Nicholas Rescher - Collected Papers - Volume 6 Studies in Social Philosophy-De Gruyter (2006)

NICHOLAS RESCHER COLLECTED PAPERS
Volume VI
Nicholas Rescher
Studies in Social Philosophy
ontos
verlag
Frankfurt I Paris I Ebikon I Lancaster I New Brunswick
Bibliographic information published by Die Deutsche Bibliothek
Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliographie;
detailed bibliographic data is available in the Internet at http://dnb.ddb.de
North and South America by

Transaction Books
Rutgers University
Piscataway, NJ 08854-8042
trans@transactionpub.com
United Kingdom, Ire Iceland, Turkey, Malta, Portugal by

Gazelle Books Services Limited
White Cross Mills
Hightown
LANCASTER, LA1 4XS
sales@gazellebooks.co.uk
Livraison pour la France et la Belgique:

Librairie Philosophique J.Vrin
6, place de la Sorbonne ; F-75005 PARIS
Tel. +33 (0)1 43 54 03 47 ; Fax +33 (0)1 43 54 48 18
www.vrin.fr

2006 ontos verlag
P.O. Box 15 41, D-63133 Heusenstamm
www.ontosverlag.com
ISBN 3-3-938793-01-5
2006
No part of this book may be reproduced, stored in retrieval systems or transmitted

in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise
without written permission from the Publisher, with the exception of any material supplied specifically for the
purpose of being entered and executed on a computer system, for exclusive use of the purchaser of the work
Printed on acid-free paper

This hardcover binding meets the International Library standard
Printed in Germany
by buch bücher dd ag
Studies in Social Philosophy
PREFACE
Chapter 1: TECHNOLOGY, COMPLEXITY,

AND SOCIAL DECISION 1
Chapter 2: RISKING DEMOCRACY 17
Chapter 3: IS CONSENSUS REQUIRED FOR A RATIONAL

SOCIAL ORDER? 31
Chapter 4: MORALITY IN GOVERNMENT AND POLITICS 45
Chapter 5: ON THE RATIONALE OF GOVERNMENTAL

REGULATION 53
Chapter 6: SOCIAL WELFARE:

SOME PHILOSOPHICAL ISSUES 63
Chapter 7: COLLECTIVE RESPONSIBILITY 77
Chapter 8: THE BELL CURVE REVISITED 93
Chapter 9: IN THE LINE OF DUTY 105
Chapter 10: TECHNOLOGICAL PROGRESS AND HUMAN

HAPPINESS 117
Chapter 11: THE SOCIAL VALUE OF A LIFE 131
Chapter 12: THE ALLOCATION OF EXOTIC

MEDICAL LIFESAVING THERAPY 143
Chapter 13: ETHICAL ISSUES REGARDING

THE DELIVERY OF HEALTH-CARE SERVICES 159
Chapter 14: MORAL ISSUES RELATING TO
THE ECONOMICS OF NEW KNOWLEDGE
IN THE BIOMEDICAL SCIENCES 171
Index 181
PREFACE
M y interest in issues of social philosophy, now dating back over forty

years, has resulted in four previous books: Distributive Justice (New
York: Bobbs Merrill, 1966), Welfare: the Social Issues in Philosophical
Perspective (Pittsburgh: University of Pittsburgh Press, 1972), Public Con-
cerns (Lanham, MI: Rowman & Littlefield, 1996), and Fairness (New
Brunswick, NJ: Transaction Publishers, 2002). Additionally, however, I
have during this time also written various particular essays on particular
problems and issues of this domain—usually in response to the needs of
some special occasion. The aim of the present volume is to collect this ma-
terial together in coordinative conjunction. The resultant book will, so I
hope, not only offer a panorama of my views on some of the key issues of
the field, but also convey a sense of the procedural ways and means by
which I think that philosophical deliberations can serve to shed some in-
structive light on such ever-controversial matters. For I am convinced in
theory and have sought to illustrate in practice that constructive thinking
on social issues calls for implementing a quantitative approach from a
moral perspective, and that neither the measurable quantities not the intan-
gible values of the situation can be overlooked in a cogent assessment of
the issues. It is my hope that these essays will confirm the justice of this
conviction.
Nicholas Rescher
Pittsburgh PA
January 2006
Chapter 1
TECHNOLOGY, COMPLEXITY, AND

SOCIAL DECISION
TECHNOLOGICAL PROGRESS MAKES LIFE MORE

COMPLICATED
W e members of Homo sapiens are amphibious creatures: we inhabit

two realms, the realm of nature and the realm of human artifice. The
former domain is one where we find matters in place; the latter is a con-
struct that we ourselves produce under the guidance of intelligence. But on
both sides alike we encounter an unfathomable complexity. Nature has lev-
els of depth exceeding the reach of our cognitive powers, and human arti-
fice also carries us ever further down the road of complexification.
Biological evolution under the aegis of natural selection makes for ever
more specialized speciation, differentially selecting for organisms that
branch off into particular varieties able to respond more effectively to the
challenges of changing environmental conditions. This process makes for
increasingly complex organisms—just as Herbert Spencer’s Law of Evolu-
tion would have it. And technological evolution exhibits exactly the same
tendency. In virtually every sphere of our human concerns we constantly
encounter new obstacles. And new problems call for doing what we were
unable to do before; they require new solutions which themselves call for
new methods, new processes, new instrumentalities. Such escalating de-
mands ongoing change of process. We are driven to devising even more
complex systems to put such increasing performative sophistication at our
disposal. We here encounter the fundamental Law of Technical Progress:
Human artifice is caught up in a complexity tropism. And the rationale of
this situation is easy to understand on principles of economy of effort, see-
ing that rational creatures are naturally inclined to try simpler solutions
first, maintaining them until such time as it becomes advantageous to re-
place them by something more complicated. In consequence, all of our
creative efforts—in material, social, and intellectual contexts alike—
Nicholas Rescher • Collected Papers VI
manifest an historical tendency of moving from the simpler to the more

complex.
People incline to think that technological progress makes life easier. It
is faster and more convenient to cross oceans by plane than by sailing ship,
to phone messages rather than mail letters, to type with word processors
than write with quills. But while all this is true enough, there is the other
side of the coin as well. In such cases, enhanced performance is acquired at
the cost of elongated training. Technological progress constantly destabi-
lizes the status quo and brings changed processes, procedures and products
in its wake. Individual actions are generally simplified at the cost of com-
plicating larger processes. For not only does the gain in efficiency that
technical progress engenders provide us with more time to do things, but it
also vastly increases the range of things that can be done and the power of
the means for their realization.
The fact is that technological progress makes life vastly more compli-
cated by widening the range of choice and opportunity. It increases the op-
erational complexity of process all about us. The driver of a horse and
buggy can afford to doze off, but the driver of a car can afford it far less,
and the pilot of a supersonic fighter plane not at all. With more sophisti-
cated camera equipment, film makers need to make many more detailed
decisions about location, perspective, lighting, timing, etc. in making their
pictures. And if we replace these human decision makers by automatic
control devices, each must be vastly more complex and sophisticated than
its more rudimentary compeers. With the growth of technology, the arti-
facts we devise threaten to become too complicated for effective human
management. Modern superjets are so complex that pilots no longer pos-
sess the cognitive skills to fly them and could not manage to do so without
the aid of computers. The problem of complexity management becomes
shifted from operating the system itself to managing the functioning of its
cybernetic governance. The drug industry is subject to far more extensive
regulations today than the entire economy was fifty years ago. In combat,
the pilot of a jet fighter makes far more decisions in five minutes than a
sailing-era ships naval captain did in a day.1
We pay for the advantages of sophisticated technology by confronting
issues of operational choice and decision that are of ever increasing com-
plexity. As technology advances, the problems of cognitive discrimina-
tion—just like those of visual discrimination—expand exponentially in line
with the proliferation of possibilities.2 And so, while technological pro-
gress—be it material or social—may indeed simplify and facilitate the per-
2
TECHNOLOGY, COMPLEXITY, AND SOCIAL DECISION
formance of particular tasks, its aggregate effect is to make large-scale

processes more complicated and difficult.
Complexity is the inseparable accompaniment of modernity. We en-
counter it throughout our science, throughout our technology, and through-
out our social and cultural environment as well. Perhaps the clearest
manifestation of this is the range of choice that confronts us on all sides in
everyday life—with sources of information, means of entertainment and
leisure activities, occupations, and even lifestyles. An ongoing prolifera-
tion in the multiplicity of cultural forms and the diversity of opportunities
is a striking and salient feature of our age, and it is an unavoidable feature
of what we are pleased to designate as “progress” that it confronts as with
an ever more complex and diversified manifold of possibilities. And the
obverse scale of such opportunity creation is the matter of opportunity cost
inherent in the fact that every opportunity that we seize also represents a
multiplicity of other opportunities forgone.
Progress itself pushes us deeper and deeper into difficulties. For every
solution opens up new problems. Such a Hydra effect is closely related to
the Principle of Question Propagation in matters of factual inquiry. The
phenomenon of the ever-continuing “birth” of new questions was first em-
phasized by Immanuel Kant, who maintained that: “The solution of any
factual (scientific) question gives rise to yet further unsolved questions.”3
This opening up of new and deeper questions in the course of our inquiries
into matters of empirical fact is something that is empirically as well estab-
lished as any encountered within our study of nature herself. The history of
science forcefully substantiates this principle of question-proliferation in
empirical inquiry. And it deserves emphasis that this phenomenon of esca-
lating diversity and complexity holds just as much for practical as for cog-
nitive problems.
But what of a “return to the past”—a nostalgia-satisfying reversion to
the conditions of an earlier, simpler age? The short answer is that this is
simply impracticable outside the limited range of museum-piece environ-
ments in isolated backwaters. For we would not have resorted to those
complications in the first place if we had not been forced to them by the
needs of the situation. After all, even in Adam Smith’s day, so he informs
us in The Wealth of Nations4 some thousand people over all had a hand in
making a woolen coat. In general, the price of a return is unaffordable. To
be sure this is a matter of how things stand in the large and in the whole. It
is possible to preserve islands of simplicity in human affairs. But this is
possible only in exceptional and highly localized conditions. Overall, the
3
later position in the realm of artifice is bound to be one where complexity

is unavoidable and uneliminable.
In his amusingly written but brilliantly perceptive book, C. N. Parkinson
took note of what might be called the managerial bloat of modern organi-
zations.5 In all sorts of enterprises as activity diminished, management
flourished. Throughout the twentieth century, the British navy had ever
more admirals as ever fewer ships were in commission. Everywhere opera-
tions downsized, management upsized.
But of course the explanation of this phenomenon is straightforwardly a
matter of complexity. To address more complex tasks more sophisticated
operations are necessary. But more sophisticated operations require more
elaborate processes and more elaborate processes require more sophisti-
cated control and thus more elaborate management. And so even where the
scale of operation is diminished as the technically more sophisticated proc-
esses grew more sophisticated and powerful, so nevertheless the problems
of planning and control grow ever more elaborate. It is not boondoggling
or diminished capacity that produces managerial bloat but the inexorable
demands of dealing with greater complexity. And so the gigantism of the
managerial machinery of the contemporary army or industrial enterprise or
university is thus no accident. Designed for affording greater control over
more complex systems, the administrative machine of such organizations is
forced to meet even greater challenges in point of performative capability.
As the functional capacity of such institutions expands under the pressure
of technological progress, the scope of management operations expands
also—and even more rapidly. With the modern technology of communica-
tion and information management, bureaucracy is thus bound to increase
irrespective of the particular operational tasks for whose management the
bureaucracy is instituted. And sometimes the functional complexities of a
system make its effective control virtually unrealizable.
PROBLEM COMPLEXITY OUTPACES SOLUTION COMPLEXITY
With the progress of science, technology, and human artifice generally,

complexity is self-potentiating because it engenders complications on the
side of problems that can only be addressed adequately through further
complication on the side of process and procedure. The increase in techni-
cal sophistication confronts us with a dynamic feedback interaction be-
tween problems and solutions that ultimately transforms each successive
solution into a generator of new problems. And these feedback effects op-
4
erate in such a way that to all intents and purposes the growth rate of the
problem domain continually outpaces that of our capacity to produce solu-
tions. Both the problems and the solutions grow more complex in the wake
of technological progress, but the crux of the matter lies in the compara-
tively greater pace of the increase in problem complexity.
Of course, technology is not just part of the problem, it affords part of
the solution as well. For while technological progress always poses new
difficulties in the management of information and the control of operating
procedures, it can of course also help with resolving issues of this sort.
Safety engineering in all its forms—redundancy provision, fault detection
sensors—together with the “cybernetic” automation of control mecha-
nisms, and above all the use of computers in information management and
decision implementation, all afford powerful resources for problem resolu-
tion in technological contexts. Process controls too can be handled in sub-
stantial measure by technological means. Interestingly enough, the
electronics in a contemporary automobile cost some two thousand dollars
more than the steel used to produce the same car.
However, there yet remains the crucial question of comparative pace.
With technological progress, which grows the faster, the manifold of prob-
lems to be resolved or the reach and power of our instrumentalities of
problem resolution? Now here it might seem that complex technology
gives the advantage to problem resolution. After all, do not the cognitive
resources that computers afford us offset the problems raised by increasing
of complexity? Alas, not really.
First of all, it has to be recognized that computers help principally with
information processing and do not equally address the problems informa-
tion acquisition. And in the course of technological progress these become
even more extensive and even more significant. Here the classic dictum
holds good: as far as the efficacy of computational information manipula-
tion is concerned, garbage in, garbage out. Moreover, the fact remains that
computers do just exactly what they are programmed to do. The level of
complexity management they are able to achieve is determined through—
and thus limited by—the levels of ingenuity and conceptual adequacy of
their programming. No central bank places unalloyed confidence in its
economic models. And there is also the problem of unforeseen and unfore-
seeable interactions within the interact fabric of the operating processes.
These “bugs” can result in malfunctions in computer operation even as
they can produce accidents in other sorts of systems. And the more elabo-
rate and complex our programs get—particularly in areas where novelty
5
and innovation are the order of the day—the larger the prospect and
chances for such mishaps.6 In every area, maiden voyages are notoriously
fertile in bringing unanticipated difficulties to light.
The fact is that as technical systems become more complex, their opera-
tion becomes even more so. And over time managerial complexity gener-
ally outpaces processual efficacy. A more elaborate repertoire always
imposes new and increasingly unmanageable difficulties in matters of op-
eration and procedure, since as already noted in the preceding chapter,
process concatenations always grow at a rate faster than the processes
themselves.
Granted, computer automated problem solving is one of the wonders of
the age. Computers fly planes, land rocket modules on the moon, win chess
competitions, develop mathematical proofs. All the same, we have to come
face to face here with what might be called a Hydra effect after the mytho-
logical monster who managed to grow several heads to take the place for
each one that was cut off. The fact is that there is a feedback symbiosis be-
tween problems and solutions which operates in such a way that the growth
of the former systematically outpaces that of the latter. Accordingly, those
sophisticated information and control technologies not so much resolve
problems of complexity as enlarge this domain by engendering complexity
problems of their own. Despite the enormous advantages that they furnish
to intellectual efforts at complexity management, computers nevertheless
do not and can not eliminate but only displace and magnify the difficulties
that we encounter throughout this sphere.
In some ways, humans are better at managing complexity than are com-
puters as we know them. For computers are programmed—their responses
to situations are automatized. Humans, on the other hand, are flexible,
spontaneous, able to innovate or to “wing it”. So when complex systems
malfunction or mis-function, as they are bound on occasion to do, humans
can come to terms such a breakdown by fair means or foul, while the com-
puter’s response would be one of “crashing”. A deficient, aberrant situation
that can conceivably be handled by people would likely as not faze a com-
puter. And the person who malfunctions into error or misunderstanding in
the management of complexity may well be able to recover where a com-
puter would plunge on into an ever deeper morass.
The long and short of it is that complexity management via computers
will not remove the obstacles to managerial effectiveness exactly because
complexity raises problems faster than it provides means for their solution.
Computers—the very instruments that enhance our capacity for complexity
6
management—widen the scope of the field of action and thereby augment

the complexity we face. The technical resources that enlarge our powers in
the area of problem resolution not only do not manage to reduce the over-
all size of the problem field that confronts us, but actually manage to
enlarge it. The fact is that technological progress engenders what might be
characterized as the “rolling snowball effect” because complexity breeds
more complexity through engendering problem-situations from which only
additional technical capacity can manage to extract us. In the natural,
unfettered course of things complexity tends to grow exponentially.
Exponential growth is the automatic result when something grows in such
a way that its size increases in proportion with the size that it has already
attained. And so if the management of complexity (be it cognitive or
operational) is to keep up, then it too must increase exponentially. Our
technological capacity in this regard must grow by exponential leaps and
bounds, passing through successively ever greater stages of capacity. The
historical course of things has certainly seen this sort of technological
escalation in complexity management, passing through successive phases
of management by individuals, by groups, by mechanical devices, by
single electronic computers, by groups or systems of electronic computers,
and possibly is yet more powerful but as yet undreamt of resources.
Eventually, no doubt, there will be an end of the line here, for it seems to
be a law of nature that all exponential growth must ultimately come to an
end. And so in the final reckoning our ability to manage ever greater
complexity will eventually become saturated.
THE INTIMIDATING IMPETUS OF THE UNKNOWN: RISK AND

DESTABILIZATION
The growing complexity that emerges in the wake of technical progress

all too often engenders unexpected difficulties. And with every practical
step that we take towards coping with these difficulties, unforeseeable con-
sequences generally arise. All along the line, the Law of Unintended Ef-
fects comes into operation. As technical sophistication increases, we
penetrate ever further into a domain where we cannot see clearly along the
road that lies ahead. We continually confront problem situations within
which we not only cannot determine optimal solutions, but where even the
identification of desirable solutions becomes problematic and imponder-
able.
7
This phenomenon has ominous implications. Physical chaos, it will be

recalled, occurs when a system functions in so volatile a way that a minute
difference in its initial condition—one that is so small as to lie beneath the
threshold of observation and perhaps even of observability—can make for
a vast difference for the outcome state. Cognitive chaos is exactly the same
sort of thing transposed to the region of information processing. It occurs
whenever a minute variation in input information can produce great differ-
ence in its inferential consequences—that is, whenever an inferential out-
come is enormously sensitive to small variations on informational input
(for example where even a “shadow of doubt” can spell the difference be-
tween guilt and innocence). The prospect of subjecting such inferential
processes to adequate cognitive control by way of systematic principled
understanding are somewhere between small and nonexistent. And it is just
here that complexity makes itself felt. For the prospect of insufficient or
misleading information grows along with system complexity and, other
things equal, sophisticated systems need to have more elaborate processes.
Prediction too becomes less practicable, save at the level of statistical in-
definiteness. The activities of a primitive tribe are easier to predict than
those of the U.S. Congress; very heavy atoms are less stable than very light
ones; strategy is easier to manage with tic-tac-toe than with chess. The
general structure of the situation is reflected in the circumstance that while
simple systems are more versatile and flexible but less effective and effi-
cient; complex systems are more specialized but less adaptable. Screws are
more complex than nails, but thereby more limited. They do a better job at
holding things together, but are less versatile (all sorts of things can be
nailed together while screws require a congenially porous material). Com-
plex systems pay for their greater efficiency and effectiveness through be-
ing more specialized, more closely attuned to the circumstantial specifics
of the case. They are, accordingly, easier to destabilize—more vulnerable
to the interventions of chaos.
As the operations of a goal-directed system of any sort become more
complex, the importance of safety engineering comes to the fore increas-
ingly the need for devising contrivances that protect against potential for
failure that is an inevitable accompaniment of increasing complexity. To
be sure, part of that growing complexity may be invested in safety assur-
ance—in building malfunction controls into the system operations and pro-
liferating back-ups and redundancies. But problems are inevitable in this
domain as well. The prospect of things going wrong will always slip
“through the cracks”, and complex systems for this very reason grow in-
8
creasingly error-prone and susceptible to malfunction. And the seriousness

of such mishaps is not only amplified by the complexity of the environ-
ment but also rendered less manageable by it.
We devise those more complex systems because we need them. But
their very existence creates new difficulties. For we soon come to depend
on them and this dependence renders us increasingly vulnerable to frustra-
tion. Complex systems are inherently less amenable to successful compre-
hension, management, and control. It is evident that where doubling the
number of steps must go right in a goal-pursuing context, we double the
prospects of something’s going wrong, other things being equal. Complex-
ity generally enlarges the prospect of system failure—and where not in
frequency, there in magnitude of effect. A washing machine can malfunc-
tion far more readily than a wash board—and with more extensive conse-
quences. There is no clearer indication of the vulnerability that complexity
engenders than the extent to which a single act of sabotage by a terrorist, or
of dishonesty by a greedy or disgruntled employee, can inflict destructive
damage on a large enterprise or an extensive organization.
This state of things engenders inevitable threats and dangers. Specifi-
cally, with the ongoing sophistication of human artifice we find that:
• with cognitive systems we face the threat of disintegration, disorgani-

zation, and cognitive dissonance;
• with technical systems we face the threat of breakdown and malfunc-

tion;
• with social systems we face the threat of gridlock and stalemate or—
at the opposite extreme—of chaos and anarchy.
Throughout the domain of human creativity—alike in matters of cognitive,

technological, and social engineering—we encounter an increasing com-
plexity that carries with it the inherent risks of system malfunction. Risk is
a natural companion to complexity. It involves the (multiplicatively inter-
active) combination of two factors: the probability of failure and the mag-
nitude of the consequences should failure occur. Increasing complexity
generally enlarges risk. It does so not necessarily by increasing the prob-
ability of failure but by increasing the negative consequences should fail-
ure occur. (Precisely when particularly of failure is diminished we tend to
9
increase our stake on success.) Increasing complexity thus tends to be as-

sociated with increasing risk.7
Our individual activity—say getting from home to work—is undoubt-
edly simplified by the technology of the automobile in contrast with that of
the horse. But the entire system of automotive transport is something vastly
more complex, penetrating pervasively into every aspect of our social and
economic life. Local simplicity rides on the back of globally systemic
complexity. The present-day motor vehicle code of U.S. states is more
complex than the whole of their transport legislation in the 1890s. And no
American who has lived through the months of the Arab Oil Embargo in
the early 1970s can fail to realize the magnitude and power of the vulner-
abilities to which such systemic complexification renders us subject.
The basic point that is at issue here is relatively straightforward. It is
that in general and as a rule complex systems are by reason of their very
complexity more expensive and more difficult to construct, operate, and
maintain. And they are also more risky to use—not necessarily by way of
an increased likelihood of malfunction but rather through the increased se-
riousness of the consequences that ensue when a malfunction occurs. In
sum, the increased complexity of our systems does not come cost-free; to
achieve its undoubted advantages we have to pay a substantial price, not in
terms of money alone, but also in terms of risk.
IMPLICATIONS OF TECHNOLOGICAL SOPHISTICATIONS
At the earlier, less sophisticated stages of technological progress, it was

easier for people to understand the implications of change. For when
changes occur in highly complex systems, the consequences are often un-
predictable. It becomes somewhere between difficult and impossible to say
in advance just what the result of modifications and innovations will be.
We all too frequently cannot see our way clearly through the accompany-
ing ramifications to grasp the implications of innovation for their manage-
ment. For effective decision making requires the timely processing of full
information. And both of these factors—readily acquiring and speedily
processing information—tend to be more difficult and cumbersome in con-
texts of complexity. And the more complex the ramifications of a choice
situation are, the more likely it is that any particular way of resolving it
will be worrisome to some of the individuals or groups who have a stake in
its outcome. Few of us have the engineering sophistication to be rocket
scientists. But, effectively none of us today have the technical sophistica-
10
tion to be social engineers. And there is a real possibility that even all of us
taken collectively do not have what it takes.
In particular, the measures that are required to cope with matters of so-
cial and economic policy become increasingly difficult and constantly
more expensive to implement and to operate. The complexities that have to
be taken account of outrun the grasp of ordinary understanding. (Think, for
example, of Hillary Clinton’s health care program, not to speak of the Fed-
eral Tax Code.) The management of America’s systems in the area medical
or social or economic processes and programs has grown so difficult
throughout the successive decades of the present century that the political
system is nowadays close to throwing up its hands in frustration.
And so the very power of technical progress brings new disabilities and
incapacities in its wake. The operational dynamics of complexity expan-
sion means that as we increase our problem-resolving capacity we will in-
evitably—preferences to the contrary notwithstanding—also increasingly
loose our grip on the overall effectiveness of problem control. All too of-
ten, no one can form an accurate picture of how proposed changes of proc-
ess and procedure in the management of complexity will work themselves
out for those involved. And even where we ourselves think that we can see
the way clear, there may well be precious few others who agree with us.
This state of affairs manifests itself in the striking and by now familiar
phenomenon that might be called the cacophony of experts. The difficulties
of rational problem solution in complex situations engenders a variety of
plausible but competingly alternative possibilities. And in the absence of a
single clean-cut resolution a dissonance of theories arise. Pundits come
upon the scene with their competing wares—each with a case that seems
plausible and persuasive but is nevertheless insufficiently clear-cut and de-
cisive to put its rivals out of business.
The destabilizing effect of technological change thus paves the way to
social discord and procedural impotence. Here sheer stagnation is the natu-
ral result of risk-aversive “better the devil we know” thinking. All too of-
ten, life in our imperfect world proceeds in such a way that to all
appearances certain abstractly desirable aims simply cannot be concretely
realized by acceptable means. Let us inspect the ramifications of this proc-
ess a bit more closely.
In an environment of increasing technological complexity we must de-
velop ever more sophisticated control processes to address new problems.
Such changes affect different people, different groups, different constitu-
encies differently. And just here the eventual effects of the measures we
11
take to address the challenges become lost in a fog of impredictability. For

every winner there are some losers and various others who—not being able
to see the way clear—come to feel threatened. When people confront more
complex problems they find it difficult and sometimes impossible, to think
their way through to satisfactory solutions. Everyone feels put to risk by
some aspect of the ill understood consequence flow of potential innova-
tion. And in the face of this perplexity, people become fearful lest any step
away from the status quo—unhappy though the existing state of things
may be—will plunge them into disaster. This sort of situation is an open
invitation to gridlock.
A RETROSPECTIVE REFLECTION
The present deliberations have revolved around four significantly prob-

lematic issues: complexity escalation; obscurity of consequences and cog-
nitive chaos; concretization quandaries; decision gridlock and
immobilization. This condition of affairs obviously has substantial and sig-
nificant implications for the general polity of social decision. In particular,
it brings to the fore the question of how to proceed sensibly in the face of a
technological progress that impels us into a sphere where the problem-field
that confronts us may come to outrun the power of our cognitive capacities
for problem resolution. For just this is the situation with which the judg-
mental principles of the case confronts us.
The instruments we forge for the solution of our problems—intellectual
and physical—all prove effective only up to a point. And as new problems
arise they will require new resources, new methods and devices. But, in the
natural evolution of things the creation of ever new and more powerful in-
struments becomes increasingly difficult. And there is every reason to
think that eventually—in the long run of things—the hurdle will be raised
to a height that we simply cannot leap. Our standard resource of problem-
solving—the use of process-modeling intelligence—may come to prove
unavailing. A point may—and foreseeably will—be reached where the fa-
miliar role of modeling and calculation will no longer prove adequate as a
problem resolving instrumentality.
Earlier in the present century, the Austrian school of economists argued
that the domain of economic phenomena is inherently of so intricate and
sophisticated a nature that such theorizing as it is practicable for us to
manage really cannot provide an adequate grasp on the phenomena—and
certainly not one powerful enough to guide our interventions in establish-
12
ing effective control. Human phenomena in the domain of social affairs are
inherently so complex, volatile, and variegated that the project of capturing
them within the confining boundaries of universal laws is unrealizable. The
complexity of the system of social processes at work in the operations of
an advanced modern economy is such that there simply is no way for us to
calculate the behavior of the system. No model that we devise will be ade-
quate to handling the requisite details. As the Austrian theorists saw it, the
very idea of a predictively adequate social science of human behavior is a
pipe dream.8
We are brought back to the “bounded rationality” that arises in decision
contexts when the complexity of a problem-situation substantially exceeds
the reasoning powers of the problem-solver. And here, often as not, the
best strategy is to let “matters run their course” and use the observation of
its processes as a guide for the formation of our policies and programs.
Where calculation based on theory is impracticable, the best we can usu-
ally do is to keep an eye on the broad tendencies of the case and let the
course of experience be our guide in responding to them.
And a strong case can be made for saying that this sort of situation
holds not just in economics but far more broadly throughout the social do-
main. For here too there are, fortunately, other ways of solving the com-
plex problems posed by modern societies than by “figuring it out” through
human artifice and calculation. One of these is to leave the solution of the
problem to “the course of events” (either in nature or in a simulation-model
of some sort), and then simply sit back and watch what happens. In cir-
cumstances that are incalculable for us because of excess complications, or
where the requisite data cannot be had on a sufficiently timely basis, such a
recourse to the practice of “watchful waiting”, of simply seeing how the
matters work themselves out when left to their own devices, is a variant
and sometimes highly useful cognitive resource. The fact is that in situa-
tions of unmanageable complexity, practice in matters of public policy is
often guided more effectively by common-stressed appraisal based on lo-
calized experimental trial-and-error than by the large-scale theorizing of an
intellectual technology unable to cope with the intricacy of interaction
feedbacks and unpredictable effects.9
13
NOTES
1
On complexity in relation to social and political issues see H. R. Kohl, The Age of
Complexity (New York: New American Library, 1965).
2
For a good survey of the issues see Klaus Mainzer, Thinking in Complexity: The
Complex Dynamics of Matter, Mind, and Mankind (Berlin, etc.: Springer Verlag,
1994).
3
Immanuel Kant, Prolegomena to any Future Metaphysic (1783), sect. 57. On the
relevant issues also see the author’s Scientific Progress (Oxford: Blackwell, 1978).
4
Adam Smith, Inquiry into the Nature and Cause of the Wealth of Nations (Dublin:
Whitestone, 1776).
5
C. Northcote Parkinson, Parkinson’s Law (Boston: Houghton Mifflin, 1957).
6
The dramatic recent failure of the U. S. Department of Internal Revenue’s effort to
computerize its operations is one particularly vivid example of this.
7
For a variety of vivid illustrations of the inherent vulnerability to failure of com-
plex systems see Charles Perrow, Normal Accidents Living with High-Risk Tech-
nologies (New York: Basic Books, 1984).
8
See K. R. Popper, Conjectures and Refutations: The Growth of Scientific Knowl-
edge (New York: Harper Torchbooks, 1965); F. A. Hayek, “The Theory of Com-
plex Phenomena” in Studies in Philosophy, Politics and Economics (London:
Routledge & Kegan Paul, 1967), pp. 22-420; F. A. Hayek, The Counter-Revolution
of Science: Studies on the Abuse of Reasons, 2nd. ed. (Indianapolis, IN: The Free
Press, 1979). On these issues see also Alexander Rosenberg, Philosophy of the
Sound Sciences (New York: Westview Press, 1988; 2nd ed. 1995).
9
This chapter was originally published under the same title in Sirkku Heleski et al.
(eds.) Taking the Liberal Challenge Seriously (Aldershot UK: Ashgate, 1997), pp.
205-18.
14
Chapter 2
RISKING DEMOCRACY
THE PERVASIVE THREAT OF GRIDLOCK
T he American political system so functions at every level, from city hall

to national legislature that all too often a small but determined minor-
ity can effectively block a measure. By exploiting alliances based on
horse-trading methods to gain allies for their cause, special interest groups
can effectively impede the adoption of measures to which they object. And
this phenomenon has become increasingly prominent over the years.
Consider, for example, something so simple as a road. Everyone may
agree that a highway is needed to connect point X and point Y. Now tech-
nical study shows that there are three ways to achieve this. But nobody
wants that road in their back yard. So three interest groups immediately
form to keep the road out of their respective proximity. If they are anything
like equally strong, each will be worried about the other two ganging up on
them. The only safe step is to band together to scotch the project as a
whole.
This is bad news, but there is even worse. For let us forget about block-
age by small groups and turn to a strictly majoritarian decision process.
For the sake of an example, consider the situation of Display 1.
___________________________________________________
Display 1
A HYPOTHETICAL SITUATION IN A MINI-ELECTORATE
People’s Acceptability Assessments
Alternative Realization
Modes X1 X2 X3 X4 X5
A1 - + + - -
A A2 + - + - -
A3 + + - - -
Note that most people favor the majority of A-modes.
A-somehow + + + - -
Not-A - - - + +
___________________________________________________
That A be realized in some way or other is (so we here suppose) favored by

the group, i.e., by the majority of those involved (viz. X1-X3, that is, 3 out
of 5). Thus on the question A-somehow vs. not-A, a decided majority is in
favor of A. But, equally, each and every one of the concrete ways of realiz-
ing A is opposed by a majority (likewise 3 out of 5). In such a situation our
mini-society finds itself in what might be called a concretization quandary:
there is no majoritatively acceptable way of reaching a majoritatively ac-
cepted goal. Gridlock situations of this sort often pose a condition in
which, despite an acknowledged abstract desirability there is no concrete
implementations found acceptable. And experience shows that this sort of
gridlock that blocks the realization of a generally desired result is in fact
all too often encountered in the political arena.
Such concretization quandaries reflect the logical impracticability of
adopting pervasively the (seemingly) natural principle of (seemingly) de-
mocratic process: Majorities represent the will of the group; if the majority
wants it done, then so be it—let it be done. For, as the previous sort of ex-
amples show, it can readily happen that a majority indeed wants A done,
yet this can be achieved only through one or another of its concrete realiza-
16
RISKING DEMOCRACY
tions Ai where a majority is against doing each and every one of the Ai.
CONCRETIZATION QUANDARIES AND DECISION GRIDLOCK
What might be called a concretization quandary arises when it is—

abstractly considered—a good idea to do A, while the only way to do so
concretely is by doing A1 or A2 or A3 (etc.), where nevertheless doing each
of these alternative Ai is a bad idea. In such situations there is no concrete
way of realizing a generically desirable objective, since this can be accom-
plished only in one or another of various particular ways, each of which
is—individually—something negative.
Think, for example, of the story of the princess whose father is a kingly
ogre who will release her from his paternal thralldom only on the condition
of her marrying the princeling of some neighboring kingdom. But it turns
out that all of the available princelings are quite ineligible: one is too ugly,
another too stupid, a third too loutish, and the like. For the princess, mar-
riage is a good idea in the abstract. Yet each of the actually available con-
crete alternatives for achieving this otherwise desirable objective is
unsuitable and unacceptable. Or again, consider the plight of the younger
son of an impoverished aristocrat. He finds himself so situated that taking
up an appropriate career is somewhere between eminently desirable and
absolutely necessary. But each of the specifically available alternatives is
infeasible: he is too cowardly for the army, too hydrophobic for the navy,
too skeptical for the church, and so on.
The schematic structure of such concretization quandary situations is
clear. The circumstance that now arises is that while realizing a goal is de-
sirable in the abstract, it nevertheless remains false that any and every con-
crete practicalization of this generic desideratum is in itself something
undesirable. Those who face such a quandary situation are emplaced in the
unhappy position where an outcome looks good—in the abstract, but only
as long as one ignores the problematic details of its concrete actualizations.
There is simply no acceptable way to get there from here.
Desiderata as such are, after all, nothing more than mere abstract
wishes—mere indefinite wants: “Would that such-and-such a condition of
things were realized.” But effective means to this realization require con-
creteness: some certain particular way for that abstract desideratum to be
concretely realized. An abstraction may be acceptable to and even desir-
able for us while nevertheless we shrink from each and every one of the
concrete states that would bring it to realization in the actually prevailing
circumstances. For it is evident that in adopting an end we do not and need
17
not necessarily endorse any of the particular means actually available for
its realization. (Recall W. W. Jacob’s classic short story, “The Monkey’s
Paw.”) In inauspicious circumstances it can, unfortunately, happen that the
cultivation of what is, in and of itself, a perfectly proper desideratum
may—in the prevailing circumstances—saddle us with collateral negativ-
ities. With such a concretization quandary, there just is no acceptable con-
cretization for that deserved end. Here the apparently best course may well
be to leave well enough alone. The princess might be better off staying
home single with Daddy Ogre.
It is easy to see how concretization quandaries impact upon the work-
ings of democracy in the context of the voting process. For situations fre-
quently arise when policy making reaches a stage where even though a
social program or public work is generally acknowledged as something
that is abstractly (or generically) desirable and desired, yet nevertheless
each and every one of the concrete ways of realizing it is deemed unac-
ceptable. Downsizing the market in illegal drugs, keeping teenage girls out
of maternity wards or reducing the exploding public expenditures for
medical services, are only a few examples of this. Here, on the question of
achieving a result R-somehow vs. maintaining not-R, a decided majority is
in favor of R. But equally, each and every one of the particular concrete
ways of realizing R is opposed by a comparable majority. This sort of
situation is something often encountered in the political arena, where we
frequently read in the press stories of the following tendency:
When asked what Congress should do about the federal deficit, two-
thirds of the voters preferred cuts in major spending programs, but this
support for spending cuts dissipated whenever it came to specific pro-
grams, with two-thirds of the voters opposed to each of the specific
ways of achieving these cuts.
Despite an accepted “Sense of the Congress” resolution, no implementing

decision can manage to gain Congressional approval. In such a situation,
society confronts a concretization quandary: there simply is no majorita-
tively acceptable way of reaching a majoritatively accepted goal. It is as
when everyone agrees that there shall be a better round from A to B but no
one agrees on the particular route. And with complexity this sort of situa-
tion proliferates because each concretization of that general desideratum
leaves many of those affected with uncertain and messy anticipations.
Gridlock is only a manifestation of a larger difficulty. It may seem sur-
prising but is nevertheless a fact of life that human development can reach,
18
RISKING DEMOCRACY
and in modern complex societies already have reached a stage in point of

