEW

When we talk about a problem we assume a solution exists. We have been brain-
washed in school: A text book presents a problem only if there is a solution (often in the
back of the book). Such books do not point out that in the living world every new solution
provided by a technology creates new problems. Public health measures cut the death rate
but have led to a global population explosion. Introduction of European agricultural
techniques in Africa produces food in the short term and desertification in the long term. It
would be more correct to state that we shift problems rather than solve them.”
“A complex problem P usually has several solution concepts S; each of these leads to several new problems
P, etc. In other words we have a tree/branch structure with levels of P and S succeeding each other indefinitely
into the future. Some Ps are of the diminishing ripple type, some have an amplifying effect (“the cure is worse
than the illness”).
a.
on well-structured technological
sys terns
b.
on complex, ill-structured systems
(technological-social-individual)
ORGANIZATIONS
Fig. 1. The technical perspective.
THE BEST SOLUTION SEARCH

Cost-benefit analysis and linear programming are typical of the search for the opti-
mal solution. It usually comes as a shock to those nurtured on this perspective that
complex living systems have not organized themselves in accordance with this principle.
As Holling [22] notes, ecological systems sacrifice efficiency for resilience, or avoidance
of failure (the fail-safe strategy familiar to engineers) for survival over failure (safe-fail
strategy). They seek to minimize the cost of failure rather than the likelihood of failure.
They strive to maximize their options rather than to confine them by selection of the best
one. They do not manage themselves by manacling themselves.” Evolution shows that
their safe-fail strategy is eminently suited to a world which is inherently unpredictable at
certain times.
REDUCTIONISM
Von Foerster’s First Law expresses the reductionist process rather well:
The more complex the problem which is being ignored, the greater are the chances for fame and success.
1171
If a system is complex, we simplify it by dividing it into subsystems. If we still cannot

handle the subsystems, we reduce it further. Finally, we arrive at problems we can solve.
Fame and success come with publishing: there is a plethora of papers which deal ele-
gantly with unimportant, even trivial, problems. The use of averages (e.g., statistical
mechanics in physics, per capita GNP in economics) and probabilities (e.g. cross-impact
analysis) is a prime example of effective reductionism. It has permitted treatment of
systems with a very large number of elements, each behaving in a unique way. In
technological forecasting there is talk of a “most probable” scenario. Either it comprises
so few elements (events and trends) that it is meaningless or so many that, even if each has
a 90% chance of occurring, the product has a very low likelihood. For example, for 20
elements each with an independent 90% probability, the scenario has only 12% likelihood.
If all alternative scenarios have a lower probability, does this make the 12% scenario a
most probable one? History is strewn with events which had a powerful impact but were
calculated to have a very low probability (e.g., the Three Mile Island accident). The
clients of technologv assessment are less interested in probabilities than in circumventing
catastrophe and in moderating effects offailure.
Another example of the simplification common in modeling is the representation of a
system as a set of elements with pairwise relationships denoting the interactions. If there
are three elements (A, B, and C), there are six possible relations (e.g., A-B, C-B). If the
pairwise qualification is removed, there are at least 49 interactions (e.g., AB-ABC, B-B,
BC-A). Anyone with a family or corporation knows that there are numerous nonpair
relationships (e.g., father to children). For a system of ten elements, this number rises to
over one million. Such calculation bares the vast simplification underlying conventional
systems analysis.
Modeling often becomes an end rather than a means. The dedicated modeler reminds
one of Pygmalion, the sculptor-king of Greek mythology. He fashioned a beautiful statue
of a girl and fell in love with it. Responding to his plea, the goddess Aphrodite brought her
to life, and he married his model. Today’s modelers, blessed with vast computer capacity,
may also become wedded to their creations: the model becomes reality. Clearly, it seems
4The dictionary tells us that to managemeans “to control or to direct the use of.” As Von Foerster notes,
this implies that management reduces the degrees of freedom of the technological system being managed. In
probing more deeply, it is ominous to find that “manage” is related to the word “manacle, a device for
confining the hands, usually consisting of two metal rings that are fastened about the wrist and joined by a metal
chain.” Much of American industry looks at technology assessment and environmental impact statements in
precisely these terms, i.e., efforts to shackle technology. [I81
to be fun for the modelers, but it also can be a nightmare for the real-world problem
solver. 5
DATA AND MODELS

Another characteristic of the science-technology world view is the use of certain
modes of inquiry 181:
1. Lockean. Empirical; agreement on observations of data; truth is experiential and

does not rest on any theoretical considerations;
2. Leihnizian. Formal model; theoretical explanation; truth is analytic and does not
rest on raw data of an external world;
3. Kantian. Theoretical model and empirical data complement each other and are
inseparable; truth is a synthesis; multiple models provide synergism (e.g., particle
and wave theories in physics).
In this world view, it is difficult to realize that, as we move beyond the pure
science-technology domain, other systems of inquiry may prove more fruitful. Following
are several candidates:
4. Hegelian. Dialectic confrontation between opposing models or plans leading to

resolution; truth is conflictual as typified in a courtroom trial;
5. Merleau-Ponty. Reality is currently shared assumptions about a specific situation;
acceptance of a new reality is negotiated out of our experience; truth is agreement
which permits action;
6. Singerian. Pragmatic metainquiring system which includes application of the
other systems as needed; the designer’s psychology and sociology inseparable
from the physical system representation; ethics swept into design;
If we concentrate on the individual, we should also mention Kant’s noumena:
7. Noumena. Reality beyond the perception of our senses, a world which we can
only intuit, to which we are linked through our unconscious mind; in such a
world, there is no temporal distinction of past, present, and future.
Thus, we see that there is much out there beyond data-based, model-based, and
complementary multimode1 systems of inquiry.
QUANTIFICATION
In ancient Greece, the Pythagoreans attempted to preserve the purity of their mathe-
matical expressions by putting to death the man who discovered incommensurables.
Today, the computer has become the ideal instrument to fuel the drive for quantification.
%ree years ago a Pentagon computerized war game resulted in such a command and control breakdown
that the entire computer-dominated Worldwide Military Command and Control System was reexamined. “The
question is whether the traditional reductionist approach to understanding can make workable such a ‘holistic’
challenge as a worldwide U.S. military command and control system.” [51]
A new version of Gresham’s Law states that “quantitative analyses tend to drive out
qualitative analyses.” Zadeh’s “fuzzy set theory” has even been developed to quantify
qualitative terms, and, in the manner of a shoehorn, squeeze them into the computer input
format.
The developed nations are, culturally speaking, measuring societies. We measure
national military strength by comparing numbers of strategic weapon systems with the
Soviet Union (ignoring the fact that accuracy is far more significant than quantity and that
the ability to destroy the enemy’s industrial capacity four times may not be better than
destroying it twice). We measure individuals by their worth in dollars and societies by
their gross national product in dollars. In the words of Yankelovich,
The first step is to measure whatever can be easily measured. This is okay as far as it goes.
The second step is to disregard that which can’t be measured, or give it an arbitrary quantitative
value. This is artificial and misleading.
The third step is to presume that what can’t be measured easily really isn’t very important. This is
blindness.
The fourth step is to say that what can’t be easily measured really doesn’t exist. This is suicide. [58]
OBJECTIVITY
The traditional assumption of objectivity on the part of scientists and technologists is
revealed more and more frequently as a myth. Churchman writes of the social sciences:
One of the most absurd myths of the social sciences is the “objectivity” that is alleged to occur in the
relation between the scientist-as-observer and the people he observes. He really thinks he can stand apart
and objectively observe how people behave, what their attitudes are. how they think, how they decide
[it is a] silly and empty claim that an observation is objective if it resides in the brain of an unbiased
observer. [7:86]
Mitroff lays to rest objectivity in its traditional meaning in the physical sciences with his
study of Apollo moon scientists [42:248]
Von Foerster, himself a cyberneticist, insists that objectivity cannot occur in the
relation between scientist-as-observer and the people he observes. The claim that the
properties of an observer must not enter into the description of his observations is non-
sense, because, without the observer, there are no descriptions. The observer’s faculty of
describing enters, by necessity, into his descriptions. [ 181
If objectivity cannot be assumed for the scientist in his proverbial ivory tower, it
would seem foolhardy indeed to carry this assumption over to technology management in
a real world setting. The real world is a complex system in which virtually everything
interacts with everything-and this includes the manager. That being the case, the choice
of model and data, of problem definition and boundaries, is always partly subjective.
AVOIDING THE INDIVIDUAL

From Adam Smith to West Churchman, there have been expressions of concern with
the danger of ignoring the individual, losing him in the aggregate view. Smith said 200
years ago:
The man of system seems to imagine that he can arrange the different members of a great society with
as much ease as the hand arranges the different pieces upon a chessboard. He does not consider that the
pieces upon the chessboard have no other principle of motion beside that which the hand impresses upon
them; but that, in the great chessboard of human society, every single piece has a principle of motion of its
own altogether different from that which the legislature might choose to impress upon it.
THE MULTIPLE PERSPECTIVE CONCEPI 283
Churchman, in his inimitable way, makes the following observations:
Economic models have to aggregate a number of things, and one of the things they aggregate is you! In
great globs you are aggregated into statistical classes
Jung says that, until you have gone through the process of individuation you will not be able to
face the social problems. You will not be able to build your models and tell the world what to do
From the perspective of the unique individual, it is not counting up how many people on this side and
how many on that side. All the global systems things go out: there are no tradeoffs in this world, in this
immense world of the inner self. All our concepts that work so well in the global world do not work in
the inner world. We have great trouble describing it very well in scientific language, but it is there,
and is important.
To be able to see the world globally, which you are going to have to be able to do, and to see it as a
world of unique individuals that is really complexity. [9:88-901
In retrospect, we encounter instance after instance where individuals were crucial in the
interaction of a technology with the society: Wernher von Braun’s leadership in rocket and
space vehicle development, Rachel Carson’s book, The Silent Spring. In these cases, we
have an impact of individuals on technology. Conversely, technology may have a tremen-
dous impact on the individual. Television ended the career of Senator Joseph McCarthy,
radio (e.g., the fireside chats) worked powerfully for President Franklin Roosevelt, and
the space program TV coverage created the astronaut as a folk hero.
THE LINEAR PERCEPTION OF TIME

The science-technology world view is concerned with physical space-time, i.e., with
time as a dimension or variable essential in grasping the dynamics of a complex system.
Distortions intrude through relativity theory, e.g., perceptions of time vary with observer
speed in an Einsteinian universe. Economists apply a discount rate to future dollars to
determine their present value; the basis traditionally is the cost of capital. Aside from such
rather mechanistic alterations, this perspective sees time as moving linearly, at a univer-
sally accepted pace determined by precise physical measurement. Thus Forrester,
Meadows, Mesarovic, and Pestel may exercise their system dynamics models over 50- or
130-year periods, but the computational time increment At is independent of society and
individuals, geographic locale, and era. We shall use the term technological time for this
case.
By contrast every individual has a very personal conception of time. Dominated by
personal time and needs, a person’s time horizon is dictated by the expected life span,
position in Maslow’s hierarchy of needs, and/or other individual considerations.
The individual applies a psychological discount rate to his perception of future
problems and opportunities which is totally distinct from the businessman’s dollar dis-
count rate based on cost of capital. The psychological discount rate means in effect that
the individual looks at the future as if through the wrong end of a telescope [31:5-61.
Distant objects appear smaller than they really are. Similarly, a problem far in the future is
of less concern to us than the same problem in the near future. Human beings apply a
discount rate to their own past and thus distort the integration of their personal experience.
Recent events tend to be overstressed in comparison to more remote ones. Evidence
indicates that the discount rate is significantly influenced by personality type (Table 1).
The variation within the personal perspective of discount rate may be a large one-a factor
that makes the selection of personalities to be interviewed critical on the time dimension.
Let us now turn from the individual to organizations or social entities. Neither
technological nor personal time prevails. Organizations usually have a longer time horizon
TABLE 1
Jung’s Typology and Time Orientations
Jung type Time focus Discounting
Sensation Present Highest

Feeling Past Selective bias
Intuition Future Ii (High/moderate)
Thinking Past-present-future Moderate
Source: [3 1: 101
than individuals; they do not expect to die like human beings. This does not mean they use
a zero discount rate, merely a lower one than individuals. The effect of different discount
rates on the perception of the increase of world population can be seen in Figure 2. Note
that at a 5% rate the crisis seems to disappear.
Social time is multigenerational. Organizations are, in fact, a curious blend of long
and short time horizons. There is the motivation of perpetuation and the pressure of
meeting next month’s payroll and protecting next year’s budget. As do individuals,
organizations have a spectrum of time horizons. Small companies contrast with large
ones, medieval Christian with modern American societies, rich European states with poor
African nations.
Thus the discount rate depends both on the conception of time and, to a lesser degree,
on organizational and individual differences (Figure 3).
These, then, are eight important paradigms which govern our usual perspective of
complex systems. They also suggest why so much systems analysis and technology
assessment is an “inside job,” performed for other analysts who share the same
paradigms and not necessarily for the audience that needs it.
Section III. Multiple Perspectives

In the preceding section we reviewed a series of paradigms integral to the way we
normally look at technology assessment. We were concerned with the “how” rather than
the “what” i.e., how we usually look at the problem rather than what we are looking at.
DISCOUNT RATE 0 DISCOUNT RATE DISCOUNT RATE

1% per year 5% per year
b
3
Nomalfzed
World
Population 2 1
(1970 = 1)
CI la IQD Iz) 47----e a0 IW II
YEARS IN FUTURE
Fig. 2. The discounting phenomenon: world population crisis. (Note: Zero discount rate case based on
Meadows’ “Limits to Growth” standard run [39:124].)
1ong hor i zon

7 ~!~~‘dk:b:~: low d i s.yuyF _
Technological time
Social time
F
affluent person
Personal time
I I poor person
Fig. 3. Relative time horizons-a schematic. (Note: For qualifications see text.)
We continue in this vein here and focus on the question: Does recent literature suggest
how we can look at the problem differently?
In a narrow sense the concept of multiple models is widely recognized. We em-
phasize that we refer to the use of multiple models simultaneously, not sequentially. Past
struggles suggest that acceptance of multiple models requires a considerable degree of
intellectual sophistication or maturity. In mathematics, the nineteenth century challenged
us with the Euclidean, Riemannian, and Lobachevskian geometries, while physics offered
the wave and particle theories as simultaneously valid.
HISTORY
AI&on’s Models [ 11
The main descriptors of the Allison models6 are shown in Table 2. Models I and II
could be built on a solid base of 1950s and 1960s scholarship. The RAND school and its
disciples produced a plethora of rational actor guidelines and case studies. The organiza-
tional process model was drawn largely from the Simon-March school of organizational
decision making [37] and from Cyert-March’s “process-oriented” theory wherein the
internal structure of the firm plays a role as significant as the market dynamics [ 121.
Similar to Allison’s questioning of the assumption of rational choice in foreign policy
making, Cyert and March challenge the axiom that profit is the only motive behind
business decisions. Instead, they argue that goals are reached through bargaining and
compromise between the major subunits within the organization. Maximization of profit is
only one consideration. In the prevailing economic theories of the firm (and Allison
Model I), decision makers are assumed to have complete information about alternatives.
Cyert and March, in contrast, posit a process of problem solving characterized by incom-
plete information and biased toward dominant perspectives within the organization.
6We prefer the word “perspective” over “model” (which Allison uses) to emphasize that we are consider-
ing different sets of paradigms or world views, not merely different mathematical formulations.
TABLE 2
Allison’s Three Models
I. II. III.
Rational actor Organizational process Bureaucratic politics
Basic unit of analysis Action as choice of total system Actions as organizational output in Action as political resultant (bargaining,
framework of present capabilities and compromise)
constraints
Organizing concepts Unitary decision maker (e.g., Constellation of loosely allied units Players (“where you stand depends on where
government) topped by leader5 you sit”)
One set of goals (e.g., national) Problems factored; power fractionated Parochial priorities and perceptions
Problem as seen by unitary Parochial priorities Goal5 include personal interests

decision-maker
Goals are constraints defining acceptable Players‘ impact based on relative power
Solution a 5teady-state choice among performance of organization
Action channels 5tructure the game
alternatives
Sequential attention to goals
Rules sanction some tactics (bargaining,
Action a rational choice based on
Standard operating procedures (SOP) coalitions, bluff) but not other5
goals/objectives, alternatives/options,
consequences, and value maximizing Programs and repertoires
selection
Avoidance of uncertainty
Problem-directed search
Central coordination and control

5
s
5
I4
Dominant inference pattern Actions are maximizing means to Behavior of organizations at time t Action resultant of bargaining game among %
achieve ends similar tot - I, at f + 1 similar to t individuals, groups E
g
General propositions Likelihood of any action results from a Standard routines: program is a cluster of Peculiar preferences and stands of individual
2
combinationof relevant values and SOPS satisfying rather than maximizing players
objectives, perceived alternative courses (first acceptable rather than best 2
Styles of play vary
of action, estimates of various sets of alternative) 8
consequences, and net valuation of each Face of issue differs from seat to seat 3
Long-range planning institutionalized,
set of consequences
then disregarded Fuzziness useful to get agreement
Increase in costs of an alternative reduces
Incremental change Focus on immediate decision rather than on
likelihood of its selection
doctrine
Tradeoffs neglected
Decrease in costs of an alternative
Views:
increases likelihood of its selection Organizational health implies growth,
imperialism Looking down+ptions
Administrative feasibility a major Looking sideways--commitment

dimension
Looking up-show of confidence
Directed change possible when
Frequent misperception
organization is in crisis
Misexpectations
Miscommunications
Source: [ 11
The foundations of Allison’s first two conceptual models (I and II) are stronger than
those of the third. Organizations are easier to analyze than individuals and permit more
ready generalization, hence propositions. (We recall the quotations from Adam Smith and
Churchman in Section II). Allison in his afterword notes that “Model III tells a fascinating
story, but is enormously complex. The information requirements are often over-
whelming” [ 1:274]. The use of a rational actor model and an organizational process
model for a case study of a 1962 presidential crisis is self-evident, considering the strong
countercurrents of the time. Rationality was the hallmark of “the best and the brightest,”
the whiz kids brought in by Secretary of Defense McNamara to the dismay of the up-
parcrtchik dominated by career civil servants, true “organization men. ” In the five years
prior to the formation of the Harvard May Group [ 1:ixl Models I and II were, so to speak,
in public confrontation all over Washington. Model III subsequently signified the recogni-
tion that two models could not encompass all the crucial aspects of the decision-making
process.
Steinhruner and Andersen

Two MIT doctoral dissertations built on the work of Allison: Steinbruner’s The
Cybernetic Theory of Decision [53] uses an analytic, a cybernetic, and a cognitive
paradigm. The latter differs most significantly from that of Allison’s Model III. It rests on
cognitive psychology, specifically three claims:
1. There are regularities in the decision process which have to do with the structure
as opposed to the content of cognitive operations;
2. The full human mental apparatus is engaged in the simplest of operations such as
direct, immediate perception;
3. Most of what happens in the human mind is not accessible to direct, conscious
experience. In other words, a great deal of information processing is conducted
independently of conscious direction.
Inferential memory, consistency, reality, simplicity, and stability provide the basis of
this paradigm.
Andersen’s perspectives are similar: rational, organizational, and cognitive [3]. A
comparison of Allison’s and Andersen’s models shows a difference in stress in the third
perspective: Allison is concerned with bargaining, Andersen with learning and mental
decision processes. Allison focuses on governmental action as a result of compromise,
perceptions, styles of play, personal goals, and rules of the game. Andersen deals with the
individual’s limited information processing capability, inferential memory, focusing
rather than scanning as a basis for choice, and small group interactions leading to a shared
reality. It is thus apparent that Andersen’s cognitive model deals not only with the
individual but with small groups.
The commonality of Allison, Steinbruner, and Anderson is that each places three
perspectives on one object of interest. It is a device also known to writers and historians.
Rashomon is a classic literary example. And historian Fernand Braudel divided his distin-
guished work on The Mediterranean into three levels: the story of man’s contact with the
physical, inanimate environment, the social history of groups, and history on the scale of
individual men. He sees historical time as geographical time, social time, and individual
time, bringing to mind our view of future time in Figure 3 [6a:20-211.
Mitroff and Blankenship

Mitroff and Blankenship developed their ideas independently of Allison. They also
conclude that the mode of inquiry for whole systems differs so radically from that of
traditional methodology that it raises severe challenges to some of our most basic tenets
underlying science. Noting that such systems are conceptualized in radically different,
even divergent, ways by different users, they develop guidelines for the analysis of
holistic systems [41:345]:
1. At least two ‘radically distinct’ disciplines of knowledge must be brought to bear;

2. At least two ‘radically distinct’ kinds of conceptualizers (personality types) must
be brought to bear;
3. At least two ‘radically distinct’ philosophical inquiry models must be brought to
bear. [This is in line with our discussion of inquiring systems in Section II.];
By radically distinct disciplines they do not mean, say, operations research and
engineering, but operations research and law. Different conceptualizers are convergers
and divergers, problem formulators and problem solvers.
OUR PROPOSED PERSPECTIVES

Before describing how we propose to look at a technology assessment (TA) we
consider what we are looking at.
What Are We Viewing?

A process such as a TA must deal not only with the technological aspect but also with
social and human facets. The technology is embedded in an environment of human
beings-it helps them, hurts them, and changes them. They, in turn, can make decisions
to develop, limit, alter, or stop the technology. Human beings are considered here in two
ways-as social entities (e.g., companies, vested interests, states) and as individuals
(e.g., specific decision makers, personal impacts).
We use the schematic in Figure 4 to discuss these sectors in terms of the elements
which they contain:
a) Technology. The starting point in TA frequently is a specific technology, e.g.,

fluoridation of water, offshore oil drilling, nuclear power plants. For this reason this
element is shaded in Figure 4. It represents what the technical expert is looking at, e.g.,
the concerns of a nuclear physicist or engineer in developing a breeder reactor or a
practical fusion energy source.
b) Physical Environmental Setting. The technical aspect of the environment of a

technology encompasses land, sea, air, and space with all its inanimate and living compo-
nents. Environmental setting here refers to such systems which physically act upon, or are
affected by, the technology. Examples are the altered ozone level in the atmosphere
resulting from chemical exhaust products and the shift of population from farm to city in
consequence of the introduction of farm tractors.
c) Sociotechnological Setting. Technologies interact with organizations and groups

in many ways. A Nuclear Regulatory Commission is established to deal with a technol-
ORGANIZATIONAL ASPECTS
------I ITECHNOPERS~NALI / I
PERSONAL ASPECTS
Fig. 4. What we are viewing in a TA.
ogy; computers alter group decision-making processes and office operations; telecom-
munications galvanize the formation of common interest groups. Cost-benefit analysis is
a technical means to evaluate some of these interactions.
d) Technopersonul Setting. Technology affects the individual and is affected by the

individual. Such interactions cannot be understood by looking at individuals only collec-
tively. Telecommunications have an impact on the human psyche; computers constitute a
prosthesis of certain parts of the brain and the impact may vary from one individual to
another. Computer output may overrule common sense because of its aura of scientific
quantification. Television creates a reality of its own. In Paddy Chayefsky’s film Ner-
work, the desperate television commentator rants: “It’s you people that are mad. You are
beginning to think that the tube is reality and you people are unreal. ”
The other side of the coin is the effect of individuals on a technology. Ralph Nader’s
Unsafe ut Any Speed and Wernher von Braun’s fascination with space vehicles affected
technological developments. The relation of Edwin Land to camera technology, and of
Dutch Kindleberger, Robert Gross, and Donald Douglas to aircraft technology, exemplify
the crucial role of individuals in technology impact, policy, and implementation.
e) Organizational Actors. Since organizations are as old as human society and much
older than science, organizational entities have existed long before cybernetic or other
theories of organization. The family, clan, village, state, church, club, union, party, and
company are all examples of organizations which interact with technology. There are
victims, beneficiaries, and regulators that become involved with the impact and control of
technology. There are primary and secondary actors. For example, in the struggle for the
Apollo program, NASA and the U.S. Air Force were primary actors, the Army and Navy
were secondary actors. The Air Force was in competition for budgets with the other
services and this competition affected its interest in the Apollo program.
The most basic societal unit, the family, has been strongly affected by technology.
Mobility has stimulated the spread of the nuclear family, isolation of the elderly, and the
rootlessness of the young.
f) Individual Actors. An individual in a crowd or an organization may be quite

distinct from the same individual by himself. As a member of a group he surrenders
responsibility, a heavy burden. But this relief is usually tempered by a loss of freedom and
of creativity.
An entrepreneur who propels a technology from concept to marketplace, a charisma-
tic politician who finds an issue in a technology, or a terrorist who uses a new technology
must be understood as individual actors. There are visionaries and realists, promoters and
obstructionists, operators and theoreticians; their activities may be as important as those of
organizations. Identification by name may clearly not be possible in the case of an
emerging technology where implementation agencies, regulators, and lead industries have
not yet crystallized, as would be the case for an existing technology. Nevertheless,
personality criteria, warning indicators, and other useful clues, as well as desiderata
pertaining to key actors should be considered in the TA, particularly in connection with
the policy and implementation aspects. The aim is analogous to that of an intelligence
organization developing a profile of a political personality (e.g., Hitler, Khomeini) to
anticipate decision patterns.
Key executives in a corporation impress their world view on the organization. When
a leader is bold and dynamic, the corporation may take considerable technological risks.7
When the leadership is conservative, the corporation shows it: investments are cautious,
flexibility is minimal, innovation is uncomfortable. Of course, when the same characteris-
tic applies to an entire industry (e.g., auto, steel, tire and rubber), we revert back to the
organization actors.
While the six elements mentioned up to this point can be discussed very meaningfully
in terms of what is being analyzed, our concern with the remaining elements really lies
much more in the how than the what. We are dealing with processes rather than objects or
entities.
g) Political Action. This element represents the well-recognized interplay between

organizations and individuals.
h) Decisions. The last element represents the direct outcome of the political activity,
e.g., a decision to adopt and implement a policy. It also signifies the indirect consequence
of the first six elements and is shaded in Figure 4 as the culmination of the cycle
originating with a. This does not mean that the process ends here. A decision may itself
affect the technology a and alter its evolution, leading to changed effects on elements b to
g, and consequently again to the decision h. A decision may alternatively affect b to g
‘Such a leader must, of course, surround himself with appropriate lieutenants. Placed in a senescent
marragement setting, the risk-taking leader may well fail and quit.
directly. In either case there is continuing change. Almost never in a TA can we consider a
sequence, e.g., u to h, as static or not subject to iteration or alteration over time. This
means that the TA should not be a one-time exercise but a continuing activity. Finally, the
values may change and alter organizations e, or new actorsf may appear following an
administration change.
Table 3 shows examples of these eight elements drawn from Lawless’ 45 case studies
of technology and social shock [30]. Not surprisingly, we have many examples in these
case descriptions of elements u, b, c, e, g, and h with few of elements d andf’. The focus
on the individual tends to be hidden or submerged by the far more easily tracked organiza-
tionally related items.
How Are We Viewing?