complexity where the prospect of finding a solution to our social problems
by consensus formation on the basis of rational calculation is simply be-
yond us.
Recent American experience affords numerous instances—health care,
social security safeguarding, and education reform among them—where
that there is on the one hand a virtually overwhelming public pressure,
duly recognized by Congress, that something be done to resolve a certain
problem, while nevertheless each of the available solutions is deemed un-
acceptable since in each case some combination of groups managing to de-
feat each one of the concretizations of that generally agreed desideratum.
Every available solution generates an opposition sufficiently powerful to
defeat it. The upshot here defeats the plausible principle that “to will the
end is to will the means” because each and every one of the particular
means to that accepted end is itself deemed unacceptable. At this stage of
our history the realities confront us with fragmented society where grid-
lock is the order of the day because powerful interest groups are able to
frustrate motion in any given particular direction.
COMPLEXITY AND ITS CONSEQUENCES
When this republic was launched and the first Congress convened in
New York City in March of 1779, there were 20 senators and 59 represen-
tatives. Today we have 100 senators and 435 representatives, an increase
of clearly substantial proportions. And in this case, at least, the advertising
slogan for the movie “Godzilla” applies: size matters. For the bigger the
group the more likely is its splintering into constituent blocks and the less
likely the chances of configuring the larger subgroup of likeminded indi-
viduals willing and able to impose a definite resolution on controversial is-
sues.
In view of the sort of gridlock situation that we have been considering,
one constantly sees our political system being hamstrung in its efforts at
problem solving. And as a result of this, the legislative branch has tended
increasingly to withdraw from decision making and to hand the problems
over to the courts.
But how do the courts actually go about deciding public issues when the
political system washes its hands of them? They, in turn, incline to “pass
the buck” to acknowledged authorities. They rely on experts to provide the
guidance that is, in the circumstance, both necessary and desirable. In-
creasingly, however, this recourse to experts itself proves unavailing. Not
19
that there aren’t any—or that those there are are unwilling to offer solu-
tions. In general there were lots of them and they all have a good deal to
say. The problem is rather one of surfeit than insufficiency. For the unfor-
tunate fact of it is that the experts almost invariably disagree. They offer us
conflicting judgments and discordant solutions that cancel each other out.
And this occurs not because the experts are incompetent but because the
problems are intractable. They are of such complexity that scientific analy-
sis and expert deliberations simply cannot settle matters.
The trouble with complexity is the difficulty it creates for decision mak-
ing by rational calculation. Consider the homely example of the young per-
son who confronts the choice between two job offers or between two
marriage partners. Here vastly many pro and con considerations can come
into it either way. There are simply too many operative factors and too
many convoluted interrelationships for the issue to be resolved by rational
calculation: there is, all too often, no rationally determinable resolution to
such a choice problem. The nature of relevant details, and the elaborate
feed-back relationships involved the intricacy of their interrelationships
prevent rational calculation from affording a viable means of resolution.
Take something as “simple” as demographic prediction. On its 80th an-
niversary in 1931 the New York Times asked various specialists to envi-
sion the world after another 80 years, in 2011. Their sociographic expert
predicted a US population of 160 million for that year. In fact it reached
260 million in 1994 and by 2011 might well reach 320 million—double
that estimate.1 Again, in the mid 1990s Americas economic gurus saw an-
nual deficits stretching as far as the eye could see, by the end of the decade
we were in surplus. In the middle of the century demographers envisioned
a population explosion of Malthusian proportions by the end of the century
the world’s population was stabilizing and many advanced countries were
facing the prospect of an imploding population and a population deficit.
The root reason for the long-term unpredictability of significant social
developments is not far to seek. For one thing, chance and chaos come into
it: the course of events over the longer term in matters of social interest
depends too much on subtle interactions which, while virtually indiscerni-
ble at present and negligible in the short term can make an enormous dif-
ference to what happens over the long term. But something deeper is also
at work. Brute contingency is the main culprit. For genuinely self-
developing systems contribute formatively to their own development over
time. Their future is not preordained by their past because novelty-
spontaneity-creativity intervenes. Such systems—whether biological, tech-
nological, or social—inevitably have aspects that are unpredictable be-
20
RISKING DEMOCRACY
cause there are always some situations to which they make an ad hoc re-
sponse and about which they simply “don’t make up their mind until they
get there,” as it were. Complexity is the inseparable accompaniment of
modernity. We encounter it throughout our science, throughout our tech-
nology, and throughout our social and cultural environment as well. Per-
haps the clearest manifestation of this is the range of choice that nowadays
confronts us on all sides in everyday life—with sources of information,
means of entertainment and leisure activities, occupations, and even life-
styles. Modern life has become vastly more complicated by widening the
range of choice and opportunity.
New-gained technical capacity accordingly brings additional prob-
lems of management in its wake. Notoriously, one virtually has to be a
rocket scientist to program one’s VCR. A modern car has many thousands
of parts, but a jet aircraft can have over four million and a space rocket
over six million. Only technical experts can carry out repairs or modifica-
tions and in this regard the era of string and sealing wax is over. Moreover,
what holds for function holds for malfunction as well. In the pre-jet era
when an airplane plummeted down from the sky, a pair of experts had
comparatively little trouble figuring out what went wrong. But as the ex-
plosion of TWA’s flight 800 demonstrated in 1996, determining the cause
of the malfunction of a system as complex as a modern high-tech aircraft
can set a team of scores into prolonged bafflement. In combat, the pilot of
a jet fighter makes far more decisions in five minutes than a sailing-era
ships naval captain did in a day.2 And so, while technological progress—
be it material or social—may indeed simplify and facilitate the perform-
ance of particular tasks, its aggregate effect is to make large-scale proc-
esses more complicated and difficult.
In his amusingly written but brilliantly perceptive book, C. N. Parkin-
son took note of what might be called the managerial bloat of modern or-
ganizations.3 In all sorts of enterprises as activity diminished, management
flourished. Throughout the twentieth century, the British navy had ever
more admirals as ever fewer ships were in commission. As operations
downsized, management upsized.
We can see the impact of complexity all about us. The computerized
control electronics in a contemporary automobile cost some two thousand
dollars more than the steel used to produce the same car. Granted, com-
puter automated problem solving is one of the wonders of the age. Com-
puters fly planes, land rocket modules on the moon, win chess
competitions, develop mathematical proofs. All the same, we have to come
face to face here with what might be called a Hydra effect after the mytho-
21
logical monster who managed to grow several heads to take the place for
each one that was cut off. The fact is that there is a feedback symbiosis be-
tween problems and solutions which operates in such a way that the
growth of the former systematically outpaces that of the latter. Accord-
ingly, those sophisticated information and control technologies not so
much resolve problems of complexity as enlarge this domain by engender-
ing complexity problems of their own. Despite the enormous advantages
that they furnish to intellectual efforts at complexity management, com-
puters nevertheless do not and can not eliminate but only displace and
magnify the difficulties that we encounter throughout this sphere.
The long and short of it, then, is that, in the presence of the intractable
problems presented to us by the highly complex social system of the mod-
ern world our technical resources are of limited utility and even experts are
not at all that much help when it is solutions that we are after. Their exper-
tise—powerful though it may be—is unable to calculate answers whenever
there are none to be found.
Examples of intractable issues that figure on the agenda of present-day
public concern are easy to come by. They include such matters as:
• how best to address the problem of illegal drugs
• what steps to take towards alleviating inner-city poverty
• how to “save” social security as the ratio of young-to-old becomes

increasingly adverse
• how best to combine the interests of job protection and environ-

mental safety
• how to bring illegal immigration under control
• how to reduce the danger of handguns to victims of accident and

criminal activity
Often as not we find that in such matters of public policy decision even
the most well-intentioned of measures result in unforeseen and unfavorable
consequences. For social medicine like chemical medicine frequently
comes with unhappy side effects. So called “reforms” all too often go
wrong. We “liberate” mental patients from institutionalization and turn
them into street people—or worse. We institute child support for the most
22
RISKING DEMOCRACY
disadvantages and destroy the family structure that exists in this sector of
society. We control immigrants and “rationalize” entry into the country
and thereby create vast hordes of illegal immigrants.
How often have we experienced the scenario of adopting expert-
recommended answers to social problems only to see them go disastrously
wrong? The fact is that in a complex modern society there is often no way
to get a rational grip on the consequences of public policy measures and
employ “scientific intelligence” to foretell the consequences. There are no
calculable solutions here—all that we ever seem to get is a clash of my ex-
perts versus your experts. Rational calculation and scientific analysis
leaves us in the lurch. The best that we can do is to feel our way cautiously
step by step—to experiment, to try plausible measures on a small scale and
see what happens, and to let experience be our guide.
REVIEW
A brief retrospect is in order. It yields a picture that has two main com-
ponents:
(1) Our political system is functioning under conditions that render it

gridlock-prone. We all too often confront a condition of affairs
where powerful minority interest groups are able—singly or in com-
bination—to prevent the legislative process from moving in any one
particular direction.
(2) The inherent complexity of issues of public policy decision is such

that we cannot count on experts to arrive at convincing resolutions
to public policy questions. The complex patterns of contemporary
policy decision are simply too complicated for resolution by the sort
of rational calculation that one can reasonably hope to obtain from
experts. Too many factors are involved in volatile interaction in too
complex interrelationships.
Just where—mere perplexity apart—does this condition of things leave

us? Clearly in something of a quandary. For when problem-solving with
respect to public issues is necessary, while nevertheless the legislative sys-
tem is unwilling and its expert-guided judicial supplement unable to pro-
vide a solution, the situation is clearly an unhappy one. Can anything be
done that is more constructive and hopeful than wringing our hands in baf-
23
flement and gnashing our teeth in frustration? Is there any practicable way
of energizing a gridlock-prone system into action?
When one sees how Congress ties itself into knots over issues like elec-
tion campaign finance reform or medical insurance rationalizations or
when one sees how state legislatures hamstring themselves over issues like
electoral reappointment or gun registration one finds oneself exclaiming:
“There has got to be a better way than this.” And there assuredly is. But to
realize it we have to take a step back and take a hard look at the first prin-
ciples that characterize democratic institutions at their most fundamental
level.
THE LESSON
Ever since the infancy of political theory in the era of Plato’s Republic,
political philosophers and theoreticians have manifested a deep aversion to
genuine democracy. Dismissing the generality of people as “the masses”
(hoi polloi—the great unwashed), they have insisted on the need to have
matters decided by wiser heads (invariably, it seems, those belonging to
people very much like themselves). No matter how loudly they enthuse
about “the people,” “the citizenry,” and “the common man” political gurus
have insisted, almost without exception, that in matters of actual decision it
is necessary to have others act on their behalf.
The great irony of the history of political thought is that democrats do
not trust democracy—or at least do not trust it in the real world as it is ac-
tually constituted. Even the most liberal of democrats shrink back when it
comes to accepting real flesh-and-blood democracy. Marxists want “power
to the people” alright, but want themselves to exercise it on their behalf.
Liberals want democracy—but only in a representative form where liber-
als carry the burden of representation. Idealistic philosophers such as John
Rawls want the fundamentals of public policy decided democratically al-
right—but by hypothetically idealized elections operating in hypothetically
idealized circumstances. The pragmatist John Dewey was less unrealistic.
But even he took the line that if electoral democracy is to function properly
then “we”—the intelligent, scientifically informed elite—must “educate
the masses” to the point where people hold the views which that educated
elite deems appropriate.
Now as I see it, what is called for on the basis of a realistic reaction to
the prevailing state of affairs is a thoroughgoing revision of this point of
view. A viable defense of democracy must be prepared to take people as it
actually finds them. The great political need—and opportunity—of our
24
RISKING DEMOCRACY
time is, I submit, the possibility of accepting a realistic rather than an ide-
alized version of democracy.
Democracy has always had a difficult time of it. It does not accord
something with our standard political categories based on the left-right
spectrum. Leftists have generally looked to elites and mistrusted the fun-
damental conservation of the silent majority. Rightists have never been
prepared to entrust traditional fundamentals to a potentially excitable
populace. But difficult times call for difficult measures. The old maxim
holds: the best way to make people trustworthy is to trust them. Conditions
may well be ripe for actually trusting the people, and to bring to realization
the idea of “power to the people”—an idea that has always been anathema
alike to the left of politics and its right.
What these considerations argue for is revolutionary in its implications
but it is not actually a revolution. It is a mere strengthening of those
mechanisms—initiative and referendum with respect to legislation and re-
call with respect to legislators—through which the “will of the people” can
achieve a more powerful and direct expression. In particular it would
greatly facilitate the problem-solving process by making it possible for is-
sues can be put on the ballot and decided by vote. All of these processes
exist to some extent in the present scheme of things, and I am suggesting
no more than that their operation be extended and strengthened whenever
possible.
To say this sort of thing is nowise to oppose exploiting the guidance of
expert opinion. By all means, let the experts study, propose, explain, argue.
Far be it from any academic to gainsay the utility of experts and deny the
necessity of making use of their worthy labors. We very much need them
to indicate alternatives, clarify issues, assess consequences, evaluate assets
and liabilities and generally working to inform the public debate on the is-
sues. But we emphatically do not need them to decide matters. By all
means let them do their work and have their say about it. But when this is
said and done, then by all means let the people decide. There is good rea-
son to think that populism is an idea whose time has come.
Clearly such a step is not an easy one. The idea of popular democracy
gives politicians fits. In the oversimplified orthography of our political
caste (and note that it is becoming a caste system, with more and more
second and third generation politicians in high public office) “populism” is
a four letter word. The idea of actual popular democracy is decidedly un-
welcome to the political establishment. For while “public opinion” inter-
ests its members deeply (why else all this money and effort spent on
polling), it interests them principally as a means towards heading off any
25
direct public involvement in the more active stages of the political process.
And in this context it should be noted that giving the general citizenry a
greater role in legislation increases the probability that there will be motion
on some issues on which the political establishment prefers immobiliza-
tion. These instance such matters as election reform with special emphasis
on spending limitations and term limits; tax reform with special emphasis
on compliance simplification and tax reduction; and very likely also some
real measure of medical insurance reform. It is likely, therefore, that one’s
reaction to the proposed measures of legislative populism will be formed
less through considerations of the general principles of political theory
than by one’s views regarding such concrete issues.
There is, of course, always the danger that some “hot button” issue may
excite public opinion and lead to problematic decisions. The traditional
theory is that the intermediation of elected representatives provides for
cooler heads that make sensible decisions. But this theory itself begs some
large questions because the sagacious and disinterested representatives that
it assumes are not all that readily available. Moreover an important point
deserves emphasis. What is being proposed here does not affect the separa-
tion of powers that is the cornerstone of our system. It leaves the executive
and the judiciary wholly unaffected, and only intrudes upon the legislative
process, the very part of our system whose operation is most direly in need
of reformation in the present condition of public affairs.
One significant asset of a more populist mode of legislation has to do
with the simple matter of numbers. The gridlocks and stalemates that so
often arise root largely in the role of special interests. And to identify the
source of difficulty here a simple prescription suffices: “Follow the
money.” Clearly, it is easier to bribe, pressurize or otherwise influence a
legislature of a couple of hundred people than an electorate of many mil-
lions (particularly where those comparatively few are perpetually cash-
hungry thanks to the costs of campaigning). Of course money can also be
expended to influence the public at large, but it is going to take a lot more
of it.
In the early days of our republic no great disadvantage attached to
stalemates. Inaction led to preservation of the status quo and in a tradi-
tional agrarian society there is nothing so bad about that. But in a complex
and dynamic modern world where matters all too often are going from bad
to worse, stability is not a particularly pleasant option, and inaction can
prove to be a disastrous course of action.
The reality of it is that in the existing condition of affairs, an important
advantage belongs to any system that produces actual decisions. When so-
26
RISKING DEMOCRACY
ciety does not settle its problems satisfactorily people naturally incline to
blame government. The result is a deep distrust and even antagonism to
our “public servants”. Gridlock and inaction in the face of pressing prob-
lems readily result in disgruntlement and disaffection. But unfairly so. It is
not the case that our public functionaries are generally corrupt or incompe-
tent. There is no proper room for a conspiratorial paranoia which holds that
our political representatives are in the pockets of some powerful conspir-
acy. The fault lies in the system and the conditions in which it has to func-
tion rather than in the persons who operate it. It is the system that needs to
be fixed and a greater infusion of democracy is in our best prospect for fix-
ing it.
Undoubtedly, a more democratic mode of procedure will also lead to
mistakes. But the choice we face is not one between some mistakes and
none; it is one between an inaction that is often the most dangerous pro-
ceeding of all and a policy that offers a chance to avert immobilization.
The reality of it is that in these matters of political decision there are no
foolproof arrangements. It goes almost without saying that the present pro-
posals will not bring utopia to realization. But be this as it may, matters
stand on a very different footing when the public makes decisions directly.
When things go wrong—when even out best conceived measures do not
deliver on their promises and live up to expectation—then in a system of
genuinely participatory decision making “we the people” will at least have
no one to blame other than ourselves.
SUMMATION
In summary review, the position of these deliberations is as follows:
1) that in the prevailing condition of things there is no reason to think

that a more democratic mode of proceeding in matters of public pol-
icy decision—one that resorts more extensively to mechanisms as
initiative referendum and recall— will do any worse than the exist-
ing modus operandi.
2) that there is good reason to think that it will help to overcome what
is a grave defeat of the present process—namely its tendency to get
bogged down in the inertia of stalemate and gridlock.
3) that from the standpoint of what is abstractly desirable—namely an
27
enhancement of a genuinely democratic process that gives people a

substantial voice in the management of their affairs—the more
overtly participatory arrangements being proposed have much to be
said on their behalf.
And on this basis, it seems plausible to suppose that the present situation is
decidedly propitious for what is simply yet another Jacksonian reformation
of our political system, a further substantial extension of popular democ-
racy.
The traditional wisdom has it that where government and policy are
concerned: “It’s all so complicated that people just cannot understand the
reasons: to let’s leave it to the experts.” This is often right in matters of de-
tail but seldom so when the big picture is concerned. There are, after all,
just too many vital matters the experts really don’t seem to do any better.
On the bigger issues of problem choice we might as well leave it to the
people—if only because it is they who, in the end, have to pay for these
choices and also learn to live with them.4
NOTES
1
See Josh Rosenthal, “Looking Forward, Looking Back: The New York Times
Magazine 1896-1996-2096.” The New York Times Magazine, September, 29, 1996,
pp. 45-46 (see p. 45).
2
On complexity in relation to social and political issues see H. R. Kohl, The Age of
Complexity (New York: New American Library, 1965) as well as the author’s
Complexity (Transaction Publishers, 1998).
3
C. Northcote Parkinson, Parkinson’s Law (Boston: Houghton Mifflin, 1957).
4
This paper was given as Distinguished Lecture in Public Affairs at the State Uni-
versity of New York at Albany on April 23, 1999. It was initially published in the
Public Affairs Quarterly, vol. 12 (1999), pp. 297-308.
28
Chapter 3
IS CONSENSUS REQUIRED FOR A

RATIONAL SOCIAL ORDER?
THE QUESTION OF PRACTICAL CONSENSUS
I t is instructive to examine some of the more or less “political” aspects of

consensuality in the context of decision making in the public forum. In
particular, it is worthwhile to consider the demand for a practical consen-
sus about what is to be done in the setting of a social group, focusing on
the question: Is consensus-seeking to be regarded as a prime imperative of
rational social policy?
In and of itself, consensus is clearly no absolute. One obviously has to
worry about what it is that people are consensing about and why it is they
are doing so. (Think of the precedent of Nazi Germany.) All the same, the
idea has been astir in some European intellectual circles for many years
now that a just and democratic society can be achieved only on the basis of
a shared social commitment to the pursuit of communal consensus. Such a
view insists that the public harmony required for the smooth functioning of
a benign social order must be rooted in an agreement on fundamentals.
Progress towards a congenial and enlightened society accordingly requires
an unfolding course of evolving consensus about the public agenda—a
substantial agreement regarding the practical question of what is to be
done.
This general line of thought traces back to Hegel, who envisioned an in-
exorable tendency towards a condition of things where all thinking people
will share a common acceptance of the manifold of truths revealed by rea-
son. Sailing in Hegel’s wake, the tradition of German social thought reach-
ing through Marx to the Frankfurt School and beyond has reinforced the
idea that the realization of a communally benign social order requires a
commitment to consensus—a shared public commitment to the idea that
the pursuit of consensus in communal affairs is a great and good thing.
This position, however, is deeply problematic. A good case can be made
out for the contrary view that a benign social order need not be committed
to the quest for consensus, but can be constituted along very different, irre-
ducibly pluralistic lines. After all, the idea that a consensus on fundamen-
tals is realistically available is in fact false with respect to most large,
complex, advanced societies and is (so we shall argue) simply not needed
for the benign and “democratic” management of communal affairs. And
even the idea that consensus is a desirable ideal is very questionable.
To be sure, the widely favored allocation of a pride of place to consen-
sus sounds benevolent, irenic, and socially delectable. Indeed, it may sound
so plausible at first hearing that it is difficult to see how a person of rea-
sonableness and goodwill could fail to go along. Nevertheless, there is
room for real doubt as to whether this utopian-sounding position makes
sense. Serious questions can be raised as to whether the best interests of a
healthy community are served by a commitment to consensus.
To begin at the end, let it be foreshadowed that the policy whose appro-
priateness will be defended here is one of a restrained dissonance based on
an acceptance of a diversity and dissensus of opinion—a benevolent (or at
any rate resigned) acceptance of the disagreement of others with a credo of
respect to beliefs and values. Such an approach envisions a posture of di-
versity conjoined with “live and let live”, taking the line that a healthy de-
mocratic social order can not only tolerate, but even—within limits—
welcome dissensus (disagreement, discord), provided that the conflicts in-
volved are kept within “reasonable bounds”. The present discussion will
accordingly maintain the merits of the consensus dispensing view that a
benign social order can be unabashedly pluralistic and based not on the
pursuit of agreement but on arrangements that provide for an acquiescence
in disagreement. This position sees as perfectly acceptable a situation that
is not one of judgmental homogeneity and uniformity, but one of a disso-
nance and diversity that is restrained to a point well short of outright con-
flict and chaos.
Dissensus has this to be said for it, at least, that it is at odds with a sti-
fling orthodoxy. A dissent-accommodating society is ipso facto pluralistic,
with all the advantages that accrue in situations where no one school of
thought is able to push the others aside. Indeed, the extent to which a soci-
ety exhibits tolerance—is willing and able to manage with a consensual di-
versity arising from free thought and expression—could be seen as a
plausible standard of merit, since a spirit of mutual acceptance and ac-
commodation is one of the hallmarks of a benign and productive social or-
der.
30
PRODUCTIVE VS. UNPRODUCTIVE MODES OF CONFLICT
It must be acknowledged, of course, that dissensus does have a negative

side. Its negativities preeminently include:
1. The danger of escalation from productive competition to destructive

conflict.
2. The possible diversion of resources (effort and energy) into poten-

tially unproductive forms of rivalry.
3. The separatist fragmentation of the community into groups estranged

from each other in a posture of mutual hostility.
4. The tendency to dismiss otherwise meritorious plans, projects, and

ideas simply because they originate from the “outside”, from a rival,
competing source.
Clearly, the story is not altogether one-sided. However, the sensible way
of handling the question of consensus vs. dissensus calls for effecting an
appropriate balance between the positive and the negative of the issue,
seeking the productive advantages of tolerating dissent while averting its
potential negativities by ad hoc mechanisms fitted to the specific circum-
stances at hand.
This said, the fact remains that it is highly problematic to maintain that
a rational public policy must be predicated on a striving for consensus.
Situations where the public good is best served by a general acquiescence
in disagreement are not only perfectly possible but also often actual. Life
being what it is, it would be too hard on all of us to be in a position where
we had to reach agreement in matters of opinion and evaluation. A society
in which the various schools of thought and opinion try to win the others
over by rational suasion is certainly superior to one in which they seek to
do so by force or intimidation. But this does not automatically make it su-
perior to one where these groups let one another alone to flourish or foun-
der in their divergent individuality. After all, the striving towards
consensus produces a sometimes debilitating uniformity of thought, and
the tolerance of diversity permits the flourishing of an often fruitful variety
of individual plans, projects, and visions. Pluralism on the other hand can
often better serve the currently prevailing interests of individuals, securing
31
for them and for their society the potential future benefits accruing from a
stimulating competition and productive rivalry.
ACQUIESCENCE AND CONTROLLED CONFLICT
Consensus by its very nature is a condition of intellectual uniformity, a

homogeneity of thought and opinion. And just herein lie some of its sig-
nificant shortcomings. For the fact is that the impulsion to consensus will
in various circumstances prove itself to be:
1. An impediment to creativity and innovation. (Settling into a consen-

sus position is a discouragement from endeavoring to outdo others
and striving to improve on their efforts by “doing one’s utmost to
excel”.)
2. An invitation to mediocrity. (By its very nature, realization of con-

sensus involves a compromise among potentially divergent tenden-
cies and thus tends to occupy “the middle ground” where people are
most easily brought together, but where, for that very reason, the
element of creative, insightful innovation is likely to be missing.)
3. A disincentive to productive effort. (One of the most powerful mo-

tives for improving the level of one’s performance is, after all, to
come under the pressure of competition and the threat of being out-
done by a rival.)
Not only is insistence on the pursuit of general consensus in practical mat-

ters and public affairs unrealistic, it is also counterproductive. For it de-
prives us of the productive stimulus of competition and the incentive of
rivalry. In many situations of human life, people are induced to make their
best effort in inquiry or creative activity through rivalry rather than emula-
tion, through differentiation rather than conformity, through a concern to
impede the folly they see all around. Productivity, creativity, and the striv-
ing for excellence are—often as not—the offspring of diversity and con-
flict.
Most human intellectual, cultural and social progress has begun with an
assault by dissident spirits against a comfortably established consensus.
The Andalusian friar Bartolomeo de las Casas upheld the human rights of
Amerindians against the consensus of Spanish conquistadores and settlers
32
alike that they were inferior beings; the 18th century American abolitionists
protested the institution of slavery in the teeth of a vast preponderance of
powerful opponents; J. S. Mill’s protest against “the subjection of women”
was a lone voice crying out in a wilderness of vociferous males. And lest it
be said that such “eccentric” but benevolent views did all eventually win
through to a general consensus of thought and universality of practice, one
can instance the teaching of Jesus whose endorsement is largely a matter of
words and not deeds. We can have no comfortable assurance about the pre-
sent—or future—consensual victory of truth, justice, and the cause of
rightness.
When we find ourselves dissenting from others, we may dislike their
opinions and disapprove of their actions—and they ours—but we can, by
and large, manage to come to terms. We can—often, at least—“get along”
with others quite adequately when we can “agree to disagree” with them or
when we can simply ignore, dismiss, and sideline our disagreements—
postponing further opposition to another day. What matters for social har-
mony is not that we agree with one another, but that each of us acquiesces
in what the other is doing, that we “live and let live”, so that we avoid let-
ting our differences become a casus belli between us. Acquiescence is the
key. And this is not a matter of approbation, but rather one of a mutual re-
straint which, even when disapproving and disagreeing, is willing (no
doubt reluctantly) to “let things be”, because the alternative—actual con-
flict or warfare—will lead to a situation that is still worse. All is well as
long as we can manage to keep our differences beneath the threshold of
outright conflict.
The crucial fact about acquiescence is that it is generally rooted not in
agreement with others but rather in a preparedness to get on without it.
What makes good practical and theoretical sense is the step of (on occa-
sion) accepting something without agreeing with it—of “going along” de-
spite disagreement—an acquiescence of diversity grounded in a resigned
toleration of the discordant views of others. The merit of such tolerance is
not (as with John Stuart Mill) that it is an interim requisite for progress to-
wards an ultimate collective realization of the truth, but simply and less
ambitiously that it is a requisite for the peace and quiet that we all require
for the effective pursuit of our own varied visions and projects.
A deep strain of utopianism runs through social contract theory, be it of
the Rawlsian form favored among North American social philosophers or
the Habermasian form in vogue on the European continent. Historical ex-
perience, empirical understanding of the human realities, and theoretical
33
analysis of our social situation, all conjoin to indicate that an insistence on

agreement among rational inquirers and problem-solving agents is simply
futile. However inconvenient for the philosophers, ample experience shows
that not only in matters of politics, art, and religion, but also in a whole
host of cognitive domains like history, economics, social science, and phi-
losophy, we shall never actually achieve a firmly secured general consen-
sus. And there is no reason to think that a benign society can exist only
where the clash of private opinions and preferences is eliminated or sup-
pressed by the processes of social coordination. A healthy social order can
perfectly well be based not on agreement but on the sort of mutual re-
straint, in which subgroups simply go their own way in the face of dissen-
sus.
Neither in intellectual nor in social contexts, after all, have we any firm
assurance that a consensual position somehow represents the objectively
correct or operationally optimal solution. To be sure, in many cases some
sort of resolution must be arrived at with respect to public issues. But we
need not agree about it: a perfectly viable result may be had simply on the
basis of a reluctant acceptance of diversity. What matters for the smooth
functioning of a social order is not that the individuals or groups that repre-
sent conflicting positions should think alike, but simply that they acquiesce
in certain shared ways of conducting the society’s affairs.
SOME OBJECTIONS
We come, at this juncture to what is a central point in the defense of our

present position. It pivots on the following objection put forward by a hy-
pothetical critic:
I agree with much of what you have said on the merits of dissensus and
diversity. But you have failed to reckon with the crucial distinction be-
tween a consensus on matters of ground level substance and a consen-
sus on matters of procedure. As you maintain, a benign social order can
indeed dispense with a substantive consensus regarding what is decided
upon. But what it indispensably requires is a procedural agreement on
modes of conflict resolution—a second-order consensus about how
those first-order issues are to be decided. If the society is to serve effec-
tively the interests of those involved, and if mutual strife and conflict—
are to be averted, there must be a consensus on process, or the validity
of the procedural ways in which these base-level resolutions are arrived
34
at. Consensus on particular decisions may be dispensable, but consensus

on the decision-making process is essential.
Despite its surface plausibility, even this more sophisticated argument for
the necessity of an at least procedural consensuality is deeply problematic.
For one thing, even where there is a consensus about process, there may
nevertheless be sharp disagreement regarding matters of implementation.
Even where people agree on, say maintenance of law and order, civility of
interaction, an equitable distribution of resources—and many other such
“procedural” principles of human action in the public domain—such pro-
cedural agreements are much too abstract to define particular public poli-
cies. (We can agree on the need for “law and order” and yet—quite
plausibly—disagree sharply on questions of civil disobedience and the lim-
its of appropriate protest.) Process consensus is a lot to ask for—but still is
not sufficient for a benign social order.
But the problem goes deeper yet. For it is simply false that procedural
agreement is indispensable for a benign social order. To manage its affairs
in a mutually acceptable way, a community needs no agreement on the
merits of those procedures as long as there is acquiescence in their opera-
tion. What matters is not that we agree on methods—I may have my favor-
ite and you yours. (I might, for example, think that the proper way to
address the issue at hand is for the electorate to decide it by referendum;
you think that the right and proper way is by a vote in the legislature.) But
as long as we both acquiesce in the established process of having the courts
decide, all is well. There is no agreement here: we emphatically do not
concur in thinking that the courts are the proper (let alone the best!) avenue
for a solution—in fact, neither of us thinks so. What we do is simply ac-
quiesce in what the courts make their decisions on the issue. What matters
for irenic conflict resolution is not second-order consensus but second-
order acquiescence. A sensible defense of acquiescence is accordingly not
predicated on ignoring the distinction between first-order substantive is-
sues and second-order methodological ones: rather it is prepared to turn
this. Distinction to its own purposes and to see it as advantageous rather
than inimical to establishing the claims of acquiescence vis-à-vis consen-
sus.
But even when we “agree to disagree” do we not in fact agree? Not
really. Or, rather, we do so in name only! An agreement to disagree is as
much an agreement as a paper dragon is a dragon—the whole point is that
there is no agreement at all here. Parties who agree to disagree do not
35
agree on anything—they simply exhibit a similarity of behavior in that

they walk away from a disagreement. They no more agree than do bounc-
ing billiard balls that move away from one another.
Another possible objection to an emphasis on acquiescence as a mecha-
nism of social decision runs as follows:
To cast acquiescence in a leading role in the management of public af-

fairs is to invite the deployment of raw power; to open the doors to co-
ercion, oppression, domination, and the subjection of the weak to
control by the strong.
But this view of the matter is simply unjust. The rational person’s ac-
quiescence is, after all, based on a cost-benefit calculation that weighs the
costs of opposition against the costs of “going along”. And to deploy raw
power is to raise the stakes—to readjust not only the benefits but also the
costs of acquiescence. As those who study revolutions soon learn, it is pre-
cisely at the point when power is made blatantly overt—when bayonets are
mounted and blood shed in the streets—that acquiescence is most gravely
endangered. It is clear that discernibly just, benign, and generally advanta-
geous arrangements will secure the acquiescence of people far more read-
ily and more extensively than those that infringe upon such obvious social
desiderata. It is quite false that an approach that roots social legitimacy in
acquiescence somehow favors oppression and injustice.
To be sure, much will depend on the sorts of people one is dealing with.
If they are unreasonably longsuffering and spineless—if they are weak-
kneed and cave in easily under pressure—then a social order based on ac-
quiescence, is one in which they indeed can be oppressed and exploited.
(But then, of course, if they are totally accommodating and yielding, a con-
sensual order based on agreement with others is also one in which their
true interests are likely to suffer.) The fact remains that sensible people are
distinctly unlikely to acquiesce in arrangements that are oppressive to
them. An acquiescence-oriented political process does not provide a ra-
tionale for domination, exploitation, oppression precisely because these are
factors in which sensible people are unlikely to acquiesce—once brought
into play they soon call forth opposition rather than accommodation. One
of the early lessons that an acquiescence-based society learns is that its eth-
ics are not smoothly viable if people are constantly testing the limits of ac-
quiescence. An emphasis on being civilized, urbane, restrained is not at
36
odds with a glaring to acquiescence but is actually conducive to the enter-

prise.
Moreover, the complaint that a polity of acquiescence inherently favors
the perpetration of injustices cannot be sustained. Acquiescence is like
agreement in this, that nobody else can do it for you. People may be able to
rearrange the conditions under which you will have to proceed in this re-
gard, but how you proceed within those conditions is always in the final
analysis up to you. As recent developments in Eastern Europe all too
clearly show, people will only acquiesce in injustice up to a certain point.
After that they turn to non-cooperation and opposition—they take up arms
against the sea of troubles or perhaps simply emigrate. The limits of acqui-
escence are finite.
Admittedly, acquiescence can be bad—it can be forced or compelled. It
is no automatic route to political legitimacy. But then of course neither is
consensus. We are always entitled to ask why people agree: is it for good
and valid reason—a concern for truth or for fairness, say—or is it because
of self-interest, conformism, constraint, or propagandism. Legitimacy is
always an additional issue: and just as it is not just consensus one wants
but a consensus that is rational and free, so it is not just acquiescence one
wants but acquiescence that is given in a way that is sensible and unco-
erced.
A POLITICAL PERSPECTIVE
A great continental divide runs across the landscape of the philosophical

tradition. On the one side lies the Platonic tradition that looks to systemic
order through a rational coordination under the aegis of universal princi-
ples. On the other the Aristotelian tradition that looks to organic balance
and an equilibration of diversity and division. The one is geared to a classi-
cism of holistic order, the other to a pluralism of countervailing checks and
balances. The one favors the rational uniformity of a harmonious consen-
sus, the other the creative diversity of a limited dissensus. The one invokes
the tidiness of theorizing reason, the other the diversified complexity of ac-
tual history.
Given this divide, European political thought since the time of the
Enlightenment has been fixated upon the idea of the “general consent” of
the people in defining a general agreement of the community (la volonté
generale) which may or may not be all that apparent to the people them-
selves (and may need to be discerned on their behalf by some particularly,
37
insightful elite). All the same, the dangers of that idea, run amuck, are ap-
parent to anyone who has looked even casually into the history of the
French Revolution.
The polity of consensus proceeds from a fundamentally socialistic
commitment to the coordination and alignment of individual action into the
uniform social order of “rationalized” central planning (albeit, no doubt, a
uniformization that is not imposed, but rather engendered—presumably—
through the “hidden hand” of an idealized rationality). Legislatures, taxing
authorities, and political theorists like to keep the affairs of the citizenry
neat and tidy. But the fact of the matter is that the impetus to public con-
sensus, agreement and concurrence of thought will not be high on the pri-
ority list of the true friends of personal freedom and liberty. And so, the
polity of pluralism abandons the goal of a monolithically unified “rational
order” for the “creative diversity” of a situation of variegated rivalry and
competition. Its political paradigm is not that a command economy with its
ideal of rationalization and uniformizing coordination, but that of a free
market with its competitive rivalry of conflicting interests. Consensuality
looks to uniformity of thought, pluralism to reciprocally fruitful harmoni-
zation of discordant elements.
Rather different sorts of policy approaches are at work in social orders
based on consensus-oriented and acquiescence-oriented principles. Con-
sensus-seeking societies will aim to maximize the number of people who
approve of what is being done; acquiescence-seeking societies seek to
minimize the number of people who disapprove very strongly of what is
being done. The one seeks actual agreement, the other seeks to avoid dis-
agreement so keen as to preclude acquiescence. The two processes sound
similar but are in actual fact quite different in spirit and in mode of opera-
tion.
The social requisite of a viable public order can thus plausibly be
viewed as lying not in the fostering of consensus, but in the forging of
conditions in which people become willing and able to acquiesce in dissen-
sus through recognizing this as a state of affairs that is not only tolerable
but even in some way beneficial. Consensus simply is not a requisite for
the prime social desideratum of having people lead lives that are at once
personally satisfying and socially constructive.
38
IS CONSENSUS A VALID IDEAL?
To be sure, in matters of practical decision at the individual and social

level consensus can be a significant desideratum. Our mind is eased if the
consulting physicians can agree in a diagnosis and cause of treatment, for
then we have done all we can to put dissonant possibilities out of range.
The climate of public opinion is healthiest where the representatives of dif-
ferent interests and points of view can reach a meeting of minds, for then
the likelihood of social conflict is minimized. But of course fate is not
generally all that cooperative; agreement in such matters is not so common
that we have a right to expect it. A well-designed social order has to be
able to make do without consensus. And so, while consensus is a desidera-
tum of sorts—something it would be nice to have if we could get it under
the right conditions, something to be welcomed where it can be found—it
is nevertheless not something whose pursuit we should insist on and persist
in. But is consensus not at least an appropriate ideal? There is good reason
to think it is not.
To begin with, one must distinguish between an ideal and an idealiza-
tion. An ideal as such belongs to the practical order. It is something that
can and perhaps should be a guide to our actual proceedings, providing a
positive goal—or at least a positive direction—for appropriate human en-
deavor. It represents a state of thing whose realization—even if only in
part—is to be evaluated positively and which should, by its very nature, be
seen as desirable. Like “liberty, equality, and fraternity”, an ideal repre-
sents a state of affairs whose pursuit in practice is to be regarded as “a
good thing”. By its very nature as such, an ideal is something towards
whose realization right-thinking people would deem it appropriate to
strive.
An idealization, on the other hand, is something quite different. It in-
volves the projection of an hypothesis that removes some limit or limita-
tion of the real (a perfectly elastic body, for example, or a utopia
comprising only of sensible and honest people). An idealization is accord-
ingly a thought-instrument—a hypothetical state of things that it may be
profitable to think about, but towards whose actual realization in practice it
may be altogether senseless to strive.
Ideals, in sum, are action-guiding, while idealizations need by no means
be so. A world of eternal springtime might be nice to have if we could get
it. But it makes no sense to expend effort and energy in this direction. A
39
positively evaluated idealization does not necessarily constitute a valid

ideal.
This distinction between ideals and idealizations bears directly and in-
formatively on the status and standing of consensus. For there is no doubt
that consensus is merely an idealization—and thus not a sensible goal and
not a plausible guide for action, seeing that it prescinds from the variety,
and dissonance that inevitably characterizes the beliefs, opinions, goals,
and values of any sizable human community. And so, the fact that consen-
sus represents an idealization does not in and of itself mean that consensus
is a valid ideal even when seen in its most positive aspect—that it is some-
thing whose pursuit in practice is reasonable and appropriate.
The notion that consensus is a valid ideal—that the endeavor to bring
about a uniformity of thought and opinion is an unqualifiedly good thing—
is deeply problematic. Consensus is not in general a goal whose pursuit
should regulate the way in which we actually proceed in the conduct of our
cognitive and practical affairs. In many contexts the interests of the entire
community are best served by a fragmentation of beliefs and values within
its ranks. For example, the social welfare of a group is usually most effec-
tively catered to when different political subunits can, though pursuing dif-
ferent policies and adopting different programs, provide testing grounds for
the evaluation of alternatives. And the communal welfare of a group is
generally more effectively served when different religious or cultural
“sects” or “schools of thought” that can provide a congenial home for indi-
viduals with different personal needs and inclination. Consensus can be the
cause of boredom, inaction, stagnation, and complacency. It can result in a
narrowing of horizons and a diminution of options that is destructively
stultifying—that substitutes bland uniformity for an envigoration variety.
The situation differs in this regard as between theoretical and practical
philosophy. A resort to idealization in theoretical philosophy—in matters
of inquiry, truth, and rationality—is something of a harmless bit of theo-
retical ornamentation. But in matters of practical philosophy idealization
can do actual harm. No doubt, ideals can be a useful motive in the direction
of positive action. But generally only as a primum mobile—an initiator. To
hold to the hard and fast, in season and out—not just as the start of the
process of decision and action but all along the line—can be dangerous and
self-defeating. By diverting our attention away from the attainable realities,
a preoccupation with the unrealizable ideal can do real damage in this do-
main where a pursuit of the unrealizable best can all too easily get in the
way of the realization of attainable betterments and impede the achieve-
40
ment of realizable positive objectives. It might be nice for me to be a good

polo player but if this lies beyond my means and talents, it would be fool-
ish to let this desideratum get in the way of my perfectly feasible and at-
tainable goal of being a good tennis player. It would be a splendid thing to
be a great artist, but for many individuals it could be counterproductive to
let this aspiration get in the way of being a good craftsman. Similarly it
might be nice to have social consensus, but it would be counterproductive
to let this get in the way of social amelioration—of effecting various
smaller, but perfectly feasible, improvements in the arrangements of a plu-
ralistically diversified society.
Insofar as consensus is something positive, then, it has to be seen in the
light of a desideratum rather than a valid aspiration of the sort at issue with
an ideal. The polity of consensus is too utopian to provide me with so use-
ful an instrument.
On this perspective, one arrives at a less tidy, less optimistic, but never-
theless truer and more realistic picture of our social condition, one that ac-
cepts without regret the dissensus of a restrained rivalry among discordant
and incompatible positions none of which is able to prevail over the rest.
The upshot is one of reciprocal acquiescence in a pluralistic community of
conflicting views—a situation in which each party is content to accept a
diversity which of itself constrains a common-sensical accommodation that
affords them the benefit of pursuing their own projects at the sole cost of
according a like privilege of others.1
NOTES
1
This chapter was originally published under the same title in Axel Wüstehube (ed.),
Pragmatische Rationalitätstheorien (Würzburg: Königshausen & Neumann, 1995),
pp. 113-24.
41
Chapter 4
MORALITY IN GOVERNMENT AND