Building on the work of Allison, Steinbruner, and Andersen, we now adapt the
concept of multiple perspectives to the study of sociotechnological systems, specifically to
technology assessment.
We propose a technicul perspective (T), an organizational perspective (0), and a
personal perspective (P). The wordperspective is used to distinguish how we are looking
from what we are looking at (i.e., an element). For example, the element may be an
organization. Using the technical perspective (T), we may see the organization as a
hierarchical structure, model it using system dynamics, and apply decision analysis or
other tools of management. With the organizational perspective (0), we see a unit, a
living system fighting competitors, cohesive or divided, a collection of baronies (di-
visions, departments) with a weak king (president, CEO) or a strong chief with loyal
lieutenants running the subunits, a powerful staff and weak line (or vice versa). The
personal perspective (P) may see the same organization as job security, an opportunity to
exert power, or a step to gain prestige.
In relating the how and the what, we posit that the use of the T perspective to study
the technical elements, the use of the 0 perspective to study the organization elements,
and the use of the P perspective to study the individual elements are vital but by no means
adequate. Any perspective may illuminate uny element. It is inconceivable that a technical
element can be understood without use of the T perspective. But the 0 and P perspectives
may add important insights. Similarly, appreciation of an organization requires an 0
perspective, but much can be gained by use of the T and P perspectives. One person may
be able to offer all three perspectives (T, 0, and P) on a problem-the rational analyst’s,
his organization’s, and his own. Or one perspective may dominate his thinking and blind
him to others. Perspectives are dynamic; they change over time. Most importantly, the
different perspectives are mutually supportive, not mutually exclusive.
The Technicul Perspective. This perspective has been the dominant one in the technology
assessment literature. The first book on the subject of TA announced that:
Technology assessment is a systems analysis approach to provide a whole conceptual framework, com-
plete both in scope and time, for decisions about the appropriate utilization of technology for social
purposes. [2 1561
The technology and its environment are viewed as a system in the now-familiar way
sketched in Section II. Terms like alternatives, tradeoffs, optimization, data, and models
suggest the rational, analytic nature of the technical perspective. The tools include proba-
bility theory, decision and cost-benefit analysis, system dynamics, and econometrics.
TABLE 3
The Eight Elements-Illustrated by
Lawless’ Case Studies
Our
element Lawless case Illustration
a) Technology DMSO (dimethyl sulfoxide) Interest revived 75 years after initial synthesis as a
suppressed wonder drug? possible means to recover useful chemicals from
wood pulp wastes. Uses were found as industrial
solvent and drug carrier.
b) Physical Setting The Donora, Pennsylvania air In October 1948 twenty deaths and nearly 6,000
pollution episode illnesses occurred due to a combination of a
prolonged temperature inversion and industrial
emmissions.
c) So&technical The nuclear power controversy The Atomic Energy Commission takes initiative in
setting facilitating nuclear power generation by cooperating
with industry in R & D.
d) Technopersonal The fluoridation controversy Dentist Frederick McKay spends his off-hours for
setting three decades on finding the cause of mottled teeth,
notes lack of decay twenty years before serious
studies were done.
e) Organizational Abuse of medical and dental X rays National Council on Radiation Protection and
actors Measurement formed in 1929 collaborated with
radiological societies and X ray industries to set first
standards for exposure.
fl Individual actors The thalidomide tragedy Dr. Frances Kelsey was new to the FDA
bureaucracy and thalidomide was her first case. She
resisted continued pressure from drug manufacturers
due to her own considerable knowledge of medicine
and pharmacology.
g) Political action The Dugway sheep kill incident Congressional hearing obtained admission by Army
that nerve gas tests killed 6,000 sheep. Admission
came after repeated denials by key army officials
and involvement of a dozen different agencies.
h) Decision NTA (nitrilotriacetic Acid) in Surgeon General acts to stop manufacturers from
detergents using NTA in detergents, based on environmental
studies which were disputed by extensive
manufacturer tests.
Source: [30]
There is much interest in organization and categorization of information, in preparing lists

and matrices, and in drawing organizational charts and digraphs. [ 10:7]
The United States as a culture is the most strongly T-oriented one in the world. We
love statistics and polls. We define quality of life and football teams in terms of numerical
indices; even Bo Derek is a “10.”
The bias toward the T perspective is seen in the Central Intelligence Agency:
Technological cleverness is the pride of the U.S. intelligence But American supremacy in technical
intelligence is profoundly misleading. It is not representative of U.S. intelligence capabilities as a whole
but stands in stark contrast. For in every other intelligence field-human spies, analysis of data collected,
and ability to conduct secret operations-the U.S. intelligence community appears to be dangerously
deficient. [54]
It is probably not a coincidence that the means of obtaining input (personal contact)
and the type of input obtained with the 0 and P perspectives are not those accessible to
surveillance satellites and other high technology sensors.
Since the T perspective is such a well-mined lode, we will focus our attention on the
other perspectives.
The Organizational Perspective. To date, the organizational perspective represents a

relative rarity in TA, but it is not unknown, as will be apparent in later discussions. The
technology is perceived from the point of view of affected and affecting organizations.
These may be formal or informal, permanent or ad hoc. If, for example, a banking
commission of senior civil servants and bank representatives is created to analyze the
impact of electronic funds transfer, it would not be unexpected to find each member using
an organizational perspective rather than a technical perspective. Although uniqueness of
the technical perspective is the rule, each organization may have a different 0 perspective
on the same sociotechnological system.
There are formal and informal groupings. The fact that a collective may not be
formally organized as a consensual group does not make its effectiveness less valid than if
the individuals were so organized. Sometimes a postulated “public attitude” can alter (or
sustain) an impact as if it were a powerful advocacy group.
The world seen from the pure 0 perspective is an orderly progression from one state
to another, with an occasional minor crisis along the way, for which experience and the
procedural manual have the answers. A process of orientation, with little concern for the
quality or the effectiveness of the end product per se, is the hallmark of 0. Rules and
procedures are there to be followed; policy is God-given and unchanging (in spite of any
evidence to the contrary). A stereotypical American, operating from the 0 perspective,
approves completely of the checks and balances in our system of government and assumes
that whatever has been done according to the rules is the right and proper thing.
Since all human beings are socialized into some sociocultural organization, this
perspective influences all actions and decisions (“Is this right?“, “Do I have a right
to?“). Perhaps the strongest argument for inclusion of this perspective is the realization
that, in the political arena, highly technical information is usually, and properly, dis-
counted in favor of social interests and considerations of values involved-and these can
never be adequately encompassed by a T perspective. Pressures emanate from institutions,
regulatory agencies, special interest groups, and mass social movements. Illumination of
the interplay of these pressures necessitates the 0 perspective.
In this perspective, there is no intensive search for analytical tools-in fact, often a
mistrust of “academic” techniques. They are viewed as either unrealistic, or unpredicta-
ble and uncontrollable.
Future problems are discounted in contrast to near-term problems, i.e., short-range
consequences for the organization and its actions are given priority. We are dealing in this
perspective with organizational or social time (see Section II).
THE MULTIPLE PERSPECTIVE CONCEFI 295
Each organizational actor is cognizant of parochial priorities and interests; they are
distinctive to his or her organization. The same applies to standard operating procedures,
and these go far toward limiting potential decision alternatives. Seen from the organiza-
tional perspective, a technology appears to create problems or solve problems for the
organization. There is concern that the technology may disturb the functioning of an
organization, i.e., become a disruptive force. Interest in a technology also is a strong
function of the amount of “noise” made about it (e.g., inquiries from “above”, mention
in the media).8 There is a strong tendency to break down problems in accordance with
organizational responsibilities.
Machiavelli, often misrepresented, was in fact vitally concerned with the soundness
of one organization, the state. As a modern executive has observed, the principles evolv-
ing from an organizational perspective in Florence, AD 1530, are readily transferable to
the corporation in New York or London AD 1980:
It means looking at the corporation in a new way: looking not through the eyes of the accountant and
systems analyst and economist and mathematician, but through those of the historian and political scien-
tist. [25:x]
Machiavelli recognizes the difference between an 0 and a P perspective. He sees

morality as necessary to guide individual conduct, but not to guide state conduct. Thus, a
diplomat, as representative of the state, is not bound by the moral code of its people: “No
good man will ever reproach another who endeavors to defend his country, whatever be
his mode of doing so.” [36] The modern executive faces the same dichotomy: He may
recommend corporate actions which he would never condone in his personal life.
The organizational perspective helps us in TA, in at least the following ways:
l Idenrifiing the pressures in support of und in opposition to the technology:

l Gaining insights into societal ability to absorb a technology+rganizational incrementalism is an
important bound;
l Increasing ability to facilitate or retard implementation of technology by understanding how to gain
organizational support;
l Drawing forth impacts not apparent with other perspectives, e.g., based on realities created within an
organization;
l Developing practical policy options (e.g., new coalitions).
The Personal Perspective. This is the most subtle and elusive perspective, the most
difficult to define. Here the world is seen through the eyes of the individual. The P
perspective should sweep in any aspects which relate individuals to the technology and its
impact and which cannot be brought out by the other perspectives. Thus we exclude from
this perspective the purely physical impacts of a technology on the individual as well as
the impact of, say, farmers on an agricultural technology. But intuition, charisma, leader-
ship, and self-interest, which may play vital roles in matters of policy and impact, may
only be evoked by the P perspective.
Freud, in his epochal Interpretation of Dreams, perceived three layers: the profes-
sional, the political, and the persona. He found the first to be the most current and
accessible; the third to be the deepest and least current and accessible [50:184]. The
similarity to our three perspectives is self-evident. As would be anticipated with
sFactors such as these suggest how the 0 perspective affects technological forecasting
Freud’s psychoarcheological conception, we shall find most difficulty with the personal
perspective.
IDENTIFYING INDIVIDUALS. There are clearly many persons who interact, directly
or indirectly, with a sociotechnological system. There are beneficiaries and victims,
builders and users, regulators and lobbyists. Then there are the “hidden movers.” These
are individuals who, from a second or third level position, pull the strings that determine
how things progress, Attention is usually so keenly focused on the behavior of the
puppets, which is overt, that the effect of the puppeteer, who is hidden from view, is
ignored. In the case of a very prominent power position, we tend to look for the power
behind the throne, especially if we suspect that the ostensible power person is not operat-
ing under his own steam. This is difficult enough, but at least in such a case-because of
the interest of historians, political analysts, etc.-the individuals surrounding the power
position are in the public eye and therefore subject to scrutiny. Still, the “gatekeeper,”
the person who controls the information flow in an organization, is often difficult to
identify.
For less publicly prominent positions, the powers behind the throne usually remain
obscure. They may not even appear on the organization chart. One way of possibly
identifying them is to look for the individuals who do the writing in an organization
(policy statements, position papers, standard operating procedures or regulations, etc.).
Their writing frequently sets the tone for the organization, or else it becomes a point of
departure for discussion within the organization. If such individuals hold their position for
any length of time, they may wield considerable influence.
Personal probing is essential in identifying key individuals. In cases of an emerging
technology, key individuals may not yet be identifiable. In such cases, types and their
characteristics must be sketched. And here, too, the P perspective is of inestimable value.
What makes a future Sammy run can often be predicted by analogy with past movers. In
fact, an interview might elicit interesting insights on how to “beat the system” or end-run
recalcitrant bureaucracies.
INTUITION, LEADERSHIP, AND SELF-INTEREST. Intuition is a well-appreciated
trait in the world of business, as the following quotes suggest. R. P. Jensen, Chairman of
General Cable Corporation:
On each decision, the mathematical analysis only got me to the point where my intuition had to take over.
L481
J. Fetzer, Chairman of Fetzer Broadcasting Company:

Walk through an office and intuition tells you if things are going well. [481
R. Siu, Management Consultant:

Effective CEOs are aware that rationality and the scientific method provide critical inputs to only one
of three crucial questions overarching key decisions. These are: a) Does it add up? b) Does it sound okay?
c) Does it feel right? Logic and science contribute primarily to the first question, less to the second, and
even less to the third. [52:851
The typical T-trained mind usually balks at the mention of intuition. However, a few
T-oriented, highly respected scientists have not been afraid to pay homage to intuition and
accept it as an important concept. Consider Jacques Hadamard’s classic, The Psychology
of Invention in the Mathematical Field. He writes:
That those sudden enlightenments which can be called inspirations cannot be produced by chance alone is
already evident there can be no doubt of the necessary intervention of some previous mental process
unknown to the inventor. in other terms, of an unconscious one. 1201
THE MULTIPLEPERSPECTIVECONCEPT 291
He quotes Helmholtz, who observed that “happy ideas” never came to him when his
mind was fatigued or when he was seated at his work table. “After the fatigue . . . has
passed away, there must come an hour of complete physical freshness before the good
ideas arrive.”
Poincare distinguishes fully conscious work, illumination (happy ideas) preceded by
incubation, and the quite peculiar process of the first sleepless night. The unconscious
appears to consist of several levels. Hadamard writes:
It is quite natural to speak of a more intuitive mind if the zone where ideas are combined is deeper, and of a
logical one if that zone is rather superficial. This manner of facing the distinction is the one I should
believe to be the most important, [20]
In the case of exceptionally intuitive minds, even important links of deduction may
remain unknown to the thinker himself who has found them. The mathematician Cardan’s
invention of imaginary numbers is a beautiful example of the use of the nonrational to leap
from one rational to another rational domain.
More recently (1980), Nobel laureate Herbert Simon and associates have explored
the differences between experts and novices in solving physics problems. They find that
the expert is mentally guided by large numbers of patterns serving as an index to relevant
parts of the knowledge store. The patterns are
rich schemata that can guide a problem’s interpretation and solution and add crucial pieces of information.
This capacity to use pattern-indexed schemata is probably a large part of what we call physical intuition.
[291
Each person has a unique set of patterns which inform his intuition. In calling on the
P perspective, we are thus augmenting the conscious, logical T process by opening
ourselves to the deeper mental levels which store patterns of great potential value.
Leadership is a quality of recognized importance in science and technology as well as
in politics. Courant in applied mathematics, Rabi in atomic structure, von Braun in
rocketry, and Ford in automobile production were leaders; disciples, students, and im-
itators built on their innovative ideas and magnified their impact.
Self-interest motivates most of us, although it is usually hidden. It may take the form
of prestige, profit, power or pleasure. For a P perspective on the engineering profession,
Florman’s delightful book, The Existential Pleasures of Engineering [16], is recom-
mended. The felt need to mask this energizing factor generates deceptions and illusions
which may be difficult to penetrate. Successful technological innovation and policy
change implementation require leadership; leadership is driven by self-interest.
Effective organizations are those that have found successful ways of making the
self-interest of the members work constructively and in unison to support the goals of the
organization.
THE ROLES OF THE P PERSPECTIVE. Specifically, we see the P perspective in four
roles:
l Understanding of the total decision process is enhanced through knowledge of

participating actors.
Political activity involves the interactions of organizations and individuals. Usually,

we cannot grasp the political process without knowing the characteristics of the individual
players, and they are illuminated by the P perspective.
Voter attitudes may be better understood by selected in-depth interviews than by
polls. The Washington Post’s Haynes Johnson blamed himself and his colleagues for
basing too much coverage of the 1980 campaign on polls, then writing learned analyses
claiming to know what was on people’s minds:
As it turned out, reporters would have been much better served by relying on their own legwork, which, in
turn, produces their own political instincts. [5.5]
l The P perspective serves as a precursor to better understanding of the 0

perspective.
The affected individuals may be considered as a group, but the impact may be
comprehensible only by dealing directly with individuals and their perspectives on a
one-to-one basis. Examples: Union reactions can be gauged better by knowing the at-
titudes of individual union members; the surprising reaction of blacks to electronic funds
transfer could be fathomed by direct dialog; the understanding of the impact of television
on children or of mobility on the aged benefits from personal contacts.
l Individuals may matter, and this perspective identifies their characteristics and
behavior.
Robert Goddard, Andrew Carnegie, and Admiral Hyman Rickover had recognizable
impact on the course of technology. Yet, we have found that there is a strong temptation
on the part of T perspective-oriented individuals to downplay the P perspective. Among
the arguments:
a. only in rare cases does the individual make a difference;

b. there are too many individuals to be considered;
c. it provides carte blanche for baseless claims;
d. the TA would become too sensititive, i.e., politicized.
We consider each point in turn.