POLITICS
PUBLIC SPHERE MORALITY
P hilosophers like to go about their work in an Olympian, detached, and

coolly analytical spirit. They can take little professional comfort in
dealing with issues that are hot-beds of emotion and acrimony on the cur-
rent scene of political controversy. Moreover, philosophers like their prob-
lems to be subtle and intricately wrought, calling for a delicate fusion of
competing considerations. They are at home, when critical intelligence is
called for, in laying bare the theoretical complexities of finely balanced
controversies. Nothing could be further from this favored terrain than the
issue of morality and ethics in government and politics. Here, only one side
of the question seems to present a live option. Like clean air or good food,
morality in public life is hard to be against.
Happily, however, there is one theoretical issue that is both central to
the problem of morality in public life and that has received considerable at-
tention at the hands of the great theoreticians of the past. Is a public figure
in government and political affairs entitled to special concessions in point
of moral criticism and ethical appraisal? Do political and governmental
agents constitute a special category in this regard?
Two sorts of arguments have been espoused in support of the view that
agents on the public stage in government and politics are not to be judged
by the same moral standards that apply generally to individuals acting on
their own behalves (in propria persona). The first of these I shall call the
Titan theory. Its main theoretical exponent was Hegel; its best publicist
was Thomas Carlyle. According to this theory, public agents are extraordi-
nary in the root sense of that term. They are cast in roles that are, as it
were, larger than life. They take greater risks, play for greater stakes and
carry larger responsibilities. Accordingly, we must not judge them by the
ordinary standards applicable to common folk. Men whose actions are
played out on the stage of world history cannot be expected to abide by the
rules and restraints that govern the acts of ordinary men. It is not so much
that they are above the petty morality of ordinary life as that they move in a
sphere that lies outside it, a sphere where different standards and criteria
are operative. Hegel himself put the issue as follows:
The deeds of great men, who are the Individuals of the World’s History, ...[are]
justified in view of that intrinsic result of which they were not conscious. ... But
looked at from this point, moral claims that are irrelevant, must not be brought into
collision with world-historical deeds and their accomplishment. The Litany of pri-
vate virtues—modesty, humility, philanthropy and forbearance-must not be raised
against them.1
The second position is what might be called the raison d’état theory. Its
main historical exponent was Machiavelli, and its paradigmatic practitioner
was reputedly Cardinal Richelieu. In contrast to the preceding theory, this
finds a fundamental difference not in the nature of the agent and his role,
but rather in the nature of the acts. Acts carried on by a public agent as
trustees for the public interest or common good are for this reason to be
placed in a special category where their performance is thereby exempt
from criticism by standards operative in the common run of cares. They
must be judged solely in terms of the public interest or common good.
Both of these theories are widely held in aversion and even—in aca-
demic circles—in derision. This seems a real mistake. For while neither
theory qualifies as totally and finally correct, nevertheless a great deal can
be said for each of them.
The Titan theory holds that public men are not to be judged by ordinary
standards. It is not that the moral and ethical aspect is irrelevant in the rea-
soned evaluation of the doings of leaders of state, but it is certainly not de-
cisive; indeed, it does not really count for much. Public men—the higher
functionaries of government and politics—play a large-scale role in our af-
fairs. They are the decision-makers, the problem solvers of a society, and
the movers of men. They function within the purposive context of the lead-
ership role and this is what defines the exact standard by which they must
be judged.
The specialist as specialist must in the first instance be judged by the
quality of his special performance. In the hospital emergency room all one
cares about is that the surgeon be (1) competent and (2) not beset by extra-
neous worries and pressures that impede the efficient exercise of his com-
petence. Whether he is a kind husband or thoughtful neighbor, whether he
44
MORALITY IN GOVERNMENT AND POLITICS
blurs his full stops when driving or exceeds the statutory limit of the catch
when fishing are now very subordinate issues.
It is hard to believe that the case can be altogether different with men of
politics and government. No amount of probity, honesty, and freedom from
breath of scandal offsets the populist mediocrity of Calvin Coolidge, the
thoroughly honest and decidedly undistinguished holder of the presidential
office.2 All his private virtues can avail but little in the scale when an as-
sessment of the quality of his leadership is at issue. Nor was any imputa-
tion of evil-doing ever cast against Benjamin Harrison, but had his failure
to secure reelection to the Senate in 1887 not propelled him into the U.S.
Presidency in 1888 no great regrets would be in order. To the utter indif-
ference of posterity, we would have simply traded a presidential mediocrity
for a competent senator. On the other side of the coin, no failure in regard
to ethical standards and moral proprieties could push people like John Ad-
ams, Andrew Jackson, Abraham Lincoln, or Franklin Delano Roosevelt off
the pedestals they occupy as colossi of American history.
In judging statesmen as statesmen, considerations of ethics and morality
carry but little weight on the evaluative scale. The issue of political
achievement is altogether predominant. This is the very large grain of truth
in the Titan theory.
STAATSRAISON
Turning now to the theory of raison d’état (Staatsraison or “national

security considerations” as we tend to call it nowadays), it is often asserted
that the operatives of government, as servants of the agency that maintains
the rule of law, must always act within the law. Those with historically
more elongated memories cannot be so sure. Two brief true stories illus-
trate why this is so.
When Henry L. Stimson took office in 1929 as Herbert Hoover’s Secre-
tary of State, he promptly closed down the ludicrously modest “Black
Chamber”, the cryptanalysis activity operated conjointly by the State and
War Departments. Supposedly Stimson acted with the remark that “Gen-
tlemen do not read one another’s mail.”3 However admirable his ethics,
this action strikes us and, indeed, Stimson himself not too long afterwards
as extremely naive and utterly unrealistic to the facts of life in international
dealings. It does not require profound knowledge of the conduct of World
War II to realize that had Stimson’s position been followed the damage to
American interests—and American lives—would have been immense.
45
Though an appeal to Realpolitik, this appears altogether convincing in the

case at hand. Thus it would seem that acts which qualify unproblematically
as ethically commendable in private transactions may to all intents and
purposes not be fitting and proper in the public theater of action.
The second story concerns the period prior to World War II when rela-
tions between the U.S. and Japan entered into a state of progressive dete-
rioration. The attention of U.S. intelligence agencies came to focus on
various Japanese consulates, in particular that in Honolulu. The telephone
lines into that Honolulu consulate must have had a large audience. In late
1941 the courts rumbled and the Justice Department enforced the wiretap
rules, with the result that on December 2, 1941 the Army and Navy wire-
taps were put away. J. Edgar Hoover was not that easily moved; the FBI’s
taps remained in operation and they soon produced significant results. On
December 5th a lowly cook in the consulate called his irate girlfriend to ex-
plain that he had been “standing her up” for the past few days because they
were busy day and night burning major documents, destroying codes and
demolishing all the cryptographic equipment save for a single code ma-
chine.
Though every security-conscious operation occasionally destroys obso-
lescent materials, here was something altogether out of the ordinary: a
crash program of systematic destruction on a near-total scale. The entire in-
telligence community in Honolulu was alerted by this rather vivid war-
signal. Roberta Wohlstetter’s masterful analysis of the warning signals
prior to the Pearl Harbor attack cites this as one of the clearest signs of
imminent hostilities actually available to the people in Hawaii.4 So far as I
have been able to determine, no one in any serious position in government
or politics ever uttered a single word of reproach against this instance of
FBI lawbreaking. The mantle of raison d’état covers this episode as se-
curely as the anxious mother wraps the new-born babe.
THE BIGGER PICTURE
Is this other than as it should be? Surely any serious person since Henry
L. Stimson has genuinely expected governments in matters of this sort to
operate wholly this side of the law. This, then, is the element of truth in the
theory of raison d’état. There are likely to be certain occasions for every
public servant when an affective discharge of his official duties will clash
with playing it strictly in accordance with the rules, and when, because of
46
the nature of the risks and the balance of costs and benefits, a persuasive
case can be made for putting the interests of the office first.
Both the Titan theory and theory of raison d’état place the acts of a
public agent on a level that can in certain cases transcend that of the ordi-
nary, everyday norms of ethics and legality. Both envisage exceptions
based on the extraordinary nature of certain cases. In each instance, the jus-
tification of this departure from the usual norms is something which, in
general, cannot be fully known at the time. The concept of justification
here operates retrospectively: It cannot be fully determined save with the
wisdom of hindsight. This is so, not because the end justifies the means,
but because the evaluative quality of the means themselves is something
whose just assessment calls for an historical and retrospective view-
frequently over a very long term. The agent in effect takes a leap into the
unknown and appeals for his justification to the course of subsequent
events. In general, this involves a factor beyond one’s control, the element
of luck (fortuna) whose role in human affairs has ever been viewed as be-
ing as important as it is imponderable. Historically, American leaders have
fared rather well in this regard, a fact attested by Bismarck’s obiter dictum
that God is on the side of children, fools, and the United States of America.
One cannot but hope that those happy times are not altogether at an end.
So far, I have tried to illustrate the contention that the issue of “ethics in
public life” is not as straightforwardly one-sided as one might think at first
view. Let us probe a little more deeply into the character of the complexity.
The intense difficulty for governmental and political figures lies in the
fundamentally dualistic role in which they tend to be cast nowadays, espe-
cially in the context of a democratic society. We expect our “public ser-
vants” to function in two very different roles: as doers and as exemplars.
The former casts public officials in the role of decision-makers, movers of
men, and problem solvers; the second casts them as models, paradigms,
and inspirers. We want our public men to be efficient in producing results
and admirable in the way in which they go about it. Can we have it both
ways; and what are we to do when we cannot have it so?
There is a marvelously amusing short story by James Thurber of 1931
vintage that points up the difficulty.5 The protagonist is a garage mechanic
named Jack Smurch, who proposes to circumnavigate the globe nonstop in
an ingeniously modified 1920 vintage monoplane. One fine day in July,
Smurch “proceeded to spit a curse of labacious juice into the still air, and
took off, carrying with him only a gallon of bootleg gin and six pounds of
salami.” There is much ballyhoo and much laughter; everyone is convinced
47
the thing is a cheap joke, a hoax, a publicity stunt. Everyone is wrong. Ten
days and one circumnavigation later, the yokel mechanic sets his plane
down at Roosevelt Field for a perfect three-point landing. Instantly he is
the hero of the decade. Unfortunately, a less Lindbergh-like hero can
hardly be imagined. Not to put too fine a point on it, Jack Smurch is a
greedy boor.
Here is the paradox: a supreme achiever on the one hand, a moral Ne-
anderthal on the other. Fortunately, the idolatrous public is offered a way
out. The people who are charged with the vain task of building a suitable
image for the hopeless Smurch manage to push him out of the 10th floor
window just before his first news conference. That is the jest of Thurber’s
story. The moral he wants to extract is clear and not all that far from the
truth. We Americans are unwilling to tolerate flawed heroes. When it tran-
spires that one of our idols is cracked, we insist on hastening it to swift de-
struction.
Does this make sense? Is it reasonable to insist uncompromisingly that
our political and governmental agents be both outstanding performers and
ethical paragons? Of course it is understandable enough that we should
want to have it both ways. After all, there is no reason for restraint at the
level of what we should ideally like to have. We want our children to be in-
tellectually bright, socially adept, and irreproachable in conduct. But in our
family affairs or our business relationships we are prepared to be realistic.
Whatever fanciful notions we entertain at the level of what we might want,
we are ready to rest satisfied with a good deal less when it comes down to
the harsh facts of what we can get. Outside the hot-house area of political
and governmental affairs we are common-sense people, not doctrinaire
idealists.
THE OUTLOOK
These considerations take us back once more to the dialectic of the real
and the ideal. It is undeniably important to have ideals, to uphold and
maintain them, and to strive to translate them into reality insofar as possi-
ble. But surely it is not treason to our ideals to recognize that their attain-
ability in the face of the recalcitrant circumstances of the real world may
have to be partial and incomplete. There is a large and important difference
between cynicism and realism, and it is one that deserves recognition and
emphasis.
48
If the history of the twentieth century has taught us only one thing, it is
this: The veneer of civilization still lies very thin over the human condi-
tion. The real conflict of politics is not that between good and bad persons,
but that between, on the one hand, those motives and tendencies of thought
which try to make one’s environing society habitable and its institutions
work for the general good, and, on the other hand, those forces which are
simply destructive, negative, and nihilistic and which invite us to strike at
whatever does not correspond to our notions of perfection. The real quality
of political leadership is revealed not by the test of adherence to the ethical
rules, but by assessing the extent to which, on balance, a contribution has
been made on the positive side of the ledger.
The big step in developing maturity and balance of judgment in human
affairs is the recognition that humanity cannot be divided into good and
bad persons, that people and issues are not just ethically black and white,
but come in all those grayish shades and gradations that lie in between. For
a complex variety of reasons, the area of public agents in politics and gov-
ernment seems to be the last domain where a democratic society is willing
to apply this maturity of balanced judgment. Regrettably, upon entering the
hall where political and governmental affairs are under discussion, we all
too often check our common sense and natural human tolerance at the
door.
Regrettably, I have no promising, concrete suggestions as to how we in
academia might be able to ameliorate this state of affairs. All I can offer is
the small but possibly useful suggestion that perhaps the most effective
way towards an education for realism is the cultivation of historical per-
spective. For this study of history and particularly its biographical dimen-
sion is a vastly useful desideratum. In addition, it would also do nothing
but good to reintroduce undergraduate ethics courses of the old style, that
is, concerned not solely with systematic abstractions, but with the substan-
tive nitty-gritty of the rational deliberation of ethical decision-making in
realistic cases.6
49
NOTES
1
G. W. F. Hegel, The Philosophy of History, tr. J. Sibree (New York: Dover, 1956),
Pt. III.
2
For a clear and by no means unsympathetic vignette of Coolidge’s administration
see Frederick Lewis Allen, Only Yesterday (New York: Harper and Row, 1931).
3
See Henry L. Stimson’s obituary in the New York Times, October 21, 1950, p. I.
4
Roberta Wohlstetter, Pearl Harbor: Warning and Decision (Stanford: Stanford
University Press, 1962).
5
James Thurber, “The Greatest Man in the World,” reprinted in E. B. White and K.
S. White (eds.), A Subtreasury of American Humor (Boston: Putnam, 1961), pp.
355-361.
6
This chapter was originally published under the same title in the Proceedings of the
American Catholic Philosophical Association, vol. 48 (1974), pp. 259-66.
50
Chapter 5
ON THE RATIONALE OF
GOVERNMENTAL REGULATION
THREE BASIC PARAMETERS
T he theoretical justification of state intervention in the sphere of indi-

vidual action and initiative depends on three basic issues: Is it right? Is
it desirable? Is it expedient? A negative response to any of these questions
would suggest that the state’s abridgement of individuals’ liberty to con-
duct their affairs in their own preferred ways is inappropriate. Rightness,
desirability, and expediency are severally necessary and jointly sufficient
to legitimate state regulation of individual affairs. Let us examine these
three parameters of the problem more closely.
The Issue of Right
The state has always impinged heavily on the lives of its individual sub-
jects. It has taxed them, impressed them into its service, dictated where
they can and cannot live, and, not infrequently, appointed the time and
means of their death. Obscure existence outside the range of the state’s
sight has usually been the best way to remain outside its reach. “Fortu-
nate,” said one of the caliphs, “is the man whom I do not know and who
does not know me.” But what is, is not necessarily right, and so the big
question remains: To what extent is it right and proper for the state to inter-
fere by way of regulation and control in the actions and dealings of its
members.
Is it right for the state to take onto itself this or that mode of regulation
or control of the activities of its citizens—to fix the rules of the road, say,
or to stipulate retirement ages, or to limit the opening hours of bars? Such
questions pose the difficult normative issue of the basis on which one is to
proceed in determining whether a state acts rightly or legitimately in adopt-
ing a certain regulative measure.
Two alternatives arise at this point. On the one hand, one can take the
view that it is somehow written large in the scheme of things that the state
should exercise certain regulative functions and these only; that this is
somehow determined by the general principles of the matter. Such a posi-
tion in effect has it that it is somehow engraved on bronze tablets delivered
from on high—to put it in caricature—what the proper regulative business
of the state actually is. The virtually insuperable difficulty here is that of
getting hold of the bronze tablets, or, rather, of establishing the general
principles that are to effect the determination at issue.
On the other hand, a more relaxed view of the matter is also possible.
That is, one can take the view that the proper business of the state is to dis-
charge whatever regulative functions it is assigned by its citizens through
the due processes of the political decision-making apparatus at their dis-
posal. If the public wants to charge public authorities with the licensing of
TV sets, the monopoly of postal services, or the monitoring of banking
transactions, then so be it; the activities and regulations at issue have
thereby automatically become part of the “proper business” of the state.
The question “Is it right and proper that the state should restrict or regulate
the activities at issue?” is on this view effectively settled when the public
uses the valid political processes at its disposal to mandate those responsi-
bilities to the state.
But this rather laissez-faire approach to the question of the right of the
state to take on certain restrictive and regulating functions does not resolve
the issue of justification. What it does is simply to displace the entire
weight of concern to the issues of desirability and expediency. Beyond
question, much can be said for the view that, if and when a viable consen-
sus of the public (acting through the channels of the duly established po-
litical process) charges the state with a certain regulative mission, then the
question of inherent rightness is closed—for this position bypasses any
problematic aspects beyond “the general will” in regard to the normative
appraisal of the legitimation of state action.1 But of course, even with such
a view the issues of “Is it desirable?” and “Is it expedient?” still remain
wide open. These considerations now come to the center of the stage.
The Issue of Desirability
The question “Is it desirable for the state to exercise a certain control-
ling or regulative function?” comes down to this: Is it manifestly in the
best (real) interests of the substantial preponderance of its members that the
52
ON THE RATIONALE OF GOVERNMENTAL REGULATION
state should assume this function—in short, is it in “the public interest”? In

concrete cases the answer to this question may sometimes be yes (e.g.,
fluoridating water supplies, licensing the construction of nuclear power sta-
tions, controlling the distribution of drugs). But sometimes it may be no
(mandating a certain retirement age) and sometimes maybe (prohibiting the
use of studded tires on private vehicles, prohibiting the acquisition of gold
or of foreign exchange by private individuals). The status of these particu-
lar examples is unimportant: they are meant to be no more than suggestive.
What matters is the general principle at issue, namely that the question of
desirability turns on the devising of a cogent case to show that a successful
carrying-through of the measure at issue will indeed serve the best interests
of a preponderant majority.
It is important, however, to recognize that even after this question of de-
sirability has been settled in the affirmative, the issue of feasibility yet re-
mains wholly untouched, for the question of desirability takes the
hypothetical form. Even if the measures at issue were implemented in an
efficient, effective, and (relatively) inexpensive manner, would worthwhile
results thereby be anticipated? And the important question yet remains
very much open, whether this hypothesis—or anything like it—is realized
or indeed realizable. It is to this concern that the issue of feasibility is ad-
dressed.
Even if it is “desirable”—in the sense of contributing to “the public in-
terest”—it is still reasonable to raise the question of whether it is actually
possible for the state to realize this desideratum. Can the mice bell the cat?
Is the state actually in a position to discharge this mission efficiently and
effectively? Prohibition was an arrangement that arguably satisfied the cri-
teria of desirability. But it soon came to grief, shattered on the rock of fea-
sibility.
This matter of efficiency and effectiveness poses large and ramified
problems. For one thing, it is important to realize that the question of effi-
ciency and effectiveness is generally not so much absolute as relative, a
matter of a comparison between alternatives; that is, the question is not just
whether the state can achieve the desired results with some degree of ade-
quacy, but whether other, less awesome mechanisms for their realization
might actually be superior in this regard. In particular, we must recognize
that potent forces are generally at work to militate against the state’s as-
sumption of a function when other, less dramatic alternatives may be sub-
stituted; this is true regardless of how desirable it is to accomplish the
53
function. These constraints are indicated by the following sorts of consid-

erations:
1. The state is an inherently sluggish and cumbersome agency that is

hard to coax initially into an area of concern, and, once settled there,
is even harder to oust when circumstances so develop that this be-
comes desirable.
2. The state is, in general, a comparatively problematic mechanism of

purpose realization, because its pursuit of any given goal is inevita-
bly fraught with political complications.
3. The actual agents through which the state effects its actions—
officialdom and its bureaucracy—are generally too insulated against
consumer feedback to emerge in fact (rather than theory) as effective
servants of the public.
4. The state operates at a disadvantage in that by proceeding within an

all-embracing monolithic framework, it loses the benefits of effi-
ciency gained through rivalry and competition among smaller units
of operation.
A large variety of considerations of this sort militate toward the conclusion

that whenever an agreed upon goal can be pursued without the state assum-
ing the responsibility of implementation, then this should be done on
grounds of feasibility.
In general and in abstracto the issue of feasibility turns on matters of
cost-effectiveness. Its governing concern is with the question, “Are the
specific measures of regulation and control so operable that the benefits at
issue with desirability are achieved at a practicable cost?” And the costs at
issue must be estimated not just in economic terms but in social terms as
well, including such matters as a diminution of the liberty of the individual.
THE ISSUE OF EXPEDIENCY AND THE RATIONALE OF STATE

INTERVENTION
We come now to the issue of expediency. At this point the salient ques-
tion is: “Is the overall costs and negative effects incurred through the
54
state’s assumption of control are significantly outweighed by the benefits

resulting from such control?” In addressing this a convincing case must be
established for believing that (1) in the absence of the particular control or
regulations there is a real risk of significant harm to individuals or of seri-
ous impairment of the public interest; (2) there is reasonable assurance that
this danger will be avoided or significantly diminished by the controls at
issue; (3) no alternative device less awesome and less cumbersome than ac-
tual state intervention is available for attaining the desiderata; and (4) the
overall costs and disabilities of implementing the particular controls and
regulations in an acceptably efficient and effective way do not exceed the
benefits being realized.
Such a process of assessment involves a number of distinct but impor-
tant factors: (1) the gravity of the potential risk; (2) the probability of the
potential risk’s realization; (3) the cost of the proposed measures (includ-
ing the realization that we may be paying an assured price for averting a
merely possible danger); and (4) the likelihood that the proposed measures
will actually avert the potential risk (and not simply deflect it into other
channels).
These considerations illuminate the difficulties inherent in establishing
a legitimating rationale for measures of state control and regulation. In-
deed, they establish a powerful presumption against all such measures. But
this presumption is certainly not indefeasible. Although the difficulties in
question are real, they are not insuperable. Consider an example of the cur-
rent standard security measures of weapons control imposed on airline pas-
sengers. They do exact from all concerned a price in terms of increased
expense, added delays, and some degree of invasion of privacy. But they
do provide a substantial benefit in terms of diminishing the inconvenience
and safety risks involved in aircraft hijackings. Accordingly, one can rea-
sonably argue in this case that the benefits outweigh the costs and that such
controls and regulations are legitimate.
On the other hand, there can be little doubt that this is not true of much
current governmental regulation in the economic and social areas. In-
stances of major expenditures in time, energy, and resources or the realiza-
tion of minor or nonexistent benefits are legion. To take just one example,
the annual cost to society of the paperwork generated by the federal gov-
ernment’s insatiable demand for information was estimated at some $100
billion in 1977. Certainly the vast bulk of this cannot be justified on cost-
benefit considerations of the sort operative in validating measures of gov-
ernmental regulation and control.
55
With respect to feasibility, it must surely be conceded that in some in-

stances we cannot be assured of knowing beforehand if the state can actu-
ally administer particular regulations or controls efficiently and effectively
—whether, for example, governmental funding of scientific research facili-
ties in colleges and universities will help or hinder the research process it is
designed to facilitate. The dangers are certainly there. A program of statu-
tory regulation often makes a leap in the dark—creating a monster that
dwarfs the evils it is designed to remedy. (Prohibition affords a good ex-
ample of this.) Whenever possible it is wisest to proceed by a slow and ten-
tatively experimental process, lest relatively minor evils be traded for even
greater ones. (And one of the real problems is that in this sphere of gov-
ernmental action a “slow and experimental process” is well-nigh impossi-
ble to realize.)
MATTERS OF PRESUMPTION
It is important to recognize that regulation and control are matters of

degree that admit of infinitely varied shadings and gradations. At the top of
the scale we have the state’s determination as to whether a certain type of
good or service should be banned altogether (the sale of fireworks, the
screening of pornographic films, the purchase of armaments by private in-
dividuals). At the next level come controls over who can provide or receive
certain goods or services (child-labor laws, licensing of professionals or
airlines or radio stations, licensing of purchases of certain “dangerous”
products, etc.). Then come regulations about conditions of operation (clos-
ing times for bars). Near the bottom come the mere requirements to pro-
vide information (regarding certain banking transactions, for example).
Now in a democratic society there must always be a strong presumption
against the control and regulation of individual action by the state. The
need for a maximum range of individual liberty is axiomatic within the
democratic ethos.
Recognition of the fact that regulation and control can be justified is
simply the first step toward recognizing that they must be justified—that
they are not proper and legitimate unless we make sure cost-benefit condi-
tions are satisfied, conditions such as those illustrated in our preceding dis-
cussion. The presumption is always against government intervention and in
favor of things being allowed to run their own course.
Accordingly, even in cases where state regulation and control is desir-
able and feasible, it is also desirable that it be accomplished at the mini-
56
mum level of intervention. Every move toward control or regulation, and

every step up the ladder of increased intervention levels must prove itself
in terms of benefits to be actually realized.
But while there is a presumption in favor of unfettered liberty, this is
certainly a defeasible presumption. We do not operate an unfettered free
market in the provision of medical services or of land utilization because
the school of bitter experience has taught us that the consequences are un-
acceptable. And this is the important factor, for it takes a clear and present
danger—a very real risk of some substantial sort—to reverse this standing
presumption in favor of “the liberty of the subject”.
THE RECENT AGGRANDIZEMENT OF STATE FUNCTIONS
Historically the state has sought to intervene in the affairs of its citizens
when necessary for: (1) the survival or aggrandizement of the state itself;
(2) the physical safety and well-being of its citizens; (3) the peaceable and
orderly settlement of interpersonal dealings; and (4) the creation of major
public works and facilities. Considerations of the first kind would yield
such measures as the impressment of seamen, conscription in time of war,
and the launching of colonies. Those of the second kind would underwrite
the quarantine of the victim of a dangerous communicable disease or the
maintenance of police forces or the construction of aqueducts. Considera-
tions of the third kind lead the state into the regulation of inheritance, the
mechanisms of commercial transaction (including, for example, the man-
agement of the coinage, the regulation of property transfers, and other as-
pects of the economic interaction of one person with another). Finally,
those of the fourth kind bring the state into the business of building roads,
constructing harbors, or exploring distant shores.
However, the past hundred years have seen a major change, with the
state assuming substantial functions beyond the four traditional areas of na-
tional security, public safety, the settlement of interpersonal disputes, and
public works. Specifically, the state has come to assume responsibility for
the economic well-being of its citizenry and for the “quality of life” in a
broader, non-economic sense that embraces various social desiderata, par-
ticularly those of promoting equality of opportunity and access.
In respect to these latter items, the question of expediency that has pre-
occupied us throughout these pages now becomes especially significant;
for this question looms extra large with respect to these “new” state func-
tions, seeing that, historically speaking, the state has not evolved with a
57
view to handling these new functions, and that it has had to develop new
and sometimes problematic devices to meet them—especially by develop-
ing a vastly enlarged and diffused regulative bureaucracy. To clarify this
issue, it is useful to look somewhat closely at a particularly important spe-
cial case: the pursuit of egalitarianism.
EGALITARIANISM AS A SPECIAL CASE
The realization of egalitarian arrangements is something that by its very

nature invites state intervention. For only action and coordination at the
grandest and most inclusive level can assume uniformity of procedure and
process. Thus, a great part of the modem state’s best efforts at control and
regulation is motivated by the pursuit of justice, fairness, equality of access
and distribution, and other such egalitarian goals.
It is at this point that the potential discord between desirability and ex-
pediency comes into prominence, for the pursuit of egalitarianism, doubt-
less laudable as an end, can run into trouble when the correlative measures
of regulation and control come to exert—as they inevitably must—a pow-
erful feedback impetus upon the process being controlled. Thus rent-
control measures can lose their effectiveness if they engender economic
distortions that eliminate the availability of rental property. Or again, there
is no point in promoting equality of access to medical services by means
that seriously diminish the quality or quantity of the services being deliv-
ered.
There is, in sum, no sense in striving for equality of access and distribu-
tion by means that degrade significantly the resources available for distri-
bution. A society can only distribute the goods and services it manages to
produce. But the egalitarian ethos of our time all too easily fails to deal re-
alistically with this factor of productivity. Consider the following two dis-
tributions of shares to individuals:
(a)
(c) (a)
Scheme I Scheme II (c)
(b) (b)
Patently Scheme I effects the fairer (indeed a perfectly fair) distribution.

From the standpoint of “justice in the narrower sense” it has the clear-cut
advantage. But that Scheme II is superior from the standpoint of the inter-
58
ests of all concerned is equally clear: everyone would be the gainer by its
adoption. Despite its less fair distribution, the advantage from the stand-
point of “justice in the wider sense” lies with this scheme. An adequate
theory for the evaluation of distributions cannot confine its attention to
fairness alone, but must also take into account the crucial factor of produc-
tion.
Examples of the sort previously indicated underline the critical impor-
tance of a principle of production for determining the expediency of regu-
latory and control measures. Due heed of these factors soon forces the
recognition that we must be prepared to acknowledge the superiority of
“unfair” distributions whose unfairness “pays for itself” by conducing to
the general advantage. We may well be prepared to tolerate discrepancies
in the fairness of a distribution in contexts where these could only be re-
moved by exacting an unreasonable price from all or most or the least
well-off of the individuals involved.
In this connection it is necessary to reemphasize the deliberation already
mentioned that the pursuit of inherently desirable ends can run into prob-
lems of expediency when it exacts too great a price in the attainment of
other desiderata. It can—and often does—happen that measures, that, in-
herently and in themselves, are incontestably desirable, create a systematic
feedback so that their operation exacts a price not commensurate with the
expected benefits. And this sort of phenomenon can totally undermine, on
grounds of expediency, the legitimacy of measures of control and regula-
tion that aim at inherently proper and desirable objectives. For as ample
experience shows, an otherwise well intentioned measure of regulation can
all too easily cause collateral damage that induces whatever good this
measure achieves.2
NOTES
1
In actuality this position is overly simplistic since its conception of rightness is too
legalistic and overlooks the underlying issue of justness. Certainly the measures of
control or regulation at issue could be unjust—in being discriminatory, for exam-
ple. For the methodological purposes of this essay we may, however, adopt a sus-
pension of disbelief in this regard.
2
This chapter was originally published under the same title in T. R. Machan and M.
B. Johnson (eds.), Rights and Regulation (Cambridge Mass: Ballinger, 1983).
59
Chapter 6
SOCIAL WELFARE:
SOME PHILOSOPHICAL ISSUES
MAN’S WELFARE
T he word welfare derives its meaning from the original sense of “having
a good trip or journey” and conveys the idea of traveling smoothly on
the road of life. The idea of the general welfare correspondingly relates to
what is for the public, or common, good, pro bono publico, as the Romans
put it. One dictionary defines welfare as the “state of faring, or doing well;
state or condition in regard to well-being: especially conditions of health,
happiness, prosperity, or the like.” On the negative side welfare contrasts
with the no longer common conception of welfare, of having one’s affairs
fare ill. It is, thus, clear that welfare is bound up with the essentials of a
man’s well-being, especially those within the range of application to which
we have become accustomed by such presently current terms as the “wel-
fare state” or a “welfare worker”.
The “welfare of a person” has a plurality of components, since welfare
is a thing of many dimensions. Preeminent among these are his physical
welfare (health), his material welfare (prosperity), and even his spiritual or
psychological welfare (state of mind or mental health). Physical health,
material circumstances, and mental and emotional well-being are the de-
finitive elements of welfare. The dictionary definition cited above is rather
misleading with respect to happiness. Happiness, clearly, is not a compo-
nent of welfare, but its goal. One feels that the man whose welfare is in
good order, who is healthy, prosperous, and secure is—or ought to be—
happy, being possessed of what general consensus regards as some of the
principal requisites of a happy life.
Significant complexity affects the concept of welfare through its psy-
chological dimension which raises a whole host of ramifications. Where
physical and material welfare alone are in question, the issue is a sub-
stantially simpler one, since these desiderata can be assured for a man in
isolation, without overt reference to his significant human interrelation-
ships. There is, however, the whole sector of a man’s relation to his fel-
lows, in the sphere of personal and close-range interactions (in family con-
tacts, professional interactions, friendships, and social interactions). As a
social animal a man’s own welfare is bound up with that of others. Welfare
is a matter of the basic requirements of well-being, and man is so consti-
tuted that he cannot achieve this condition without reference to the condi-
tion of others.
YARDSTICKS FOR WELFARE
It seems to me important to stress that the determination of the extent to

which the welfare of a person is achieved is an objective issue. This leaves
open the question of the criteria to be used—the yardsticks for measuring
welfare. It is clear that the multidimensionality of welfare will necessitate
use of a variety of criteria.
Before one considers this issue itself, a preliminary remark is in order.
Satisfaction of these various criteria is not part of the meaning of welfare,
any more than the meaning of acid or of “pressure front” has a part in the
behavior of the litmus paper or of the barometer we use for its determina-
tion. The criteria are not definitional components of the meaning of wel-
fare, but instead they represent tests for assessing the degree of a person’s
welfare.
The principal criteria by reference to which the degree of realization of
a man’s welfare can be measured are as follows:
A. Physical Health
1. life expectancy
2. physical condition (vigor, adequacy of functioning, absence of

pain, etc.)
3. external normality (absence of disfigurement or disability)
B. Mental Health
1. capacity to act effectively along chosen lines in chosen directions

(outside controlled, limited, or institutional contexts)
62
SOCIAL WELFARE: SOME PHILOSOPHICAL ISSUES
2. satisfaction with himself and his circumstances (absence of ab-

normal fears, animosities, etc.)
C. Material Prosperity
1. Income
2. negotiable wealth (money, gold, negotiable securities)
3. possession of goods (fixed and moveable)
D. Environmental Assets
1. availability of goods
2. availability of services
3. availability of amenities
4. quality of the environment (physical and natural, cultural, eco-

nomic, public health, etc.)
A long commentary could—and ideally should—be written about each of

these items, but we shall simply note them and pass on.
WHO IS THE JUDGE?
A key issue regarding welfare is posed by the question of who is best

qualified to make assessments about welfare. Who is the best judge of a
person’s welfare? The exploration of this problem will make possible the
clarification of basic aspects of welfare itself and will at the same time
point the way towards some of the very difficult issues involved with the
concept of welfare.
There is present in most of us a deep-rooted feeling that no one can bet-
ter judge the considerations involved in a person’s affairs—and above all
those regarding his own being—than the man himself. This sentiment is
supported and fortified by extensive exposure to the liberal democratic
principle that people are themselves necessarily the best judges of their in-
terests.
63
Fundamental to this issue is a distinction between subjective states on

the one hand and objective conditions on the other. Whether a man “feels
cold” or “is worried” or “feels well and healthy” are all subjective matters,
in contrast to whether it “is cold” (relative to what the man is accustomed
to) or whether he “has cause for worry” or whether he “is actually in good
health”, all of which are objective conditions that can obtain without refer-
ence to anybody’s subjective feelings. Now this distinction is critical for
our present problem. A subjective state is best judged by a person him-
self—no one else is in a better position to make judgments about it. But
whether that person finds himself in a certain objective condition is best
judged by persons with the clearest heads and the greatest quantity of rele-
vant information at their disposal, and this need by no means be the subject
himself.
A person’s welfare is a matter of objective conditions, not of his subjec-
tive state. His physical condition, his material prosperity, his mental
health—all these are objective circumstances that can be as well (or better)
known to others as to himself—his family doctor, his legal advisor, etc.
For the condition of his welfare—unlike a fleeting feeling of contentment
or happiness or elation—is not a subjective psychological state that the
subject himself is, in the very nature of things, best fitted to judge. Welfare
is not a matter of the psychological feelings but a question of the extent to
which certain objective circumstances are realized—circumstances gener-
ally regarded representing requisites for the achievement of happiness in
the contemporary life environment.
A clear sign that people are not to be regarded as automatically the best-
qualified judges of their own welfare is provided by the case of the
intellectually immature or abnormal—viz., children and madmen. It would
be patently inappropriate to regard all such people as invariably competent
to judge in matters of their own welfare. And this fact shows, at any rate,
that it is not in the inevitable nature of things that people should be
qualified to make judgments in this area. (Note that there is no doubt that
children and madmen are the best judges about whether they feel
discomfort or pleasure, and whether they are happy or miserable.)
The fact is that judgments of welfare are matters of (objective) knowl-
edge and not matters of (subjective) feeling. Thus, whether a person’s wel-
fare is in better—or worse—shape than it used to be is an issue about
which others may be better informed than he—his physician, his attorney,
his accountant, etc. Welfare is a matter of objectively determinable circum-
stances, and thus makes room for the entry of expertise.
64
What has been said here regarding the welfare of an individual holds
true also for that of a group. The members of a group need not themselves
be infallible authorities in matters of their own welfare. The “foreign” doc-
tor or economist or anthropologist may well know better than the “natives”
themselves that existing conditions are somehow detrimental to their wel-
fare and that their state of welfare would be increased by some change in
their modus operandi.
Recognizing that the external observer may be in a better position to
make an assessment of welfare than the subject himself, it must be stressed
that it is the subject himself, however, that is the center about which these
considerations revolve. Welfare considerations are not to be made from
some abstract, depersonalized point of view. Welfare certainly has its idio-
syncratic involvements—the subject’s own tastes, inclinations, personality
make-up, physical constitution, etc., are indispensable reference points for
welfare considerations. Welfare judgments will thus be not subjective, but
subject oriented: they require objective information about the subject. In
this regard welfare judgments are typified by medical judgments regarding
one’s state of health. Just as the patient himself is the definitive authority
about “how he feels”, so the subject is the definitive authority about
whether or not “he feels content and happy.” But this subjective element
settles the question of welfare no more than that of health. Neither in re-
gard to state of health nor in regard to state of welfare are feelings defini-
tive indicators of objective conditions. In both cases expert information
provides the crucial basis for judgment. But this information is, in large
measure, not a matter of information of general and universalizable sort,
but one of specific data regarding the characteristic make-up of the particu-
lar individual at issue.
Dangers inhere in welfare judgments made by outsiders on a subject’s
behalf. The dictator, the foreign sociologist, and the government official
are all prone to make horrible blunders when they tell themselves (and oth-
ers): “I know what’s best for the people here.” This misconception of fact
can lead readily to cynical abuse or to myopic paternalism. There is no
doubt that the man who makes decisions regarding the welfare of others is
in a position of power, and it is but too well known that power corrupts.
But to say all this is to say no more than that welfare judgments—like all
judgments—can be made badly. We insist upon our paradigm of the medi-
cal or economic or legal counselor. Here too, mistakes can doubtless oc-
cur—but in many or most cases are no more likely to come from a well-
trained, well-informed, and systematic outsider than from the subject him-
65
self.
RESPONSIBILITY FOR WELFARE
Who—apart from the man himself—is responsible for a person’s wel-

fare? It is clear that other people can be both legally and morally responsi-
ble for a person’s welfare. The basis for this statement can be natural: as a
parent is responsible for the welfare of a young child, or the child for that
of an aged parent. Or the basis can be contractual: captain, for crew and
passengers; general, for his troops. There are also borderline cases, e.g., a
host and his guest. In final analysis the relationship is always contractual in
its generic nature—either by an overt contract or by a “social contract”,
given implicitly by the rules of the society. Even the scope and nature of
parental responsibility are defined by societal arrangements.
One important question in this area is: Under what circumstances and to
what extent are others (morally) entitled to interfere when a person ne-
glects his own welfare? Obviously such entitlement exists in a substantial
way in the child-parent case—at any rate in our culture. But what of other
cases? The answer seems in general to reside in the principle that persons
are entitled to interfere in regard to someone’s welfare when his welfare is
systematically connected with theirs so that his neglect of his own welfare
endangers their interests (their welfare, their rights, etc.). The key example
here is represented by the collective welfare issues, where one man’s wel-
fare is bound up with that of others. Examples include the man who ne-
glects his health on a crowded ship or who lets his house become a fire
hazard in a crowded town.
Such collective welfare issues are especially prominent in modern soci-
ety. We are all living on a crowded ship. This fact puts great stress on the
collective sector of welfare. The conditions and circumstances of our
common environment are clearly an area of legitimate public responsibil-
ity—air- and water-pollution control, public health, crime control, etc. And
the area at issue here not only includes the prevention of negative but also
the promotion of positive aspects of natural and created environment—e.g.,
through the provision of services like television, transportation, and com-
munications.
What demands for the promotion of his welfare can an individual rea-
sonably make upon his society? No a priori answer is possible. The range
of legitimate demands depends crucially upon: (1) custom, and (2) law.
These criteria create a basis for legitimate expectation and claims. Thus,
66
the answer to our question hinges in significant measure upon what the so-
ciety decides it should be. But this decision is not just a matter of fiat. It is
definitely limited by what society can afford. But not just this, but (as was
already said) the society’s own decisions and decrees may well limit (re-
strict) society’s commitment, not because of stinginess, but because it es-
pouses a conception of “the good life” for man that itself delimits the range
of what others can and should do for a person’s welfare and to what extent
he should “fend for himself”. A society thus has at least two legitimate
grounds for not doing something on behalf of one of its members’ welfare:
• that it cannot afford the requisite action;
• that taking the requisite action would be against the best interests of
people in general (by destroying initiative, rendering people too de-
pendent, impeding or undermining socially desirable behavior pat-
terns, etc.).
When an individual makes otherwise reasonable-seeming demands upon

the society to act in support of his welfare, the society can legitimately de-
cline on either of these grounds.
How far is a society responsible for welfare of its constituent individu-
als? It is certainly not reasonable to view society as standing in loco par-
entis to people, with virtually open-ended responsibilities (at least not with
the parental concept predominant in the West). In this sphere society
should at most play the role of a parent of last resort: relation-less orphans
or those who pose an excessive or impracticable burden for their natural
guardians—the insane, handicapped, or illegitimate.
It is advisable to differentiate between first-line and second-line respon-
sibility for welfare. The person who has first-line responsibility is himself
to assure welfare of his subject; the person with second-line responsibility
merely helps his subject to care for his own welfare, which may mean the
creation of opportunities, a climate for initiative, a fruitful environment,
etc. A parent has first-line responsibility, but a society in general has only
second-line responsibility for the welfare of its members (except perhaps
for special categories—the insane, disabled veterans, etc.).
67
GOALS BEYOND WELFARE
A good part of the discussion prior to this point has been devoted to an
attempt to clarify the nature of welfare. The time has now come to criticize
it, to confront the question: Is welfare all? The issue of social goals beyond
welfare deserves careful consideration. Does the welfare-centered society
represent an ultimate ideal, or is it to be viewed as an interim stage enroute
to more significant objectives?
It is important to recognize that welfare has a certain minimality about
it. The components of welfare represent great, indeed almost indispensable
assets to “the good life”, but all the same, they furnish no more than the
beginnings of such a life. The good life is something whose range extends
far beyond the core issue of welfare. The man whose cultural horizons are
narrow, whose physical environment is unattractive, or whose government
is despotical may not actually suffer in any of the dimensions of his wel-
fare—indeed, he personally may conceivably even be every bit as “happy”
as though he lived otherwise. Nevertheless we could not qualify his as be-
ing “the good life”. Welfare is only the foundation of such a life; it is not
the structure itself. Physical health, material circumstances, and mental and
emotional well-being are enormous—perhaps even indispensable—aids
toward a meaningful and satisfying life, but they are not in themselves suf-
ficient for this purpose. The components of the good life must extend far
beyond considerations of welfare.
What things are in a person’s best interests? Certainly his own welfare.
And unquestionably also the welfare of others—his own family, his col-
leagues, his fellow citizens, and so on toward mankind at large. But a per-
son’s best interests go beyond the issue of welfare for self and others. As
already mentioned, they include such matters as educational and cultural
opportunities and attainments, attractiveness of physical surroundings,
pleasantness of natural environs, and openness and orderliness of social
and political arrangements—all these represent areas of a man’s best inter-
ests that go beyond the minimalities of welfare for himself and others.
Certainly in the case of an individual it is obviously appropriate and de-
sirable that he should have life goals that extend beyond his welfare and
that of his kindred. Then one contemplates the wide range of desirable hu-
man achievements and accomplishments it is quite clear that most of them
do not lie within the restricted confines of welfare. A man’s being educated
(along other than vocational lines), for example, or his taking an intelligent
interest in the arts are not matters that affect his welfare, but do all the
68
same represent important human desiderata. Of the seven deadly sins

(pride, greed, lust, anger, gluttony, envy, and sloth) at most three—viz.,
greed, lust, and gluttony—have a direct and immediate connection with
material welfare.
The situation with regard to social welfare is exactly analogous. Welfare
has to do with the material requisites for a satisfying life. But this very fo-
cus on the material is the key indicator of the limited and restricted scope
of the whole issue of welfare. A people can enjoy a good standard of liv-
ing, material prosperity, a good standard of public health, etc. and yet be
lacking in education, culture, artistic and scientific creativity, and the good
manners and decencies that make for a civilized life. A society whose wel-
fare is in a salubrious condition can all the same lead a low, degraded, and
in some respects impoverished existence. To set up welfare as the social
goal is folly. Welfare is all very nice as far as it goes, but it is certainly in-
complete. The Biblical dictum that “man doth not live by bread alone” can
be applied to welfare also. (Though, of course, this is not to say that he
who lacks bread—or other requisites of welfare—is in a position to devote
himself to these “higher” matters.)
Again, for the individual person there are, or should be, important goals
beyond welfare. Examples of such goals include attaining the respect of his
fellows and the love of some among them; pride of achievement in some
areas of activity, vocational or avocational; and appreciation of man’s ac-
complishments in art and science. The achievement of such goals lays the
basis for a legitimate view of oneself as a unit of worth. And a significant
and general lack in these regards is indicative, not necessarily of any dimi-
nution of welfare, but of an impoverishment of spirit. In consequence peo-
ple as individuals have (i.e., can, should, and do have) a wide spectrum of
trans-welfare goals—goals the attainment or progress toward which is
definitely to be viewed as “in the interest” of a person, even though it need
not be his welfare that suffers by a deficiency in this regard.
In a wholly parallel way there are legitimate goals for a society that ex-
tend well beyond the region of the welfare of its members: in the cultiva-
tion of literary, artistic, and scientific creativity and appreciation; in the
forging of an attractive and comprehensively pleasant life environment; in
the cultivation of enjoyable human interactions; and in preserving and en-
hancing the appreciation of its historical, cultural, and intellectual heritage.
A society can and should look beyond the physically comfortable life
comprising its welfare to the fostering of those conditions and circum-
stances that make life rich, full, and rewarding. The material factors at is-
69
sue in welfare are only the springboard from which a society should move
on to more deeply meaningful objectives.
From this standpoint it becomes obvious what answer is to be given to
our question of whether the welfare-centered society is an optimal Utopia
or whether it has intrinsic weaknesses and shortcomings. The position we
have been developing points immediately to the second situation. The vi-
sion of the welfare-centered society does not extend far enough. The con-
cern with welfare is all very well as far as it goes, but welfare is myopic in
failing to see beyond the range of immediate human needs—perhaps even
“needs” that have been substantially escalated in an affluent society—to
the less mundane, but in the long run no less important desiderata that lie
beyond.
We have only given some random examples of legitimate trans-welfare
goals. Broadly speaking, all these have related to matters of excellence, of
creative achievement, and the enhancement of the quality of human life.
Now it might be objected: “Granted that these trans-welfare objectives are
of value in the long run. But in the here and now, in the short run, they are
luxuries. What concern has a society with the pursuit of these trans-welfare
goals until after the first item of its agenda—the welfare of its members—
has been properly attended to?” To this critic we reply that his reasoning is
misleading and mistaken. There are priorities that cannot be translated into
sequences. A physician cannot tell a man: “Postpone exercise until after
your diet has been perfected.” It makes no sense to advise a student to put
French off until after he is perfect in Latin. A society must be prepared to
cultivate the pursuit of and the taste for excellence even amidst material
adversity; otherwise the taste and the capacity for its achievement will both
be dead before the day of material prosperity arrives. The plant that is not
nurtured through the winter cannot flourish in the springtime. The trans-
welfare goals cannot be made over into post-welfare goals because such a
temporally sequential ordering of priorities can render their realization very
difficult or even impossible.
THE POST-WELFARE STATE
In the present condition of affluence of the technologically advanced,

modern societies we can already discern the making of the post-welfare
state, able, if willing, to devote a substantial fraction of its resources and
energies toward the realization of what I have characterized as trans-
welfare objectives. Such a state will differ from the welfare state as it has
70
developed in Great Britain, Sweden, and even the United States in primar-
ily three ways as regards its basic value structure:
• The prime focus of emphasis will be shifted away from catering to

the basic human needs of the welfare complex to the forging of more
pleasant operating conditions for the conduct of life under the
crowded circumstances of modern times (especially in the urban cen-
ters).
• A higher priority will be given to goals of the cultural-creative sec-

tor. It will come to be recognized in a more emphatic way than ever
before that society has a commitment to and responsibility for the
level of culture, higher education, and the development of the arts.
Society and its instrumentalities will take an increased interest in the
institutions dedicated to promoting appreciation of the arts, the pres-
ervation of antiquities, etc. There will be a renewed stress on excel-
lence and creativity.
• The great surge of nationwide uniformity typical of the mass orient-

edness of the welfare state (mass production, mass media, mass poli-
tics) will give way in the face of tendencies of fragmentation. There
will be renewed emphasis on regionalisms, parochialisms,
“participatory democracy”, and decentralized group action of many
sorts. There will be a renewed surge of diversity-in-unity and
cohesiveness-in-variety. The post-welfare society will combine unity
with diversification, and will be a genuine unum e pluribus.
These value reorientations have far-reaching implications for the political

modus operandi of the post-welfare state. The welfare state as it has devel-
oped in the European-North American setting is based upon a comprehen-
sive hierarchical bureaucracy and a highly centralized control structure.
Unified coordination and control have been virtually inevitable because of
the nature of the problem with which the welfare state has been primarily
designed to cope. Three factors have been especially significant here:
• The bulk of the problems confronted by the welfare state has been
economic in nature or in origin, and a nation’s economy lends itself
with relative ease to centralized direction and control. (In this regard
71
a nation’s economic life stands in contrast to its socio-psychological

and its cultural dimensions.)
• Prominent among the most fundamental problems of the welfare

state have been the issues of economic and political egalitarianism.
The results of equality and across-the-board equity can clearly be a
centralized uniformitarianism.
• The issues with which the welfare state has been concerned have
been ones that can be tackled effectively by the legislative process.
They revolve about issues of control and are, thus, of such a kind as
to be responsive to centralized direction.
By contrast the critical issues confronting the post-welfare state will be of a

different order. The problems that confront the society it serves will, as we
have submitted, in the main lie, not in the economic sector, but in that of
the cultural and socio-psychological aspects of life. As such they are most
effectively and naturally amenable to a decentralized treatment better
adapted to accommodate regional and communal differences. In conse-
quence much of the structure of the post-welfare state will become multi-
cellular rather than monolithically centralized. The “table of organization”
of the many sectors of the post-welfare state will look like a spider’s web
rather than an inverted tree. A substantial realignment of the channels in
which power and authority run will be called for, involving a marked local-
ization of control—a shift away from the pattern of control imposed
downward from on high in the centralized hierarchy, toward a relocation of
authority in local units with the resultant capacity for the implementation
of local initiative. The individual member of the post-welfare state will
have to be closer—and will feel closer—to the effectively operative
sources of authority and control—so close that he can touch their levers
with his own hands. This is the pattern (already to be discerned in a rudi-
mentary way) in present towards a shift from the bureaucratic centralism of
the welfare state towards the “participatory democracy” of its post-welfare
successor.
The welfare state has occupied a unique historical position. The princi-
pal social goals of the era, the main issues on the agenda of the society
(economic growth, economic justice, political and economic egalitarian-
ism, the forging of adequate “social services”) were such as to be effi-
ciently and effectively handled at the level of the state. But in the more
72
affluent and socially more complex environment of the post-welfare state,

this circumstance no longer obtains. The “issues of the day” will in large
measure be such as to lie outside the sphere of effective state action. The
state as such will become increasingly less relevant to the society’s prob-
lems, and the society will readjust itself to ways of achieving its goals in
ways other than through the instrumentality of the state. This is not to say,
of course, that the state will “wither away” in the manner familiar in the
dogmas of theoretical communism. The state and its historical functions
will certainly survive. Just as the traditional security functions of the state
survived (and expanded vastly) with the development of the welfare state,
so the established welfare function of the state will survive (and conceiva-
bly expand) in the era of the post-welfare state. The point is that the social
environment of the future will see the evolution of new forms of social ac-
tion—largely outside the sphere of the state, to a degree that calls for inno-
vative structures—to cope with the “problems of the day”. The public
agenda of the post-welfare state—its schedule of burning issues—will be
such that the instrumentalities for effective handling call for a combination
of (1) a revision of decentralized redeployment of state activities together
with (2) the evolution of new, increasingly influential and important non-
state approaches.
A major—perhaps the main—problem (or, rather, family of problems)
that faces the post-welfare state is that of untying the knot of the city, of
devising the means to tame and civilize urban life. Everyone is familiar
with the catalogue of problems: dirt, noise, physical ugliness, crowding,
crime, friction in interpersonal contacts. And we all know the results: per-
sonal demoralization, group tensions, and the social alienation that comes
about when people interact with others not as people but as parts of a sys-
tem. A prime task of the post-welfare society is to devise means of secur-
ing pleasantness for urban life—to restore convenience to city living, to
make the city physically attractive, and (as far as possible) to break it down
into units of human scale, to facilitate civility in interpersonal dealings.
Here, then, we have one typical example of the reorientation of the fo-
cus of society’s attention that will characterize the situation of the post-
welfare state. In the era of the welfare state, society’s attention has been di-
rected to the basic considerations needed to make life viable on a mass ba-
sis in the context of a developed industrial economy. With the post-welfare
state the prime task will be that of making the circumstances of life richer
and fuller for a mass population in the context of a “post-industrial” econ-
omy oriented heavily toward the production of services rather than goods.
73
In brief, the welfare state sees its problem in terms of creating a public con-
text that makes life in a mass society more secure; the post-welfare state
faces the problem of creating a public context that makes it more pleasant.1
NOTES
1
This chapter was originally published in John D. David (ed.), Value and Valuation
(Knoxville: University of Tennessee Press, 1972), pp. 221-32.
74
Chapter 7
COLLECTIVE RESPONSIBILITY
THE ISSUE
I s the U.S. military at large to blame for the massacre at My Lai? Do

Americans in general deserve reproach for the plight of the country’s In-
dian reservations? Whom can one fault for the decline of civility in Amer-
ica’s cities or the slippage of mathematical competence in its high schools?
It is problems of this sort that are at issue in the present deliberations,
whose focal questions are as follows: What is involved in a group’s being
responsible for producing a collective result? What conditions must be sat-
isfied for it to be appropriate for us to praise or blame a group for some re-
sult of its doings?
In addressing this problem, the natural step is to begin by basing our
understanding of group responsibility upon that of individual responsibil-
ity. Now here, at the level of individuals, responsibility for a result is
clearly a matter of producing the result through one’s own deliberate
agency—barring the intrusion of defeating aberrations such, for example,
as duress or deceit. For responsibility to enter in, it is not enough that that
untoward result be produced, as the causal result of an individual’s actions,
it must also be intended in some appropriate fashion. Carrying this idea
over to groups, one thus needs to address two pivotal issues: group agency
and group intention. Unfortunately, neither is anything like as simple and
straightforward as one might wish.
PROBLEMS OF GROUP AGENCY
The “responsibility” at the focus of the present discussion will be of the

sort that opens the door to evaluative and normative appraisals, so that
praise and blame will be pivotal considerations.1 The merely causal “re-
sponsibility” of productive contribution is of course neutral in this re-
gard—a merely necessary but not sufficient condition. For authentic
responsibility some element of intentionality must always attend to causal

participation—the factor of productive intentionality is crucial.
All the same, causal or productive involvement is an important part of
the picture. And as regards causality, the first thing to note is that a group
can bring it about that a certain result obtains without any member of that
group having any significant or substantial contributory relationship to the
production of that result as such. Thus if every dweller in a town adores his
neighbor’s cats and hates his dogs, they can collectively produce the result
that all the town’s cats are loved and all its dogs hated, even though this
may occur in such a way that no one ever desires—or even contemplates—
this result. And the same sort of thing happens if every driver happens to
be out on the road and a traffic jam results. Moreover, in such a case no
one contributes more than a minute share to the result, and sometimes, in-
terestingly, the smaller the individual shares—the more people on the road,
say—the worse the result. Such situations are legion; groups regularly
manage to produce results with which, as such, their members have little or
nothing to do.
The fact is that when a group collectively produces a certain result, none
of its individual agents need do anything that bears significantly on that re-
sult as such. Indeed, often this result is something the individuals could not
facilitate if they wanted to—which they well may not. Consider: “The affi-
davits made by Tom and Jerry created a conflict of testimony.” Clearly nei-
ther agent—acting on his own—did anything that was in and of itself
conflict-engendering. Again, suppose that a hardware store carries ten
hammers and that ten customers come along and buy them. Among them
they have “exhausted the store’s supply of hammers”. But of course noth-
ing of this sort figures in any of their thoughts or actions. In all such cases
it makes no visible difference as far as the individual agent is concerned
whether that contributory act obtains in isolation or in an unwitting, merely
fortuitous concert with other agents for the production of an untoward ag-
gregate results. The inherent status of contributing individuals’ doings are
clearly not affected one way or the other by the essentially accidental cir-
cumstances of context. And even should it eventuate that the collectivity is
a disaster, the fact remains that the individual’s act may well be perfectly
innocuous—absent any or all personal foresight and understanding in rela-
tion to the aggregate end-product. Neither in production nor in contempla-
tion need a group’s individual members play a significant role in such
cases.
76
The circumstance that an individual’s act constitutes a contributing part

of a certain overall result represents a feature of that act all right—but a
contextual feature of it. And this context is in general something over
which the individual in question has no control nor even significantly pro-
ductive influence. (Think, for example, of the “wave” created by spectators
at a sporting event, where each individual simply stands up and raises his
or her arms—or, even more drastically, think of what happens when a for-
eign expression enters a language as a “loan word”.) Moreover, the con-
tributory aspect of the action may well lie entirely outside the individual’s
awareness. (That individual’s contribution to the wave may simply be an
aping of his neighbor in a mere social conformity.)
PROBLEMS OF GROUP INTENTIONS
Suppose that someone comes across a nearly unconscious sufferer on a

country lane. After providing an effective but rather slow-acting medical
remedy, he goes off to get further assistance. Then another person comes
along and does exactly the same. But these remedial doses, though indi-
vidually helpful and appropriate, nevertheless create a jointly fatal over-
dose. The two people acting together have, in effect, caused the sufferer’s
demise by their well-intentioned actions. And this is the polar opposite to
what they intended or expected. This “Good Samaritan Mishap” illustrates
the sort of thing that can—and often does—happen with interactively pro-
duced results, where product and intention can readily diverge.
For intention and responsibility, then, distributive activities that collec-
tively engender a result by way of causal production are clearly not
enough. Overt purposiveness in relation to this result must be added. Sup-
pose that X wants p, Y wants q, and that they act accordingly. Joined to-
gether the group X-and-Y wants p-and-q. Yet neither X nor Y may want this
outcome or anything like it. (Indeed, wanting it may be senseless, as when
q happens to be not-p.) Accordingly, the overall upshot may well be some-
thing seen by both as eminently undesirable, as when X wants to kill Y, and
Y to kill X. With groups of agents, the separate intentions of individuals
cannot simply be combined. Here wholes that have features which, as far
as intentionality is concerned, are nowise mere sums of the parts.
Group responsibility clearly calls for coordination and depends on the
extent to which the group acts as a unit within which the actions of indi-
viduals are concerted. With the product of a merely fortuitous confluence
influence of individual actions (e.g., a bank run), group responsibility is
77
clearly out of the picture. The element of collaborative coordination—of a

“conspiracy”, so to speak, in producing the result—is an indispensable
requisite. The difference between acting as a coordinated group and acting
as only a collection of disjoint and disaggregated individuals is crucial
here.
With group responsibility as with individual responsibility, this factor of
intention is critical. The weaker the element of intentionality, the weaker
that of responsibility. Thus consider the series
• active and deliberate participation in the production of a result
• passive agreement in its production
• detached spectatorship
• reluctant acquiescence (going along)
Responsibility clearly fades away as we move down the line here. The
closer the “degree of association” of the individual with the production of a
collective result, the greater the responsibility (and the greater the extent of
blame or credit).
From the causal point of view, individual contributions appear to sum
up productively. The individual agents make their causal contributions to
the whole, and the whole consists of the sum of the parts—causally speak-
ing. But intentions certainly do not work this way. Groups regularly bring
things about that none of their members plan, intend, or indeed even ever
envision. Every cowboy just wants to kill his few buffalo—no one con-
templates extermination of the species. There is no way to sum up individ-
ual intentions to underwrite the imputation of an aggregate intention vis à
vis the overall result. The denizens of the city of London rebuilt their city
after the great fire of 1661, and the inhabitants of Charleston, South Caro-
lina did the same after the catastrophic destruction wrought by Hurricane
Hugo in 1991. Collectively they accomplished the task, but distributively
each property owner simply addressed the problems of his or her own
situation. Among those who were active in the rebuilding, virtually no one
had any intentions in regard to the bigger object—the intentions of each
were for the most part focused on the particular micro-task at hand. The
overall macro-achievement lay outside the reach of anyone’s intention. A
group’s interactively produced macro-results are all too frequently uncon-
78
templated, unforeseen, unplanned, and even undesired by many, most, or

even all of that group’s individual members.
And so, while the responsibility or intention of groups must indeed de-
rive from and inhere in that of its constituent individuals, it will actually do
so only in special circumstances. The coordinative factor of an at least vi-
carious consent must be there: absent individual participation (of an at least
statistical, majoritarian sort), there must at least be centralization through
representation. The fact of the matter is that without an appropriate coordi-
nation of individual intentions it makes no sense to impute intention to a
group.
It makes sense to say that a collectivity wants or intends something only
when we have either
i. consensual subscription: when the macro-objective at issue is some-

thing that the generality or the substantial majority of group members
want as such, thus rendering it into an object of the collective volonté
générale, so to speak.
or
ii. representative endorsement: when the duly delegated representatives

of the group’s members duly agree to producing the result in ques-
tion.
To be sure, distributive coincidence is not sufficient to yield the sort of in-

tent at issue. The element of collectivity must be present. And there must
be the right sort of coordination. The members of a board may distribu-
tively happen to be of one mind in all wanting the chairman dead, but this
does not mean that the one who goes and shoots him is implementing a
group consensus. Their “wanting him dead” does not come to “wanting
him killed”, let alone “wanting him killed by X.” Here those who have re-
mained inert have certainly not “agreed” to the murder simply in view of
their (perhaps reprehensible) attitude. The members may “want him dead”
but do not agree to his being rendered so by the sort of action at issue. For
collective responsibility the group must constitute something of a “moral
person” with a collective unity of mind.
And the conjunctivity at issue means that whenever group responsibility
does in fact exist, there must be responsible individuals: group responsibil-
ity cannot exist without individual responsibility. And this need for a
proper grounding in the responsibility of individuals means that it cannot
79
happen that a group does something wrong without there being culpable
individuals at whose door some of the blame can be laid. (Note that it is
crucial for the tenability of this statement that it reads “something wrong”
and not merely “something bad”.) Group responsibility must have a root-
ing in the responsibilities of individuals and cannot manage to exist with-
out this.
All the same, collective responsibility is just exactly that—collective. It
emphatically does not function distributively—it cannot automatically be
projected upon the individuals who constitute that collectivity. We can in-
deed reason from “Tom, Dick, and Harry talked about mathematics” to
“Tom and Dick talked about mathematics.” But we can no more reason
from “Tom, Dick, and Harry carried the piano upstairs” to “Tom and Dick
carried the piano upstairs” than we can reason from “Tom, Dick, and Harry
filled up the sofa” to “Tom and Dick filled up the sofa.” Only in very spe-
cial cases will the doings of collectivities project down to their component
units.2
CONSEQUENCES FOR COLLECTIVE RESPONSIBILITY
With these considerations about causation and consent in mind, we can

profitably return to our initial question. When the individual actions of di-
verse agents collectively issue in an overall result, there is, by hypothesis, a
collective causality in point of productivity. But does moral responsibility
follow from this? Is there also automatically room for guilt and credit,
praise and blame, laudation and reprehension?
Given the aforementioned complexities of collective intentionability,
the answer is clearly a negative here. For this reason, it would be folly to
argue: The individually intentional actions of the members of a group pro-
duced a certain result, hence the members of the group are individually re-
sponsible for that result. Exactly as in the accidental overdosing case
considered above, no single agent in a group whose acts are conjointly dis-
astrous need do anything wrong or blameworthy.
Without the requisite coordination of individual intentions through con-
sensus or delegative authorization it makes no sense to speak of group in-
tentions. So-called “guilt by association”—by merely being a member of a
group that collectively produces a bad result—is just not enough to estab-
lish a valid imputation of responsibility. In a bank run every depositor just
wants his own money from the bank—no one foresees or intends the ruin
and bankruptcy that ensues. But who is responsible—“everybody and no-
80
body”—and so in the final analysis no one. Except in special conditions

and circumstances—when the proper sort of coordination obtains—the in-
tentions of individuals simply do not aggregate into some sort of group
product: they do not somehow blend together into a group intention. And
where intention is absent, there too responsibility is missing. When a group
produces an unintended disaster, the situation as regards culpability is ex-
actly the same as when an individual produces a wholly unintentional dis-
aster—to wit there simply is no culpability. Without a normative responsi-
bility that transcends more causality, there can be no actual guilt. When
this sort of thing happens, we can regret but cannot reproach.
The pivotal fact in this connection is that responsibility for the collec-
tive transgressions of groups can be projected down upon its component
individuals only in special conditions. For only where group malfeasance
indeed roots in the informed consent of individuals through coordination or
delegation can those individuals be held to blame. And even then only sub-
ject to limits. For to be culpable, individuals must form part of that consen-
sus and be party to its consent. The defenses “I was opposed to it” and “I
cast my vote against it” must be allowed their due weight in matters of in-
dividual exculpation.
Escaping group responsibility by dissociation is thus possible through-
out the spectrum from explicit abstention to actual opposition. The ideo-
logical impetus of the ethically inspired resisters to Nazism is a case in
point. Their actual efforts were futile and unavailing (perhaps often even
incompetent). But symbolically they are of the greatest significance in pro-
viding a highly visible token of the fact that the German people in general
did not go along here, so that the moral properties require an explicit line
to be maintained between Germans and Nazis.
On the other hand, however, individuals can indeed be held responsible
for group actions even if in a strict sense they “didn’t have anything to do”
with the malfeasance at issue. For when their intentionability is betokened
by way of consent or consensus, then those evil acts which its agents (few
though they be) actually perform on behalf and under the consensual aegis
of the wider group will also fall into the responsibility sphere of its indi-
vidual members. Association is not enough to establish responsibility, but
the sort of association involved in being an “accessory” of sorts—be it
consensual or delegative—can prove sufficient. And even tacit consent can
do so. The acquiescing citizenry is indeed responsible for the authorized
actions of its duly delegated agents.
81
THE LEGAL ASPECT: MORAL VS. LEGAL RESPONSIBILITY
A thorny question now arises. If an appropriately structured coordina-

tive basis in individual responsibility is indeed required for the moral re-
sponsibility of a group, then what of collectively produced outrages that
are uncoordinated?
Two contrasting positions are possible here. At the one pole there is the
“Protestant” position that individuals are the prime (perhaps even the sole!)
bearers of responsibility. The position is that agents stand on their own feet
in matters of evaluation appraisal. And so, when a group produces a collec-
tive result, then its individual agents are responsible only for their own in-
dividual acts—and so only for their free and intended individual
contributions. They are responsible for and creditable with only those
negativities and positivities that they themselves engender through their
own suitably deliberate actions. Contextual considerations can be dis-
missed from the moral point of view. In practical effect we can simply for-
get about group responsibility as a distinct issue: moral deliberation can be
limited to the domain of individuals; group evaluations are at most statisti-
cal summaries.
At the other pole is the “Hebraic” position that the community is the
prime (perhaps even the sole!) bearer of responsibility for collectively pro-
duced results—that what the group collectively does can and must be laid
unavoidably at its collective door in point of responsibility. (To be sure,
whether this collective responsibility can then be downloaded upon the
constituent individual of the community still remains as an additional and
potentially controversial question, but the responsibility of the community-
as-a-whole remains in any event—irrespective of how we answer this addi-
tional question.) In effect, the group is thus treated as a responsible indi-
vidual in its own right in a way that is essentially independent of the
responsibilities of its members. Groups are morally autonomous: they
stand on a collective potting—the idea is simply rejected that the responsi-
ble actions of groups must inhere in or derive from its individual members.
We thus come to the question: As regards specifically moral responsi-
bility, which is the right line to take here: the “Protestant” or the “He-
braic”? At this point the question of law vs. ethics comes to the fore.
To start with, it deserves to be noted that in actual practice we generally
allow legality and morality to go separate ways. And we do so for very
practical reasons. Some examples will help to make this point.
82
Strictly speaking, the person who drives home drunk after the office
party and has the good luck not to have an accident that injures others is in
exactly the same moral position as the person who fails to be so lucky.3
But legally there is all the difference in the world here: legally the one is
(so we suppose) guilty while the other is altogether guilt-free. Here the le-
gal standing of the two is thus very different. The law is concerned with
actual results in a way that morality is not. Again, military law holds the
commander responsible for mishaps for which there may well be no actual
causal responsibility at the personal level. So here there will be situations
where one will be morally innocent but legally culpable. And again, in
group punishment situations, one charges the “innocent” members of the
group with an onus of responsibility that has no moral basis. Here too the
law can reflect the society’s pragmatic care for results in a way that by-
passes moral complications.
The fact is that the law is part of a system of social contract that has
other fish to fry than that of fixing moral culpability, and in consequence it
often insists on beneficial overall results at the expense of strict justice.
The pivotal point is that legal and administrative systems embody a con-
cern for certain social desiderata distinct from strict justice per se. This cir-
cumstance makes for a crucial difference between moral and legal
responsibility.4
From the moral point of view the proper line will have to be that the
source and basis of responsibility is always with individuals. Thus moral
responsibility belongs to groups only insofar as the individuals are suitably
active within them. Groups can only bear responsibility derivatively—
either by way of aggregation (consensus) or by way of delegation (via rep-
resentation). And moral responsibility is in a way inalienable. It remains
with those causally contributing individuals even when they transmit it to
the group with which they act. But with legal responsibility, the situation is
different. Groups can be legal persons and thus bear legal responsibility.
For legal responsibility is alienable and capable of transfer and delegation.
Forming a corporation (or “legal person”) or imposing a collective sanction
on a criminal group or a destructive society makes perfectly good sense.
In this contrast between moral and legal responsibility we thus find a re-
flection of the contrast between the aforementioned “Protestant” and “He-
braic” positions on responsibility. And as far as moral wrongdoing is
concerned, the “Protestant” position is surely plausible: the moral culpabil-
ity of groups must inhere in that of their individuals—with the result that
83
there can be collective acts that are unfortunate and regrettable, but yet not
wrong, owing to the absence of any personal wrongdoing.
SOME LESSONS
An instructive lesson emerges from these deliberations. If as was in-

sisted above (i) the only avenue to group responsibility/reprehension is in-
deed via group intention, and moreover (ii) group intention requires
coordination—either through an aggregation afforded by consensus or
through a centralized delegation to representative deciders—then it follows
that (iii) there will unavoidably be many instances where group-
engendered outrages will “fall between the stools” as far as responsibility
and reprehension goes. And this means that there will thus be group-
engendered catastrophes responsibility for which admits of no specific al-
location to individuals, since it is a “merely fortuitous confluence of indi-
vidual actions” that does the mischief.
Who, then, can be held responsible for the carnage on America’s roads,
the poor performance of its schools, the decay of its social conscience and
its public civility? Clearly only those who bear some sort of immediate re-
sponsibility. Reproaching the group-as-a-whole makes no sense here, see-
ing that the requisite element of interpersonal coordination is lacking. Such
aggregate negativities are the confluence of the uncoordinated and dis-
jointed actions of innumerable individuals. They result from the causal
contributions of many but the intentionality of none. And this lack of inten-
tionality precludes the availability of actual culprits.5
Those aggregate effects which come to be realized through the vicissi-
tudes of context lying beyond the ken and control of the individuals in-
volved have to be viewed as “accidents of circumstance” with respect to
which the chain of moral responsibility is severed en route to an aggregate
causal result. As the case of the “Good Samaritan” overdose illustrates,
those terrible overall results have come about through the intentional do-
ings of individuals all right, but there is nothing intentional about them as
such. The only relevant intentions were fragmentary and disjointed—and
so for this very reason were the responsibilities of individuals. That overall
catastrophe was never envisioned—let alone intended—by anybody. The
responsible actions of people produce disaster, all right, but a disaster
which, as such, is detached from the responsibility of individuals (in any
sense over and above the causal).
84
Thus the perhaps unwelcome fact of the matter is that individually the
actions of people can still be blameless—and perhaps praiseworthy—even
in cases where the collective, combined result of their actions is a disaster.
(The destabilizing rush of people to the side of the boat from which a cry
of “save me” emanates is perhaps something of an example.) When such
aggregated mishaps occur as a causal result of people’s unconcerted ac-
tions, there is nothing that those disaggregated individuals can be held cul-
pable for. There is no occasion for blame allocation: here collective actions
can engender aggregate outrages that are entirely culprit-free as far as indi-
vidual agents are concerned. The prospect of a lack of any suitable basis
for attributing a communal intent means that the action of groups can
sometimes produce terrible results for which there is neither collective re-
sponsibility nor individual fault.
What we have here is a fact of life that moral philosophy and common
sense alike simply have to take in stride. For in this regard, group causation
is like nature causation—the group in effect acts like an unmanageable
natural force rather than a voluntary personal agent. There is no sense in
blaming the chair that we stumble over in the dark. When uncooperative
nature produces a bad result, there simply is no one who can plausibly be
asked to bear the burden of reproach. We regret the result but cannot find
someone to blame for it. It’s just “one of those #!@?* things” that we have
to come to terms with in a difficult world. And much the same thing has to
be said when unhappy aggregate effects come about through the disaggre-
gated actions of members of a group. Morally, each individual can, should,
and must bear responsibility for his or her own individual acts and inten-
tions. But, to reemphasize, the aggregate effect—however unfortunate—
may prove to be just one of those unfortunate things.6
This, at least, is how matters stand from the moral point of view. And
there is a significant lesson here. Causal and moral responsibility behave
very differently in situations of collectivity. By hypothesis, an agent whose
intended actions play a contributing part on the side of causal production
will thereby and for this very reason bear a share of causal responsibility in
relation to the overall product. But of course moral responsibility is not
like that; it is not simply a matter of aggregation. For here the whole can be
less than the sum of its parts—or more. No causal collective results can ex-
ist without individual causal contributions. But collective morally negative
or positive results can indeed emerge in situations where individuals make
no personal contributions of a morally positive or negative coloration.
85
But what about group responsibility for faults of omission—for culpable

inaction? The pivotal considerations here are a failure to act in the presence
of opportunity, when this matter of opportunity pivots on (i) the existence
of a suitable occasion where action is called for, (ii) an awareness that this
is so, be it by the group at large or those duly responsible for the conduct
of its affairs, and (iii) the availability of the requisite means for action. On
this basis, the culpable inaction of groups is again something substantially
analogous to the situation with regard to individuals. Still, when groups
neglect doing something that they (morally speaking) ought to do—that is,
to do as a group—can the blame for these omissions be laid at the door of
its individuals? The answer is YES-BUT, namely: but only under special
conditions obtaining when that group neglect is the result of a culpable (or
“inexcusable”) oversight on the part of individuals. If, as maintained
above, it is indeed the case that group responsibility must inhere in indi-
vidual responsibility and cannot exist without it, then this will also hold in-
sofar as responsibility for omission is concerned. (Q: But just which
individuals are responsible for the group’s default? A: Exactly those whose
intentionality was causally involved. Intentional causation is again the
crux—absent the usual array of responsibility deflectors.)
A REVIEW OF THE ARGUMENT
The preceding deliberations, though brief, tell a rather complicated

story. It is accordingly useful to pass the overall argument in review:
1. Responsibility presupposes (i) productive agent-causality, and (ii)

deliberate intentions.
2. The productive causality of groups can issue from the entirely dis-
connected and uncoordinated agency of its constituent individuals.
There are accordingly two modes of group-productivity: (i) the actu-
ally coordinated, and (ii) the uncoordinated and “accidentally” con-
fluent as it were.
3. For the intentionality of group products there must be a coordinative

synthesis of the individuals’ intentions. Such coordinative cohesion
can take two forms: the informally consensual or the formally repre-
sentational. In the former case we have the explicit agreement of (at
least most of) the members; in the latter case we have the imputed
86
consent of the members at issue via the mediation of representational

institutions.
4. Where the products of group activity are concerned, it only makes

sense to speak of group intentions in the case of coordinated produc-
tions. Without the synthesis or unification of actions there is no
meaningful collective intention.
5. An absence of intentionality means that group actions may be disas-

trous without there being any wrongdoing on the part of individuals.
6. However, group actions cannot be wrong apart from individual

wrongdoing. For—
7. Intention—and thus responsibility—must initiate with individuals.