a) The criterion as to whether an individual makes a difference is often difficult to
apply. There are obvious situations, but more frequently it is possible to provide a
meaningful answer only in hindsight.
Suppose there had been no Wright brothers. Someone else would have initiated
powered flight, possibly in Germany or France. Would it have made a difference? The
American aerospace industry might have been a German or French one. It frequently does
matter who is first.
b) This argument assumes that a P perspective is useful only if a very large number of
individuals with distinct perspectives is included. This is analogous to the contention that
a decision maker cannot make a decision unless he has obtained input from every affected
individual. Recognizing that this is impractical, the decision maker does not jump to the
other extreme and argue that he should not talk to any affected party.
c) It may be argued that our ability to assess a technology is very modest even with
the perspective which is most “scientific.” After all, the T perspective involves conjec-
tures about consequences, assumptions about models, and at best partial evidence.
The lack of rigor characterizing P provides the analyst leeway for all kinds of
gratuitous interpretations and attributions. When it comes to assessing, by way of anticipa-
tion, the influence of an individual actor on the development of a technology (or, more
generally, on historical processes), the criterion of “reasonableness of conjecture” should

be applied. Certain wild speculations should be eschewed, even though it can be demon-
strated that there are historical precedents for the projected scenario. Because Caligula
appointed a horse to the position of consul, we are not justified to spin out extraordinary
possibilities of a similar nature. Unfortunately history is a cautionary tale of the chame-
leon nature of the reasonableness criterion itself. Hitler seemed eminently reasonable to a
majority of German voters in the 1933 election.
A priori, one cannot argue that multiple perspectives are invariably superior to a
single T perspective. Obviously, three poorly done perspectives will be inferior to one
excellent T perspective, as in aerial photography three overexposed shots from different
angles will tell us less than one sharp shot. The answer lies with the quality of the multiple
perspectives, i.e., the integrity, appropriateness, and experience of the assessment team.
On a ceteris paribus basis multiple perspectives should be preferred to single perspectives.
d) The argument assumes that the TA can be apolitical and operate effectively at a
macrolevel where individuals are seen far below as so many ants. Those most comfortable
with the T perspective are also most likely to be made nervous by the political sensitivity
of the perspectives which are most appropriate to what Allison calls bureaucratic politics
(Table 2) and what is shown as political activity in Figure 4. They nearly always prefer to
shun these aspects, although doing so may strip the TA of much of its value and submerge
some of the very things that should be brought to the surface. A noncontroversial TA is
also likely to be a TA of very modest usefulness.
l Communication of complex problems and issues is easier through the personal

perspective.
Novelists and playwrights express social issues very effectively through individuals.
Shaw uses the unique Lisa Doolittle and her father to portray class problems in England;
Arthur Miller uses Willie Loman to describe the American urge to be well-liked.
Hauptmann’s Rose Bernd, Ibsen’s The Wild Duck, Solzhenitsyn’s Gulag Archipelago,
Burgess’ 1985, and television’s Upstairs, Downstairs are other striking illustrations of the
use of the P perspective to communicate social concepts.
The rare individual who can incorporate in his own thinking the appropriate balance
among the T, 0, and P perspectives can develop a TA as a one-person effort. Burby’s The
Great American Motion Sickness, is, in effect, a TA. It is merely necessary for a member
of Congress to read this book, which the government did not need to fund, to be quite well
informed about transportation technology. It might be more effective to support a good
communicator for one or two years to write a book on technology, approaching the subject
from different perspectives, than to fund a conventional TA team.
T + 0 + p: A SINGERIAN INQUIRING SYSTEM

Table 4 summarizes the distinctive characteristics of our perspectives and Figure 5
relates them to our discussion in Section II; it should be compared with Figure 1. The
Multiple Perspective Concept may be seen as a Singerian inquiring system in Church-
man’s terms [8]:
l It is a metainquiring system, i.e., it includes all the other inquiring systems (data,
model, dialectic, etc.):
TABLE 4
Multiple Perspectives
Technical (T) Organizational (0) Personal (P)
Weltanschauung Science-technology Social infrastructure IndividuatioMhe self
Characteristics Cause-effect Cause-effect and challenge-response Challenge-response
Objective Objective and subjective Subjective
Problem solving Problem avoidance or delegation Leaders and followers

abstraction and modeling Agenda (“problem of the moment”) Game-in-progress for most
Analysis Experience, credentials Need for beliefs, certainties
Prediction Recognition of partial unpredictability Status improvement or maintenance

Long range planning often ritualized Focus on simplistic hypotheses rather than scanning
many
Optimization Satisticing (first acceptable, rather than best, option) Creativity and vision by few-‘&the yogi”
Incremental change, slow adaptation Each construes attributes of others
Parochial priorities Partial rationality

Loyalty Inner world
Use of averages, probabilities Standard operating procedures Learning and experience

tradeoffs Factoring/fractionating problems Cope with few alternatives or variables only
Compromise and bargaining Filter out images inconsistent with past experience
Product Process Impact

Action or implementation Maslow hierarchy of needs or
Power/influence/dominance-“the commissar”
Uncertainties noted: many Avoidance of uncertainty Fear of change and unknown

caveats
Preferred Inquiring Lockeandata Hegeliandialectic Intuitiorr-noumena

System Leibnizian-model Merleau-Ponty-negotiated reality Individual reality
Kantian-multimodal Electicism-experiential
Time Concept Technological time Social time Personal time

Minimal discounting Moderate discounting High discounting
NICAL PERSPECTIVE Multiple Perspectives

e)
ibniz)
\
MULTIPLE MDELS (K
ORGANIZATIONS
"Crowds and Power"
PERSPECTIVES
Fig. 5
It is pragmatic, i.e., the truth content is relative to the overall goals and objectives
of the inquiry;
No single aspect has anybndamental priority over any of the other aspects;
It takes holistic thinking so seriously that it constantly attempts to sweep in new
components; it is in fact nonterminating and explicitly concerned with the future;
It postulates that the system designer is a fundamental part of the system: his
psychology and sociology are inseparable from the system’s physical representa-
lion.
A subtle, but important, advantage in the use of such a pragmatic Singerian approach
is the automatic reduction in the reliance on shaky theoretical analysis in decision making.
The strength of the quasitheological faith in models developed via the T perspective is
awe-inspiring and frightening. Multiple perspectives should minimize this self-delusion
since the T perspective no longer dominates the analysis.
The schematic in Figure 5 shows a single vertex for the T perspective and multiple
vertices for 0 and P. We do not consider the differences between disciplines (e.g.,
aeronautical engineering, electronics engineering, economics) as reflecting different in-
quiring systems. They all use the same data/model based paradigms (see Section II)-the
same mode of perceiving-but they do look at different parts of the problem (e.g., use
different variables and mathematical models). The aeronautical engineer, electronics en-
gineer, and economist may look at the same aircraft design and focus on different prob-
lems (e.g., lift, avionics, cost per mile). But they all use the T perspective: They develop
models (wind tunnel or computer), and seek to quantify and optimize. This is denoted in
Figure 6a by use of a single vertex for all three. Putting it simply, their “how” is the same
but their “what” differs from one person or organization to another.
The crucial differences between perspectives T on the one hand and O/P on the other
may be explored further. In Figure 6b we show by a slight modification of the figure the
quasicontinuous range of perspectives between individuals and organizations. There are
El as distinguished from multiple

discial ines--an example
aircraft design
1111 discipline perspec tive
personal-organizational continuum
Fig. 6. Multiple perspectives

THE MULTIPLE PERSPECTIVE CONCEm 303
various individuals, informal groupings, more formal small and large organizations and
social entities. Each has a unique Weltanschauung, a different perspective.
FURTHER COMMENTS
Another Catch 22
In Section I we noted a Catch 22 situation. We can now add another. Impact and
policy analysis are essential to TA and the typical TA is expected to deal with impacts and
policy decisions which lie in the future, often five to twenty years away.
The Multiple Perspective Concept leads us to recognition of a predicament. Each
perspective tends to have a different planning horizon (Figure 3 and Table 4). Combining
this characteristic with our basic schematic, Figure 5, we can represent the dilemma as
shown in Figure 7. Thinking of the perspectives as beams of light directed at the problem,
we find that each beam has a different illuminating distance (like high and low headlights
in an automobile). So we arrive at the catch:
Multiple perspectives are critical in developing insights on complex socio-

technological problems;
The differential horizons of the three perspectives make it hard to deal with the
long-term impacts and policy questions vital to many TAs.
Even the longest beam, representing the T perspective in Figure 7, is often overesti-
mated. Consider technological and economic forecasting, for example. Ascher, in his
careful analysis of many academic, government, and industry forecasts, finds a consistent
Fii. 7. Different perspectives-different planning horizons.

pattern: the more distant the forecast target date, the less accurate the forecast [6: 1991. He
correlates this tendency with the core assumptions underlying the forecast. They are the
major determinants of forecast accuracy and much more crucial than sophistication of the
forecasting model. Since the core assumptions usually degrade with increasing time
horizon, they constrain the T perspective horizon more than the methodology-oriented
rational actor generally assumes.
We draw two implications:
1. TAs are inherently difficult to do well when they must deal with long-term
impacts and policies;
2. The most effective applications of multiple perspectives are likely to be in
sociotechnological systems for which the focus is on near- and mid-range prob-
lems, i.e., a time horizon encompassed by the 0 and P perspectives.
Point 1) explains the frustrations experienced by TA practitioners in the past decade.

It tells us why the TA frequently assesses the technology (and its implementation) but does
not assess the impacts. We also recall Peter Drucker’s admonition:
The future impact of new technology is almost always beyond anybody’s imagination. At the same
time, a technology impact that the “expert” foresees almost never actually occurs. [ 131.
On the other hand, he insists that:

Technology monitoring is a serious, an important, indeed a vital task. [ 131.
The difference is the horizon--predicting long-term complex system behavior versus

monitoring what is going on now with a young technology.
The most important advice now given to the technology assessor-never consider
your analysis as complete, keep iterating or revisiting and “burnishing” the work at
subsequent time points---can be interpreted as a means to deal with this dilemma. Refer-
ring to Figure 7, we recognize that the situation is dynamic. The 0 and P perspectives
move in closer as time passes and will begin to illuminate the system.
Point 2) is hardly a revelation to top level corporate and governmental decision
makers. In fact, it is reflected in the discounting which is so pervasive in the business and
political decision process (Figure 2). But it does explain the frustrations of corporate and
governmental long-range planners. Either they conscientiously delve into long-range as-
pects via the T perspective and find their reports unread and unused, or they concentrate
on near-term policy and implementation decisions to find their products becoming con-
troversial and politically sensitive.
At the same time we draw some positive inferences:
1, There are in any society farsighted individuals (who may as leaders also inspire
organizations and social entities to be farsighted). History has given us a Pericles,
a Washington, a Lincoln, and a Churchill-men of vision. Churchill, for exam-
ple, was thinking far beyond victory during the last years of World War II. His
planning horizon encompassed the postwar Soviet threat at a time when nearly
everyone was only concerned with the great on-going war (see Section IV).s We
also find men with vision as pioneers in industry-they see a new product line or
gNot surprisingly, individuals like Churchill, who use a lower-than-average discount rate in looking forward
to the future, also apply a lower-than-average discount rate to the past. Hence they have a more profound sense
of history.
company image before others in the organization. And they recognize the long-
term impacts of their decisions. Ideally the TA team should seek out such indi-
viduals; unfortunately they are only rarely available. But the team leader should
certainly make a conscious effort to include, as team members of consultants,
individuals with strong 0 and P perspective orientations and a longer-than-
average time horizon.
2. There are many situations where a near or midrange time horizon provides a
sound basis for policy decisions. The multiple perspective concept should ulti-
mately leave its deepest imprint there as it bridges the gap between the model
world and the real world.
Choosing the Team

An individual steeped throughout his working life in one set of paradigms and trained
to suppress the others will find it difficult to apply another in developing a new perspec-
tive. Thus the scientist or technologist is accustomed to T and may be unable to do justice
to 0 or P. To a man with a hammer the world appears to be full of nails ready to be hit. We
find that a T trained person will see, or assume, the world to be populated by rational
actors and afflicted with problems to be solved by data and model based techniques. Such
an individual often views the 0 perspective as mere detail and the P perspective as dirty
politics. The 0 and P perspectives similarly have preconceptions about the T per-
spective-usually viewing it as naive. The mutual misperceptions are shown in Figure 8.
Great care must be exercised in choosing TA team members: the usual TA partici-
pants are not ideal in developing 0 and P perspectives. Either team members must be
chosen to balance the needed characteristics or the available T participants must be trained
to develop the desired inquiring procedures and sensitivities.
Let us first dispense with the idea of interdisciplinarity. In the scientist-technologist
world view dominated by reductionism, compartmentalization by disciplines is natural.
Frequently, interdisciplinarity is interpreted as the participation of experts from several
disciplines, e.g., engineering, economics, and mathematics. In any case, all such disci-
plines use the same T perspective, and so will their joint effort. We might ask next
whether a systems analyst or an operations analyst solves our problems. Such a person is,
after all, interdisciplinary according to most definitions. The answer here also is negative.
I I
wants a smooth 0 is part of the game
based on SOP P is a personnel problem

>
Fig. 8. The basis for miscommunications. (Source: [563.)

Such an individual may indeed be interdisciplinary but he is still anchored in the same T
perspective.
An effective 0 or P perspective team member brings to bear a very different back-
ground than that of a T perspective member, e.g., law, journalism, acting, or administra-
tion. We know that a lawyer is at home with dialectic inquiring systems, an investigative
reporter works on hunches. Such an individual may be aided by having a high “social
intelligence quotient.” [4].
Archer [5] reports that women are generally more adept than men in this regard and
suggests that female intuition is not a myth. This observation is supported by Kaje [26] on
futures research and by our experience with male and female interviewers. We suspect
that many male academics, particularly those with science or technology training, tend to
be quite weak in social intelligence. This, in turn, is reflected in the bias toward the T
perspective.
Communicutions
Communication techniques for 0 and P perspectives (either for input or output) are
by no means identical to those associated with the T perspective.
Input. Jay observes that standard documentation is not the place to determine the foci of
power in an organization:
Real power does not lie in documents and memos outlining your terms of reference and area of jurisdic-
tion: it lies in what you can achieve in practice. The boss’s secretary can wield great power, like the king’s
mistress, without any authority at all-r at least not the sort you can show anybody. [25: 1421
Interviews play a dominant role in 0 and P, in contrast to the reports and surveys
used for T. Other methods such as participant observation, guided group dialog, and
open-ended simulation may also prove valuable in gaining an understanding of organiza-
tions and individuals. Future applications of 0 and P perspectives should explore these
possibilities more fully.
The key to comprehending individuals in P perspectives is the empathic capacity of
the interviewer or observer. No one individual is capable of empathy with everyone.
There are inevitably psychosocial barriers between certain people, no matter how skilled or
naturally intuitive a given interviewer or observer may be.
Empathic capacity in the interview situation is, in part, a function of the respondent’s
perception of the interviewer as nonthreatening and in part a matter of the ego strength and
self sufficiency (in emotional terms) of the interviewer. In certain circumstances, neurotic
behavior can produce close empathic feelings+specially if the neuroses complement
each other-but, in such cases, the degree of comprehension will be distorted by the
interviewer’s own feelings and emotional reactions. Unless the interviewer can both gain
insight and preserve the ability to dissociate himself from the client, the product will be a set
of idiosyncratic projections on the part of the interviewer, not a useful understanding of
the repondent.
Thus, we have a prescription nearly impossible to execute well in interviews. But
some talent can go a long way, and our experience over the past four years shows that it
exists most often in individuals who are not focused strongly on the T perspective.
Outpur. We take for granted the technical report format for technology assessments.
It is standard for the T perspective but not necessarily appropriate for the 0 and P
perspectives. The latter in particular suggests a unique contextual point of view which
lends itself to experiential formulation and personalized presentation. A scenario, a
vignette, or a story may crystallize the image or convey the tone far better than a conven-
THE MULTIPLE PERSPECTIVE CONCER 307
tional format. It can also serve as a vehicle to communicate the main ideas of the entire TA
(all perspectives) to an audience which cannot deal with, or is not satisfied with, the
tee hnical report format.
Table 5 suggests some of the possible matches of perspective and means of presenta-
tion.
In the applications, we have experimented with several of these vehicles for the 0
and P perspectives [34]. The oral briefing needs further explication here, however. Al-
most any report can be transformed into an oral briefing. It is included in Table 5 for 0
and P to highlight the problem of the sensitivity of these perspectives. It may be advisable
to present important insights orally but not in writing.
Unquestionably, these perspectives pose problems not of concern with the T perspec-
tive. They reflect the political nature of the TA and must be faced, just as they are by the
journalist.
There are several alternatives:
Partly oral and partly written report:

Two-part written report (one part for distribution by client, another for internal
use by client);
Use of quotes but without attribution;
Use of sensitive material by the TA team as input only with implications presented
in the report:
Integration of insights obtained from different perspectives before the final report
preparation:
Transformation of sensitive material into$ctional format.
Changing Perspectives
Not only are there many 0 and P perspectives, but any one such perspective may
shift over time. In other words, the picture should be viewed as dynamic rather than static.
Aging tends to make the individual more conservative, less of a risk taker. Furthermore,
the individual assumes different roles at different times. The balance among the perspec-
tives should not be viewed as a fixed or rigid one. A technology which does not involve
human beings should not require 0 or P perspectives, but such a technology does not
exist. And if it did, technology assessment would be superfluous.
The apportionment of effort among the three perspectives cannot be disposed of by a
simple formula. It is a function not only of the technology, but also of the political
framework of the TA. As a first approximation, an equal three-part split in effort among
the three perspectives seems eminently reasonable.
TABLE 5
Some Means to Communicate Perspectives
T 0 P
Technical report fl v
Scenario/vignette v w ti
Transcript of interviews rc v
Summaries of interviews J v
Oral briefing rc v r/
Story/play r/
Noret For illustrations see [341

Integration of Perspectives
Pro and Con. We must deal with two basic questions:
1. Do the perspectives work at cross-purposes, or do they reinforce each other?

2. How are they to be integrated, and by whom?
Consider a corporate executive faced with a complex sociotechnological decision. He

obtains a T perspective from his engineers and systems analysts, an 0 perspective from his
department heads, and a P perspective from workers and staff whom he queries infor-
mally. There may well be conflicting input to him as well as reinforcement. Similarly, the
various witnesses at a trial provide different perspectives to the jury. The process of
integration is not merely one of assembling a composite picture from jigsaw puzzle pieces,
or ironing out contradictions by some rules of thumb, or arranging the information hierar-
chically. Integration resembles the task of conceptualizing a three-dimensional object
from a series of one-dimensional descriptions and two-dimensional drawings. Alterna-
tively, we can think of the process in terms of the integration of stimuli from the left and
right hemispheres in the brain. The perspectives cross-cue each other. Most decision
makers find it quite difficult to describe this process explicitly, although they do it every
day. It is often just as difficult to reconstruct how a jury arrived at a decision. The fact that
in both cases effective decision making does occur suggests that the integration process in
our situation be left to the user or decision maker. Like prosecuting and defense attorneys
at a trial, the assessor may point to cross-cuing instances, but the decisive linkages must
be made by the decision maker or other user. Providing an integrated perspective to the
client means weighing and interpreting, making value judgments and choices that, in most
cases, are better left to the decision maker.
An even stronger argument against integration by the analyst can be made if there is a
priori no well-defined user, as in the case of a technology assessment done for the
National Science Foundation. Subsequent users will, indeed must, use different weighting
and cross-cuing.
An argument for integration by the analyst is the possibility that an inexperienced
client may misuse the perspectives. For example, he or she might choose a best perspec-
tive and ignore the others. Or, the analysis might be thrown out because there is disagree-
ment among perspectives. (We recall that a technologist is not used to dialectic inquiring
systems.)
Comprehensiveness and Definiteness? Two words which should be taboo in any discus-
sion of TA are “comprehensive” and “definitive,” as in:
a TA should be “comprehensive,” or
the described perspectives are “definitive.”
Only a T perspective mind is likely to use these words. Anyone who has tried to
determine the number of interactions possible with a small number of elements in a system
will understand the unreality of comprehensiveness. lo So will anyone who has studied the
impacts of technologies in history.
‘OSee Section II
THE MULTIPLE PERSPECTIVE CONCEFI 309
The changeability of perspectives has been noted. Individuals and organizations

change perspectives over time, key actors enter and leave the stage. The changing setting
has, of course, been recognized by TA practitioners in the advice to periodically iterate a
given TA and consider it an ongoing activity.
Finally, we express a nagging worry. The Singerian inquiring system implied by the
use of multiple perspectives understandably jars those (scientists and technologists) accus-
tomed to Lockean (data based), Leibnizian (model based), and Kantian (data plus model
or multimodel) inquiring systems. And this leaves us with the residual fear that the 0 and
P perspectives may be unconsciously transformed into T perspectives. This concern is best
alleviated by a) creating an interparadigmatic, rather than interdisciplinary, team at the
beginning of a TA, and b) discouraging academic efforts to create formal theories and
standard routines for the application of multiple perspectives.
Perhaps the ultimate value of the multiple perspective concept is to give us a better
balance. The paradigms of science and technology may have dazzled us with their suc-
cess, but they do not carry us very far into the realm of human affairs. Just as three-
dimensional space helps us understand the limitations of one-dimensional space, so the 0
and P perspectives make us see the limitations of the T perspective and deal with them
constructively.
Section IV. Illustrations

The examples span a wide spectrum of uses of multiple perspectives: from the Civil
War to the future, from business to government, and from simple technology (rifles and
hydropower) to sophisticated systems (atomic bombs and advanced electronic compo-
nents). l1
MILITARY TECHNOLOGY
l Introduction of a new ship concept into the U.S. Navy (1868-1869) [43]
The USS Wampanoag was a 4200 ton “advanced technology” destroyer built for the
U.S. Navy and commissioned in 1868. She had sails and a steam engine, was fast (over 17
knots), maneuverable, and stable.
T Sea trials prove her to be a magnificent technical achievement. All specs

are exceeded; she is superior to similar class ships in any navy at this
time.
0 In 1869, a Naval Board examines the ship and comes to the conclusion
that it is a “sad and signal failure.” A typical argument in their report:
“Lounging through the watches of a steamer or acting as firemen and
coal heavers will not produce in a seaman that combination of boldness,
strength, and skill which characterized the American sailor of an elder
day.” As Elting Morison observes: “What these officers were saying
was that Wampanoag was a destructive energy in their society .
“For more details see [34]

They perceived that a machine tends to create its own environment and
draw men into it . I don’t admire their solution, but I respect their
awareness that they had a problem. [It] is not primarily engineering or
scientific in character. It’s simply human.”
A decision which appeared “stupid” from the T perspective becomes reasonable when
seen from the 0 perspective.
l The M-16 rifle: the long struggle to provide the U.S. Army with a superior rife
(originally called AR-15)
Army Ordnance has been furnishing rifles to the Army for over a century. In the
195Os, its Springfield Arsenal developed the M- 14 rifle.
T At about the same time, Armalite Corporation’s Eugene Stoner de-

signed the AR-15. It used smaller and faster bullets and had a plastic
stock (therefore, was considerably lighter). Extensive Army field tests
and analyses led to the conclusion that the AR- 15 was superior to the
M- 14. World Wars I and II showed that few soldiers in combat fire their
rifles at targets over 300 yards distant. They want lethality, which is
enhanced by bullet speed more than by caliber size. U.S. troops in
Vietnam requested the AR-15; General LeMay declared it standard for
the U.S. Air Force (1962). Tests by the Advanced Research Projects
Agency (not part of the Army) concluded “in overall squad kill poten-
tial, the AR- 15 is up to 5 times as effective as the M- 14 rifle. [The
AR- 151 can be produced with less difficulty, to a higher quality, and at a
lower cost than the M-14 rifle. In reliability, durability, ruggedness,
performance under adverse conditions, and ease of maintenance, the
AR- 15 is a significant improvement over any of the standard weapons,
including the M-14 rifle.” [38:84]
Army Ordnance, responsible for providing small arms, strenuously opposed the
AR-15.
0 The Civil War inaugurated the era of the bullet. The myth of the western
sharpshooter and the formation of the National Rifle Association re-
flected development of a marksmanship tradition which infused the
Army Ordnance Corps. A marksman firing in long-range rifle competi-
tion needs a large, heavy round to maximize steadiness in flight and
minimize sensitivity to wind. The marksmanship tradition led Army
Ordnance to the M- 1 in World War II and then to the M- 14.
P Stoner tried hard to market his rifle. Opposition to Army Ordnance

appeared within the Army-Colonel Neilsen of the Infantry Board, for
example, helped to persuade that group to favor smaller caliber rifles
(based partly on his personal World War II experience). Secretary of
Defense McNamara was strongly oriented to the T perspective and
forced new analyses which negated validity of Ordnance test results.

His modus operandi was not that of a compromising politician.
Ordnance ordered modifications to the AR-15 (now M-16), which “had very little to do
with . . warfare but quite a lot to do with settling organizational scores.” [ 15:77] Thus,
the M-16 became unreliable in combat in Vietnam. We note that the rational actor’s
selection of the best system hardly reflects the real-life decision process in which the 0
perspective is central. Parochial objectives of organizational actors may play a crucial
role. The Army does not operate as a unitary decision maker; individuals play a vital role
in the process. Overt and hidden motivations for actions must be clearly distinguished.
l The decision to use the atomic bomb: June I, I945
Truman has been President for only seven weeks, following the death of Franklin
Roosevelt. Germany has tendered its unconditional surrender and Japan is almost a de-
feated nation. The atomic bomb has been developed in great secrecy and is considered the
greatest military technological innovation of modern times.
T There are five alternatives: a) conventional bombing and blockade of

Japan, b) invasion, c) dropping of the atomic bomb on Japan without
warning, d) dropping the bomb on Japan after warning, and e) dropping
the bomb on an uninhabited island. Analysis shows clearly that c) is the
preference if the objective is to obtain unconditional surrender with a
minimum cost of Allied lives and minimum destruction of Japan.
0 Nobel Laureate P. M. S. Blackett recalls: “The dropping of the atomic

bomb was not so much the last military act of the Second World War, as
the first major operation of the cold diplomatic war with Russia.”
[2:232] With Germany’s defeat, each country’s parochial priorities,
perceptions, and interests shift. Russia seeks expansion in Europe and
the Far East. Great Britain, Churchill in particular, is concerned about a
Russian threat in Europe.
At a different level, nonuse of the bomb raises organizational fears,
since vast funding was obtained without formal congressional approval.
The Manhattan Project director: “If we succeed, they won’t investigate
us at all; if we fail, Congress for a long time won’t investigate anything
else.” [2:28] Admiral Leahy: “I know FDR would have used it in a
minute to prove that he had not wasted two billion dollars.” [49:57]
P The key individuals believed deeply in the rightness of unconditional

surrender. Truman was new, an outsider to the 12-year FDR establish-
ment. He could not challenge policies and fiefdoms so early in his
presidency. He had to look bold and decisive. A decision not to use the
bomb would appear as a sign of weakness to historians (and Truman had
a strong sense of history).
Thus the 0 and P perspectives reinforce the T perspective.