Groups can achieve such a condition only “by derivation”—that is,
via the mechanics of consensuality and/or delegated consent. Ac-
cordingly—
8. Group intention/responsibility therefore exists only with coordinated

group products produced under conditions of a synthesis of individ-
ual intentions via consensus or delegation. Then and only then is it
proper to project group responsibility unto its component individu-
als—and only to the extent that their own intention were causally in-
volved.
CONSEQUENCES
One significant consequence of these deliberations is that group respon-

sibility can lapse into nothingness where rogue regimes abrogate the nor-
mal processes of representative government. The people of Uganda cannot
be held responsible for the excesses of Idi Amin’s regime, nor the people
of the U.S.S.R. for those of Stalin, whose transformations of governmental
organism through state terrorism they neither foresaw nor endorsed. On the
other hand, the responsibility of Americans for the atomic bombing of Ja-
pan in WWII or for the defoliation of crops and the destruction of villages
in the Vietnam war cannot be denied—at any rate not on this basis of
flawed intentionality. (But, as noted above, this collective responsibility of
the group does not automatically devolve upon its individual members.)
87
At this point an interesting question arises. For group intentions we

have appealed to two coordinative factors: distributive consensus and cen-
tralized consent through representational institutions. But what if these two
get out of joint with the one pointing in one way and the other in another
one. What becomes of group responsibility when the group’s legally con-
stituted representatives act in the face of a general consensus to the con-
trary? For example, should one impute to the people of Britain credit for
abolishing the death penalty or reprehension for tolerating fox hunting,
seeing that Parliament’s position on these issues is decidedly out of phase
with the opinion polls?
There is—there can be—no simple answer here. What we have is yet
another illustration of the fact that conceptual tidiness cannot be secured in
a difficult world. In such situations we confront complex questions that re-
quire complex answers.
APPENDIX: COLLECTIVE CREDIT
The preceding deliberations have mainly taken into view the negative
side of evaluative responsibility in regard to blame and guilt. But of course
there is also the positive side of praise and credit. And it is—and should
be—reasonably clear that the overall situation here must be regarded as be-
ing substantially analogous in this regard.
However, the analogy is not complete. An important and interesting dif-
ference arises. When several individuals actively collaborate in doing
something bad (say in a murder conspiracy), then each of them is stan-
dardly credited with—that is, bears legal and moral blame for—the produc-
tion of that negative result: each of them is regarded as being “guilty of
murder”.7 But when individuals actively collaborate in the production of
something good (say in making a scientific discovery or in establishing a
museum), then they are credited only with their own particular identifiable
contribution. And there is a sound rationale for this. It would seem that the
difference in treatment here lies in the practicalities of the matter rather
than in purely abstractly theoretical considerations. In our laws and social
moves we systematically seek to discourage individual participation in the
doing of bad things and to encourage the efforts of individuals toward the
doing of good. And these desiderata are clearly reflected in the disanalogy
at issue.
Let us explore this aspect of the issue a bit further. Anyone who has
ever worked on a crossword puzzle with a collaborator realizes that here
88
the whole is greater than the sum of the parts—that the collaboration syn-
ergy of people working together is something superior to the mere compi-
lation of their separate achievements. In such collaborative problem-
solving situations we encounter the synergetic potentiation of teamwork.
The work of one member helps to potentiate that of another. In joining
forces the group members pave the way to an entirely new level of
achievement. But in such cases to whom does the credit for this advance-
ment—this “collaborative surplus”—properly belong? When individuals
cooperate collectively in the production of something positive, the “sur-
plus” of achievement over and above what individuals accomplish on their
own will generally be allocated proportion with the extent to which they
make their individual contributions. And for a good practical reason. For it
is clear that such a principle of procedure will provide a maximum of rea-
sonable encouragement. And so there is a lesson here. Our standards of
justice are often the fruit of sensible practical considerations, and even eth-
ics is not beyond the reach of practicality.8
NOTES
1
Legal or institutional responsibility is something else again—something rather dif-
ferent from the moral. The captain is “responsible” for what happens on the ship,
the officer is “responsible” for the acts of subordinates. But what is at issue here
involves a rather different use of the term.
2
Compare G. J. Massey, “Tom, Dick, and Harry, and all the King’s Men,” American
Philosophical Quarterly, vol. 13 (1976), pp. 89-107.
3
On this feature of “moral luck” see the author’s treatment in Luck (New York:
Farrar-Straus-Giroux, 1995).
4
On these issues see Nicholas Rescher and Carey B. Joynt, “Evidence in History
and the Law,” The Journal of Philosophy, vol. 56 (1959), pp. 561-78.
5
Note that so long as we refuse to project group responsibility unto the constituent
individuals, then—even in the “Hebraic” case—we are confronted with the anoma-
lous upshot that a group can, through the deliberate actions of the individuals in-
volved, produce a terrible result for which as regards individuals “no one is to
blame” through lack of the right sort of intent in the point of individuals.
89
6
I am grateful to my colleague David Gauthier for helpful commentary on this issue.
7
To be sure, the presence of some degree of active participation is a crucial factor.
Mere membership—wholly passive and inert—in a group that is collectively re-
sponsible does not as such contribute to the individual’s moral blame or credit. And
so the terrible things done by the Nazis at large detract nothing from the credit of
Schindler, the Nazi.
8
This chapter was initially published under the same title in the Journal of Social
Philosophy, vol. 29 (1998), pp. 410-17.
90
Chapter 8
THE BELL CURVE REVISITED
THE BACKGROUND
T he 1994 book by Herrnstein and Murray on The Bell Curve enjoyed

such a success by scandal that one virtually had to be living on another
planet not to have heard of it.1 Now it is difficult to summarize an 800 page
book in a single paragraph. But in telegraphic outline the main theses are
these: That American society is increasingly stratified into socio-economic
classes by training and education. That the available evidence indicates a
person’s capacity to benefit from training and education hinges on their
I.Q. That I.Q. is in large measure genetically determined, and is thus im-
pervious to the sorts of educational and environmental arrangements that
are amenable to public-policy management. That this situation is creating a
permanent and (in an economy of increasing technological sophistication)
growing under-class trapped at the bottom of the socio-economic heap.
That this (largely genetically determined) underclass is heavily if not pre-
dominantly populated by blacks. Moreover, the book holds that I.Q. also
predicts crime rates, welfare dependence, poor parenting, and civic irre-
sponsibility. And in marshalling these statistics the authors have preempted
their race-conscious critics by looking solely at the situation for whites.
The Bell Curve has found an immense and immensely hostile reception
throughout the academic and public press, and has propelled its surviving
author into media stardom (Herrnstein having died as the book was being
published). The book has driven politically correct intellectuals up the wall
and set liberal pundits to frothing at the mouth, with the one issue of the
New Republic publishing no less than seventeen commissioned rebuttals.
All concerned—from academic critics to talk-show hosts—agree in regard-
ing The Bell Curve as an blitzkrieg attack upon the whole spectrum of edu-
cational and social service policies that have been pursued in the U.S.A. at
least since the days of Lyndon Johnson’s “Great Society” program.
This controversy has such substantial ramifications into questions of
scientific method, social choice theory, and public philosophy that it well
deserves the more dedicated attention of philosophers. And when looked at

in the cooler, clearer light that often prevails at this Olympian attitude, the
matter assumes a rather different, less spectacularly ominous look.
It must be stressed at the outset that this paper will be concerned not
with personalities but with methodologies. I have no inside knowledge of
the authors’ political attitudes nor do I care to speculate about their mo-
tives. These deliberations will address only one sector of their discussion,
namely the socio-statistical analyses that are at issue in much of the body
and also the main title of the book.
This said, let us go back to square one and begin with some methodo-
logical considerations.
In considering how a certain measurable characteristic of individuals is
distributed in two human groups (of possibly different size), we may well
encounter situations where this distribution occurs in a significantly dis-
crepant way, as a schematic of the following sort indicates.
40
30
% of 20 group X group Y
individuals at
this magnitude 10
magnitude of a certain characteristic
The comparative distribution of the height (or the shoe size) of Polynesian
and Mongolian males might be an example. Such a difference in the statis-
tical distribution of certain characteristics reflects what one might call a
bell-curve discrepancy between the two groups.
In such cases, the one group seen in the statistical aggregate, drops off
more rapidly than the other as the magnitude in question increases, yield-
ing to that other a significant statistical advantage—or disadvantage, de-
pending on whether the characteristic in question is seen as positive or as
negative.
Suppose now that one is confronted with a characteristic of individuals
that clearly represents a positive personal asset and moreover is something
that is measurable, biologically determined, and unresponsive to social
manipulation. And suppose further that this feature is distributed unequally
between two social groups in the way indicated by such statistical discrep-
ancies. Then what follows?
92
What follows in the case where the characteristic feature in question is

human intelligence as exhibited by whites and blacks and purportedly
measured by I.Q. tests, is a media donnybrook of the proportions ensuing
upon publication of The Bell Curve. And what we have been witnessing in
this reaction is a paroxysm of denial—denial that there is such a things as
“intelligence”, that it is measurable, that it is biologically determined, and
that it is unresponsive to management by social policy measures. And de-
nial is not the whole of it; there is a good deal of denunciation too. The so-
ciometric researches described by these investigators have been
condemned as pseudoscience2 or—somewhat more vigorously—as “a
scabrous piece of racial pornography masquerading as serious scholar-
ship”3 and their contentions have been characterized as “odious”.4 The
president of Rutgers became entangled in a public relations blizzard to save
his position for having said that some students lack the (quote) “genetic he-
reditary background to score well on college admission tests.” He excul-
pated himself with the explanation that he had been thinking and talking
about reviews of The Bell Curve. But he went on to explain carefully that
“he and his wife had both refused to read the book itself because they con-
sidered its basic premise as morally wrong.”5
MISTAKEN APPREHENSIONS
The reason for this massively antagonistic media response in our era of
political correctness is the widely shared sentiment that all manner of
harmful social-policy consequences must ensue if the basic thesis is
granted—that is, if it were conceded that intelligence exists as a measur-
able characteristic that is biologically determined, substantially unrespon-
sive to social manipulation, and unequally distributed as between white
and blacks, with only one black in six scoring above the average for whites
on I.Q. tests. The fact, however, is that any close inspection of the matter
shows that this bugaboo of horrendous consequences is an extremely dubi-
ous proposition. To bring this to light clearly, it is useful to consider a case
of substantially the same structure that carries a far lower emotional
charge.
The comparison case that is worth pondering is that of the male/female
divide in the context of longevity. Let us take it as conceded—as is surely
plausible on the presently available evidence—that the life expectancy of
males and females represents a desirable feature that is, moreover, measur-
able, biologically determined, and significantly unresponsive to (further)
93
social-policy manipulation in already advanced technological societies.

And let it further be granted that—as all the available evidence indicates—
women are advantaged over men in point of this particular asset, with only
one man in three or four surviving the average lifespan for females. Does
this situation of itself warrant any conclusions whatsoever as regards the
institution of social policies that bear differentially on the interests of men
and women?
Look at the possibilities. Does the existence of a bell-curve discrepancy
in life expectancy as between men and women imply that society should
invest more in providing for the health of men than in providing for the
health of women? Does it mean that, since men have fewer years to enjoy
life’s benefits, something should be done to advantage them in some com-
pensatory way? Does it mean that, since women have more inactive years
per year of productive life than men do, their annual claims upon socially
distributed benefits are diminished? The answer here is clearly: None of the
above. The fact that women live longer than men in and of itself has no
implication whatsoever for how the goods or bads of a society should be
distributed between them. It is—or should be—perfectly clear that in the
case of the differential longevity of men and women there just are no valid
public policy considerations that would unfairly advantage the one group
as against the other. Whether you are an egalitarian who wants to have eve-
ryone treated just alike or a chauvinist who thinks that one group should be
advantaged over the other—either way, that bell-curve discrepancy in and
of itself does nothing to aid or injure your cause.
Even if one were to grant by way of fundamental policy that equality of
treatment is a prime social desideratum and that the society should do what
it can to see to it that men and women are placed into a like condition in
point of societally distributed assets and liabilities, the response to a bell-
curve discrepancy remains wholly undetermined. It is left wholly indeter-
minate, for example, if one is somehow to improve the condition of men or
somehow to diminish that of women. And either way, the question of
which ones is left wholly untouched.
At this point we must plunge into the sorts of distinctions of which phi-
losophers are notoriously fond, but which are—nevertheless—important
for cogent thinking about social issues.
The pivotal fact is that different discourses are at issue with policy and
with statistics. Sociostatistics address the question: What has been happen-
ing? Policy deliberations address the question: What should we be doing in
the future? Totally different matters are at stake—though of course there
94
will be links of relevancy for beings whose views of the future must be
conditioned by their experience of the past. Those links of relevancy, how-
ever, are mediated not by facts but by desires. What has been cannot of it-
self determine what should be. The long and short of it is that social
statistics as such are barren of implications for the merit of policy innova-
tions. They are substantially ineffectual in the support of policy decisions,
which must, is the end, be addressed by deliberations that transcend the
level of statistics.
Statistics show that very young and very old drivers are more accident
prone than middle-aged ones. Does this mean that they should drive less
often? Only if you believe that each driver should only be allowed to im-
pose a fixed amount of risks upon the other members of the community.
Does it mean that they should pay more for their drivers license? Only if
you believe that individuals should compensate the community in line with
risks imposed. Does it mean that they should be kept off the roads? Only if
you think that drivers should not be allowed to put people to above-average
risks.
In and of themselves, such statistical disparities have no policy implica-
tions. Only in conjunction with an actual policy principle—be it explicit or
implicit, tacit or overt—can they provide for policy implications. They can
only function as minor premisses in arguments with a policy major. And in
any such argument it is that policy major which carries the weight and does
the real work.
The fact is that there are a myriad different ways of differentiating
population subgroups by biological/genetic features: not just male/female
or Asian/Caucasian, but by blood type, by height, by hair color, and what
have you. And there are myriad ways of specifying measurable and geneti-
cally based characteristics: not only life expectancy and (possibly) I.Q., but
athletic capability (at running, jumping), performance talents (at memori-
zation, chess, arithmetical computation), vulnerabilities and susceptibilities
(to diseases, hair loss, seasickness), need and requirements (for food, sleep)
and so on. There are likely to be many innate and acquired bell-curve dis-
crepancies between different groups. But in no case does the discrepancy
as such—in and of itself—carry any public policy mandates in its wake.
No doubt, variously differentiated groupings can or do exhibit variously
differentiated characteristics at the statistical level. But such statistical dif-
ferentiations in and of themselves are sterile of policy implication as far as
the treatment of individuals is concerned.
95
Accordingly, even conceding a bell-curve discrepancy between two so-

cial groups cannot in and of itself sustain the weight of positive or negative
conclusions about public policies regarding their differential treatment.
Even if one were to grant the contested point at issue—that whites and
blacks exhibit a bell-curve discrepancy in point of heritable intelligence—
the fact remains that no definite public policy conclusions could be drawn
from it. It does not follow that black children should be schooled differen-
tially; it does not follow that less money should be spent, per capita, on
their education; it does not follow that they should not be bussed to the
suburbs; and above all it certainly does not follow that schools should be
excused from an obligation to make the best and the most of inherent po-
tential (however limited) of each and every child that is committed to their
charge. The long and short of it is that nothing of this sort would follow,
and that anyone who was minded to argue for such a conclusion cannot
simply base it upon that bell-curve discrepancy, but would have to bring
very different, non-statistical and deeply problematic policy considerations
to bear.
Take just one example. Herrnstein and Murray write:
The earliest returns on Head Start were exhilarating. A few months spent by pre-
schoolers in the first summer program seemed to be producing considerable I.Q.
gains—as much as 10 points. ... [But soon] experts were noting the dreaded “fade-
out”, the gradual convergence in the test scores of children who participated in the
program with comparable children who had not. ... [Further study showed that the
cognitive benefits] picked up in the first grade of school are largely gone by the
third grade. By the sixth grade they have vanished entirely. (p. 403)
Let us assume for the sake of discussion that the facts are exactly as repre-
sented here. Does it follow that Head Start was a worthless program and
thereby institute a policy of abandonment? Not at all! On the average, chil-
dren go to school for roughly twelve years. Functioning at a higher level of
performance for even only three years means doing better for one-quarter
of one’s school time. Who is to say that this gain is not worthwhile—even
if the enhancement is transitory. (In a life whose terminus is the grave there
are, after all, no permanent victories.)
Even if it were so that one sector of the population—however defined—
were on genetic grounds to be less educable than another, does it follow
that we should not do everything that we can affordably and effectively can
do to assist them in reaching whatever intellectual potential may be there?
Should society invest in people only where it can expect ample returns?
96
Where is it written that social policy must operate on cost-benefit princi-

ples rather than by striving for a kinder, gentler quality of life? To ask
these questions—or at any rate to ask them of good-hearted people—is vir-
tually to answer them.
Be this as it may, mere bell-curve discrepancy cannot sustain policy
conclusion. Statistics can certainly inform us about the values of social pa-
rameters. And when we are very clever and very lucky statistics can also
inform us about causal relationships among parameter values. But that is as
far as it goes. Statistics can tell us what is happening and can sometimes
even enable us to explain why it is happening. But questions about what
ought to happen—about the direction we should go to change what is in
process of happening—are something else again. Those policy issues
themselves cannot be settled by an appeal to social statistics. Statistics in
and of themselves are unable to carry the weight of policy. And it is easy to
see why this should be so. Social statistics tell us where we are or have
been; they have to do with what has been and case in the past-&-present.
But policies are a function of where we would like to go. They have to do
with what ought to be in the future. And there simply is not way—that is,
no way that is itself free from policy involvements—to get from the former
to the latter. As any traveler knows, the journey of the past cannot of itself
determine the itinerary of the future.
THE GAP BETWEEN STATISTICS AND POLICY
Statistics describe facts; policies prescribe actions. Policies tell us what

we are to do about the facts. One cannot frame sensible policies without
reference to facts. But the statistical facts themselves are inert in relation to
socio-political policy determinations; the motive power must have come
from policy considerations as such. It is not that statistical discrepancies
are totally disconnected from public policy issues, but rather the connec-
tion is one that can itself only be mediated by policy considerations.
What social statistics can achieve—and the most that they can
achieve—is to exhibit a consonance or dissonance between a policy and its
professional rationale. If a policy is instituted on the basis of its bringing
about a certain result, then statistics can (at least in ideal conditions) show
that implementing and following that policy has proved effective—or inef-
fective—in realizing that result. But a discovery of this nature does not of
itself validate or invalidate that policy as such. After all, even in the case of
a severe disparity between policy and professed rationale, we are left with
97
an open choice between abandoning the policy or abandoning the ration-

ale—that is, going out and finding ourselves another one. (The policy of
maintaining basic human rights did not go to the wall once people ceased
thinking that this represented something inalienably ordained for people by
their divine creator.) Moreover, even if those policies have not succeeded
under the historic conditions of the past, this does not mean that they will
not succeed under the changed conditions of the future. And, more cru-
cially yet, they cannot possibly show that even when these old policies
have yielded disappointing results, their alternatives would not have re-
sulted in a still worse situation.
It is worth considering exactly where the position of these present delib-
erations stands. The fact is that it stands in a distinctly dangerous place,
caught up in a cross-fire between warring opponents. The one side argues:
• Sociostatistical claims have certain policy implications.

• The latter are unacceptable.
Therefore: The former must therefore be rejected as invalid.
Their opponents argue:
• Sociostatistical claims have certain policy implications.

• The former must be accepted.
Therefore: The latter must be accepted as well, like it or not.
Both lines of reasoning pivot on that initial premiss, from which the one
side proceed by modus tollens and the other by modus ponens. But this
premiss is simply false. And in rejecting the premiss, the present analysis
in effect says: A plague on both your houses. From the angle of present de-
liberations, both the The Bell Curve authors and their critics get it wrong in
their haste to get from statistics to policy.
To be sure, such an even-handed negative position will make enemies
all around. It is a message nobody wants to hear. It is clearly not going to
win any popularity contest. But is it incorrect?
The usual line of attack on The Bell Curve—general abuse aside—is to
question the adequacy of its statistical methods for establishing the causal
connections that are claimed to obtain between race and intelligence. But
here we have been taking the very different line of simply conceding the
claimed relationship at issue for the sake of discussion, and then pressing
the question: What operative difference does it make? And the answer is
98
that it makes no difference at all—that the policy issues stand after exactly
as before. Those sociostatistical claims may be a starting point for delibera-
tion but they do not prejudice its end.
No doubt this line will leave many people dissatisfied. We Americans
like to shift policy questions to the sidelines, letting them be subordinate to
the rulings of judges, say, or the findings of statisticians. For this means
that we ourselves—and our political representatives as well—do not have
to confront them directly and saves us the anguish of dealing in the public
forum with the normative questions of values and goals that are involved.
“What sorts of people do we want to be?”, “What sort of society do we
want to create for ourselves?”, and “How much are we willing to pay for
realizing these desiderata?” Questions like these make us distinctly uncom-
fortable. As a pluralistic society striving for social equilibrium, we do not
like to address issues that divide us. We would much prefer to transmute
such value questions into empirical matters of social statistics. But in the
final analysis this convenient escape just isn’t an available option.
CONCLUDING OBSERVATIONS
The present deliberations are themselves certainly not statistical but

methodological. And as such they self-consistently can—and indeed do—
have a policy bearing upon the treatment of one particular group, namely
social scientists and, more specifically, those social statisticians who study
group-differential conditions.
The public, the media, and all too often their own colleagues (who
ought to know better!) treat such investigators as pariahs. Their work—it is
often said—brings aid and comfort to those who espouse policies prejudi-
cial to the just claims of particular social groups. This charge is unreason-
able and unsustainable.
A group’s existing features are what they are, and no matter what this
may be, we need have no hesitancy about looking the facts of the matter in
the eye. No socially deleterious policies or practices can inhere in such
facts. If men are more given to crimes of violence than women, we will
need more men’s prisons than women’s prisons if the criminal justice sys-
tem is to treat men and women alike. But the crux of that public policy de-
cision (“build more men’s prisons than women’s”) lies not with these
statistics but with the ideology of equal treatment and its implementation.
That policy is certainly not dictated by the social statistics of criminality
themselves. It cannot be overemphasized that the social policies we adopt
99
in the treatment of others cannot inhere in those statistical discrepancies.

Other, more powerfully normative and more directly policy-geared consid-
erations will have to be brought into operation.
The fact is that reasonable people of whatever allegiance need not be
apprehensive that social statistics will sustain and nourish an unpalatable
policy agenda, because the fact is that statistics cannot sustain and nourish
any policies at all. When statistical disparities are invoked to support ques-
tionable public policies, the fault and error lies not in those social statistics
but in the substantive ideological assumptions that are—all too often—
inappropriately superimposed upon them.
Accordingly, it makes good rational sense to let the social statisticians
get on with their inquiries without fear of reproach let alone intimidation.
Policy aficionados—be they of left or of right orientation—actually have
nothing to fear from their labors. One can no more shape policies from sta-
tistics than one can build houses of air: either way, additional containers
will be needed. In policy-supportive argumentation, the weight will have to
be borne not by social statistics but by other, very different and emphati-
cally non-statistical considerations. The lesson is clear. Social policy issues
must be argued on a social policy basis. In this regard, the sociometrician
is an innocent bystander caught up in somebody else’s fight.
People who condemn sociostatistical research on grounds of the poten-
tial politico-social harmfulness of its policy implications like to fancy that
they are on the side of the angels. They are not. Quite the reverse: their
stance does very real damage to the democratic process and seriously un-
dermines its operation. For suppose that they were right and that statistics
were in the driver’s seat as a determining factor in resolving public policy
questions. Note what this would mean: For one thing, policy deliberations
would then be transmuted into a matter of conjuring with social statistics,
averting any real policy debate through the consequence that policy issues
do not have to be confronted head-on. For another thing, the advocacy of
controversial policy options would now effectively be hidden away behind
a screen of social science statistics, with the wolf of policy advocacy con-
cealed under the sheep’s clothing of empirical inquiry. If policy were car-
ried along in the wake of statistics, then the work of public policy
determination would be shifted from the arena of political debate into the
hands of the technical experts who preside over sociostatistical inquires.
Clearly none of this can be seen as supportive of a democratic ethos of
public decision making.
100
To be sure, this procession of unhappy consequences does not mean that

the claim that statistics governs policy is wrong. (That would have to be
argued differently—as we have indeed been doing here.) But it does mean
that the exponents of this position cannot legitimately wrap themselves in
the cloak of unmixed virtue and good will.6
NOTES
1
Richard J. Herrnstein and Charles Murray, The Bell Curve: Intelligence and Class
Structure in American Life (New York: The Free Press, 1994). For the period up to
the end of 1994, the data base DIALOG yields some 250 newspaper items relating
to the book.
2
Brent Staples, Editorial page of The New York Times, Friday 28 October 1994, p. 3.
3
See the op-ed piece by Bob Herbert in The New York Times for 20 October 1994
*(p. 27). See also the feature article on Murray in the New York Times Magazine
for Sunday October 9, 1994. A Scientific American article (by Tim Beardsley) stig-
matizes the book as “a tendentious tome [that] abuses science to promote far-right
policies,” (vol. 272, January 1995, p. 16).
4
E. D. Hirsch Jr. “Good Genes, Bad Schools,” discussion of The Bell Curve, The
New York Times, Op. Ed. page, Saturday, 29 October 1994.
5
The New York Times, Monday, February 1, 1995, p. 8. Political ineptitude is appar-
ently not this gentleman’s only shortcoming.
6
This chapter was originally published under the same title in The Philosophical
Quarterly, vol. 9 (1995), pp. 321-30.
101
Chapter 9
IN THE LINE OF DUTY
THE COMPLEXITY OF MILITARY

OBLIGATION
O ur theme will be the complexity of military obligation. It must be said

straightaway, however, that this matter of obligation of members of
the profession of arms is something that differs only in degree, and not in
kind, from the complexity of human obligation in general. After all, any
adult human being bears a substantial diversity of obligations—to one’s
employer, to one’s client, to one’s family, to one’s friends, and so on. All
of us owe debts of duty to various individuals and groups—obligations that
root in the nature of the different relationships that exist among people.
They result from the sort of role that each one of us plays in the lives and
affairs of others—be it as their employee, their attorney, their brother, their
friend, their neighbor, and so on. In joining the profession of arms one
takes on further special obligations—though, of course, one does thereby
not exit from the events of obligations in which all people are caught up
and.
The obligations that people have are highly differential in nature. Some
are legal, some moral, some merely social and rooted in the custom of the
country. But each such sort of obligation has a certain force and legitimacy
of its own. And all are important to maintaining a healthy and fully devel-
oped human existence. No one of them predominates to the exclusion of all
the rest. For example, someone who says to all others “I shall only give
you what is legally your due, nothing else matters to me” and proceeds to
treat all people on this basis—parent, child, friend, and colleague alike—is
literally inhuman. In theory, this sort of thing is possible. One could write a
story about such an individual. But the picture one would draw in the
course of this story would be that of a monster, not of a person.
Some obligations are assumed voluntarily, others are involuntary, and
still others are mixed. Your obligations to your parents are involuntary;
those to your spouse or employer are assumed voluntary; those to your co-
workers or your fellow citizens are mixed—because while you certainly do

not choose them, you could, in principle, arrange to have others. Neverthe-
less, be they voluntary or not, all those various obligations are perfectly
real and genuine. The obligations that one finds in place are just as authen-
tic as those that one puts in place. And one certainly does not eliminate ob-
ligations by thinking them inconvenient and wishing they were not there.
Moreover, in taking on new roles and entering into new obligations you
do not in general shed the old ones. When you change employers you do,
to be sure, exchange your erstwhile employment-related obligations for
others, but you do not thereby cancel your obligations to a spouse or to
your children. When you become a parent, you do not lose your obligations
to your siblings or your fellow citizens. Our duties in life do not cancel one
another; they generally become combined and superimposed upon each
other.
The coexistence of obligations means that they can compete and con-
flict. You cannot devote to your children the time you spend with a sick
neighbor in the hospital. You cannot appropriately dedicate to your friends
and relations the efforts and energies you owe to your employer. It is a fact
of life that the obligations that we have will often compete and must, in
such situations, be balanced off against each other.
In this life we are all servants of many masters. We have obligations to
ourselves and to the other people who play a role in our lives. And it is not
always easy to harmonize them—to figure out, say, what is, in the circum-
stances, due to Caesar and what is due to God. The complexity of obliga-
tion is a phenomenon that pervades human life. It is something that every
responsible adult has to learn to deal with.
But while this is a general phenomenon, it is one that is particularly
acute for members of the military profession. And the reason for this is
simple and straightforward. It lies in the fact that the obligation of the mili-
tary person is—like that of a spouse or a parent—particularly large and
pervasive in its scope. In the military, one’s duty to one’s country, to one’s
service, and one’s unit looms very large indeed. It does not end with the
time of day, the location of one’s placement, or the extent of one’s re-
sources. When the going gets too tough, the discontented spouse can get a
divorce, the dissatisfied employee can quit the job, but the reluctant sol-
dier, sailor, or aviator has no easy recourse.
All the same, the obligation of the military person, wide and deep
though it is, does not become all absorbing. It does not automatically over-
104
IN THE LINE OF DUTY
power and abrogate all other obligations. Many of them remain in place—
crowded into a narrower space, perhaps, but still very much present.
The member of the profession of arms accordingly has many foci of ob-
ligation. I shall limit myself here to considering just five of them: the chain
of command, the service, the nation, civilization, and humanity at large.
Each of these has its own characteristics and its own ramifications.
THE CHAIN OF COMMAND
Let me begin with the chain of command. For the military person, the
first and most obvious source of obligation is clearly those duties that come
into being through the injunctions of those whom the military system has
emplaced in positions of authority. The duty of obedience to legal com-
mand—personal preferences notwithstanding—is the most basic and de-
finitive commitment of those who belong to the profession of arms.
Irrespective of whether one’s affiliation is voluntarily self-generated (via
enlistment) or generated through the action of one’s fellow citizens (via a
draft), one is by law, custom, and hoary practice bound to this chain of
command, and in consequence subject to the separate and stringent code of
justice to which the military stands bound.
However, if one’s obligation to the chain of command were all—if it
cancelled all of one’s other obligations—then the life of a member of the
military profession would be much simpler than it is. It is precisely be-
cause this most characteristic and weighty of soldierly duties is not all that
there is to it, that the complexity of military obligations arises.
THE SERVICE
Beyond the chain of command lies first of all “the service”—the entire
organization in its historical unity—to which the chain of command gives a
concrete temporary embodiment. When it does its work well and serves its
function appropriately, the chain of command generally implements the
best interests of the service. But every once in a while things go wrong.
And then the conscientious professional has to ask him or herself the diffi-
cult question of whether the particular orders at hand or those larger inter-
ests are to prevail. The legendary supply sergeant on Wake Island who,
adhering to standing regulations, is reported to have made his troops sign
receipts for the live ammunition being issued them as the Japanese were
105
storming ashore would be a clear example of someone who didn’t get their
competing duties properly prioritized.
THE NATION, THE CIVILIZATION, AND HUMANITY AT LARGE
And beyond the service there loom the large issues of one’s country,
one’s civilization, and also humanity at large. In being a member of the
profession of arms, one does not cease to be a citizen, a responsible person,
or a human being. And at those levels of relationship issues of potentially
conflicting obligations also arise. Even in the heat of war there are some
sorts of militarily useful things that are just not done. (One does not, for
example, clear minefields by marching enemy civilians across them—or,
for that matter, even prisoners of war.)
I would like to illustrate these complexities of military obligation with
some concrete examples. Specifically, we shall consider four episodes
which clearly bring to view the intricacies of military obligation. These
episodes have been chosen, quite deliberately, from the pages of German
rather than American military experience. For when a situation comes too
close to home, our personal loyalties and ideological leanings prevent our
being able to view the matter with the objectivity and detachment that is
needed in order to see clearly and objectively just what the issues are.
Geographic and cultural distance improves the prospects of making a de-
tached and dispassionate appraisal of the fundamental principles at stake.
THE GENERAL AND THE DANISH JEWS
In the middle of World War II, the German occupation forces in Den-
mark were commanded by Infantry-General (General der Infanterie)
Heinrich von Henneken, a professional soldier of proven ability. His was
not, however, an autonomous command, for his military jurisdiction was
subordinate on all matters relating to specifically political affairs to the
primacy of the Ambassador. Since Denmark came under German control
by political surrender rather than military conquest, the power of control
was vested in the Foreign Office in Berlin, to which, in effect, von Henne-
ken was ultimately answerable.
In September of 1943, orders went forth from Berlin decreeing that a
state of emergency should be declared in Denmark and all Jews rounded up
for deportation. On 23rd September, von Henneken requested a postpone-
ment of this operation. Pleading a shortage of manpower support, he sim-
106
IN THE LINE OF DUTY
ply dug in his heels. When Berlin persisted, he strongly represented the
impracticability and the undesirability of any Army involvement in the op-
eration, insisting that this would “injure the prestige of the Wehrmacht at
home and in foreign countries.” Instructions to the contrary notwithstand-
ing, he refused the use of military personnel in the round-up and generally
minimized cooperation. His intransigence created a delay during which
most Danish Jews were successfully evacuated to Sweden.
The case of von Henneken affords a clear and striking illustration of a
soldier who put the best interests and good repute of his service ahead of
the demands of the chain of command. As best as one can tell, von Henne-
ken’s motivation was—as far as external appearance went, at any rate—not
so much a humanitarian compassion for the fate of the victims, as a sense
of the inappropriateness of involving the military in their victimization. His
concern was for the good name and repute of his service, taking the posi-
tion that there are certain sorts of things that a self-respecting army just
does not do. As far as von Henneken was concerned, the best interests of
the service took priority over any mere mechanical obedience to instruc-
tions from higher authority. Fortunately for him, despite the irritation of
Nazi authorities in Berlin, friends in high places were able to protect von
Henneken against unpleasant personal consequences—although this was
something of which he had no prior assurance at the time of action.1
THE ARMY AND THE COUNTRY
Let us next turn to another, rather different sort of episode.

In September 1990, General Augusto Pinochet, Chile’s former long-
term head, and then still continuing Army commander, set off an interna-
tional furor by his remarks in an address to the Rotary club in Santiago.2
Using strong and biting terms, he characterized the post-war transforma-
tion of the German military as an unmitigated disaster. According to Pino-
chet, the democratization of the German army and its acceptance of the
individual rights of its members had brought a once proud military
organization near to ruin. Pinochet viewed with alarm the prospect that
there might be Chilean disciples of the generals who built the new German
army in a way that undermined the long-standing Prussian tradition of
strict discipline, unquestioning obedience, and absence of discussion.
(After all, the Chilean military inclines to view their army as the last
Prussian-style army still in existence.) Pinochet accordingly castigated the
German commanders “who betrayed that institution by trying to convert it
into an army of inferior values.” No doubt echoing a body of sentiment in
107
ferior values.” No doubt echoing a body of sentiment in the German mili-

tary itself, Pinochet took the line that from a strictly service point of view,
the “democratization” of the military is a step towards its emasculation. Pi-
nochet characterized the officers responsible for the changes as “the big-
gest traitors Germany has had with respect to its army.”
Pinochet’s remarks understandably provoked an immediate reaction
from the German government—and from the German military as well.
Their spokesmen responded by stressing the idea that: “The soldier of to-
day has to be a citizen.” And they observed that the change in the German
military system since the war was intended “to harmonize obedience and
military discipline with the principles and values of liberty and human dig-
nity.”
As the German military spokesmen emphasized in response to Pino-
chet’s strictures, a duty to the service and its strictly military values has to
be balanced and coordinated—in contemporary Western democracies at
any rate—with a duty to the nation and to the values that its military is in-
stituted to defend. Against Pinochet’s narrow professional traditionalism,
the German military spokesmen gave voice to a widespread acknowledge-
ment that the interests and values of the country and its citizenry have to
play a significant role in the management of military affairs. As they repre-
sented the matter, the modern soldier’s duty to the civic values of the
community is not less significant than their duty to strictly military values
of the service.
THE GENERAL THE CITY—IS PARIS BURNING?
Sometimes members of the profession of arms come up against obliga-

tions of even a more far-reaching and—so to speak—ominous sort. This is
exemplified by our next illustrative episode.3
In 1944, the German commander of the Paris region was one General
Dietrich von Choltitz. He was a Prussian general of the old school. A third
generation professional soldier, he had served in the airborne infantry with
impressive success. He had led the unit that made the initial thrust into
Holland and had devastated Rotterdam. Much further from home, he had
commanded the regiment that took the Russian stronghold of Sebastopol in
the Crimea.
As the allies were now converging on Paris, Hitler ordered an all out,
house-to-house defense of the city. The high command, the OKW under
108
IN THE LINE OF DUTY
von Rundstedt, put demolition teams at von Choltitz’ disposal and ordered
him to mount a Stalingrad-like defense of the city.
As the man on the spot, von Choltitz saw the matter in a different light.
He realized that an all-out defense of Paris would lead to an immense loss
of life, since there was no prospect of evacuating the civilian population.
He also believed that an effective effort to hold the city was ultimately im-
practicable, and that a house-to-house defense could achieve no significant
goal beyond a modest delay of the Allies’ overall advance, and that any se-
rious steps in this direction would result in the physical destruction of
Europe’s most beautiful city. His considered professional judgment was
that no valid military purpose could be served by an all-out effort to defend
Paris. But orders are orders. Von Choltitz was in a deep and painful di-
lemma. And he resolved it in a bold and radical way.
Von Choltitz simply cut the chain of command. He ignored his orders,
contacted the allies, and arranged to surrender the city with minimal resis-
tance.
There is little doubt that von Choltitz’ actions verged on the outright
treasonable. And yet it could be argued that he was, in a way, justified—
that his obligation to the chain of command was outweighed by obligations
and interests of a higher order. Von Choltitz had made a very hard and very
dangerous decision. As he was led away to internment by his allied captors
unknowing Frenchmen spat at him in the street. They would have done
better to build him a statue.4
Von Choltitz realized that nations and peoples survive wars. Win or
lose, Germans and French are destined to share the land-mass of Europe as
neighbors. And he recognized that an utterly pointless destruction of peo-
ple and cultural treasures would poison the atmosphere and render peaceful
coexistence difficult if not impossible for many years to come. Form one
point of view—that of the chain of command—we could say that von
Choltitz failed in his duty: perhaps even that he was a traitor. And yet it is
difficult to deny that a responsible member of the profession of arms has
obligations that extend beyond the chain of command, obligations to sup-
port the best interests of his country—and sometimes even the higher in-
terests encompassed in the values that the traditions of the country profess.
THE STAUFFENBERG CONSPIRACY
This brings me to my fourth and last episode. It is undoubtedly the most

drastic one—the 1944 attempt to assassinate Hitler.
109
I don’t want to take time to describe the details—the story of Colonel

Count Stauffenberg and his associates is too well known. The bomb that
blew up on the wrong side of the conference table’s supporting pedestal,
the rallying of the shaken Führer’s authority, the collapse of the conspiracy
in the face of the opposition of military loyalists, the swift and cruel end of
the conspirators and their sympathizers—all these are circumstances that
do not need elaboration here. Rather what matters for present purposes is
that this was not a matter of the rising of an outraged populace or of a
counterstroke launched by disaffected opposition politicians, but a conspir-
acy launched from deep within the traditional beliefs and values of the
German military establishment.
The conspirators worked under difficult conditions. In some ways, they
were less than highly competent, in others, they were just plain unlucky.
But the important consideration, for my present purposes, is that these offi-
cers were actuated not by considerations of personal gain or political ad-
vantage, but acted out of a genuine—and surely not altogether
misguided—sense of duty to the German army, the German nation, and the
German people. While these officers indeed violated their pledge of loyalty
to Reich and Führer, they nevertheless acted out a deep sense of obligation
to their service, their country, their national traditions, and their higher du-
ties.5
The episodes we have just considered illustrate the complexity and po-
tential conflicts of the obligations of those who exercise military authority.
In von Hannecken’s case, he had to choose between his good name as a
soldier and the honor of his service on the one hand, and obedience to law-
ful orders on the other. Again, those German officers who responded to Pi-
nochet placed the health of their service political values of their country
ahead of the traditional values of its military establishment. Von Choltitz in
Paris found the awesome choice between his straightforward duties as an
officer on the one side, and a commitment to the long-range best interests
of his country and his dedication to civilized values on the other. Finally,
von Stauffenberg and his co-conspirators put their lives on the line in plac-
ing their soldier’s honor ahead of their soldier’s oath, and setting what they
saw as the best interests of their service and of their country ahead of un-
questioning obedience to constituted authority.
All of these cases exemplify in a particularly dramatic way the agoniz-
ing situations arising when an officer faces a complex choice among con-
flicting obligations in situations fraught with difficulty and danger.
110
IN THE LINE OF DUTY
No doubt the particular episodes we have considered involve rather ex-

treme situations of conflicts of duty. But precisely because of the extremity
of such cases, they serve to highlight sharply the complexity of obligation
and the potential conflict of duty that—in some less dramatic way—faces
virtually every responsible member of the profession of arms.
Most of you, perhaps all of you, will be lucky and will never have to
face such conflict situations at a high level of visibility and historical
significance, in circumstances where the pull of conflicting duty to chain of
command, to country, to civilization—and indeed even to oneself as an in-
dividual human being—come into conflict with one another. But every
military officer faces less dramatic cases of this general sort of situation
some of the time. This simply goes with the territory of being responsible
for matters connected with the well-being of the lives of other people.
The fact is, that at one point or another, every officer faces difficult
choices among competing obligations. For the exercise of command re-
sponsibilities over the actions and situations of other people can all too eas-
ily create a condition of conflicting obligations—though (mercifully)
usually of a minor and undramatic kind. But even in their milder forms,
such experiences will bring home a realization that the complexity of mili-
tary obligation is not an abstract theoretical exercise, but something real
that grabs you in the heart, the chest, and the stomach.
You may well wonder what advice I have to offer on this subject. And
to my embarrassment, I must confess that I have no very satisfactory re-
sponse. One reasonable tactic is to think of someone you admire very
much as being placed in the same sort of situation, and then ask: “What
would he or she do in my place?” Another tactic is to reverse the roles.
Following the approach of the philosophy Immanuel Kant, one would ask,
“If I myself were not personally involved but were charge with laying
down a general rule that others would be called on to follow in such cases,
what would I then stipulate?” In this sort of way one could, as it were, cre-
ate one’s own guidelines for choice in situations of conflicting obligation.
These different foci of obligation—one’s lawful orders, one’s service,
one’s country, one’s values—represent valid and appropriate commitments
upon every officer—commitments that make very real and pressing claims,
be they legal or moral. When these claims clash with one another, there
just are no easy solutions and no mechanical formulas for working out the
answer. Like any mature individual in a responsible position in life, an of-
ficer needs the ability to resolve conflicts thoughtfully—to examine an is-
111
sue on every relevant side and to give each element of a difficult and con-
troversial situation its due.
The conflicting claims that arise from complex obligations must be con-
fronted and weighed and balanced and resolved. All one can do is to work
things out in one’s heart and mind as thoughtfully and conscientiously as
one can—and then do what one has to do. There are no simple, automatic
rules, and to respond in a simple-minded and automatic way to problems of
the sort at issue is to court disaster. One has no alternative but to worry and
fret and sweat.
But there is one point I would like to emphasize particularly. It is this:
When one experiences the responsibilities incumbent upon a military offi-
cer, one does not cease to be a citizen and one does not cease to be a hu-
man being.
This may perhaps sound rather simplistic and naive. But is has substan-
tial consequences. For it means that the officer, though indeed just another
link in the chain of command, is nevertheless not a mere automaton re-
sponding to the will of others, but continues bearing responsibility for one-
self as a person. “Following orders” is a crucial part of the soldier’s code,
but it is not the only one.
The officer cannot look simply to the wishes of a superior, or simply to
the practice of the group, to what “the others” are doing, but continues as
an individual, as a citizen and as a human being—as someone who must to
his or her own self be true, as someone who must act as a person in the
light of his or her own values.
A fundamental recognition of right and wrong—a conscience, in
short—is what makes one into a responsible person. It is a resource that
every officer does and should have. It is what marks one as a responsible
individual agent, capable of being answerable for his or her acts, even in
the face of contrary custom and group disapproval. It is a manifestation of
that most admirable and awesome human quality—the willingness to as-
sume responsibility and stand by one’s obligations as one sees them even
where risks for oneself are involved. The fact is that in the larger scheme of
things, the quality of moral courage is no less desirable and no less admi-
rable in a military officer than the quality of physical courage.6
112
IN THE LINE OF DUTY
NOTES
1
For a fuller account of this episode see Raul Helberg, The Destruction of the Euro-
pean Jews, Vol. 2 (New York, 1985), pp. 558-68.
2
For a detailed report on this episode see the 8 September, 1990 dispatch from
Santiago filed by Shirley Christian of the New York Times, and published in that
newspaper on 9th September, 1990 under the title “Pinochet Irks the West Germans
With a Potshot at Their Military.”
3
The episode is described in detail in Larry Collins and Dominique Lapierre, Is
Paris Burning? (New York: Simon & Schuster, 1965.)
4
Von Choltitz survived to write his memoirs, Soldat unter Soldaten (Konstanz: Eu-
ropa Verlag, 1951).
5
A useful source of information and appraisal is Hans-Adolf Jacobsen (ed.), July 20,
1944: The German Opposition to Hitler as Viewed by Foreign Historians (Bonn:
Press and Information Office of the Federal Government, 199)
6
This chapter is a revised version of an address on “The Complexity of Military Ob-
ligation,” presented to the junior class of the Air Force Academy in Colorado
Springs on 15th November 1990. The address was published by the Air Force
Academy as the third Joseph A. Reich Sr. Distinguished Lecture on War, Morality,
and the Military Profession. It has reprinted in C. J. Fiearrotta (ed.), The Leader’s
Imperative (West Lafayette, Ind.: Purdue University Press, 2001), pp. 243-54.
113
Chapter 10
TECHNOLOGICAL PROGRESS AND

HUMAN HAPPINESS
STAGESETTING
D oes increased knowledge of nature and technological mastery over it

enhance man’s happiness and satisfaction, or is what we honorifi-
cally—nay almost reverently—characterize as “progress” really irrelevant
to this key aspect of the human condition?
This question lies at the dead center of any examination of the relation-
ship between technology and humanistic concerns. It goes to the very heart
of the matter—the linkage between man’s knowledge and “mastery over
nature” on the one side, and his humane life-world of thought and feeling
on the other. It is a question which theoreticians of science and technology
generally ignore. But humanists have often touched on it. (It may be
viewed as a key issue in Goethe’s Faust, for example.)
Writing in 1920, the able British historian of progress J. B. Bury painted
a picture in the following terms:
The very increase of “material ease” seemed unavoidably to involve conditions-

inconsistent with universal happiness; and the communications which linked the
peoples of the world together modified the methods of warfare instead of bringing
peace ... (The modern) triumphs of the advance of man’s aims hardly seemed to
endanger the conclusion that, while knowledge is indefinitely progressive, there is
no good reason for sanguine hopes that the condition of man is “perfectible” or that
universal happiness is attainable.
This quote provides an appropriate stagesetting for deliberating about

the implications of the impressive modern growth in our technological
competence for human happiness and the tenor of the condition of man.
THE HISTORICAL DIMENSION
Let us look briefly at the historical dimension of this issue. The question
of the reality and significance of progress has been debated since the
“quarrel between the ancient and the moderns” regarding the relative im-
portance of the wisdom of classical antiquity as compared with modern
learning—was launched in the late Renaissance. Now at the dawn of mod-
ern science in the 17th century, the leading figures from Bacon to Leibniz
all took a highly optimistic view. Man’s knowledge was about to enter a
new era, and his circumstances and conditions of life would become trans-
formed in consequence. Consider a typical passage from Leibniz:
I believe that one of the biggest reasons for this negligence (of science and its ap-
plications) is the despair of improving matters and the very bad opinion entertained
of human nature ... But ... would it not be fitting at least to make a trial of our
power before despairing of success? Do we not see every day new discoveries not
only in the arts but also in science and in medicine? Why should it not be possible
to secure some considerable relief from our troubles. I shall be told that so many
centuries had worked fruitlessly. But considering the matter more closely, we see
that the majority of those who dealt with the sciences have simply copied from one
another or amused themselves. It is almost a disgrace to nature that so few have
truly worked to make discoveries; we owe nearly everything we know ... to a hand-
ful of persons … I do believe that if a great Monarch would make some powerful
effort, or if a considerable number of individuals of ability were freed from other
concerns to take up the required labor, that we could make great progress in a short
time, and even enjoy the fruits of our labors ourselves.1
Such a perspective typifies the 17th century view of the potential of scien-
tific and technical progress for making rapid and substantial improvements
on the human condition.
By the 19th century the bloom of ameliorative hopefulness was defi-
nitely beginning to fade. The lines of thought worked by Malthus and
Darwin introduced a new element of competition, struggle, and the pres-
sure of man against man in rivalry for the bounties of nature. The idea that
scientific and technological progress would result in enhanced human sat-
isfaction/contentment/happiness came to be seriously questioned. Writing
around 1860, the shrewd German philosopher Hermann Lotze said:
The innumerable individual steps of progress in knowledge and capability which

have unquestionably been made as regards this production and management of ex-
ternal goods, have as yet by no means become combined, so as to form a general
advance in the happiness of life ... Each step of progress with the increase of
116
TECHNOLOGICAL PROGRESS AND HUMAN HAPPINESS
2
strength it brings, brings also a corresponding increase of pressure.
Thus already over a century ago, thoughtful minds were beginning to doubt
that man’s technical progress provides him with a royal route to happiness.
SOME DISTINCTIONS
Before turning to this issue, one or two important preliminary points

must be gotten out of the way. For one thing, it is necessary to heed the
important distinction between negative and positive benefits. A negative
benefit is the removal or diminution of something bad. (It is illustrated in
caricature by the story of the man who liked to knock his head against the
wall because it felt so good when he stopped.) A positive benefit on the
other hand is one which the addition of something that is good in its own
right rather than by way of contrast with an unfortunate predecessor.
Now there is no doubt that the state of human well-being has, or can be,
enormously improved by science and technology as far as negative benefits
are concerned. There can be no question but that technical progress has
enormously reduced human misery and suffering. Consider a few in-
stances:
1. medicine (the prevention of childhood diseases, through inoculation,

anesthetics, plastic and restorative surgery, etc.);
2. waste disposal and sanitation;
3. temperature control (heating and air conditioning).
It would be easy to multiply examples of this sort many times over.

But the fact remains that such diminutions of the bad does not add up to
a good: that the lessening of suffering and discomfort does not produce a
positive condition like pleasure or joy or happiness or the like. Pleasure is
not the mere absence of pain, nor joy the absence of sorrow. The removal
of the negative does not create a positive—though, to be sure, it abolishes
an obstacle in the way of positivity. And so the immense potential of mo-
dem science and technology for the alleviation of suffering does not auto-
matically qualify it as a fountain of happiness.
Moreover, in various ways science and technology have created a set-
ting for life which is counterproductive from the angle of happiness. One
117
instance is modern military technology and life under the shadow of the
atomic sword. Another example is the overcrowding of human popula-
tions—the product largely of modern medical, hygienic, and agricultural
technology. There is organizational centralization that has put all of us at
risk as victims of disgruntled employees, irate consumers, disaffected citi-
zens, political terrorists, and other devotees of direct action against inno-
cent bystanders as a means to the realization of their own objectives. This
list of such happiness-counterproductive areas of modern technology is
easily prolonged. But here I want to make a larger and perhaps foolhardy
assumption. For I am simply going to adopt the somewhat optimistic
stance that all such problems which science and technology has created,
science and technology can also resolve. And accordingly I am going to
leave this negative aspect of the situation wholly out of account, and to
look at the situation if best comes to best, so to speak.
The question to be faced here is accordingly this: even if we view the
consequences of science and technology for the human condition in the
most rosy light and look on them in their most favorable setting—not
Ethiopia, say, or India, but the USA and Western Europe where the most
advantageous and least problematic conditions have prevailed—is it really
clear that science and technology have wrought benign effects upon the
condition of human happiness, viewed in its positive aspect?
TRANSITION: THE SOCIOLOGICAL PERSPECTIVE
To this point we have considered what might be called the philosophical

background of the issue. Let us now move off in a different direction.
The issue nowadays has a new dimension. In the past, discussion has
proceeded on a speculative basis, and the participants were principally phi-
losophers and philosophically inclined historians of students of social af-
fairs. But nowadays the sociologists and social psychologists have taken
over. It thus becomes possible to bring statistical data to bear and to look at
the empirical situation. We need no longer speculate about the relation be-
tween progress and happiness. We can “go out into the field” and find out
how things go in what tough-minded social scientists like to refer to as “the
real world”. That is, we can proceed by means of questionnaires and the
whole paraphernalia of empirical social science. When we do this, all but
the most cynical among us are in for some surprises.
If the thesis that increased physical well-being brings increased happi-
ness were correct, one would certainly expect Americans to regard them-
118
selves as substantially happier today than ever before. This expectation is

certainly not realized by the facts.
The contrast is a striking one here. Substantial majorities envisage a
course of substantial improvement in terms of material and intellectual at-
tainments. Nevertheless sizable pluralities take the view that our situation
is deteriorating as regards “inner happiness” or “peace of mind,” and—
interestingly enough—the more highly educated the group being sampled,
the more emphatic this sentiment becomes. The upshot of such question-
naires indicates that in fact a substantial majority of people incline to the
view that there is a negative correlation between progress and happiness.3
Such evidence, to be sure, relates to the subjective impression of the
people interviewed.4 But there are also relevant data of a more objective
kind that indicate a failure of Americans to achieve a higher plateau of per-
sonal happiness in the wake of substantial progress in the area of social
welfare. For one thing, the suicide rate per one hundred thousand popula-
tion per annum has hovered with remarkable stability in the eleven-plus-or-
minus-one-half region ever since World War II. Moreover, since 1945 a
steadily increasing number of Americans are being admitted to mental
hospitals, and, on the average, are spending an increasingly long stay there.
And statistical indicators of this sort are readily matched by a vast body of
other, more subjective psychiatric data. Moreover, even political observers,
who certainly have their hand on the nation’s pulse, have begun to be con-
cerned over our inability to translate augmented personal affluence into in-
creases in happiness. Thus Richard Nixon in his first State of the Union
message said: “Never has a nation seemed to have had more and enjoyed
less.” And in his book Rich Man, Poor Man5 Henry Miller, chief of the
Population Division of the Bureau of the Census, observes that: “We seem
to be getting richer and richer in the number of things we own and poorer
in our ability to enjoy them.”
THE PREFERENCE FOR THE PRESENT
Now given such extensive—and continuing—indications that people’s

happiness is on the wane, it would seem clear that people would prefer the
circumstances of the bygone era. In the face of a widespread consensus that
Americans were happier a generation or so ago, it would seemingly follow
that people would hanker after “the good old days”. One would expect to
find that many or most people would prefer to have lived in this bygone,
happier time. When things are seen as going downhill, one would surely
119
prefer the past.

So, indeed, it might well appear. But the actual fact is just the reverse of
this expectation: the statistics actually obtained in the field indicate that
this expectation is altogether wrong. Here we have something of a paradox.
On the one hand, people incline to believe that “things are going to the
dogs”; on the other hand, people evince no real preference for “the good
old days”. And these findings are altogether typical of findings obtained
over the past generation. Invariably, Americans reject the would-rather-
have-lived-then-than-now option by a ratio of better than two to one.
EXPLAINING THE PARADOX: THE ROLE OF THE SUBJECTIVE
How can this paradox be explained?

This is an issue one can only approach on the basis of conjecture and
guess work. But a pretty plausible account can be developed along these
lines. The key lies in the consideration that satisfaction and happiness are
subjective issues that turn on subjective factors.
The desired account can, it would seem, be given in something like the
following terms: an individual’s assessment of his happiness is a matter of
his personal and idiosyncratic perception of the extent to which the condi-
tions and circumstances of his life meet his needs and aspirations. And
here we enter the area of “felt sufficiency” and “felt insufficiency”. A per-
son can quite meaningfully say, “I realize full well that, by prevailing stan-
dards, I have no good reason to be happy and satisfied with my existing
circumstances, but all the same I am perfectly happy and quite contented.”
Or, on the other hand, he may conceivably (and perhaps more plausibly)
say, “I know full well that I have every reason for being happy, but all the
same I am extremely discontented and dissatisfied.” We are dealing with
strictly personal evaluations.
In this context one is carried back to the old proportion from the school
of Epicurus in antiquity:6
Degree of satisfaction
The man who personal vision of happiness calls for yachts and polo ponies
will be malcontent in circumstances many of us would regard as idyllic. He
who asks but little may be blissful in humble circumstances. It is all a mat-
ter of how high one aspires in point of one’s expectations and aspirations.7
The issue of expectations deserves a closer look. People’s expectations
120
tend to be geared to the record of their past experience. And when im-
provements are subject to the limits of finitude which generally prevail in
human affairs, a situation of the following results:
___________________________________________________
Display 3
EXPECTATIONS OF FUTURE IMPROVEMENT
Level of Expected level of

achievement achievement
Actual level of
achievement
Present Envisaged future Time
Time
___________________________________________________
The phenomenon of deceleration is obvious here. We have here the

usual configuration of an S-shaped, sigmoid curve 01 development. Now
when we extrapolate past experience in a situation like this, we see that the
result is inevitably such that our expectations outstrip our attainments. The
inescapable result is one of dissatisfaction. Sure—things get “better”—
objectively speaking—but they don’t get better fast enough to meet our
subjective expectations.
On this basis, it becomes possible to provide a readily intelligible ac-
count for the—on first view, startling—phenomenon of increasing discon-
tent in the present era of improving personal prosperity and increasing
public care for private welfare. What we are facing is an escalation of ex-
pectations, a raising of the levels of expectations with correspondingly in-
creased aspirations in the demands that people make upon the
circumstances and conditions of their lives. With respect to the requisites
of happiness, we are in the midst of a “revolution of rising expectations”, a
revolution that not only affects the man at the bottom, but operates
throughout, to the very “top of the heap”. And—as our Epicurean propor-
121
tion shows, when increased expectations outstrip actual attainments—even

significantly growing attainments—the result is a net decrease in satisfac-
tion.8
The paradox mooted above is readily resolved on the basis of these
considerations. In the past people were happier because their achievements
lived up to—or exceeded—their expectations. With us—even though our
level of achievement is higher (and therefore our demands greater)—a
lower degree of satisfaction is bound to result because of a greater shortfall
from expectations. But nevertheless we will not want to exchange our cir-
cumstances for the subjectively happier (but objectively less well off) cir-
cumstances of the past. It would seem that Americans have come to require
more of life to achieve a given level of happiness. Their view seems to be:
“To be sure, given what little people asked of life in those ‘simpler’ days,
what they had was quite sufficient to render them happy, or at any rate
happier than we are today—we who have more than they. But of course
we, with our present expectations, would not be very happy in their
shoes.”9
CONCLUSION
A review of some of the main points at issue from our deliberations is

now in order.
(1) Thinking about happiness and technological progress tends to be dis-

torted by a deep-rooted tendency in men to think well of the past:
“the good old days” (les bons vieux temps—every language has the
phrase; nostalgia is universal). Cualquiera tiempo pasado fue meior
says the old Spanish proverb. The Romans had a word for it: glorifi-
ers of the past (laudatores temporis acti). This is simply an amusing
fact that tends to color our thinking about this issue, and that we have
to learn to discount for if we are going to think realistically.
(2) Contentment and satisfaction seem to depend on very basic elements

of the human condition, factors which our technical progress leaves
largely untouched and which actually do not admit of ready manipu-
lation.
(3) There is what might be called the Fundamental Paradox of Pro-

gress: Progress produces dissatisfaction because it inflates expecta-
122
tions faster than it can actually meet them. And this is virtually inevi-
table because the faster the expectations actually are met, the faster
they escalate.
And at this stage it becomes necessary to dwell on dangers of turning

against reason. Science and technology just cannot deliver on the $64,000
question of human satisfaction and happiness because—in the final analy-
sis—they simply do not furnish the stuff of which real happiness is made.
And here lies the slippery slope of the dangerous descent from anti-
scientism to anti-intellectualism to irrationalism. Only reason and intelli-
gence can solve our problems: If we turn against them, we are lost.
And many people seem prepared to turn away from reason and rational-
ity. To quote from one well-selling recent book (significantly entitled Sci-
ence is a Sacred Cow by Anthony Standen):10
Modern life in this country is highly unnatural. Machines, telephones, radios, vi-
tamin pills, subways, cars, trams, airplanes, elevators, injections, television ... all
products of science, and all intended individually to help us, collectively carry us
day and night and drive us to stomach ulcers or the psychiatric ward. (p. 173)
Such a point of view marks a significant and increasingly prominent ten-

dency of thought. In the 1920s and 30s thoughtful and socially concerned
people looked on science and technology as man’s best hope and friend.
Exactly the same sort of people nowadays unhesitatingly dismiss this view
as hopelessly naive. Indeed science and technology are nowadays often
seen as “the enemy” of all that is good and humane. Thus even so informed
a thinker as the distinguished bacteriologist René Dubos in his recent book
Reason Awake11 draws the contrast between the past, when man was
threatened by natural forces he could not control, and the present, when our
most potent fears are engendered by the malign effects—or side effects—
of science and technology.
Surely, great dangers loom ahead along this road. Science, technology,
and education in general nowadays present the nation with an enormous
bill in terms of human and material resources. As long as people maintain
the illusion that they are a royal road to human contentment, they will foot
this bill willingly. But what if such disillusionment reaches serious propor-
tions not just with respect to science, but the whole area of the life of the
mind?
Science and technology will not—cannot—produce the millennium.
And yet in a crowded world of very limited resources we cannot create an
123
adequate setting for human life without them. To recognize that scientific
rationality is not a sufficient condition for human happiness is one thing—
and represents a true insight. However, to reject it as not being a necessary
condition of human welfare would be a gross mistake. It makes no sense to
join the cult of anti-reason in turning our backs on science and technology.
The poor workman always blames his tools; but in this context the difficul-
ties lie not with the tools but in our capacity to make intelligent use of
them.
Also, perhaps most seriously, it is worth dwelling on the dangers of an
inflation of expectations. Throughout the history of this country, each gen-
eration has addressed itself to life on the premiss that the conditions and
circumstances of its children would be better than its own. Our faith in
“progress” runs deep. What is life going to be like when this expectation is
abandoned—or even reversed? Turning expectations around in a zero-
growth world is no easy matter. It will be a very difficult thing to get peo-
ple who have been taught that every day in every way things are getting
better and better to accept the idea that the millennium is not around the
corner. There is no need to elaborate upon the whole collage of grumbling,
discontent, search for scapegoats, political extremism, and so on, that lies
in this direction.
Now if the recent escalation of expectations in regard to the requisites
of happiness were to continue unabated, then a tragic time of reckoning
will come. But man is a creature that learns by experience, and a harsh cur-
riculum of unpleasantly monitory experiences lies ahead.
* * *
Finally, let us look back to the initial question: Does scientific and tech-
nological progress promote human happiness? Regrettably, that the an-
swer will have to be no. For something akin to a principle of the
conservation of negativity seems to be operative in human affairs. It is a
cruel “fact of life” that the achievement of real progress need not be ac-
companied by any commensurate satisfaction. And there is nothing per-
verse about this: it is all very “natural”. Man (as we know him in the West)
tends to be a creature of discontent—be they divine or otherwise. The im-
minent goal once achieved, he simply raises his level of expectation and
presses onward to the next goal under the goad of renewed discontent.
One result of this tendency—a result that may properly be viewed as
unfortunate—is what might be characterized as the phenomenon of he-
124
donic discounting. This is best explained by an analogy. It is a familiar

commonplace that the stock market primarily responds not to the present
economic facts but to anticipations of the future. Making present allowance
for foreseeable future economic improvements(or declines), the market has
already discounted them by anticipation when they become a reality, and
so under-reacts to or even ignores major achievements when they occur. A
parallel phenomenon operates in the context of foreseeable improvements
in the conditions of human life: a similar undervaluation of realized
achievements in the light of prior expectations. Having expected as much
(or generally more), we simply refuse to value very real achievements at
their own true worth. Once progress is achieved, it becomes discounted as
regards its real contribution to happiness: by the time an achievement is
made, we have already “raised our sights” in anticipation of its successors.
The considerations in this discussion point to the ironic conclusion that ad-
vances have in the past—through their promotion of an escalation of ex-
pectations—been self-defeating from the standpoint of promoting
happiness. The progress that has been made—real though it is—has never-
theless tended to bring in its wake a diminution rather than an increase in
“the general happiness” of Americans.
Humanists are—or should be—lovers of reason, and one’s every ra-
tional bone cries out that, ideally speaking, people ought to be happier as
their conditions of life improve. But recalcitrant circumstances of the real
world indicate that they do not in fact become so. And seemingly for a
deep-seated reason. As concerns happiness, progress sets a self-defeating
cycle into action:
Improvement → Aroused Expectations → Disappointment
It seems that we must bring ourselves to realize—more in sorrow than an-

ger—that it is a forlorn hope to expect technological progress to make a
major contribution to human happiness, taken in its positive aspect.12
125
NOTES
1
Leibniz: Selections, ed. by P.P. Weiner (New York: 1951), pp. 584-585.
2
Quoted in J.B. Bury, The Idea of Progress (London: 1920).
3
The discussion of this section and its immediate successor draws on the author’s
book Welfare: The Social Issues in Philosophical Perspective (Pittsburgh: 1972).
4
Of course, judgments of this sort—even about oneself—are notoriously problem-
atic:
It is hard enough to know whether one is happy or unhappy now, and still harder to
compare the relative happiness of different times of one’s life; the utmost that can
be said is that we are fairly happy so long as we are not distinctly aware of being
miserable.—Samuel Butler, The Way of All Flesh.
5
New York: Thomas Y. Crowell, 1971.
6
A. Campbell, P.E. Converse, and W. Rodgers, The Quality of American Life,
makes use of this Epicurean formula and offers detailed support for it. Some other
empirical studies regarding this bit of speculative philosophy as to the relationship
between expectation and (probable) achievement are: Arnold Thomsen: “Expecta-
tion in Relation to Achievement and Happiness,” Journal of Abnormal Social Psy-
chology, vol. 38 (1943): pp. 58-73. Other related discussions and further references
are given in James G. March and Herbert A. Simon, Organizations (New York:
Wiley, 1958); Richard M. Cyert and James G. March, Behavioral Theory of the
Firm (Englewood Cliffs, N.J.: Prentice Hall, 1963); T. Costello and S. Zalkinf,
Psychology in Administration (Englewood Cliffs, N.J.: Prentice Hall, 1963); see pt.
2, “Needs, Motives, and Goals.” It is worth noting that often one finds “aspiration”
in place of “expectation” in the denominator of the basic proposition. The differ-
ence is important but subtle. The enterprising person may aspire to more than he
expects to realize; the all out optimist may expect to realize more than what he as-
pires to.
7
Rousseau’s Emile—works this line of thought extensively: “True happiness con-
sists in decreasing the difference between our desires and our powers in establish-
ing a perfect equilibrium between the power and the will ... (M)isery consists not in
the lack of things but in the needs which they empire.” (I owe this reference to Pe-
ter Hare.)
8
An important lesson lurks in this finding, to wit, that consideration of only the idio-
syncratic happiness of a society’s members is a poor measure of its attainments in
the area of social welfare. It would only be a good measure in a society whose ex-
126
manner, that is, in a gradualistic pattern that did not automatically leap beyond in-
creasing attainments.
9
For an interesting discussion of cognate issues see Philip Brickman and Donald T.
Campbell, “Hedonic Relativism and Planning the Good Society,” in M.H. Appley
(ed.) Adaptation Level Theory (New York and London, 1971), pp. 287-302. One of
the interesting points of. this discussion is its conclusion that “there may be no way
to permanently increase the total of one’s pleasure except by getting off the he-
donic treadmill entirely” (p. 300).
10
A good deal of recent antiscientism is surveyed in Bernard Dixon, What is Science
For? (London, 1973; reprinted as a Pelican Book in 1976.)
11
New York, Columbia University Press, 1970.
12
This chapter is an abridged version of a paper originally published in Philosophical
Exchange, vol. 2 (1978), pp. 64-79.
127
Chapter 11
THE SOCIAL VALUE OF A LIFE
THE COST-BENEFIT APPROACH TO RISK REDUCTION
I n the social management of risks many public-sector decisions hinge on

the issues of determining if various measures to reduce risk to life are
worthwhile and of comparing their relative value. On the surface, such
problems seem to have a straightforward solution: Estimate how many
lives (whole or fractional) the measure may be expected to save, and then
multiply the resulting number by the “value of a life”. Accordingly, we are
told that decisions about public expenditures relating to such matters as
compensation for system-failure or mandating safety devices or managing
hazardous facilities, implicitly rest on assessing the value of human lives in
essentially economic terms. Decisions as to what sorts of measures are
worth-while here are seen as simply a matter of comparing the cost of the
protective measure on the one hand with the product of lives saved by “the
value of a life” on the other. Consider just one example of this line of
thought:
A few year ago Business Week reported an economic analysis of what is

called “underride” protection for a truck. When a truck stops unexpect-
edly, a car behind is may run under its rear, making the accident more
likely to result in the death of the riders in the car. The underride
protection on a truck would keep the car from going underneath.
Following an accident in which two people were killed, a federal
agency determined that it would cost only a few hundred dollars a year
per truck to install the device and maintain it. That does not seem like
much. But not too many accidents like this happen, and another federal
agency had added up the costs and calculated that to install the safety
device on all trucks would cost a total of about $5 million for each life
saved, an amount considerably more, they figured, than the $200,000 or
so that society could reasonably be expected to pay. From an economic
standpoint, the agency said, “Rulemaking on underride guards is not
point, the agency said, “Rulemaking on underride guards is not attrac-

tive.”1
Discussions of such issues often contemplate a decision-theoretic ap-

proach to risk assessment throughout the whole range of life-threatening
situations, maintaining that a quantitative determination of “the value of a
life” can and must be appraised. But unfortunately, this immerses the
whole issue in darkness rather than light. For this matter of determining
“the value of a life” embarks us upon a vain and Quixotic quest for some-
thing that no one has ever been able to determine to other people’s satisfac-
tion.
THE “SOCIAL VALUE OF A LIFE” AS PROBLEMATIC
One must, of course, assume that people will set the value of their own
lives very high indeed. To be sure, even for the individual this value will
not be infinite. Most or many people would, quite likely, be prepared to
give up life for things held dearer yet. And in fact all of us are prepared to
put our lives at risk for various objectives. Thus one recent study, which
examined salary as a function of occupational risk, concluded that a pre-
mium of about $200 a year was sufficient to induce workers in risky occu-
pations to accept an increase of .001 in the annual probability of accidental
death, a finding the authors took to indicate a life-valuation of some
$200,000.2 Still, the fundamental issue in the context of public policy is not
what value individuals set on their own lives but what an individual’s life
is worth to “the society”, to the whole environing community or body pub-
lic.
At this stage we encounter various suggestions. In particular, it has been
proposed that the social value of a life should be measured:3
1. In terms of social investment. This is a matter of evaluating the tal-

ent, resources, and services (doctors, nurses, school, etc.) that have
been dedicated to the nurturing and training of individuals.
2. In terms of social contributions. This is a matter of evaluating—

perhaps by way of estimates of lost future income or of lost produc-
tion—the productive contributions to be expected from the economic
activities of the individual.
130
3. In terms of society’s expenditures in such preventive or compensat-

ing matters as safety measures, preventive health maintenance, in-
surance, or the like.
However, any such yardstick is highly problematic and betokens a some-

what counterintuitive slant on the issue. Each particular alternative engen-
ders anomalies by leading to unacceptable consequences. As regards
investment, one could argue that society makes the greatest investment in
criminals and those who need institutional support. Contributions, on the
other hand, militate towards a highly elitist criterion, obviously favoring
the intelligent or talented over their less gifted compatriots. Expenditures
reflect the affluence of society and the drift of political fashion.
To be sure, society has one familiar and readily available device for put-
ting a price on things—the market. But this instrumentality is unavailable
in the present case. With the ending of slavery there is—and should be—
no market in human lives.
Other possibilities for life-evaluation can be envisaged. But the upshot
is always the same. All such proposals yield highly questionable and coun-
terintuitive results in various concrete applications. To be sure, in some
cases external circumstances may enforce an artificial answer where natu-
ral conditions do not provide one. If a judge is to make a monetary award
in the wake of an accidental loss of life, he must fix upon some figure or
other. (Here future income is a not unreasonable standard insofar as the ob-
ject of the exercise is to provide compensation for the survivors.) But all
such arrangements are artificial, ad hoc contrivances that do not admit of
plausible generalization.
THE “SOCIAL VALUE OF A LIFE” AS A CHIMERA
The idea that there is such a quantity as the “social value of a life” is
deeply mistaken. There is no such thing as a well-defined quantity out
there waiting to be measured. The question of “the value of a life” pushes
beyond the proper limits of cost-benefit analysis in its insistence on quanti-
fying something that is inherently unquantifiable. The “value of a life” to
society (as measured of the worth to society of preserving a person’s life)
is a chimera. It approaches the issue on the basis of the presupposition that
such a quantity actually exists in a stable and well-defined manner. And
this is far from being the case. To be sure, in this case as in other cases of
negativity-evaluation people could certainly decide to settle the issue by
131
fiat, fixing by a decision upon a value left undefined in the objective nature
of things. But even a cursory look at the facts of the matter shows that this
is something that, in the present domain of consideration, they are in fact
wholly unwilling to do.
The “value of a life” approach to risk calculation is predicated in the
mistaken supposition that the question “How much is it worth to society to
prevent the death of a person?” has a uniform and determinate answer. But
it does not. The question is ill defined. For it runs recklessly past the cru-
cial issue of death by what means?
The important fact is that society values death-avoidance differentially.
Death by industrial accident is one sort of thing, by murder another, by a
sporting misadvantage a third. (Individuals do much the same thing—being
killed in an auto collision is one thing, being murdered by one’s own chil-
dren another!) The question “How much is it worthwhile to spend to re-
duce (by a fixed amount) the risk of a person’s death?” does not have a
unique answer—the “reasonable” answer will vary with the mode of death
and manner of risk at issue. Threat of death is not a fixed and homogene-
ous item: we appraise different modes of death differently. Note for one
thing, how differently we appraise it in relation to the particular sort of
causal mechanism at issue. Thus occurrence of n deaths as the result of a
freakish cold-spell would be seen as a negativity entirely different from the
same n deaths if resultant from an accident in a nuclear power plant. The
qualitative features of life-threatening eventuations are crucial to their ap-
praisal as negativities. For example the threat-proximity of a given cause of
death—in terms of the issue of “there but for the accidents of fate I go my-
self”—will enter in. Major hotel fires accordingly create a more substantial
response in public concern than dam collapses. When fatality-causing ac-
cidents involve people who are isolated from the rest of society, public re-
sponse is considerably dampened. The temporal pattern of responses to the
recent failure of the Teton Dam in Idaho, which caused considerable loss
of life, indicates that even such major disasters are quickly forgotten—in
contrast to say a nuclear reactor accident or a failed vaccination program.
Again, the issue of voluntariness is crucial—is the mode of death wholly
voluntary (suicide), or with an element of decisional risk-taking (swim-
ming accident), or wholly unvoluntary (incurable disease)? While the ac-
tual loss is the same in any case, such qualitative features of its mode of
eventuation make all the difference. And this is by no means something ir-
rational. For example, actions which can issue in involuntary death are
quite rightly perceived as posing graver threats to the community.
132
Consider the data regarding fatal accident frequency rates set out in
Display 1. If society viewed life-loss in terms of some uniform planning-
factor, it is clear that it would act to restrict socially unnecessary risks to
life in a much more extensive way. Horse racing and professional boxing
would long ago have gone the way of cock-fighting. Would-be canoers and
mountain climbers would find themselves hedged in by innumerable regu-
lations and restrictions. But that’s just not the way it is. In the context of an
individual’s risk-taking it is his own personal value-system that is determi-
native, in the context of public policy it is the matter of group attitudes as
expressed through public decision-making that alone can (and should) de-
termine the “social value of a life”.
_______________________________________________
Display 1
FATAL ACCIDENT FREQUENCY RATES
(accidental deaths per hundred million hours of exposure)
OCCUPATIONAL
Industry generally 4
Steel industry 8
Coal Mining 40
Construction 67
Air Crew 250
Professional Boxing 7,000
Racing Jockeys (National Hunt) 50,000
NON-OCCUPATIONAL
Staying at home 3
Traveling by car 57
Bicycling 96
Traveling by air 240
Motorcycling 660
Canoeing 1,000
Mountain Climbing 4,000
SOURCE: S. B. Gibson, “Risk Criteria in Hazard Analysis,” Chemical Engineering

Progress, vol. 92 (1976), pp. 59-62. (Data for the UK for the early 1970s.)
___________________________________________________
133
In actuality, society views various threats to life in a highly differential

way. What concerns it is not loss of life or value of life per se, but the
qualitative character of the threat to life. Evaluation turns crucially on the
exact nature of the death being risked or the mode by which it is risked. In
the eight years of active U.S. involvement in the Vietnam war (1965-73),
some 60,000 Americans were killed and our society came close to being
rent at the seams. Yet some 60,000 Americans are killed in automobile ac-
cidents every year and yet the political system fails to bat an eyelash. The
fact is that people see the answer to the question of “How much is it worth
expending to reduce the risk of fatality at issue?” to differ drastically from
case to case.
An approach geared to finding “the value of a life” is misguided be-
cause it sees uniformity where there is ineliminable diversity. What counts
is not just whether X is killed but how X is killed: not just the seriousness
of the risk but its detailed qualitative character becomes a paramount con-
sideration. This is manifested by the vast differences in public concern
about and public reaction to various sorts of death. This circumstance is
which strongly suggests that the public’s concern for a mode of death (as
measured by its “newsworthiness”) varies inversely with its frequency—to
the extent to which it has become “old hat” and thus familiar and accepted
as “a normal fact of life”. There is clearly an enormous variation in the
proportion of people who are “shocked and distressed” at the occurrence of
various deaths. The “newsworthiness” of various types of fatalities tends to
vary inversely with their frequency—to the extent to which it has become
“old hat” and thus familiar and acceptable as “a normal fact of life”. We
tent to react very differently to auto accidents or industrial accidents on the
one hand and parenticides or terrorist killings on the other. To say that
“dead is dead” is to oversimplify and distort the social realities.
All the evidence we can gather indicates that the value society attaches
to reducing risks of death is not something merely quantitative and statisti-
cal but calls for detailed attention to such qualitative aspects as the nature
of the death being risked or the mode by which it is risked. The “social
cost” of different forms of death is very different.
If this were the end of the matter, the situation would be simpler than it
actually is. One would simply say: “Well, so there just is no one single
quantity: ‘the values of a life’; there are several such quantities, varying
with the particular life-threat at issue.” But even this will not do. For there
yet remain still further qualitative distinctions of context that complicate
134
this notion beyond reasonable manageability. Any conscientious attempt at

defining a specific “value of life” will ultimately come to naught—it will
die the death of a thousand cuts through an endless proliferation of distinc-
tions and aspects.
For example, the notion of a fixed value of a life is also geared to mis-
taken preconceptions in another sort of way. It is bound up with the idea
that a social loss through multiple deaths (with n victims) is not too far re-
moved from the product n times “the value of life”. But this seems by no
means to be the case. In an interesting recent study, Baruch Fiuschhoff and
his colleagues asked subjects to specify one of the three alternatives of
Display 2 as representing their judgment as to how society should evaluate
life-loss in multi-fatality situations.4 Curve 2 was the predominant choice,
selected by well over half the subjects. The loss of a life is clearly seen by
people not as a fixed quantity, but as something highly context-dependent,
involving, among other things, the question of how may further lives are
being lost in the context at issue. When life loss is part of a wider disaster
it is the disastrousness of this development—the over-all life-loss—that
becomes the predominant consideration in the assessment. One might, in
fact, envisage a kind of Richter scale as in Display 3 to underwrite meas-
urement-comparisons for man’s fatal encounters with man of nature, but
with the “social cost” of each eventuation seen as separated from its prede-
cessor by some multiplicative factor greater than a mere order of magni-
tude.
___________________________________________________
Display 2
ALTERNATIVE SOCIAL-COST CURVES FOR LIFE-LOSS IN

MULTI-FATALITY SITUATIONS
Social cost Social Cost Social Cost
012••••N 012••••N 012••••N