OTHER FEDERAL ILLUSTRATIONS
l The Federal bureaucracy
Charles Peters’ and Henry Kissinger’s observations on the operation of the federal
government suggest that the 0 perspective is essential to an understanding of the decision
setting.
0 On shadow organizations:
Clubs are just part of a larger social bond that exists everywhere but is
especially prevalent in Washington, where private life is so much an
extension of professional life. This bond is the “survival network” .
From the day [government] people arrive in Washington they are busy
building networks of people who will assure their survival in power.
[4.5:5]
On protection of turf:
The Firemen First Principle . . when faced with budget cuts, the
bureaucrat . chops where it will hurt constituents the most, not the
least. The howls of protest wil then force the cuts to be restored. [45:40]
On power:
In a recent survey [lobbyists] rated the congressional staffs as their
number one lobbying target (by contrast, the White House ranked
sixth). [45: 1151
Kissinger on Secretary of Defense Laird:
I eventually learned that it was safest to begin a battle with Laird by
closing off all his bureaucratic or congressional escape routes, provided
I could figure them out. Only then would I broach substance. [28]
l Inflation
Konrad Kellen is not satisfied with the conventional Washington economists’ theoret-
ical T perspective or the bureaucrats’ regulatory 0 perspective on inflation. He asks: “Is
inflation fun?”
P The true secret of inflation is that it is pleasure at a price . . Spending

isfun.. . the nagging feeling of guilt has been eliminated . . You not
only spend more, you also earn more, or so it seems.
Such collective vice is not curable by economic means; it has deep
psychological roots.
It may have started as a set of adverse economic developments, just
as a person may have started drinking because he was fired. But after
several years of drinking, his alcoholism-not his having been fired-
will be his problem. [27: l-51
HYDROPOWER
l Construction of a hydroelectric facility

THE MULTIPLE PERSPECTIVE CONCEF’I- 313
The environmentally benign nature of hydropower and the projected energy shortage
make hydropower a particularly attractive source. A utility’s recent proposal to build a
hydroelectric facility is illuminated by the use of all three perspectives.
T From 1958 to 1980 five major engineering and cost-benefit studies are
undertaken and four alternatives developed. The construction of a 60
megawatt facility is the preferred choice.
0 The change in presidency of the company is accompanied by major

shifts in organization and decision processes. The engineering and rate
departments do not maintain their traditional dominance. A corporate
planning department is formed and the position of Vice President for
Human Resources created. The decision process on this project reflects
these organizational changes. A citizens’ task force is formed to involve
the community in the decision. With fishermen who speak up strongly
against the project dominating the meetings, the result is cancellation of
the project by the company.
P The new president is an outsider, having a background of public rela-

tions and finance. He has a very different style than his predecessor.
The changes weaken the Old Guard; younger persons are promoted.
The new decision process would have been very difficult, if not im-
possible, to implement without these changes.
HEALTH CARE
l Perinatal Regionalization [56]
“Perinatal” refers to the period surrounding birth-prenatal, natal, and postnatal

care, family planning and genetic counseling. “Regionalization” refers to coordination of
institutions and personnel providing perinatal care services in a given area (both public
and private). The aim is to match health risk level with appropriate services to insure the
best outcome for mother and infant with efficient use of resources. (The U.S. ranks 17th
in the world in birth-related mortality). The region studied lies within a large metropolitan
county, and the majority of its inhabitants are Black and Hispanic.
T This perspective, in a well-ordered approach, systematically sets out to

present its case, implement a plan of action, and evaluate the results.
Project staff contribute by establishing system linkages and procedures.
What has been gained over five years? What ongoing costs need to be
picked up and are worth it? Cost-benefit indicators and performance
measures are determined. In and of itself, however, the perspective
misses many role aspects of the staff and critical elements in gaining
entry to organizations as well as inducing change. Nor is it attuned
to such factors as “climate” and “image.”
0 The intricacies of breaking through the traditional roles, established

procedures, and drives devoted to other priorities; of breaking across
professions and institutions; and of involving the interrelation of per-

sonal, economic, and positional factors are emphasized. Project staff
are sensitive to their roles in “others’ territory.” An attempt to change
the practices of a hospital unit may be challenged by the unit not just
because its head is paranoid, incompetent, or resistant to change. The
prerogative of change may belong to someone who, if bypassed, can
destabilize the multilayer network of authority, status, and working
relationships. On the other hand, these very linkages make possible
radical change. Those sensitive to organizational functioning, e.g.,
managers and union representatives, are likely to do well in interpreting
the information obtained by this perspective. There is a disparity be-
tween the institution that exists on paper and the functioning reality. For
example, the department in the county hospital is desperately short of
faculty, but the books show capacity staffing. The professionally
socialized nursing role is very different from actual nursing practice in
an overcrowded, understaffed hospital. “Turf” is a very important
consideration (administrators, doctors, registered and other nurses,
etc.).
P This perspective provides the most immediate grasp of the essential and
unique world of the participants, but is the most elusive to deal with.
The role of the staff as conveyors and developers of an image and their
development of others in this context is quite different than their role in
T and 0 perspectives.
Typical P issues: Who are the movers? How can the image of
regionalization be developed by various interests and help to congeal
them? For example, it should reflect sensitivity to Black professionals
and their unique and controversial role in the community.
The cross-cuing in the process of integrating perspectives is demonstrated in Figure 9. For

example, taking the P perspective as the context, the T perspective serves as a kickoff, a
statement of need, and the 0 perspective presents the real life lay-of-the-land which
contains the potential of transformation.
T PERSPECTIVE 0 PERSPECTIVE P PERSPECTIVE
Need, task, plan Organlzatlon Position Imap
(effectiveness, efficiency) (orientation. dynamics, (reality, uniqueness, power)

str tegyj
1 f J
Organlration (0) Role Market Conditions Kickoff Potential in
(implementation, feasibility) CT) the current setting

(0)
J t 1
lndlvldual Positions (P) Policy Plan Create Market. Products
(motivation. acceptrbility) (T) (P)
Fig. 9. Cross-cuing of perspectives.

THE MULTIPLE PERSPECTIVE CONCEF’I- 315
ELECTRONIC FUNDS TRANSFER
l A Retrospective Look at two Studies
In 1975 Arthur D. Little, Inc. (ADL) undertook a technology assessment of Elec-

tronic Funds Transfer for the National Science Foundation [35]. Shortly thereafter the
Congress created an independent commission to study the same subject and make recom-
mendations for administrative action and legislation [44]. The ADL study was aimed at
the concerned public-an “off-the-shelf supermarket ” item; the Commission Report was
done for the President and Congress.
Electronic funds transfer (EFT) includes a number of functions: services provided via
teller machines, automated clearing houses (including direct deposits and telephonic bill
paying), as well as check safekeeping, corporate-to-corporate transactions, etc. and
point-of-sale terminals.
A schematic comparison of the studies is shown in Figure 10. Both focused on the
organizational aspects of EFT, but the perspectives were somewhat different. The analysts
of ADL used more of the T perspective, the senior bureaucrats and vested interest repre-
sentatives comprising the National Commission emphasized the 0 perspective. The ADL
team was concerned with the “audit trail,” i.e., the ability to trace the results back to
inputs and thus assure replicability. The Commission members hammered out a set of
recommendations through confrontation, negotiation, and compromise. ADL’s T perspec-
tive brought out the uncertainties; the Commission’s 0 perspective tended to resolve them
to obtain agreement. In hindsight, the emphasis on organizations was sound, and the
reports appear very reasonable five years later. However, it may be argued that the
long-term impacts of other technologies on EFT may need to be examined in greater
depth: rapid fund movement and the reverberations of monetary shocks, financial system
controllability by an oligarchy, sophisticated computer crime, and the shifting turf of
banks and high-technology companies may prove significant. [ 1 l]
THE COMMERCIALIZATION OF GUAYULE
l A Technology Assessment (1980)
Guayule is a rubber-producing plant suitable to growth in semidesert areas such as

Northeastern Mexico and the Southwestern U.S. It is a substitute for the natural rubber
obtained from the hevea tree. The Office of Arid Lands Studies at the University of
Arizona has, with the support of the Midwest Research Institute, done a technology
assessment for the National Science Foundation. [19] It is a T perspective on the
technological and economic aspects of the commercialization problem. Multiple perspec-
tives broaden the view.
T Guayule rubber has been tested and found quite satisfactory as a re-
placement for hevea rubber. A worldwide shortfall of natural rubber of
2.4-4.9 million metric tons is anticipated by 2000. Assuming a
surprise-free scenario, 100% of the shortfall can be made up with
guayule by 1991. The impacts on the Southwest economy are likely to
be modestly beneficial. [ 191
A.D. Little
Technology
Assessment
aspects studied
(e.g., home computer)
National
Commission
Report
//,Ierspectives used
\
::::::
LOCUS of problems
-'**-' (elements)
w
P
Fig. 10 Schematic of perspectives: Electronic Funds Transfer Reports.
0 There is a network of researchers which has been keeping interest in

guayule alive for decades. But this is not sufficient to get commerciali-
zation off the ground. A coalition with adequate leverage is needed; a
leader or power broker may be the key to its creation. Since rubber is a
strategic material in the event of a national crisis, both the Department
of Defense and Federal Emergency Management Agency could become
the core of this coalition. The tire and rubber companies are financially
in a weak position and managerially very conservative, hence doing
defensive research and hoping the government will take the risk in the
early stages of commercialization, Alternatively, there may be oppor-
tunities for more aggressive outsiders to jump in.
P Each actor sees his own prospective behavior differently than the others
perceive it. Each sees the others’ perceptions as distorted; each therefore
interprets a given act differently. This situation discourages formation
of intersectoral coalitions and encourages intrasectoral ones. None of
the present actors yet sees enough in it for him to have a feasible action
strategy. There is a crucial difference between those who are emotion-
ally tied to guayule and those who must make commercialization deci-
sions (advocates versus judges).
ON-SITE SOLID WASTE TREATMENT
l Another Technology Assessment (1981)
Modern urbanized societies have become completely accustomed to centralized

sewer-transported systems for solid household waste disposal and to large-scale treatment
facilities. The high sewer costs and lavish use of water in this process have raised serious
questions when seen in the context of potential federal sewer funding reductions and water
shortages. Decentralized systems are available and worthy of consideration, particularly
for small communities. There are three types of on-site systems: biological, incinerating,
and aerobic. Stanford University has undertaken a technology assessment which might be
described in our terms as a T perspective on the decision process for on-site waste
technology implementation [24]. Multiple perspectives illuminate other factors.
T The decision analysis approach of [24] recognizes different types of

individuals+conoman, environman, nominalman, and analyticman.
These idealized types are defined by a small number of variables relat-
ing to cost. Organizations make decisions based on a costing model and
cost-benefit calculations modified by a “responsiveness parameter. ”
Decision analysis provides a sophisticated technique which indicates
what ought to be good decisions in light of the quantitative information
available.
0 There is an absence of strong vested interest support; even environmen-

talists have higher priorities. Health departments favor centralized sys-
tems because they appear easier to monitor. Heavy federal funding
support of centralized sewer systems is also a major factor. Real estate
developers and builders are the potential nucleus of a coalition to push
on-site systems. However, they are too atomized at present, and there is
little incentive. (A cutoff of federal sewer support could provide the
spark.) Normally, they will not take risks, but will wait until such
fixtures can be bought at most plumbing supply houses.
P The engineering schools tend to bias the student very strongly toward
centralized systems, viewing on-site systems as temporary expedients.
Most people have a “flush and that’s the end of it” attitude. There is a
strong dislike of the maintenance which is required with on-site sys-
tems: “You’ve got to babysit the toilets.”
THE RANGE OF APPLICATIONS

The first classroom exercises using multiple perspectives were carried out by
Linstone in January-March 1977 in the University of Washington’s Program for the
Social Management of Technology. Students individually produced interesting term pa-
pers using the three perspectives on subjects such as the West Seattle freeway extension,
the cruise missile, and a local methadone project.
The Multiple Perspective Concept has now become the heart of the last quarter of the
basic three-quarter graduate course, The Systems Approach, in the Systems Science PhD
Program at Portland State University.
We have a clear intimation in this section that technology assessment, although the
focus of the present project, is merely one realm of multiple perspective applications. As a
further example, consider risk analysis. Table 6 shows how the three perspectives illumi-
nate different views of risk. The Three Mile Island nuclear accident investigation recog-
nized the inadequacy of the T perspective in risk analysis:
A truly unexpected result came out of the Kemeny Commission’s study of Three Mile Island. A group that
set out to investigate a technology ended up talking about people. In the Commission’s own words, “It
became clear that the fundamental problems were people-related problems.” [ 141
Even technological forecasting itself may benefit from the use of multiple perspec-
tives. Normative forecasts supposedly express future needs. But the indicated needs may
reflect the 0 perspective of the client organization. A case in point is the defense estab-
lishment. From 1959 to 1970, Linstone directed four corporate planning needs analyses in
the area of national defense and space programs. By the time of the fourth of these
privately funded projects, the T perspective analysts had made an important discovery:
“The gap between what is needed and what is marketable means that a ‘needs analysis’ is,
in fact, a mirage.” [47: 1161 We buy what we like and are comfortable with, not necessar-
TABLE 6
Physical Risk as Seen from the
T, 0, and P Perspectives
T 0 P
Probabilistic Threat to product line, company image Type of danger (cancer as the “plague”
of this century)
Actuarial Abilitv to avoid publicity
. .
Time for consequences to materialize
Expected value Politial sensitivity (voter anger)
(discounting)
calculations
Ease of litigation
Popular image of horror (nuclear
Fail-safe concept
Ease of shifting or spreading blame accident)
Margin of safety
Anonymity of decision makers Personal experience (survival of a hotel
design
tire)
Ease of changing decisions
Applicability of
Age of individual
precedent designs Socioeconomic status as determinant of
risk acceptance (life “guaranteed Popularization of risk (e.g., nuclear
safe” in U.S.) plants and The China SyndromeJ
Cost of recall Ethical overlaps (e.g., religion and gene

splicing)
Standard operating procedures
Archetypal values (e.g.. virility and
Access to expertise on specific risk
guns, fertility and water)
ily what we need. Pilots want aircraft to fly so that an all-missile force would be organiza-
tionally unpalatable. Advancement in the Navy traditionally requires experience in com-
manding ships. Thus, the availability of an adequate number of ships is an important
consideration in developing normative forecasts. We are dealing here with an 0 perspec-
tive very similar to that which Morison recognized in the case of the U.S.S. Wampanoag.
A priority needs list based solely on a T perspective looks very different from one
based on both T and 0 perspectives. The T perspective takes account of Richard III’s
lament “For the want of a nail the shoe was lost . . .” so that the list would include
nonglamor items such as changes in training and communications equipment. A list based
on T and 0, however, features glamor items in the firepower and vehicle areas-new
aircraft and ships, and more exotic weapons.
Thus, the organizational perspective explains constraints and core assumptions which
strongly affect the technological forecasts. Standard operating procedures, morale needs,
as well as incrementalism and tradition determine the forecasts in subtle ways. l2 Even the
personal perspective comes into play. A technology-trained analyst nearly always moves
toward greater technological sophistication in addressing future needs. It is rare to find the
option of less sophistication, of fewer technical approaches, given equal consideration by
a technologist. We see the result in the imbalance in the intelligence field [54], in the
unending efforts to advance technological solutions in the Vietnam conflict, and in the
reliance on technical analyses in strategic military planning [ 151. The application of
multiple perspectives sweeps in considerations which can more clearly reveal the core
assumptions and avoid self-delusion in technological forecasting.
Finally, there is no mason to assume that only our own social setting lends itself to
the application of the Multiple Perspectives Concept. Workshops held in Mexico and
Chile (1981) show that non-U.S. practitioners grasp the concept readily and can use it in
very different cultural contexts.
Section V. Guidelines
In this final section we translate lessons learned into guidelines for the multiple
perspective user. While primary attention is devoted to technology assessment, most of
the discussion is applicable to sociotechnological systems generally, whether in the con-
text of government policy making, corporate strategic planning, regional development, or
systems implementation.
FORM THE TEAM TO ASSURE AN INTERPARADIGMATIC MIX RATHER THAN
MERELY AN INTERDISCIPLINARY MIX (EXAMPLE: ENGINEER PLUS LAWYER
PLUS WRITER PLUS BUSINESSMAN).
In Section III the question of personnel selection was addressed. In particular, the
common misconception that an interdisciplinary mix of professionals (such as an en-
gineer, an economist, a sociologist, and a systems analyst) is most suitable as a TA
team was exposed. l3 For multiple perspectives, we need an interparadigmatic mix,
“In cakes where a forecast is done by one organization A for another B and B imposes few constraints, the
biases of A may be crucial to the forecast. Examples: A’s ingrained modeling tools, its previous studies, and the
nature of its other current clients.
13By extension, a team composed only of academics is likely to be a disaster.
i.e., individuals who have been nurtured on different inquiring systems. Thus, one of
the most important members of the National Hydropower technology assessment team
was a lawyer; he has a very different sense of regulatory issues than does an engineer.
We are not assuming either is a priori familiar with the regulations; rather, they tend to
emphasize different perspectives (0 and T, respectively).
Formation of an interparadigmatic team for TA is not the end, but only the
beginning of the team design task. The most obvious characteristic of the different
perspectives, indeed the reason they are collectively more valuable than they are indi-
vidually, is that they typically conflict with each other; through their interaction and
conflict resolution, they lead to the generation of insights that would not have emerged
otherwise. Consequently, team process design, which anticipates the conflicts and
facilitates their interaction. must be addressed also.
CONSIDER A ONE- OR TWO-PERSON EFFORT OVER A LONGER TIME PERIOD

AS AN ALTERNATIVE TO A LARGER EFFORT CONCENTRATED IN A
SHORTER TIME.
A question which must be addressed is the distribution of effort. The answer may, of
course, be fixed by the available budget/time stipulations. Let us assume this is not the
case. There are individuals who have an appreciation and balance of T, 0, and P
perspectives. Da Vinci was a technologist and artist, H. G. Wells was a mix of scientist
and writer. Many of the most successful modern executives are well-balanced mixes of
the perspectives. If such a special person can be found, he or she may be as valuable or
more valuable than a team in producing a useful technology assessment. Another
possibility is a two-person effort. If both individuals are well balanced in perspectives
they may work independently. Alternatively, one could use a team of two persons very
different in their perspective emphasis (probably one T and the other 0).
AIM FOR DEEP TRANSDISCIPLINARY IMMERSION IN THE SUBJECT MATTER

OVER CONSIDERABLE TIME. PERSPECTIVES CANNOT BE MEANINGFULLY
DEVELOPED BY THROWING TOGETHER A SET OF TECHNOLOGISTS,
ECONOMISTS, AND SYSTEMS ANALYSTS FOR A FAST-PACED, DATA-
GORGED, AND MODEL-DRIVEN ANALYSIS DONE IN A PRESSURE COOKER.
The best TAs are characterized by a true immersion of the TA team in the problem. One
has to live with the issues for some time. A way to do this is to be an insider, observing
activities from day to day, identifying actors, and isolating false issues. The insider
position is more significant for the 0 and P perspectives than for the T perspective.
With the advantages comes the disadvantage of greater political sensitivity, a subject
which will be considered below. Another argument for time is the desirability of
iteration.
STRIVE FOR A BALANCE AMONG THE T, 0, AND P PERSPECTIVES.

It is not possible to prescribe a priori the relative effort to be allocated to T, 0, and P
perspectives. In the absence of any other information, we would recommend a partition
into equal parts, i.e., one-third of the total person-months on each type. It may be a
crude rule, but it is almost certainly better than the 90% T perspective approach so
frequently seen. With experience, the practitioner will develop a sense for the appro-
priate balance in any given case. A newly emerging technology (e.g., recombinant
DNA) may call for more emphasis on the T and 0 perspectives, while a current
technology is likely to produce richer 0 and P perspectives (as, for example, guayule).
AS A USEFUL FIRST STEP IN A TA, DEVELOP AN ISSUE PAPER GIVING A

FIRST APPROXIMATION TO THE CRITICAL ISSUES AND IDENTIFYING KEY
ACTORS AND PERSPECTIVES.
A good way to minimize imbalance in the assessment is to begin the project with a first
cut at the issues. This prevents moving far along the path which later turns out to be
tangential. It also helps to identify the key actors and perspectives early. The multiple
perspectives aid in limiting the number of issues and preventing an unmanageable
collection. The perception of issues may still change later, of course.
ONCE THE ISSUES INDICATE THE KEY TECHNOLOGICAL, SOCIAL, AND IN-
DIVIDUAL ELEMENTS, SELECT THE PERSPECTIVES. THERE WILL PROB-
ABLY BE SEVERAL 0 AND P PERSPECTIVES SIGNIFICANT FOR POLICY FOR-
MULATION AND THE DECISION PROCESS. RECOGNIZE THAT YOU WILL
PROBABLY NOT CAPTURE ALL OF THEM. REMEMBER THAT CROSS-CUING
AMONG PERSPECTIVES CAN PROVE VERY VALUABLE.
The practitioner will find that, as he moves from T to 0 to P, he follows a path not
dissimilar to that of Freud, penetrating from the professional to the deeper “political”
to the still deeper “persona” level. It is a difficult challenge, but the transdisciplinary
mix should help. In an emerging technology, the P perspective can bring out differences
and similarities in beliefs among technologists about the paths of development and
consequences. As the technology matures, the emphasis of the P perspective would
shift to an understanding of the values of a greater circle of participants. An important
aspect of the value of multiple perspectives is their cross-cuing interaction [Figure 93.
DO NOT CONFUSE THE WHAT WITH THE HOW.
Recall the discussion in Section III, which differentiated how we are looking and whar
we are looking at. Looking at an organization does not imply an 0 perspective. We may
perceive any element from any perspective. The how and what could in some situations
be interpreted as strategy and concern or as means and ends, respectively.
UNDERSTAND THE QUASICONTINUOUS RANGE OF PERSPECTIVES FROM
THE PERSONAL (P) TO THE LARGE FORMAL ORGANIZATION (0) AND DO
NOT BE PUT OFF BY QUESTIONS OF CATEGORIZATION.
The important point is to select a relevant and representative set of perspectives in the
spectrum shown in Figure 6b. Seek perspectives which are likely to be a) significant in
terms of policy formulation and decision process and b) honest representations.
RECOGNIZE THE NONUNIQUENESS OF THE 0 AND P PERSPECTIVES.
An organization may have one perspective of a problem for external use and another for
internal use. Perspectives are dynamic and often shift over time. One individual’s
perspective may become that of an organization later. It is important to keep in mind at
all times that the paradigms basic to the T perspective are not suitable to 0 and P. The
replicability which we take for granted with T is a case in point. A jury trial is not
replicable, even if there is a retrial. Nor are significant, nonroutine executive decisions
replicable. Similarly, with our 0 and P perspectives we rarely have the luxury of
replicability. Use cross-checks where possible, but remember that different perspec-
tives may well yield contradictory images.
RECOGNIZE THE PERSISTENT DILEMMA OF LONG-RANGE ANALYSES.
USE ALL THREE PERSPECTIVES (T, 0, AND P) AS FAR OUT IN TIME AS
THEY JOINTLY REACH.
KEEP REVISITING THE TECHNOLOGY ASSESSMENT, AND NEVER CON-

SIDER IT COMPLETE. REVISE AND REFOCUS ALL PERSPECTIVES AT
INTERVALS.
KEEP EYE AND EAR OPEN TO LOCATE INDIVIDUALS WITH STRONG 0
AND P PERSPECTIVE ORIENTATIONS AND A LONGER-THAN-AVERAGE
TIME HORIZON.
In section III, we agonized over the difficulty in dealing with long-range impacts and
policy questions. The different horizons associated with the three perspectives (Figure
3) laid bare the dilemma. We cannot escape the fact that technology assessments are
difficult to perform well to the degree they must evaluate impacts and policies at distant
target dates. The T perspective then has trouble with core assumptions; the 0 perspec-
tive is accustomed to disregard long-range planning (Table 4); and the P perspective has
difficulty envisoning a future which involves changes in many variables at the same
time. l4 The result in a technology assessment is usually either a brave, quixotic effort to
rely on a long-range T perspective which falls very wide of the mark, or a shift of
attention from distant impacts to closer-in aspects such as implementation strategies and
paths. It is here that the TA has the strongest links to policy making and that the 0 and P
perspectives provide the most potent support. In seeking individuals compatible with
the third part of the above guideline statement, one should not confuse them with
faddists: today’s upper middle class “cause” may have little to do with tomorrow’s key
issues.
USE SKILLED INTERVIEWERS TO DEVELOP THE 0 AND P PERSPECTIVES.
YOU CANNOT RELY ON CONVENTIONAL DOCUMENTATION.
We would like the subject to speak for himself or herself without our providing some
clue as to what we want to hear. We want to understand and see things from the other
person’s point of view. Ethnographers and other skilled field investigators are quite
sensitive to this problem and have taken steps to reduce, control, or at least identify the
sources of bias. Regardless of whether we conduct several independent assessments,
corroborate our findings with participant observation, or do repetitive intensive inter-
viewing with the same informant or extensive interviewing with a variety of partici-
pants, there is bound to be some mixing of the observer and the observed. The assessor
at least has to be clear about which perspective he or she is using in the field, in the
subsequent analysis, and in the writeup. A superior interviewer has the ability to catch a
mood and tone, to hear the unsaid.
Care must be taken to preserve anonymity (i.e., no attribution) or other constraints
imposed by the interviewee, and ethical questions may arise to confound the inter-
viewer. He or she may be subject to manipulation, being “used” to plant ideas or
becoming brainwashed, slowly turning into an advocate or apologist.
To design a field strategy for using multiple perspectives, the first question to ask
is: Who cares about this technology? Individuals and organizations who care and are
concerned about a technology are more likely to come into the political arena than those
who are indifferent. In the early stages of the investigation, the assessor does not have
to worry about whether that concern is supportive or hostile, only that it is a likely
indicator of future activity and behavior.
‘*Most individuals picture readily only one or two changes.