# of lives lost # of lives lost # of lives lost
at one time at one time at one time
Curve 1 Curve 2 Curve 3
___________________________________________________
135
An approach geared to finding “the value of a life” is accordingly mis-

guided, because the matter of “the value of a life” is something highly dif-
ferential with respect to such qualitative considerations as the particular
threat to life that is at issue, and many other matters as well. And here we
come up against society’s decision—an essentially political decision—how
much it is worthwhile to expend on controlling this or that particular sort
of risk to life.
To suppose such a thing as “the value of a life” is to hypostatize some-
thing that simply fails to be there in the objective nature of things. It is like
supposing that there is such a thing as “the price of an item in a department
store”. The issue hinges on too many things to yield something that is well-
defined at that level of generality. The very conception is based on a mis-
taken supposition of context-irrelevancy—that because something (the cost
of a pound of food, the weight of a cubic foot of metal, the value of a hu-
man life) is defined in this or that particular circumstance or situation it is
meaningful at a more general level. It illicitly absolutizes something that at
best exists in a way that involves an endlessly varied spectrum of context-
dependency.
___________________________________________________
Display 3
THE SCALE OF FATAL ENCOUNTERS
No. Killed Sample Event

(order of magnitude)
10˚ = 1 a murder or assassination

101 a serious auto collision
102 a major plane crash or a holiday weekend on U.S.
roads
103 a major flood or earthquake
104 a Napoleonic battle
105 a medieval plague
106 a major famine
107 World Wars I, II;
the 1918-19 influenza pandemic
108 World War III
___________________________________________________
136
Society can—in particular contexts—put a price of life. (In awarding

compensation to the estate of a victim of culpable negligence, a judge does
just exactly that.) But this is something that does not generalize—it has no
potential for carry-over into other, substantially different situations. The
judge is making decision and not an estimate of a preexisting—stable and
well-defined—quantity. Facing the variation among existing values-of-life,
one set of authors summarizes that “a definitive value may still elude us.”5
Our analysis suggests that this does not do justice to the complexities of
the case. To reemphasize: there simply is not well-defined quantity there
waiting to be measured.
CONCLUSION
An approach geared to finding “the value of a life” is accordingly mis-

guided. For the fact is that the “value of a life” is something highly differ-
ential with respect to nature of threat.
Is it unreasonable or even irrational that different life threats should be
seen in these very different lights? By no means. As we have already seen
in a wide variety of contexts, the way in which a “given result” is reached
can (quite properly) affect one’s valuation of the outcome. Deaths by nu-
clear accident just are seen by people as particularly horrendous in relation
to other causes of fatality. Evaluation looks not just to ends but to means as
well. And there is no reason why this could and should not be so in the
present case as well. There is no reasonable basis for the charge of “incon-
sistency” or “irrationality” in treating different sorts of threats to life dif-
ferently despite the sameness of the life threatened. To think otherwise is
to admit the gravely mistaken supposition that uniformity is something
normal, natural and appropriate in cases of this sort.
It is important to stress that the infeasibility of measuring “the value of
a life” is emphatically not as damaging to the prospect of rational risk-
management as might on first thought seem to be the case. Consider the
following example:
U.S. health authorities ... [face the fact that] the bill for a [heart] trans-
plant, nearly always paid by private plans of public funds, ranges from
$30,000 to 190,000 [and] postoperative ambulatory care costs $2,500 a
year. The trustees of Boston’s Massachusetts General Hospital this year
voted against starting a transplant program partly because they reckoned
137
that each patient would consume as much of the hospital resources as

eight open heart operations.6
Note that the question here does not come in the form of whether it is
worthwhile to save a life at a conceivable cost of say, 200,000 U.S. dollars,
but “rather in the form of the comparative question”. Given that only cer-
tain limited resources are available for heart surgery, “are they better in-
vested in this or that form of lifesaving therapy?” The reasoning becomes a
matter of comparison—of dominance-argumentation. Whenever the matter
can be cast in this comparative forms—as it frequently can—the risk-
management decisions become wholly freed from any supposed need to
measure “the value of a life”. Thus the whole spectrum of cost-benefit
questions of the form “If [a specific amount of] additional money were to
be allocated to disease control programs which program would show the
highest payoff in terms of lives saved and disability prevented per dollar
spent?”7 can be handled largely on such comparison/dominance princi-
ples—no life-valuation in economic terms is requisite at all.
Our inability to measure “the value of a life” need not impede rational
deliberation about the management of life-threatening risks, provided suf-
ficient care is taken to set the questions up in an appropriately manageable
form. An orthodox decision-theory approach that stands committed to the
standard reliance on quantitative measures, is—fortunately—not our only
resource for rational deliberation. To say this is not, of course, to say the
matter of risk-management falls outside the economics altogether, but only
that the economics of the enterprise is bound to be subtle and complex—
and fraught with evaluative considerations, which is to say political ones.
These objections to a mathematicized decision-theory analysis that pro-
ceeds in terms of “the value of a life” are not a matter of sentimental objec-
tions to hard-headed thinking, arising from an emotional antipathy to
assessing the value of human life in economic terms. Be such an attitude
justified or not, it is not the point of our present deliberations. Our point is
the perfectly hard-headed one that the issue has an inner complexity that
the value-of-a-life approach simply cannot capture.8
The key question is that of a society’s ability and willingness to pay for
safety for minimizing the risk of life-loss. The resolution of such issues is a
matter of social values channeled through the political process—a matter
of decision not of measurement, of actual social-policy formulation rather
than of mere fact-finding. This is true, in particular, when the issue is not
of the form “Is a certain fixed volume of resources more efficiently in-
138
vested in this form of life-protection or in that one?”, but takes the form
“How much by way of resources should we invest in a given form of life
protection (as opposed, say, to some mode of life-enhancement)?” The po-
litical process cannot avoid the hard choices of decision by invoking some
technical expert’s assessment of “the value of a life”, it must address the
issues head-on, without the comforting cushioning of a supposedly imper-
sonal technical mechanism able to avert direct confrontation with difficult
normative issues.9
NOTES
1
“The Cost-Benefits of Saving Lives,” in Social Accounting: Theory, Issues and
Cases ed. by Lee. J. Seidler and Lynn L. Seidler (Los Angeles: Melville, 1975), p.
319. The account is based on an article in Business Week (October,14, 1972), p. 41.
2
R. Thaler and S. Rosen, “The Value of Saving a Life: Evidence from the Labor
Market” in N. Terlechyj (ed.) Household Production and Consumption (New York:
Columbia University Press, 1976). A replication of this study under somewhat dif-
ferent assumptions increased this figure by one order of magnitude. (See E. Rap-
poport, unpublished Ph.D. dissertation, University of California as Los Angeles,
1977.) The linearity assumption involved in such calculations is clearly something
questionable. The man who accepts a 1% chance of death for $10,000 may well
balk at accepting $1,000,000 for certain death.
3
The following may be consulted regarding these issues: E. J. Mishan, “Evaluation
of Life and Limb: A Theoretical Approach,” Journal of Political Economy, vol. 79
(1971), pp. 687-705; Seidler and Seidler (eds.), op. cit.; R. Zeckhauser, “Proce-
dures for Valuing Lives,” Public Policy, vol. 23 (1975), pp. 419-464; Ralph Estes,
Corporate Social Accounting (New York: Wiley, 1976), pp. 136ff.
4
Baruch Fischhoff, et. al., “Knowing What You Want: Measuring Labile Values” in
T. Wallsten (ed.), Cognitive Process in Choice and Decision Behavior (Hillsdale,
NJ: Erlbaum, 1980).
5
Baruch Fischhoff, et. al., “What Risks Are Acceptable?” Environment, vol. 21
(1979), pp. 17-38 (see p. 19).
6
Time Magazine, November 17, 1980; p. 92.
139
7
Robert N. Grosse, “Problems of Resource Allocation in Health” in The Analysis
and Evaluation of Public Expenditures: The PPB System (Washington, Subcom-
mittee on economy in Government of the Joint Economic Committee, U.S. Con-
gress 1969), vol. 3, pp. 1208-1225. See also this author's Effectiveness in Saving
Lives as a Resource Allocation Criteria” in Seidler and Seidler, op. cit., pp. 330-
346.
8
For an interesting discussion of very different aspects of the issue see T. C. Schel-
ling, “The Life You Save May Be Your Own” in S. B. Chase Jr. (ed.), Problems in
Public Expenditure Analysis (Washington, D.C.: The Brookings Institute, 1966),
pp. 127-162.
9
This chapter was originally published in M. Bradie and K. Sayre (eds.), Reason
and Decision (Bowling Green, OH: Bowling Green Studies in Applied Philosophy,
1982).
140
Chapter 12
THE ALLOCATION OF EXOTIC

MEDICAL LIFESAVING THERAPY
THE PROBLEM
T echnological progress has in recent years transformed the limits of the

possible in medical therapy. However, the elevated state of sophistica-
tion of modern medical technology has brought the economists’ classic
problem of scarcity in its wake as an unfortunate side product. The enor-
mously sophisticated and complex equipment and the highly trained teams
of experts requisite for its utilization are scarce resources in relation to po-
tential demand. The administrators of the great medical institutions that
preside over these scarce resources thus come to be faced increasingly with
the awesome choice: Whose life to save?
A (somewhat hypothetical) paradigm example of this problem may be
sketched within the following set of definitive assumptions: We suppose
that persons in some particular medically morbid condition are “mortally
affected”: It is virtually certain that they will die within a short time period
(say ninety days). We assume that some very complex course of treatment
(e.g., a heart transplant) represents a substantial probability of life prolon-
gation for persons in this mortally afflicted condition. We assume that the
facilities available in terms of human resources, mechanical instrumentali-
ties, and requisite materials (e.g., hearts in the case of a heart transplant)
make it possible to give a certain treatment—this “exotic (medical) lifesav-
ing therapy”, or ELT for short—to a certain, relatively small number of
people. And finally we assume that a substantially greater pool of people in
the mortally afflicted condition is at hand. The problem then may be for-
mulated as follows: How is one to select within the pool of afflicted pa-
tients the ones to be given the ELT treatment in question; how to select
those “whose lives are to be saved”? Faced with many candidates for an
ELT process that can be made available to only a few, doctors and medical
administrators confront the decision of who is to be given a chance at sur-
vival and who is, in effect, to be condemned to die.
As has already been implied, the “heroic” variety of spare-part surgery

can pretty well be assimilated to this paradigm. One can foresee the time
when heart transplantation, for example, will have become pretty much a
routine medical procedure, albeit on a very limited basis, since a cardiac
surgeon with the technical competence to transplant hearts can operate at
best a rather small number of times each week and the elaborate facilities
for such operations will most probably exist on a modest scale. Moreover,
in “spare-part” surgery there is always the problem of availability of the
“spare parts” themselves. A report in one British newspaper gives the fol-
lowing picture: “Of the 150,000 who die of heart disease each year [in the
U.K.], Mr. Donald Longmore, research surgeon at the National Heart Hos-
pital [in London], estimates that 22,000 might be eligible for heart surgery.
Another 30,000 would need heart and lung transplants. But there are
probably only between 7,000 and 14,000 potential donors a year.”1 Envis-
aging this situation in which at the very most something like one in four
heart-malfunction victims can be saved, we clearly confront a problem in
ELT allocation.
A perhaps even more drastic case in point is afforded by long-term
hemodialysis, an ongoing process by which a complex device—an “artifi-
cial kidney machine”—is used periodically in cases of chronic renal failure
to substitute for a non-functional kidney in “cleaning” potential poisons
from the blood. Only a few major institutions have chronic hemodialysis
units, whose complex operation is an extremely expensive proposition. For
the present and the foreseeable future the situation is that “the number of
places available for chronic haemodialysis is hopelessly inadequate.”2
The traditional medical ethos has insulated the physician against facing
the very existence of this problem. When swearing the Hippocratic Oath,
he commits himself to work for the benefit of the sick in “whatsoever
house I enter.”3 In taking this stance, the physician substantially renounces
the explicit choice of saving certain lives rather than others. Of course,
doctors have always in fact had to face such choices on the battlefield or in
times of disaster, but there the issue had to be resolved hurriedly, under
pressure, and in circumstances in which the very nature of the case effec-
tively precluded calm deliberation by the decision maker as well as criti-
cism by others. In sharp contrast, however, cases of the type we have
postulated in the present discussion arise predictably, and represent choices
to be made deliberately and “in cold blood”.
It is, to begin with, appropriate to remark that this problem is not fun-
damentally a medical problem. For when there are sufficiently many af-
142
THE ALLOCATION OF EXOTIC MEDICAL LIFESAVING THERAPY
flicted candidates for ELT then—so we may assume—there will also be

more than enough for whom the purely medical grounds for ELT allocation
are decisively strong in any individual case, and just about equally strong
throughout the group. But in this circumstance a selection of some afflicted
patients over and against others cannot ex hypothesi be made on the basis
of purely medical considerations.
The selection problem, as we have said, is in substantial measure not a
medical one. It is a problem for medical men, which must somehow be
solved by them, but that does not make it a medical issue—any more than
the problem of hospital building is a medical issue. As a problem it belongs
to the category of philosophical problems—specifically a problem of moral
philosophy or ethics. Structurally, it bears a substantial kinship with those
issues in this field that revolve about the notorious whom-to-save-on-the-
lifeboat and whom-to-throw-to-the-wolves-pursuing-the-sled questions.
But whereas questions of this just indicated sort are artificial, hypothetical,
and far-fetched, the ELT issue poses a genuine policy question for the re-
sponsible administrators in medical institutions, indeed a question that
threatens to become commonplace in the foreseeable future.
Now what the medical administrator needs to have, and what the phi-
losopher is presumably ex officio in a position to help in providing, is a
body of rational guidelines for making choices in these literally life-or-
death situations. This is an issue in which many interested parties have a
substantial stake, including the responsible decision maker who wants to
satisfy his conscience that he is acting in a reasonable way. Moreover, the
family and associates of the man who is turned away—to say nothing of
the man himself—have the right to an acceptable explanation. And indeed
even the general public wants to know that what is being done is fitting and
proper. All of these interested parties are entitled to insist that a reasonable
code of operating principles provides a defensible rationale for making the
life-and-death choices involved in ELT.
THE TWO TYPES OF CRITERIA
Two distinguishable types of criteria are bound up in the issue of mak-

ing ELT choices. We shall call these Criteria of Inclusion and Criteria of
Comparison, respectively. The distinction at issue here requires some ex-
planation. We can think of the selection as being made by a two-stage
process: (1) the selection from among all possible candidates (by a suitable
screening process) of a group to be taken under serious consideration as
143
candidates for therapy, and then (2) the actual singling out, within this
group, of the particular individuals to whom therapy is to be given. Thus
the first process narrows down the range of comparative choice by elimi-
nating en bloc whole categories of potential candidates. The second proc-
ess calls for a more refined, case-by-case comparison of those candidates
that remain. By means of the first set of criteria one forms a selection
group; by means of the second set, an actual selection is made within this
group.
Thus what we shall call a “selection system” for the choice of patients
to receive therapy of the ELT type will consist of criteria of these two
kinds. Such a system will be acceptable only when the reasonableness of
its component criteria can be established.
ESSENTIAL FEATURES OF AN ACCEPTABLE ELT SELECTION

SYSTEM
To qualify as reasonable, an ELT selection must meet two important

“regulative” requirements: it must be simple enough to be readily intelligi-
ble, and it must be plausible, that is, patently reasonable in a way that can
be apprehended easily and without involving ramified subtleties. Those
medical administrators responsible for ELT choices must follow a modus
operandi that virtually all the people involved can readily understand to be
acceptable (at a reasonable level of generality, at any rate). Appearances
are critically important here. It is not enough that the choice be made in a
justifiable way, it must be possible for otherwise untutored people to “see”
(i.e., understand without elaborate teaching or indoctrination) that it is jus-
tified, insofar as any mode of procedure can be justified in cases of this
sort.
One “constitutive” requirement is obviously an essential feature of a
reasonable selection system: all of its component criteria—those of inclu-
sion and those of comparison alike—must be reasonable in the sense of be-
ing rationally defensible. The ramifications of this requirement call for
detailed consideration. But one of its aspects should be noted without fur-
ther ado: it must be fair—it must treat relevantly like cases alike, leaving
no room for “influence” or favoritism, etc.
144
THE BASIC SCREENING STAGE:

CRITERIA OF INCLUSION (AND EXCLUSION)
Three sorts of considerations are prominent among the plausible criteria

of inclusion/exclusion at the basic screening stage: the constituency factor,
the progress-of-science factor, and the prospect-of-success factor.
A. The Constituency Factor
It is a “fact of life” that ELT can be available only in the institutional

setting of a hospital or medical institute or the like. Such institutions gener-
ally have normal clientele boundaries. A veterans’ hospital will not con-
cern itself primarily with treating non-veterans, a children’s hospital cannot
be expected to accommodate the “senior citizen”, an army hospital can re-
gard college professors as outside its sphere. Sometimes the boundaries are
geographic—a state hospital may admit only residents of a certain state.
(There are, of course, indefensible constituency principles—say race or re-
ligion, party membership, or ability to pay; and there are cases of border-
line legitimacy, e.g., sex.4) A medical institution is justified in considering
for ELT only persons within its own constituency, provided this constitu-
ency is constituted upon a defensible basis. Thus the hemodialysis selec-
tion committee in Seattle “agreed to consider only those applications who
were residents of the state of Washington. ... They justified this stand on
the grounds that since the basic research ... had been done at ... a state-
supported institution—the people whose taxes had paid for the research
should be its first beneficiaries.”5
While thus insisting that constituency considerations represent a valid
and legitimate factor in ELT selection, I do feel there is much to be said for
minimizing their role in life-or-death cases. Indeed a refusal to recognize
them at all is a significant part of medical tradition, going back to the very
oath of Hippocrates. They represent a departure from the ideal arising with
the institutionalization of medicine, moving it away from its original status
as an art practiced by an individual practitioner.
B. The Progress-of-Science Factor
The needs of medical research can provide a second valid principle of in-
clusion. The research interests of the medical staff in relation to the spe-
145
cific nature of the cases at issue is a significant consideration. It may be

important for the progress of medical science-and thus of potential benefit
to many persons in the future—to determine how effective the ELT at issue
is with diabetics or persons over sixty or with a negative RH factor. Con-
siderations of this sort represent another type of legitimate factor in ELT
selection.
A very definitely borderline case under this head would revolve around
the question of a patient’s willingness to pay, not in monetary terms, but in
offering himself as an experimental subject, or by contracting to return at
designated times for a series of tests substantially unrelated to his own
health, but yielding data of importance to medical knowledge in general.
C. The Prospect-of-Success Factor
It may be that while the ELT at issue is not without some effectiveness
in general, it has been established to be highly effective only with patients
in certain specific categories (e.g., females under forty of a specific blood
type). This difference in effectiveness—in the absolute or in the probability
of success—is (we assume) so marked as to constitute virtually a differ-
ence in kind rather than in degree. In this case, it would be perfectly legiti-
mate to adopt the general rule of making the ELT at issue available only or
primarily to persons in this substantial-promise-of-success category. (It is
on grounds of this sort that young children and persons over fifty are gen-
erally ruled out as candidates for hemodialysis.)
We have maintained that the three factors of constituency, progress of
science, and prospect of success represent legitimate criteria of inclusion
for ELT selection. But it remains to examine the considerations which le-
gitimate them. The legitimating factors are in the final analysis practical or
pragmatic in nature. From the practical angle it is advantageous—indeed to
some extent necessary—that the arrangements governing medical institu-
tions should embody certain constituency principles. It makes good prag-
matic and utilitarian sense that progress-of-science considerations should
be operative here. And, finally, the practical aspect is reinforced by a
whole host of other considerations—including moral ones—in supporting
the prospect-of-success criterion. The workings of each of these factors are
of course conditioned by the ever-present element of limited availability.
They are operative only in this context; that is, prospect of success is a le-
gitimate consideration at all only because we are dealing with a situation of
scarcity.
146
THE FINAL SELECTION STAGE: CRITERIA OF SELECTION
Five sorts of elements must, as we see it, figure primarily among the
plausible criteria of selection that are to be brought to bear in further
screening the group constituted after application of the criteria of inclusion:
the relative-likelihood-of-success factor, the life-expectancy factor, the
family role factor, the potential-contributions factor, and the services-
rendered factor. The first two represent the biomedical aspect, the second
three the social aspect.
A. The Relative-Likelihood-of-Success Factor
It is clear that the relative likelihood of success is a legitimate and ap-

propriate factor in making a selection within the group of qualified patients
that are to receive ELT. This is obviously one of the considerations that
must count very significantly in a reasonable selection procedure.
The present criterion is of course closely related to item C of the preced-
ing section. There we were concerned with prospect-of-success considera-
tions categorically and en bloc. Here at present they come into play in a
particularized case-by-case comparison among individuals. If the therapy
at issue is not a once-and-for-all proposition and requires ongoing treat-
ment, cognate considerations must be brought in. Thus, for example, in the
case of a chronic ELT procedure such as hemodialysis it would clearly
make sense to give priority to patients with a potentially reversible condi-
tion (who would thus need treatment for only a fraction of their remaining
lives).
B. The Life-Expectancy Factor
Even if the ELT is “successful” in the patient’s case he may, considering

his age and/or other aspects of his general medical condition, look forward
to only a very short probable future life. This is obviously another factor
that must be taken into account.
C. The Family Role Factor
A person’s life is a thing of importance not only to himself but to oth-

ers—friends, associates, neighbors, colleagues, etc. But his (or her) rela-
147
tionship to his immediate family is a thing of unique intimacy and signifi-

cance. The nature of his relationship to his wife, children, and parents and
the issue of their financial and psychological dependence upon him are ob-
viously matters that deserve to be given weight in the ELT selection proc-
ess. Other things being anything like equal, the mother of minor children
must take priority over the middle-aged bachelor.
D. The Potential Future-Contributions Factor (Prospective Service)
In “choosing to save” one life rather than another, “the society,” through
the mediation of the particular medical institution in question—which
should certainly look upon itself as a trustee for the social interest—is
clearly warranted in considering the likely pattern of future services to be
rendered by the patient (adequate recovery assumed), considering his age,
talent, training, and past record of performance. In its allocations of ELT,
society “invests” a scarce resource in one person as against another and is
thus entitled to look to the probable prospective “return” on its investment.
It may well be that a thoroughly egalitarian society is reluctant to put
someone’s social contribution into the scale in situations of the sort at is-
sue. One popular article states that “the most difficult standard would be
the candidate’s value to society,” and goes on to quote someone who said:
“You can’t just pick a brilliant painter over a laborer. The average citizen
would be quickly eliminated.”6 But what if it were not a brilliant painter
but a brilliant surgeon or medical researcher that was at issue? One won-
ders if the author of the obiter dictum that one “can’t just pick” would still
feel equally sure of his ground. In any case, the fact that the standard is dif-
ficult to apply is certainly no reason for not attempting to apply it. The
problem of ELT selection is inevitably burdened with difficult standards.
Some might feel that in assessing a patient’s value to society one should
ask not only who if permitted to continue living can make the greatest con-
tribution to society in some creative or constructive way, but also who by
dying would leave behind the greatest burden on society in assuming the
discharge of their residual responsibilities.7 Certainly the philosophical
utilitarian would give equal weight to both these considerations. Just here
is where I would part ways with orthodox utilitarianism. For—though this
is not the place to do so—I should be prepared to argue that a civilized so-
ciety has an obligation to promote the furtherance of positive achievements
in cultural and related areas even if this means the assumption of certain
added burdens.8
148
E. The Past Services-Rendered Factor (Retrospective Service)
A person’s services to another person or group have always been taken to

constitute a valid basis for a claim upon this person or group—of course a
moral and not necessarily a legal claim. Society’s obligation for the recog-
nition and reward of services rendered—an obligation whose discharge is
also very possibly conducive to self-interest in the long run—is thus an-
other factor to be taken into account. This should be viewed as a morally
necessary correlative of the previously considered factor of prospective
service. It would be morally indefensible of society in effect to say: “Never
mind about services you rendered yesterday—it is only the services to be
rendered tomorrow that will count with us today.” We live in very future-
oriented times, constantly preoccupied in a distinctly utilitarian way with
future satisfactions. And this disinclines us to give much recognition to
past services. But parity considerations of the sort just adduced indicate
that such recognition should be given on grounds of equity. No doubt a jus-
tification for giving weight to services rendered can also be attempted
along utilitarian lines. (“The reward of past services rendered spurs people
on to greater future efforts and is thus socially advantageous in the long-
run future.”) In saying that past services should be counted “on grounds of
equity”—rather than “on grounds of utility”—I take the view that even if
this utilitarian defense could somehow be shown to be fallacious, I should
still be prepared to maintain the propriety of taking services rendered into
account. The position does not rest on a utilitarian basis and so would not
collapse with the removal of such a basis.9
As we have said, these five factors fall into three groups: the biomedical
factors A and B, the familial factor C, and the social factors D and E. With
items A and B the need for a detailed analysis of the medical considerations
comes to the fore. The age of the patient, his medical history, his physical
and psychological condition, his specific disease, etc., will all need to be
taken into exact account. These biomedical factors represent technical is-
sues: they call for the physicians’ expert judgment and the medical statisti-
cians’ hard data. And they are ethically uncontroversial factors—their
legitimacy and appropriateness are evident from the very nature of the
case.
Greater problems arise with the familial and social factors. They involve
intangibles that are difficult to judge. How is one to develop subcriteria for
weighing the relative social contributions of (say) an architect or a librarian
149
or a mother of young children? And they involve highly problematic is-

sues. (For example, should good moral character be rated a plus and bad a
minus in judging services rendered?) And there is something strikingly un-
pleasant in grappling with issues of this sort for people brought up in times
greatly inclined towards maxims of the type “Judge not!” and “Live and let
live!” All the same, in the situation that concerns us here such distasteful
problems must be faced, since a failure to choose to save some is tanta-
mount to sentencing all. Unpleasant choices are intrinsic to the problem of
ELT selection; they are of the very essence of the matter.10
But is reference to all these factors indeed inevitable? The justification
for taking account of the medical factors is pretty obvious. But why should
the social aspect of services rendered and to be rendered be taken into ac-
count at all? The answer is that they must be taken into account not from
the medical but from the ethical point of view. Despite disagreement on
many fundamental issues, moral philosophers of the present day are pretty
well in consensus that the justification of human actions is to be sought
largely and primarily—if not exclusively—in the principles of utility and
of justice.11 But utility requires reference of services to be rendered and
justice calls for a recognition of services that have been rendered. Moral
considerations would thus demand recognition of these two factors. (This,
of course, still leaves open the question of whether the point of view pro-
vides a valid basis of action: Why base one’s actions upon moral princi-
ples?—or, to put it bluntly—Why be moral? The present paper is, however,
hardly the place to grapple with so fundamental an issue, which has been
canvassed in the literature of philosophical ethics since Plato.)
THE INHERENT IMPERFECTION (NON-OPTIMALITY) OF ANY

SELECTION SYSTEM
Our discussion to this point of the design of a selection system for ELT
has left a gap that is a very fundamental and serious omission. We have ar-
gued that five factors must be taken into substantial and explicit account.
A. Relative likelihood of success. Is the chance of the treatment’s being

“successful” to be rated as high, good, average, etc.?12
B. Expectancy of future life. Assuming the “success” of the treatment,

how much longer does the patient stand a good chance (75 percent or
150
better) of living—considering his age and general condition?
C. Family role. To what extent does the patient have responsibilities to

others in his immediate family?
D. Social contributions rendered. Are the patient’s past services to his

society outstanding, substantial, average, etc.?
E. Social contributions to be rendered. Considering his age, talents,

training, and past record of performance, is there a substantial prob-
ability that the patient will—adequate recovery being assumed—
render in the future services to his society that can be characterized
as outstanding, substantial, average, etc.?
This list is dearly insufficient for the construction of a reasonable selection

system, since that would require not only that these factors be taken into
account (somehow or other), but—going beyond this—would specify a
specific set of procedures for taking account of them. The specific proce-
dures that would constitute such a system would have to take account of
the interrelationship of these factors (e.g., B and E), and to set out exact
guidelines as to the relevant weight that is to be given to each of them. This
is something our discussion has not as yet considered.
In fact, I should want to maintain that there is no such thing here as a
single rationally superior selection system. The position of affairs seems to
me to be something like this: (1) It is necessary (for reasons already can-
vassed) to have a system, and to have a system that is rationally defensible,
and (2) to be rationally defensible, this system must take the factors A-E
into substantial and explicit account. But (3) the exact manner in which a
rationally defensible system takes account of these factors cannot be fixed
in anyone specific way on the basis of general considerations. Any of the
variety of ways that give A-E “their due” will be acceptable and viable.
One cannot hope to find within this range of workable systems some one
that is optimal in relation to the alternatives. There is no system that does
“the (uniquely) best”—only a variety of systems that do “as well as one
can expect to do” in cases of this sort.
The situation is structurally very much akin to that of rules of partition
of an estate among the relations of a decedent. It is important that there be
such rules. And it is reasonable that spouse, children, parents, siblings, etc.,
be taken account of in these rules. But the question of the exact method of
151
division—say that when the decedent has neither living spouse nor living
children then his estate is to be divided, dividing 60 percent between par-
ents, 40 percent between siblings versus dividing 90 percent between par-
ents, 10 percent between siblings—cannot be settled on the basis of any
general abstract considerations of reasonableness. Within broad limits, a
variety of resolutions are all perfectly acceptable—so that no one proce-
dure can justifiably be regarded as “the (uniquely) best” because it is supe-
rior to all others.13
A POSSIBLE BASIS FOR A REASONABLE SELECTION SYSTEM
Having said that there is no such thing as the optimal selection system
for ELT, I want now to sketch out the broad features of what I would re-
gard as one acceptable system.
The basis for the system would be a point rating. The scoring here at is-
sue would give roughly equal weight to the medical considerations (A and
B) in comparison with the extramedical considerations (C = family role, D
= services rendered, and E = services to be rendered), also giving roughly
equal weight to the three items involved here (C, D, and E). The result of
such a scoring procedure would provide the essential starting point of our
ELT selection mechanism. I deliberately say “starting point” because it
seems to me that one should not follow the results of this scoring in an
automatic way. I would propose that the actual selection should only be
guided but not actually be dictated by this scoring procedure, along lines
now to be explained.
THE DESIRABILITY OF INTRODUCING AN ELEMENT OF

CHANCE
The detailed procedure I would propose—not of course as optimal (for

reasons we have seen), but as eminently acceptable—would combine the
scoring procedure just discussed with an element of chance. The resulting
selection system would function as follows:
1. First the criteria of inclusion of Section 4 above would be applied to

constitute a first phase selection group—which (we shall suppose) is
substantially larger than the number n of persons who can actually be
accommodated with ELT.
2. Next the criteria of selection of Section 5 are brought to bear via a
152
scoring procedure of the type described in Section 8. On this basis a

second phase selection group is constituted which is only somewhat
larger—say by a third or a half-than the critical number n at issue.
3. If this second phase selection group is relatively homogeneous as re-

gards rating by the scoring procedure—that is, if there are no really
major disparities within this group (as would be likely if the initial
group was significantly larger than n) —then the final selection is
made by random selection of n persons from within this group.
This introduction of the element of chance—in what could be dramatized

as a “lottery of life and death”—must be justified. The fact is that such a
procedure would bring with it three substantial advantages:
First, as we have argued above (in Section 7), any acceptable selection
system is inherently non-optimal. The introduction of the element of
chance prevents the results that life-and-death choices are made by the
automatic application of an admittedly imperfect selection method.
Second, a recourse to chance would doubtless make matters easier for
the rejected patient and those who have a specific interest in him. It would
surely be quite hard for them to accept his exclusion by relatively mechani-
cal application of objective criteria in whose implementation subjective
judgment is involved. But the circumstances of life have conditioned us to
accept the workings of chance and to tolerate the element of luck (good or
bad): human life is an inherently contingent process. Nobody, after all, has
an absolute right to ELT—but most of us would feel that we have “every
bit as much right” to it as anyone else in significantly similar circum-
stances. The introduction of the element of chance assures a like handling
of like cases over the widest possible area that seems reasonable in the cir-
cumstances.
Third (and perhaps least), such a recourse to random selection does
much to relieve the administrators of the selection system of the awesome
burden of ultimate and absolute responsibility.
These three considerations would seem to build up a substantial case for
introducing the element of chance into the mechanism of the system for
ELT selection in a way limited and circumscribed by other weightier con-
siderations, along some such lines as those set forth above.14
It should be recognized that this injection of man-made chance supple-
ments the element of natural chance that is present inevitably and in any
case (apart from the role of chance in singling out certain persons as vic-
153
tims for the affliction at issue). As F. M. Parsons has observed: “any va-
cancies [in an ELT program-specifically hemodialysis] will be filled im-
mediately by the first suitable patients, even though their claims for therapy
may subsequently prove less than those of other patients refused later. “15
Life is a chancy business and even the most rational of human arrange-
ments can cover this over to a very limited extent at best.16
NOTES
1
Christine Doyle, “Spare-Part Hear Surgeons Worried by Their Success,” The Ob-
server, May 12, 1968.
2
J. D. N. Nabarro, “Selection of Patients for Haemodialysis,” British Medical Jour-
nal (March 11, 1967), p. 623. Although several thousand patients die in the U.K.
each year from renal failure there are about thirty new cases per million of popula-
tion—only 10 percent of these can for the foreseeable future be accommodated
with chronic hemodialysis. Kidney transplantation—itself a very tricky proce-
dure—cannot make a more than minor contribution here. As this article goes to
press, I learn that patients can be maintained in home dialysis at an operating cost
about half that of maintaining them in a hospital dialysis unit (roughly an $8,000
minimum). In the United States, around 7,000 patients with terminal uremia who
could benefit from hemodialysis evolve yearly. As of mid-1968, some 1,000 of
these can be accommodated in existing hospital units. By June 1967, a world-wide
total of some 120 patients were in treatment by home dialysis. (Data from a forth-
coming paper, “Home Dialysis,” by C. M. Conty and H. V. Murdaugh. See also R.
A. Baillod et al., “Overnight Haemodialysis in the Home,” Proceedings of the
European Dialysis and Transplant Association, VI [1965], pp. 99 ff.).
3
For the Hippocratic Oath see Hippocrates: Works (London: Loeb Classical Library,
1959), I, p. 298.
4
Another example of borderline legitimacy is posed by an endowment “with strings
attached,” e.g.: “In accepting this legacy the hospital agrees to admit and provide
all needed treatment for any direct descendant of myself, its founder.”
5
Shana Alexander, “They Decide Who Lives, Who Dies,” Life, 53 (November 9,
1962), pp. 102-125 (see p. 107).
154
6
Lawrence Lader, “Who Has the Right To Live?” Good Housekeeping (January
1968), p. 144.
7
This approach could thus be continued to embrace the previous factor, that of fam-
ily role, the preceding item C.
8
Moreover a doctrinaire utilitarian would presumably be willing to withdraw a con-
tinuing mode of ELT such as hemodialysis from a patient to make room for a more
promising candidate who came to view at a later stage and who could not otherwise
be accommodated. I should be unwilling to adopt this course, partly on grounds of
utility (with a view to the demoralization of insecurity), partly on the non-utilitarian
ground that a “moral commitment” has been made and must be honored.
9
Of course the difficult question remains of the relative weight that should be given
to prospective and retrospective service in cases where these factors conflict. There
is good reason to treat them as a pair.
10
This in the symposium on “Selection of Patients for Haemodialysis,” British Medi-
cal Journal (March 11, 1967), pp. 622-624. F. M. Parsons writes: “But other forms
of selecting patients [distinct from first come, first served] are suspect in my view
if they imply evaluation of man by man. What criteria could be used? Who could
justify a claim that the life of a mayor would be more valuable than that of the
humblest citizen of his borough? Whatever we may think as individuals none of us
is indispensable.” But having just set out this hard-line view he immediately backs
away from it: “On the other hand, to assume that there was little to choose between
Alexander Fleming and Adolf Hitler ... would be nonsense, and we should be naive
if we were to pretend that we could not be influenced by their achievements and
characters if we had to choose between the two of them. Whether we like it or not
we cannot escape the fact that this kind of selection for long-term haemodialysis
will be required until very large sums of money become available for equipment
and services [so that everyone who needs treatment can be accommodated].”
11
The relative fundamentality of these principles is, however, a substantially disputed
issue.
12
In the case of an ongoing treatment involving complex procedure and dietary and
other mode-of-life restrictions-and chronic hemodialysis definitely falls into this
category—the patient’s psychological make-up, his willpower to “stick with it” in
the face of substantial discouragements—will obviously also be a substantial factor
here. The man who gives up, takes not his life alone, but (figuratively speaking)
also that of the person he replaced in the treatment schedule.
13
To say that acceptable solutions can range over broad limits is not to say that there
are no limits at all. It is an obviously intriguing and fundamental problem to raise
155
the question of the factors that set these limits. This complex issue cannot be dealt
with adequately here. Suffice it to say that considerations regarding precedent and
people’s expectations, factors of social utility, and matters of fairness and sense of
justice all come into play.
14
One writer has mooted the suggestion that: “Perhaps the right thing to do, difficult
as it may be to accept, is to select [for hemodialysis] from among the medical and
psychologically qualified patients on a strictly random basis” (S. Gorovitz, “Ethics
and the Allocation of Medical Resources,” Medical Research Engineering, [1966],
p. 7). Outright random selection would, however, seem indefensible because of its
refusal to give weight to considerations that, under the circumstances, deserve to be
given weight. The proposed procedure of superimposing a certain degree of ran-
domness upon the rational-choice criteria would seem to combine the advantages
of the two without importing the worst defects of either.
15
“Selection of Patients for Haemodialysis,” op. cit., p. 623. The question of whether
a patient for chronic treatment should ever be terminated from the program (say if
he contracts cancer) poses a variety of difficult ethical problems with which we
need not at present concern our selves. But it does seem plausible to take the
(somewhat anti-utilitarian) view that a patient should not be terminated simply be-
cause a “better qualified” patient comes along later on. It would seem that a quasi-
contractual relationship has been created through established expectations and re-
ciprocal understandings, and that the situation is in this regard akin to that of the
man who, having undertaken to sell his house to one buyer, cannot afterward uni-
laterally undo this arrangement to sell it to a higher bidder who “needs it worse”
(thus maximizing the over-all utility).
16
This chapter was originally published under the same title in Ethics, vol. 79 (1969),
pp. 173-86. Reprinted over a dozen times since its initial appearance, this paper of
mine caused a greater stir than any other thanks to its timely coincidence with a ris-
ing wave of interest in issues of medical ethics.
156
Chapter 13
ETHICAL ISSUES REGARDING

THE DELIVERY OF HEALTH-CARE
SERVICES
INTRODUCTION
W e are often “of two minds” about something in life. Conflicting fac-
tors pull us in opposite directions and “we can see it both ways,” so
to speak. We feel ambivalent because there is a good deal to be said on
each of two diametrically opposed sides. This produces the sort of situation
which Marxist theoreticians liked to refer to as a contradiction. In such
cases the operation of conflicting forces impels us towards divergent and
incompatible directions.
This paper will describe and examine three such conflicts or contradic-
tions in the ethics of health care. These are:
1. the system vs. the individual
2. quality vs. equality
3. the present vs. the future
Each of these conflicts indicates deep ethical problems that have very prac-
tical and pressing implications for policy and procedure.
THE SYSTEM VS. THE INDIVIDUAL
The first tension to be considered relates to the issue of the allocation of

responsibility for health care. Is it to be the individual himself who is pri-
marily and predominantly responsible for seeing that his health-care ser-
vices are adequately provided for in “the system” that society might
institute through design or inadvertence? Such questions of responsibility

clearly pose paradigmatically ethical issues.
In its present-day setting, this problem cannot be handled in the abstract
on the basis of general principle alone. We must begin with a closer look at
the realities of the situational context, and examine the actualities of the
situation regarding threats to health and the delivery of health-care ser-
vices.
What is striking to anyone who studies the historical statistics of the
causes of death in the U.S. is the massive role among these causes of con-
ditions traceable to what may simply and honestly be characterized as bad
personal habits. Consider the state of affairs portrayed in Table 1.
___________________________________________________
Table 1
DEATH RATES FOR SELECTED CAUSES: 1972