THE MULTIPLE PERSPECTIVE CONCEm 323
The assessor does not start trying to create a scientific random sample but more of
a snowball sample, going from one key informant to another on the basis of leads. For
future technology, the assessor is likely to go to the scientists or the designers of the
technology and then proceed from producers to likely policy makers, consumers, and
other groups who may act on, or be affected by, the technology. With an existing or
near-term technology, it is possible to start with any concerned informant, and the more
knowledgeable the better. Within the resource constraints of the project, it is preferable
to be expansive in choosing people to interview or otherwise observe. Some people
from analogous fields may have useful perspectives, and others may represent views
which will become relevant in the future.
ADAPT THE MEDIUM TO THE MESSAGE, I.E., FIT THE MODE OF COMMUNI-
CATION TO THE PERSPECTIVE.
Different perspectives invite different modes of communication. For communicating
the 0 perspective, consider the use of excerpts or summaries of interviews, oral brief-
ings, scenarios, and vignettes together with the conventional type of report. For the P
perspective, all of the above, with the exception of the conventional report, merit
consideration.
A talented communicator can also use the P perspective to get all of the perspec-
tives (T, 0, and P) across to a wide audience. However, only rarely is this a practical
option; outstanding writers are unavailable and use of media is constrained by the
project budget.
RECOGNIZE THE POLITICAL SENSITIVITY OF THE 0 AND P PERSPECTIVES.
CAREFULLY EVALUATE HOW TO COMMUNICATE INSIGHTS IN AN INEVI-
TABLY POLITICAL SETTING.
Only the T perspective has the luxury of being neutral and politically harmless, but the
price is high: it may also prove rather useless. The 0 and P perspectives reflect the
realities of human beings, encompassing both the “original sins” of greed, crime, and
war, and the “divine touch” of creativity, leadership, and concerted action pro bono
publico. Political sensitivity may not arise as a problem in purely technological activi-
ties such as systems design. But it is inescapable in technology assessment, strategic
planning, and decision analysis. There are no simple answers; each case has unique
aspects. Some alternatives which a multiple perspective team should consider are listed
in Section III.
IN MOST CASES, LEAVE THE INTEGRATION OF THE PERSPECTIVES TO THE
USER OR DECISION MAKER, BUT DO POINT OUT CROSS-CUING LINKS
AMONG THEM.
Since a client or decision maker may integrate the perspectives in a very personal,
unique way, it is usually preferable to provide the analysis output to him or her without
trying to prejudge the integration process.
The assessor should usually maintain the integrity of each of the multiple perspec-
tives in the initial analysis and writing stages, producing T, 0, and P documents. In this
way, it is possible for insights to emerge which originate from each of the perspectives.
However, cross-cuing among the perspectives is of potential significance, and the
assessor should examine and exhibit such linkages. Finally, the use of multiple perspec-
tives makes communication with policy makers easier because the assessor soon ap-
preciates the orientation and language of various policy makers. In a sense, multiple
perspectives enhance our knowledge of the audience and the kinds of communication
which are appropriate to the specific situation.
REMEMBER THAT MULTIPLE PERSPECTIVES CAN ADD INSIGHTS TO DECI-
SION ANALYSES FOR A WIDE SPECTRUM OF SOCIOTECHNOLOGICAL SYS-
TEMS (NOT ONLY TECHNOLOGY ASSESSMENTS).
The illustrations in Section IV cover not only technology assessments, but military and
corporate planning as well as health care delivery.
References
1. Allison, G., Esser~cr ofDecision: E.xp/aining the Cuban Missile Crists, Boston, Mass.: Little, Brown, &
Company, 1971.
2. Amrine, M., The Great Decision, New York: G.P. Putnam’s, 1959.
3. Andersen, D. F., Muthematicul Models ond Decision Making iu Bureaucracies: A Case Story Told from
Three Points of View, Ph.D. Dissertation. MIT, August 1977.
4. Archer, D., Howl to Expand Your S/Q (Sociul Intelligence Quotient), New York: M. Evans, 1980.
5. Archer, D., quoted in Time September 8, 1980, p. 101.
6. Ascher, W., Forecasting: At1 Approisol fir Poliq-Mnker cmd P/urmrr.s. Baltimore: The Johns Hopkins
University Press, 1978.
6a.Braude1, F., The Mediterranean and the Mediterranean World in the Age of Philip II, translated from the
French, New York: Harper & Row Publishers, Vol. I, 1972.
7. Churchman, C. W., Challert~r ro Retrsorl. New York: McGraw Hill, 1968, p. 86.
8. Churchman, C. W., The Design ofInquiring Systems, New York: Basic Books, 1971.
9. Churchman, C. W., A Philosophy for Planning, in Linstone. H. A., and Simmonds, W. H. C. (eds.),
Furures Research: New Directions, Reading, Mass.: Addison-Wesley Publishing Company, 1977.
IO. Coates, J., Technology Assessment-A Tool Kit, Chemtech. June 1976, 372-383.
11. Colton, K. W., and Kraemer, K. L., Computers and Bunking, New York: Plenum Press, 1980.
12. Cyert, R. M., and March, J. G., A Behavioral Theory ofthe Firm, Englewood Cliffs, N.J.: Prentice-Hall,
1963.
13. Drucker, P., New Technology: Predicting Its Impact is Perilous and Futile, New York Times April 8, 1973,
Sec. 3, p. 1.
14. Editorial, Washingron Post Oct. 3 I, 1979.
15. Fallows,J., National Defense, New York: Random House, 1981.
16. Florman, S. C., The Existenticrl Plrrrsrrres qf’Errgirzerriryg, New York: St. Martins Press. 1976.
17. van Foerster, H., Responsibilities of competence, Jo~~T~IN/ ~~fC~/~er~tetrct, 2 (2): I, 1972.
18. van Foerster, H., The curious behavior of complex systems: lessons from biology. In H. A. Linstone and
W. H. C. Simmonds (eds.), Futures Research: New Directions, Reading, Mass.: Addison-Wesley Publiah-
ing Company, 1977, pp. 107-108.
19. Foster, K., et al., A Sociote~hrlo/ogi~u/ Sur\,ey of Guoyule Rubber Cornmer~icrlizotion. Office of Arid Lands
Studies, University of Arizona, Tucson, Arizona, and Midwest Research Institute, Kansas City, Missouri,
April 1980.
20. Hadamard, J., The Psychology qf/n\,ention in the Mathematical Fie/d. Princeton, N.J.: Princeton Univer-
sity Press, 1945.
21. Hetman, F., .Socie+ und the Assessment of‘ Technology. Paris: Organization for Economic Cooperation and
Development, 1973.
22. Hollmg, C. S., The curious behavior of complex systems: lessons from ecology. In H. A. Linstone and
W. H. C. Simmonds (eds.), Futures Resetrrch: New Directions, Reading, Mass.: Addison-Wesley Publish-
ing Company, 1977, p, 129.
23. Hoes, I. R., Societal aspects of technology assessment, Technological Forecosting and Social Change 13,
1979.
24. Howard, R., et al., An Application of Decision Analysis to Technology Assessment ofDecentralized Waste
Treorment Methods, Engineering Economic Systems Dept., Stanford University, April 1981.
25. Jay, A., Management and Machiu~,e//i. New York: Holt, Rinehart, and Winston, 1968.
26. Kaje, R., Bringing the Feminine Into Forecasting, Linstone, H. A.. and Simmonds, W. H. C. (eds.),
Futures Research: New Directions, Reading, Mass.: Addison-Wesley Publishing Company, 1977.
27. Kellen, K., Is Inflation Fun? Technological Forecasting and Social Change 17, 1980, l-5.
28. Kissinger, H., The Whire House Years, excerpted in Time October 8, 1979.
29. Larkin, J., McDermott, J., Simon, D. P., and Simon, H. A., Expert and Novice Performance in Solving
Physics Problems, Science, 208, 1980, 1335-1342.
30. Lawless, E. W., Technology and Social Shock, New Brunswick, N.J.: Rutgers University Press, 1977.
31. Linstone, H. A., and Simmonds, W. H. C., (eds.), Futures Research: New Directions, Reading, Mass.:
Addison-Wesley Publishing Company, 1977.
32. Linstone, H. A., et al., The Use of Structural Modeling for Technology Assessment, 2 volumes, Futures
Research Institute, Portland State University Report 78-1, 1978.
33. Linstone, H. A., et al., The use of structural modeling for technology assessment, Technological Forecast-
ing and Social Change 14: 291-328, 1979.
34. Linstone, H. A., et al., The Mulfiple Perspective Concept, Futures Research Institute, Portland State
University Report 81-1, 1981.
35. Little, A. D., Inc., The Consequences of Electronic Funds Transfer: A Technology Assessment of Move-
ment Toward a Less Cash/Less Check Society, prepared for the National Science Foundation, GPO,
Washington, D.C., 1975.
36. Machiavelli, N., History of Florence, 2, London: 1851.
37. March, J. G., and Simon, H., Organizations, New York: Wiley, 1958.
38. McNaugher, T. L., Marksmanship, McNamara, and the M-16 Rifle: Innovation in Military Organizations,
Public Policy 28 (1): Winter 1980.
39. Meadows, D., et al., The Limits fo Growth, New York: Universe Books, 1972.
40. Miser, H. J., Operations research and systems analysis, Science, July 4, 1980.
41. Mitroff, I. I., and Blankenship, L. V., On the Methodology of the Holistic Experiment: An Approach to
the Conceptualization of Large-Scale Social Experiments, Technological Forecasring and Social Change 4,
1972, 339-353.
42. Mitroff, I. I., The Subjective Side of Science, New York: Elsevier-North Holland, 1974, p. 248.
43. Morison, E. E., Men, Machines, andModern Times, Cambridge, Mass.: The MIT Press, 1966.
44. National Commission on Electronic Funds Transfer, EFI and the Public Interest, Washington, D.C., 1975.
45. Peters, C., How Washington Really Works, Reading, Mass.: Addison-Wesley Publishing Company, 1980.
46. Porter, A. L., et al., A Guidebook for Technology Assessment and Impact Analysis, New York: North
Holland, 1980.
47. Project Mirage 85, Vol. III, DPR-87, Corporate Development Planning Dept., Lockheed Aircraft Corpora-
tion, Burbank, Calif., April 1970.
48. Rowan, R., Those Business Hunches Are More Than Blind Faith, Fortune April 23, 1979, p 112.
49. Schoenberger, W. S., Decision of Destiny, Athens, Ohio: Ohio University Press, 1970.
50. Schorske, C. E., Fin-de-Siecle Vienna: Politics and Culture, New York: Vintage Books, 1981.
51. Science, October 24, 1980, p. 410.
52. Siu, R. G. H., Management and the Art of Chinese Baseball, Sloan Managemenr Review, 19 (3), MIT,
Spring 1978.
53. Steinbruner, J. D., The Cybernetic Theory of Decision, Princeton, N.J.: Princeton University Press, 1974.
53a.Thompson, W. I., At the Edge qfHistory. New York: Harper & Row, 197 1,
54. Toth, R., Los Angeles Times December 30, 1980, p. 1.
55. Time December 8, 1980, p. 101.
56. Umbdenstock, L., The Perinatal Regionalization Project: A Study in Form and Development, Systems
Science Ph.D. Dissertation, Portland State University, Portland, Oregon, 1981.
57. Ways, M., The Road to 1977, Fortune January 1967, p. 4.
58. Yankelovich, D., quoted in Smith, A. (pseudonym), Supermoney, New York: Random House, 1972, pp.
271-272.

Harold A Linstone Multiple Perspective Theory

Uploaded by

Copyright:

Available Formats

Harold A Linstone Multiple Perspective Theory

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Harold A Linstone Multiple Perspective Theory

Uploaded by

Copyright:

Available Formats

TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE 20, 275-325 (1981)

The Multiple Perspective Concept

HAROLD A. LINSTONE ET AL. 1

@ H.A. Linstone, 1981

Linstone’s twenty years of systems analysis activity-in industry (corporate plan-

Decision making inherently involves organizations and individuals. Their perspec-

l Ill-structured nature of the problem (typically sociotechnological systems);

Section II. Our Usual Perspective and Its Limitations

In a technology assessment, as in other technical studies, systems analysis tools are

Richard M. Cyert, co-author of A Rrhu~~iuml Theor! cfrhr Firm, and President of

Wall Street Journal, April 7. 1980

THE PROBLEM-SOLUTION VIEW

Fig. 1. The technical perspective.

THE BEST SOLUTION SEARCH

If a system is complex, we simplify it by dividing it into subsystems. If we still cannot

DATA AND MODELS

1. Lockean. Empirical; agreement on observations of data; truth is experiential and

4. Hegelian. Dialectic confrontation between opposing models or plans leading to

If we concentrate on the individual, we should also mention Kant’s noumena:

AVOIDING THE INDIVIDUAL

Churchman, in his inimitable way, makes the following observations:

THE LINEAR PERCEPTION OF TIME

Jung type Time focus Discounting

Sensation Present Highest

Section III. Multiple Perspectives

DISCOUNT RATE 0 DISCOUNT RATE DISCOUNT RATE

CI la IQD Iz) 47----e a0 IW II

1ong hor i zon

Problem as seen by unitary Parochial priorities Goal5 include personal interests

Central coordination and control

Administrative feasibility a major Looking sideways--commitment

Steinhruner and Andersen

Mitroff and Blankenship

1. At least two ‘radically distinct’ disciplines of knowledge must be brought to bear;

OUR PROPOSED PERSPECTIVES

What Are We Viewing?

a) Technology. The starting point in TA frequently is a specific technology, e.g.,

b) Physical Environmental Setting. The technical aspect of the environment of a

c) Sociotechnological Setting. Technologies interact with organizations and groups

Fig. 4. What we are viewing in a TA.

d) Technopersonul Setting. Technology affects the individual and is affected by the

f) Individual Actors. An individual in a crowd or an organization may be quite

g) Political Action. This element represents the well-recognized interplay between

How Are We Viewing?

There is much interest in organization and categorization of information, in preparing lists

The Organizational Perspective. To date, the organizational perspective represents a

Machiavelli recognizes the difference between an 0 and a P perspective. He sees

l Idenrifiing the pressures in support of und in opposition to the technology:

J. Fetzer, Chairman of Fetzer Broadcasting Company:

R. Siu, Management Consultant:

l Understanding of the total decision process is enhanced through knowledge of

Political activity involves the interactions of organizations and individuals. Usually,

l The P perspective serves as a precursor to better understanding of the 0

a. only in rare cases does the individual make a difference;

We consider each point in turn.

generally, on historical processes), the criterion of “reasonableness of conjecture” should

l Communication of complex problems and issues is easier through the personal

T + 0 + p: A SINGERIAN INQUIRING SYSTEM

Technical (T) Organizational (0) Personal (P)

Weltanschauung Science-technology Social infrastructure IndividuatioMhe self