(PER 100,000 POPULATION)
I. MEDICAL
1. Major cardiovascular diseases 494

2. Malignant neoplasms 167
3. Influenza and pneumonia 29
4. Diabetes mellitus 19
5. Cirrhosis of the liver 16
6. Certain diseases of early infancy 16
7. Bronchitis, emphysema, and asthma 16
II. NON-MEDICAL
8. Motor vehicle accidents 27

9. Other accidents 27
10. Suicide 12
11. Violent crime 9
Date from Statistical Abstract of the United States: 1974

(Washington, 1974), p. 62
___________________________________________________
158
ETHICAL ISSUES REGARDING THE DELIVERY OF HEALTH-CARE SERVICES
Such statistics indicate a tendency for death rates to increase markedly

from conditions traceable to the impact over many years of “bad” personal
habits: smoking, drinking (and drugs), inactivity, and poor diet contribute
significantly to some of these items—drinking for example, especially to
item 8 (two-thirds), and one-half of all homicides and one-third of all sui-
cides in the U.S. We all know the propaganda about these “diseases of af-
fluence” as they have been called. Middle-aged men who are 20% above
normal weight run 2-3 times the risk of fatal heart attack. About 75% of
lung cancer is caused by smoking. Death rates for heart attacks in men
range from 50 to 200 percent higher among cigarette smokers than among
non-smokers depending on age and the amount smoked. But I shall not
prolong this statistical litany.
As one examines the major elements of the casual situation regarding
the big killers that represent the major threats to health, the consideration is
forcibly brought home to us that an increasingly improminent role is
played by those factors (communicable diseases and non-automotive acci-
dents) over which the individual has relatively little control, and an
increasingly prominent role is played by those factors over which the
individual himself in fact has substantial control. Like it or not, we are
driven towards the conclusion that in very large measure it is the individual
himself who alone can and should take the responsibility for delivering the
great bulk of health-care services he stands in need of.
This line of thought indicates the unrealism of some of the current dis-
cussions on health-care delivery. And too often one hears the issues posed
in the format of a Victorian melodrama. We have the villain (the existing
institutional system), the hapless heroine (the poor person being denied
health-maintaining services), the hero (the state that forces the villain’s
hand in doing the heroine justice), the appreciative bystander (the cheering
public). The monitory example along the lines of the poor man with ap-
pendicitis being turned away from a mercenary hospital to perish miserably
in the streets. This sort of picture cannot withstand critical scrutiny. Even a
cursory look at the fact of the situation is enough to show up the unrealism
of it all as a pattern for the problems of distributive justice in the medical
area.
If this perspective is anything like correct, it bears some very uncom-
fortable implications. What we’re addicted to—and pretty competent at—is
handling issues that revolve about technological solutions and systems-
design problem-solving. We are fascinated by problems that admit of
economico-political solutions and all for expanded funding and administra-
159
tive rationalization. What engages our attention are things like getting an
ambulance service better able to care for heart-attack victims or the really
big issues of distributive imbalance: geographical imbalance (getting more
doctors into the rural areas) and service imbalance (fewer narrow special-
ists and more doctors in family practice and other primary care activities).
And there is no question that such problems are important and difficult.
But important though they are, they yet remain second-order-,or even,
fourth-order-considerations in the larger health-care picture.
We have to face up to the sobering but inescapable fact that no foresee-
able improvement in medical practice or in the distribution of medical ser-
vices could make an impact on the health of Americans that would amount
to more than a minute fraction of the improvement that could be wrought
by the cultivation of better personal habits.
The major issues—I would like to suggest—have to do with matters of
a rather different sort where the deficiencies lie in the sphere of the indi-
vidual rather than that of “the system”. And just this is the source of the
discomfort. For the real issues turn on questions of the sensible attitude and
intelligent action of the level of the individual. They root in the facts of
human heedlessness and stupidity. It is clear that such factors are compara-
tively intractable and do not lend themselves to monetary or administrative
resolution. The basic issue is one of the relative effectiveness of public ac-
tion vs. private values in grappling with the issues. It would be a grave—
even if comforting—illusion that a combination of funding and statutory
manipulation will prove of substantial avail here. For what we most ur-
gently need is not socialized medicine, or an ampler infusion of funds into
the medical area, or a better distributive system of medical services, but a
larger dose of old-fashioned morality or even plain common sense. Unfor-
tunately, the latter items are a lot harder to come by than the former.
QUALITY VS. EQUALITY
Let me now turn to the next tension or “contradiction” I would like to

consider—that of quality vs. equality.
In the modus operandi of a system for providing health-care services
this fundamental tension crops up from many points of view. One faces in
various ways the choice between equal care vs. a superior system for pro-
viding care in general. It is a question of a smaller pie divided equally vs. a
larger pie where everyone is better off but the divisions are unequal:
160
Let me give a very concrete example of this. I have long admired the Na-
tional Health Service of Great Britain as an intelligent and efficient com-
ponent of a combination of public and private medicine, a two track system
of publically and privately financed medical care both functioning at
what—under the realistic circumstances at issue—deserves to be consid-
ered a very high level. But some parts of this combination reflect a highly
pragmatic compromise made when Parliament instituted the NHS in the
post-war period: the allocation of some private beds in the public hospitals
and the provision of some supporting services (x-ray facilities, laboratory
facilities, etc.) by the public to the private center—in both cases at charges
that more than meet the actual expenses involved. This pragmatic com-
promise has in fact proved very advantageous for public-sector medicine.
By exacting a partial subsidy for the private sector it has made the public
system superior to what it otherwise would be. And by helping to make
medical practice more attractive it has helped to recruit and retain practi-
tioners for British medicine.
On the other hand, there is no question that the two-tier system pro-
duces inequities. The private patient has a wider range of choice of physi-
cians and services at his disposal than the public health patients do. And
above all the private patient can get himself into an operating theater for a
nonemergency procedure in short order as compared with the many months
that a NHS patient would spend cooling his heels on a long waiting list.
We here arrive at what is a rather general issue of distributive justice
throughout many contexts, namely the issue of more vs. more equal. There
are various situations where a situation that is unfair may nevertheless not
be unjust in a deeper sense provided the unfairness can be shown to work
itself out of the general advantage. We face the deep question of whether
equality is not bought at too dear a price if it can only be had by compro-
mising the qualities of the services available to everybody. The example I
161
have adduced of the tension in the delivery of medical services between the
systematic quality of service vs. equality of access seems to present a vivid
illustration of such a situation.
THE PRESENT VS. THE FUTURE
The third and final example of a tension or “contradiction” is that be-

tween the demands of the present and those of the future: the claims of the
living vs. those of the yet unborn. The issue here placed under this rubric is
the problem of the costs and the benefits of medical research.
Historically, this was a non-issue. Even a single generation ago, the
amount of money and talent invested in medical research was a trivial
quantity.
For example, less money was spent on polio research in the entire pre-
war generation 1915-1945 than in any pair of years during the decade after
1948.
Throughout the early 1960s substantially more was spent on medical re-
search than on nursing home care throughout the U.S.
In the early 1970s we spent annually on medical research and develop-
ment an amount standing at some 6% of the total costs of hospital care and
at more than 12% of the sum-total of physicians’ services.
And in recent years the growth of medical R&D expenditures has been
particularly dramatic, to the point where these costs are a substantial share
of the over-all pie of total health costs. Medical R&D outlays have in-
creased from a 3% slice of a $26 billion pie in 1960 to a 4% slice of a $94
billion pie in 1973—an impressive quadrupling from $.85 billion in 1960
to 3.5 billion in 1973.1
A very general and fundamental point about the development of science
is forcibly illustrated by biomedical research rather than violated by it. In
the course of scientific progress one solves the relatively easy and straight-
forward problems first and puts the relatively more complex and intracta-
ble ones off to the future. As time goes on, problems become more
difficult.
The medical area affords an interesting illustration of the extent to
which latter-day problems tend to be more intractable than earlier ones and
demand for their solution a vastly greater resource-investment. The histori-
cal predominance of the diseases that represent the major killers is set out
in Table 2.
162
___________________________________________________
Table 2
THE KILLER DISEASES*
Percentage of deaths
Cause of Death occurring are caused by
1900 1930 1960
Heart disease and stroke 15.4 35.9 54.0

Influenza and pneumonia 11.8 9.1 3.9
Tuberculosis 11.3 6.3 0.6
Gastritis, duodenitis, enteritis
And colitis 8.3 2.3 0.5
Children’s ailments (measles,
Diphtheria, whooping cough) 3.8 1.1 0
Cancer 3.7 8.6 15.6
Typhoid and para-typhoid fever 1.8 0.4 0
* Data from Economic Costs of Cardiovascular Disease and Cancer,1962

(Washington, 1965; Public Health Service Publication No 947-5), Table I.
___________________________________________________
The record of success is doubtless impressive: more than half of the big
killer-diseases of 1900 have virtually been eliminated as serious threats.
But what is significant from our angle is the greater intractability of the
problems that remain. Finding a cure for TB or gastritis or diphtheria is
still small potatoes compared with finding a cure for today’s big killers. In
1962, a total of $1,032 x 106 was spent on medical research in the U.S.,
distributed as follows.2
$117 x 106 Cardiovascular Disease

$128 x 106 Cancer
$787 x 106 Other
The U.S. was spending more money (and effort) on cancer in 1962 than it
was spending on all of medical research in 1950—and (arguably) more
163
than had been spent on all of medical research in the history of mankind
until 1940. Phenomenally, this massive expenditure has more than doubled
to $2,277 x 106 by 1972 and America is currently spending on medical re-
search an amount that is somewhat over 4% of what private consumers are
expending on medical care.3 The scale of this research effort is truly im-
pressive. Of course, one tackles the easier problems first—that goes with-
out saying. But the striking thing is the extent to which the later problems
become more difficult and demand ever increasing levels of effort for their
resolution—in the bio-medical area exactly as elsewhere in natural science.
It is sobering to contemplate the vast efforts and expenditures of present-
day drug research when one considers that the basic research that led to the
discovery of penicillin was a shoe-strong operation whose whole cost did
not come to more than $20,000.
We thus find ourselves entering into the characteristic condition of a
classic law of diminishing returns: costs rise, the significance of results
diminish on the average, the waiting times for significant results increase
dramatically. Many observers have commented on the ironic fact that, as
one acute British writer recently put it:
It is precisely during the last two decades when scientific medicine is alleged to
have blossomed and when the quality of resources allocated to medical care has
rapidly increased-that the decline in mortality that has been associated with indus-
trialization has tapered off to virtual zero.4
The lesson of these considerations for our purposes is contained in two

facts. (1) Research is getting so expensive as actually to direct significant
resources from delivery of health-care. (2) Progress is so slow and the pay-
off on research such a long term issue that the benefit of research may
never redound to those actually making the investment but that only their
historical successors will be the real beneficiaries.
Consider the increased waiting-times involved between the initiation of
an intensive research effort and its successful issue in decisive preventative
or curative instrumentalities. For the communicable diseases that were big
killers at the turn-of-the-century this was a matter of a few years and some
tens of thousands of dollars; for polio it was a matter of a couple of dec-
ades and some $25 million; for cancer the escalation may well reach bil-
lions of dollars stretched over a whole century.
Such a massive increase in research costs and elongation of gestation-
period exemplifies the operation of my third “tension” or contradiction, the
claims of the present vs. those of the future. There can be no question that
164
the massive requirement of medical research for the investment of material

resources and skilled manpower leads to a substantial curtailment of our
capacity to deliver health-care services here-and-now.
And there is another ramification of this theme that deserves at least
passing mention. As medicine advances into regions of ever-growing com-
plexity and sophistication, the demands on talent, human services, and
equipment grow to a point where the economics of therapy become un-
manageable. The patient himself has already arrived at this juncture of
economic incapacity in more cases than one likes to think of. And the time
when third-party insurers will get there is not all that far away. More and
more the real question does not revolve around the technical issue of what
we can do, but the economic issue of what we can afford to do. This whole
area of choices is heavily laden with ethical ramifications.
CONCLUSION
Just one methodological point deserves note in this connection. The

present discussion has in large measure been woven around statistical
themes. And this is quite deliberate. There is a very strong temptation in
the domain of biomedical ethics to focus on exotic matters that appeal to
the imagination. Accordingly there is a noticeable tendency—quite pro-
nounced among newspaper columnists but also present among my philoso-
phical colleagues—a tendency to dwell on theoretically intriguing, exotic
issues like psychosurgery or humanoid cloning. However much such issues
may appeal to the science-fiction fan inherent in each one of us, they are
not now (and are not likely soon to become) an actually pressing issue in
day-to-day medical affairs. To focus attention on such exotic matters is to
become entrapped in a kind of unrealism that abandons the really impor-
tant issues in turning away from the practical concerns of “the real world”.
Our deliberation here accordingly focused upon statistically prominent
phenomena indicative of problems that have an interest transcending the
perspective of the armchair theoretician. Specifically, they have addressed
three conflicting tensions in the area of the delivery of health-care services:
(1) the extent to which responsibility can be imputed to “the system” for
delivering medical services vs. the extent to which it does and should
rest with the individual.
165
(2) the relative desirability of enhancing the over-all quality vs. equality-
of-access to medical services.
(3) the claims of the present vs. those of the future as an issue posed by
the escalating resource-demands of biomedical research.
Each of these cases has the classic form of a moral dilemma—in each in-
stance we are torn in a conflict of two discordant desiderata: All three of
these issues (the individual vs. the system, quality vs. equality, the present
vs. the future) pose problems of a very general sort that arise as ethical is-
sues in many areas—for example, in economic policy. But they seem to
pose particularly relevant and increasingly pressing problems in the medi-
cal context because the stakes are so high, since the very physical well-
being and indeed the lives of people are at stake.
Some pretty clear morals emerge more or less spontaneously once the
issues are laid out:
• Our concern for the delivery of health care services has been too sys-
tem-oriented. People as individuals can and should carry a far heav-
ier burden of responsibility here.
• The desideratum of equality—no matter how legitimate a value in its

own right—should not be promoted in so doctrinaire and unthinking
a way as to dilute the capacity of the health-care system to serve the
community at large.
• We should acknowledge as real and authentic the claims presently

made upon us by the concerns of the future through the mediation of
research. And this should be done with open eyes, with the realiza-
tion that these claims may become very large indeed if substantial
progress is to be made as one penetrates further into the area of di-
minishing returns.
And overall, these lessons reflect the conviction that one is not likely to ar-
rive at sensible solutions to difficult problems in the absence of critical
analysis of their nature.5
166
NOTES
1
Data from B.C. Cooper, N.L. Worthington and P.A. Piro, “National Health Expen-
ditures: 1929-1973” (Washington, D.C., 1974; Social Security Bulletin, February
1974 issue).
2
Ibid.
3
B.C. Cooper, et al., op cit.
4
John Pawles, “On the Limitations of Modern Medicine,” Science, Medicine and
Man, vol I (1973), pp 1-30 (see p 2).
5
This chapter was originally published under the same title in Connecticut Medicine,
vol. 41 (1977), pp. 501-06. In writing this lecture I have been greatly helped by the
suggestions and materials provided by Dorothy P. Rife (Deputy Assistant Commis-
sioner, Social Security Administration), Solomon Schrezder (Acting Associate Di-
rector for Program Planning and Evaluation, National Institutes of Health), and
Herbert P. Woolley (Economist, National Institutes of Health).
167
Chapter 14
MORAL ISSUES RELATING TO THE

ECONOMICS OF NEW KNOWLEDGE IN
THE BIOMEDICAL SCIENCES
PLANCK’S PRINCIPLE OF INCREASING EFFORT
O ver the past few generations there has been an exponentially increas-
ing investment of human and material resources in scientific inquiry,
particularly medical research. Nevertheless, there remains the economists’
uncompromising question regarding the actual structure of the relationship
between resource investment and product output. In particular, the problem
arises of whether, as science progresses, a fixed amount of effort continues
to yield uniformly significant results, or whether a process of declining
yields is operative in this respect. Even greatly increasing resource invest-
ments will fail to generate a corresponding increase in output if the unit
cost of production is rising.
A great deal of impressionistic and anecdotal evidence certainly points
towards the increasing costs of high-level science. Scientists frequently
complain that “all the easy researches have been done”.1 The need for in-
creasing specialization and division of labor is but one indication of this. A
devotee of scientific biography can easily note the disparity between the
immense output and diversified fertility in the productive careers of the
scientific collosi of earlier days and the more modest scope of the
achievements of their latter-day successors.
It is clearly implausible to interpret this shrinkage as indicating that the
days of greatness are over and that our contemporaries are men endowed
with less brain power or with diminished capacity for hard work. Rather,
one is drawn towards the very different conclusion that the work is simply
getting harder. This poses the prospect that a stonier soil is being farmed—
one where comparable effort simply can no longer yield comparable re-
turns.2 The successive victories of science, like the battles of Marlborough,
are won only at an ever-mounting cost.
There are, it would seem, substantial grounds for agreement with Max
Planck’s appraisal of the situation:
To be sure, with every advance [in science] the difficulty of the task is
increased; ever larger demands are made on the achievements of re-
searchers, and the need for a suitable division of labor becomes more
pressing.3
The underlined thesis of this quotation will here be characterized as

Planck’s Principle of Increasing Effort. We shall interpret this as asserting
that successive substantial discoveries become more and more expensive
over the course of time in terms of the investment of talent, manpower, and
material resources—that scientific work is subject to an escalation of costs.
Accordingly, we construe this principle of increasing effort in specifically
the following sense: As science progresses within any of its established
branches, there is a marked increase in the over-all resource cost of realiz-
ing scientific findings of a given level of intrinsic significance (by essen-
tially absolutist standards of importance).
The medical field affords a particularly vivid illustration of the extent to
which latter-day problems tend to be more intractable than earlier ones and
demand for their solution a vastly greater resource investment. The histori-
cal record of the success of modern medicine is doubtlessly impressive:
more than half of the big killer-diseases of 1900 have virtually been elimi-
nated as serious threats. But what is significant from our angle is the
greater intractability of the problems that remain. Finding a cure for tuber-
culosis, gastritis, or diphtheria was simple compared to finding a cure for
today’s catastrophic diseases. But the remarkable aspect of this phenome-
non is its indication of the extent to which the later problems become more
difficult and demand ever increasing levels of effort for their resolution—
in the biomedical field as in natural science. The operation of a principle of
cost-escalation has become strikingly manifest in the contemporary medi-
cal research.
FIRST THINGS FIRST: TECHNOLOGICAL ESCALATION
Nature inexorably exacts a dramatically increasing effort in terms of

enhanced sophistication in data-deployment for revealing her ‘secrets’ and
accordingly becomes less and less yielding to the efforts of our inquiry at
given fixed levels of information-gathering technique.4 This accounts for
170
MORAL ISSUES RELATING TO THE ECONOMICS OF NEW KNOWLEDGE IN THE
BIOMEDICAL SCIENCES
the continual introduction of more and more sophisticated technology into

inquiry in natural science and the subsequent increasingly extensive reli-
ance upon it.
In natural science we do the easy things first. The very structure of sci-
entific inquiry forces us into the situation reminiscent of an arms race:
technological escalation where the frontier-equipment of today’s research
becomes the museum-piece of tomorrow under the relentless grip of tech-
nical obsolescence. It would thus be as futile to follow the authors into
hankering after the days of “string and sealing-wax” apparatus as it would
be to join Talleyrand in lamenting the lost douceur de la vie of the old re-
gime. There is no point in blaming human foibles or administrative ar-
rangements for a circumstance that is built into the very structure of
investigation in natural science (realizing, to be sure, that the facts that
make bigness a necessary condition of significant progress do not establish
it as sufficient). The enormous power or sensitivity or complexity deployed
in present-day experimental science has not been sought for its own sake,
but rather because the research frontier has moved on into an area where
this sophistication is the indispensable requisite of on-going progress.
A homely fishing analogy of Sir Arthur Eddington’s is useful here.5 He
saw the experimentalist as a “trawler”—that is, as one who trawls nature
with the “net” of his equipment for detection and observation. Now sup-
pose (says Eddington) that fishermen trawl the seas using a fish-net of two-
inch mesh. Then fish of a smaller size will simply go uncaught. And the
theorists who analyze the experimentalists’ catch will have an incomplete
and distorted view of aquatic life. Only by improving our observational
means for “trawling” nature can the existing limitations be mitigated.
Technological escalation is an imperative of scientific progress. The range
of phenomena to which the old data-technology gives access is soon ex-
hausted. And the same is generally true with respect to the range of theory
testing and confirmation that the old phenomena can underwrite.
Once all the findings accessible at a given state-of-the-art level of inves-
tigative technology have been realized, one must continually move on to a
more expensive level. An ongoing enhancement in the quality and quantity
of the data-input requires more accurate measurements, more extreme tem-
peratures, higher voltages, more intricate combinations, etc. On such a
view, the phenomenon of cost-escalation is explained through a combina-
tion of the finitude of the body of first-rate results realizable within a given
level of investigative technology, together with a continual (geometric) in-
crease in the resource—costs of pushing from one level to the next.
171
This fundamental phenomenon of technological escalation, operative

throughout the natural sciences and certainly in the medical field, serves to
account for the recent rise of high-technology medicine that is a character-
istic feature of present-day frontiers of research and therapy.
COST-ESCALATION IN RESEARCH
The issue I primarily want to pose in this context comes in the more
comfortable form of a resource-allocation question. Specifically, it is the
problem of cost-benefits in medical research.
Historically, this was not an issue. Even a single generation ago, the
amount of money and talent invested in medical research was a trivial fac-
tor in the wider economic scheme of things. For example, less money was
spent on polio research in the entire pre-war generation 1915-1945 than in
the years during the decade after 1948. And throughout the early 1960s
substantially more was spent on medical research than on nursing home
care throughout the U.S. In the early 1970s we spent annually on medical
research and development an amount standing at some 6 percent of the to-
tal costs of hospital care and at more than 12 percent of the sum-total of
physicians’ services.
In recent years, the growth of medical research and development expen-
ditures has been particularly dramatic, to the point where these costs are a
substantial share of the over-all pie of total health costs. Medical research
and development outlays have increased from a 3 percent slice of a $26 bil-
lion pie in 1960 to a 4 percent slice of a $94 billion pie in 1973 and has
proceeding apace since then.
A very general and fundamental point about the development of science
is forcibly illustrated by biomedical research rather than violated by it. In
the course of scientific progress one solves the relatively easy and straight-
forward problems first and delays the relatively more complex and intrac-
table ones. As time goes on, problems become more difficult and
expensive to resolve.
In the final analysis, there is nothing unique to natural sciences as re-
gards the basic principles operative here. For we are concerned with an en-
deavor to push a technology to the limits of its capacity, and one knows
from innumerable cases there is an analogous cost-increase in any situation
where technology is used to press towards any natural limit, even in situa-
tions quite outside the strictly scientific context.
The record of recent success in medical research is without doubt im-
172
BIOMEDICAL SCIENCES
pressive: more than half of the catastrophic diseases of 1900 have virtually
been eliminated as serious threats. Yet a significant number of these re-
main, and the cost of eliminating them is enormous. It is sobering to con-
template the vast efforts and expenditures of present-day drug research
when one considers that the basic research that led to the discovery of
penicillin was a shoe-string operation whose whole cost did not come to
more than $20,000.
THE SPILL-OVER FROM RESEARCH TO THERAPY:

THE RISE OR HIGH-TECHNOLOGY MEDICINE
The movement from inquiry to application, from medical research to

medical treatment, represents a sure and fairly swift transition. Technologi-
cal escalation of research is thus accompanied by technological escalation
of therapy. The development of scientific medicine over the last 100 years
has moved pari passu with technological revolution of breath-taking rapid-
ity and scope. The devices used in contemporary medical treatment and the
instrumentalities used to produce the materia medica of contemporary
therapy are clearly among the most elaborate creations of a technological
civilization. Increasingly, and dramatically, medicine has become a high-
technology enterprise.
And, of course, sophisticated technology is expensive technology, both
as regards the processes of its production and the quality of the manpower
inputs for its effective utilization. The scientific implications of the proce-
dures of modem medical research have run parallel with the skill-utilizing
sophistication of the delivery system through which the products of mod-
ern medicine reach their consumers.
The consequences of this situation are all too well known. The costli-
ness of modern high technology medicine is an important issue of social
concern and political controversy.
And once the issue of costs rises to such prominence, that of cost-
effectiveness is not far behind.
Indeed some observers see the situation not in terms of a slow-down but
in terms of a stoppage. The eminent Australian immunologist Sir
MacFarlane Burnett (Nobel Laureate in Medicine, 1960) after surveying
work in the biological sciences concludes:
None of my juniors seems to be as worried as I am, that the contribution

of laboratory science to medicine has virtually come to an end. The
173
biomedical sciences all continue to provide fascinating employment for

those active in research, and sometimes enthralling reading for those
like me who are no longer at the bench but can still appreciate a fine
piece of work. But the detail of an RNA phage’s chemical structure, the
place of cyc1ostomes in the evolution of immunity or the production of
antibody in test tubes are typical of today’s topics in biological re-
search. Almost none of modern basic research in the medical sciences
has any direct or indirect bearing on the prevention of disease or on the
improvement of medical care.6
It has become questionable whether massive effort and expenditure in re-

search and technologically sophisticated therapy is able to yield a com-
mensurate result in terms of life-prolongation.
The interesting question of costs and returns thus comes to assume a
particularly problematic and embarrassing form.
THE MORAL DIMENSION
The title of my essay spoke of not only the “economics” of new knowl-
edge in the biomedical area but of the “moral issues” that arise in this con-
nection. So far I have addressed myself primarily to the former topic. It is
time now to turn to the latter.
The issue gains its moral dimension through the inevitable fact of the
finitude of our resources. In the past, the biomedical sciences have flour-
ished at a logarithmic growth-rate whose ever-accelerating increases can-
not be projected into the future. Take the cost of delivering medical
services in the U.S.A., for example. If these costs continue to rise at to-
day’s rate, by the year 2020 A.D. they would amount to the whole of our
gross national product.
We are entering an era of zero growth—which means an era of effective
scarcity in many regards. (For conditions of scarcity exist whenever one’s
resources, however ample, are inadequate to do everything that one would
like to do.) At this stage forced choices come upon us and we must select
priorities. Issues of a fundamentally ethical and moral character now come
to be confronted.
And here I reach the core thesis of my discussion: the fact that the very
economics of modern high-technology medicine itself poses moral prob-
lems of the most acute and difficult variety. It is, of course, easy to say “all
these are merely economic questions—questions of the allocation of re-
174
BIOMEDICAL SCIENCES
sources.” But it is clear that these economic issues are fraught with moral
and ethical ramifications. The crucial issue throughout is not one of effi-
cient means to agreed ends, but of the very ends and objectives of the en-
terprise. Let us consider some aspects of this question of ends.
Throughout recent times American medicine has been governed by a
simple precept: “The patient deserves the best”—where the medical needs
of people are concerned nothing less than the very best is viewed as mini-
mally acceptable.
This reflects an attitude which I find admirable and which I largely
share. But its application is nevertheless highly problematic.
The situation in contemporary medicine is reminiscent of the late Ro-
man empire—strength at the outer frontiers and a great multitude of weak-
nesses and problems in the less exotic regions closer to home.
The economics of the situation puts some very difficult and uncomfort-
able questions on the agenda:
• Considering the great expense in manpower and resources involved,

should there be a major redeployment from research to therapy?
• Should we redeploy resources from complex and expensive high-

technology intervention to lower-technology therapy and, above all,
to preventive medicine?
• Should we abandon the idea that “only the very best is good
enough”? And at just what stage are we towards an optimizing ra-
tionale in the delivery of health care?
The economics of the evolution of new knowledge in the biomedical

sciences has put before us some very difficult and painful choices—
choices fraught with far-reaching moral ramifications.
How are such choices to be resolved? What rationale can be developed
for making them one way rather than another? The appropriate standard is
perhaps relatively clear. It is the minimization of human suffering, on bal-
ance and across the whole spectrum, over the foreseeable future. It is a
matter of augmenting the range of human life to the greatest extent com-
patible with a minimal quality—roughly the standard of utilitarianism. And
when (as with the question of research expenditures) this standard dictates
the payment of a greater cost now for the sake of a greater benefit in the fu-
ture, we must not shrink from a willingness to pay it.
175
To be sure, all these questions of social ethics and public policy are
highly controversial.
It is none too soon to begin to debate the issues at the level of general
theory before they bewilder us at the level of detailed practice. It is well to
clarify the theoretical issues before the push and shove of political contro-
versy blocks any prospect of calm and informative debate. The questions
of public policy that arise in this area are not minor epiphenomena but cut
through to the very heart of things. The traditional stance of political liber-
alism has always been oriented towards the rights and claims of the indi-
vidual. Its line is that the individual is supreme, and that his interests are
the ultimate arbiter of public policy which no purported wider public needs
and interests can properly overrule. Opposed to this individualistic liberal-
ism stands the tradition of holistic concern for “the good of the whole” so
prominent in Hegel and the political philosophies that represent the evolu-
tion of the tradition he set afoot. At this level, it seems, we are going to
have to face some very far-reaching issues of distributive justice in the de-
livery of health care services. These questions relate to the third point
above, that the claims of the individual are absolute and that only the best
is acceptable in our treatment of individuals. For we are rapidly entering an
era where the best is only attainable for some at the expense of failure to
realize a system that serves the best interests of the wider community.
In closing I would like to say something about the political ramifica-
tions of the cost-effectiveness considerations adduced above. In that fasci-
nating government publication Science Indicators7 we find the question
“Do you feel that science and technology change things too fast, too
slowly, or just about right?” Some 53 percent of the respondents answered
this question with: too slowly. On the other hand, there was also the ques-
tion: “In which of the areas listed (below) would you most like (and least
like) to have your taxes spent for science and technology?” Topping the list
with 69 percent here was “improving health care”. People seem to feel that
improvements in the health care area are needed, are worth paying taxes
for, but are forthcoming at too slow a rate.
Now where will we stand when it becomes clear that ever more massive
expenditures yield ever lessening returns in actual improvement? Clearly
there is a real potential danger here—the danger of the demagogy of unmet
expectations and scapegoatism—a revulsion not just against science and
technology, but against reason and the life of the mind. It is none too soon
to start worrying about what will happen when people come to realize in-
creasingly—and with growing disillusionment—that spend though we will
176
BIOMEDICAL SCIENCES
and must on research in science and medicine, the age of swift and massive
returns is coming to an end.8
NOTES
1
The sentiment is not new. George Gore vainly lambasted it 100 years ago: “Noth-
ing can be more puerile than the complaints sometimes made by certain cultivators
of a science, that it is very difficult to make discoveries now that the soil has been
exhausted, whereas they were so easily made when the ground was first broken ...”
The Art of Scientific Discovery (London: Longmans, Green and Co., 1878), p. 21.
2
This prospect seems borne out also by a phenomenon whose reality, though it is
difficult to substantiate, is nevertheless unquestionable—namely, that there are
nowadays not a few authentic scientific geniuses of the first order who never man-
aged to produce first-magnitude results. (For obvious reasons, one is hesitant to cite
examples.)
3
See Max Planck: Vorträge und Erinnerungen, (S. Hirzel Verlag, Stuttgart 1949).
Shabecoff, P.: 1978, “Runaway Medical Costs”, New York Times, May 12, 1978, p.
A-18.
4
For further detail see Nicholas Rescher, Scientific Progress (Oxford: Blackwell,
1978).
5
A. S. Eddington: The Nature of the Physical World, (AMS Press, New York 1978,
Reprint of the 1928 edn.).
6
MacFarlane Burnett, Genes, Dreams, and Realities (New York: Basic Books,
1971), p. 42.
7
U.S. Government Printing Office.: 1975, Science Indicators—1974, publication
No. 038-000-00253-8, Washington, D.C.
8
This chapter initially appeared under the same title in W. B. Baudson et. al. (eds.),
New Knowledge in the Biomedical Sciences (Dordrecht and London: D. Reidel,
1982), pp. 135-45.
177
Name Index
Adams, John, 45
Alexander, Shana, 154n5
Allen, Frederick Lewis, 50n2
Bacon, Francis,116
Baillod, R. A., 154n2
Beardsley, Tim, 101n3
Brickman, Philip, 127n9
Burnett, MacFarlane, 173, 177n6
Bury, J. B., 115, 126n1
Butler, Samuel, 126n4
Campbell, Angus, 126n6

Campbell, Donald T., 127n9
Carlyle, Thomas, 43
de las Casas, Barolomeo, 32
Chitian, Shirley, 113n2
von Choltitz, Dietrich, 108-10, 113n4
Clinton, Hillary, 11
Collins, Larry, 113n3
Conty, C. M., 154n2
Converse, Philip E., 126n6
Coolidge, Calvin, 45
Cooper, B. C., 167n1, 167n3
Costello, Timothy W., 126n6
Crowell, Thomas Y., 126n5
Cyert, Richard M., 126n6
Dewey, John, 24
Dixon, Bernard, 127n10
Doyle, Christine, 154n1
Dubos, René, 123
Eddington, Arthur, 171, 177n5

Estes, Ralph, 139n3
Fischhoff, Baruch, 139n4, 139n5

Fleming, Alexander, 155n10
Gauthier, David, 90n6

Gibson, S. B., 133
Goethe, Wolfgang, 115
Gore, George, 177n1
Gorovitz, Samuel, 156n14
Grosse, Robert N., 140n7
Hare, Peter, 126n7

Harrison, Benjamin, 45
Hegel, G. W. F., 29, 43, 44, 50n1, 176
Helberg, Raul, 113n1
von Henneken, Heinrich, 106-10
Herbert, Bob, 101n3
Herrnstein, Richard J., 91, 96, 101n1
Heyek, F. A., 14n8
Hirsch, E. D., Jr., 101n4
Hitler, Adolf, 108-09, 155n10
Hoover, Herbert, 45
Hoover, J. Edgar, 45
Jackson, Andrew, 45
Jacob, W. W., 18
Johnson, Lyndon B., 91
Joynt, Carey B., 89n4
Kant, Immanuel, 3, 14n3, 111

Kohl, H. R., 14n1, 28n2
Lader, Lawrence, 155n6

Lapierre, Dominique, 113n3
Leibniz, G. W., 116
Lincoln, Abraham, 45
Longmore, Donald, 142
Lotze, Hermann, 116
Machiavelli, Niccolo, 44
Mainzer,Klaus, 14n2
Malthus, Thomas, 116
March, James G., 126n6
Marx, Karl, 29
Massey, G. J., 89n2
Mill, John Stuart, 33
Miller, Henry, 119
Mishan, E. J., 139n3
Murdaugh, H. V., 154n2
Murray, Charles, 91, 96, 101n1, 101n3
180
Nabarro, J. D. N., 154n2
Nixon, Richard, 119
Parkinson. C. Northcote, 4, 14n5, 21, 28n3

Parsons, F. M., 154, 155n10
Pawles, John, 167n4
Perrow, Charles, 14n7
Pinochet, Augusto, 107-10
Piro, P. a., 167n1
Planck, Max, 170, 177n3
Plato, 24, 150
Popper, Karl R., 14n8
Rappoport, E., 139n2

Rawls, John, 24
Rescher, Nicholas, 89n4, 177n4
Rife, Dorothy P., 167n5
Rodgers, Willard L., 126n6
Roosevelt, Franklin Delano, 45
Rosen, Sherwin, 139n2
Rosenberg, Alexander, 14n87
Rosenthal, Josh, 28n1
Rousseau, J. J., 126n7
Schelling, T. C., 140n8

Schrezder, Solomon, 167n5
Simon, Herbert A., 126n6
Smith, Adam, 3, 14n4
Spencer, Herbert, 1
Stalin, Joseph, 87
Standen, Anthony, 123
Staples, Brent, 101n2
Stauffenberg, Klaus Schenck, 110
Stimson, Henry L., 45-46, 50n3
Thaler, Richard H., 139n2

Thomsen, Arnold, 126n6
Thurber, James, 47, 50n5
Wohlstetter, Roberta, 46, 50n4

Woolley, Herbert P., 167n5
Worthington, N. L., 167n1
Zalking, Sheldon S., 126n6

Zeckhauser, Richard, 139n3
181

Nicholas Rescher - Collected Papers - Volume 6 Studies in Social Philosophy-De Gruyter (2006)

Uploaded by

Document Information

Original Title

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

Nicholas Rescher - Collected Papers - Volume 6 Studies in Social Philosophy-De Gruyter (2006)

Uploaded by

Copyright:

Available Formats

NICHOLAS RESCHER COLLECTED PAPERS

Studies in Social Philosophy

North and South America by

United Kingdom, Ire Iceland, Turkey, Malta, Portugal by

Livraison pour la France et la Belgique:

No part of this book may be reproduced, stored in retrieval systems or transmitted

Printed on acid-free paper

Chapter 1: TECHNOLOGY, COMPLEXITY,

Chapter 2: RISKING DEMOCRACY 17

Chapter 3: IS CONSENSUS REQUIRED FOR A RATIONAL

Chapter 4: MORALITY IN GOVERNMENT AND POLITICS 45

Chapter 5: ON THE RATIONALE OF GOVERNMENTAL

Chapter 6: SOCIAL WELFARE:

Chapter 7: COLLECTIVE RESPONSIBILITY 77

Chapter 8: THE BELL CURVE REVISITED 93

Chapter 9: IN THE LINE OF DUTY 105

Chapter 10: TECHNOLOGICAL PROGRESS AND HUMAN

Chapter 11: THE SOCIAL VALUE OF A LIFE 131

Chapter 12: THE ALLOCATION OF EXOTIC

Chapter 13: ETHICAL ISSUES REGARDING

M y interest in issues of social philosophy, now dating back over forty

TECHNOLOGY, COMPLEXITY, AND

TECHNOLOGICAL PROGRESS MAKES LIFE MORE

W e members of Homo sapiens are amphibious creatures: we inhabit

manifest an historical tendency of moving from the simpler to the more

formance of particular tasks, its aggregate effect is to make large-scale

later position in the realm of artifice is bound to be one where complexity

PROBLEM COMPLEXITY OUTPACES SOLUTION COMPLEXITY

With the progress of science, technology, and human artifice generally,

management—widen the scope of the field of action and thereby augment

THE INTIMIDATING IMPETUS OF THE UNKNOWN: RISK AND

The growing complexity that emerges in the wake of technical progress

This phenomenon has ominous implications. Physical chaos, it will be

creasingly error-prone and susceptible to malfunction. And the seriousness

• with cognitive systems we face the threat of disintegration, disorgani-

• with technical systems we face the threat of breakdown and malfunc-

Throughout the domain of human creativity—alike in matters of cognitive,

increase our stake on success.) Increasing complexity thus tends to be as-

IMPLICATIONS OF TECHNOLOGICAL SOPHISTICATIONS

At the earlier, less sophisticated stages of technological progress, it was

take to address the challenges become lost in a fog of impredictability. For

The present deliberations have revolved around four significantly prob-

THE PERVASIVE THREAT OF GRIDLOCK

T he American political system so functions at every level, from city hall

A HYPOTHETICAL SITUATION IN A MINI-ELECTORATE

People’s Acceptability Assessments

Note that most people favor the majority of A-modes.

That A be realized in some way or other is (so we here suppose) favored by

CONCRETIZATION QUANDARIES AND DECISION GRIDLOCK

What might be called a concretization quandary arises when it is—

Despite an accepted “Sense of the Congress” resolution, no implementing

and in modern complex societies already have reached a stage in point of

COMPLEXITY AND ITS CONSEQUENCES

• how best to address the problem of illegal drugs

• what steps to take towards alleviating inner-city poverty

• how to “save” social security as the ratio of young-to-old becomes

• how best to combine the interests of job protection and environ-

• how to bring illegal immigration under control

• how to reduce the danger of handguns to victims of accident and

(1) Our political system is functioning under conditions that render it

(2) The inherent complexity of issues of public policy decision is such

Just where—mere perplexity apart—does this condition of things leave

In summary review, the position of these deliberations is as follows: