Bureaucracy in Modern Society (Peter M. Blau)

4
OM/y
9/\.'% STUDIES
iI r r
IN SOCIOLOGY
Bureaucracy
in Modern Society
by P E T E R M . B L A U , University of Chicago
With a foreword by
CHARLES H. PAGE, Provost, University of

California, Santa Cruz
Random House
NEW YORK
FIFTEENTH 1>1unT1nG, NOVEMBER 1966
© COPYRIGHT, 1956, BY RANDOM HOUSE, INC.
All rights reserved under International and Pan-American

Copyright Con ii s. P bush d i New Y re by Ra d m
House, Inc., and in Toronto, Canada, by Random House of
Canada, Limited.
LIBRARY OF CONGRESS CATALOG CARD NUMBER: 56-7690
MANUFACTURED IN THE UNITED STATES OF AMERICA

FOREWORD
THE MOUNTING interest of social scientists in the study of

the structure and dynamics of bureaucracy has several
sources. Most apparent is the unprecedented growth in
modem society of large-scale formal organizations within
which must be developed hierarchical administrative # g
operating social machinery, if their tasks are to be achieved.
The pacesetters, of course, are big business and industry,
big government, massive armed forces, and, in recent years,
big labor; but bureaucracy's features mark more and more
areas of modern life, including, for example, many associ-
ations devoted to education, scientific and scholarly pur-
_
suits, religion,
. .. social welfare, and recreation. These facts
. .. .
of changing social organization are inescapable for social

scientist and layman alike. _. an
. ..
. . ..
The moral and political implications of these facts are

a second source of interest in bureaucracy, W coal scien-
tists, no less than philosophy m
anti artists an,d~ many less
articulate witnesses-and bureaucracy's
multisided thrust, are often deeply concerned with tile pre-
sumed dangers of standardization and routinization, Wh:-
personality and interchangeability, of bigness itself. These
traits of bureaucracy are viewed in many quarters as an
imposing threat to freedom, individualism, and sponta-
neity, cherished values in a liberal society. Such anxieties
are not relieved, necessarily, by the recognition of the enor-
mous accomplishments of man's "greatest social invention."
But many social scientists are, or become, social tech-
nicians. Managerial preoccupation with the Improvement
of organizational e§'icieney--whether the goal is the pro-
duction of automobiles or the training of combat troops
5
6 BUREAUCRACY IN MODERN SOCIETY
or the provision of social services or the education of young

people-encourages the recruitment of personnel equipped
with social research skills. And here is a third source of
growing interest in bureaucracy. For these recruits are not
only social technicians, providing immediately useful in-
formation for the managers of men and machines. In many
cases, they maintain their scientific role, producing empiri-
cal studies, the findings of which, of course, may often
serve managerial interests, but may also contribute to accu-
mulative social theory. Thus the Hawthorne Western Elec-
tric studies, a landmark investigation of the 1930s, helped
to revise the theoretical model of bureaucracy by demon-
strating and documenting the role of various social factors
in the operations of the plant, including heretofore obscure
functions of informal groups and relationships. This study
and subsequent reports of the informal and relatively
spontaneous features of bureaucracy--in factories, govern-
mental bureaus, military units, and elsewhere--have altered
the ever formal model by establishing the positive func-
tional contributions (not merely the dysfunctions) of "bu-
reaucracy's other face."
Its informal face was known to Max Weber, although
that great theorist's ideal scheme strongly accents bureauc-
racy's formal components. Weber's theoretical formulations
in this subject, supported by the general prestige of his
writings among American sociologists in recent years, con-
stitute a further source of widespread interest in bureauc-
racy.
These several influences are evidenced in a rapidly ex-
panding literature. Recent publications by Reinhard Bendix,
Peter Blau, Robert Dub if, Alvin W. Gouldner, S. M.
Lipset, Herbert A. Simon, and philip Selznick (among
others) represent efforts to test and refine theoretical propo-
sitions on the basis of empirical research. In 1952, R. K.
Merton and his colleagues published a source book of
Foreword '7
readings,* containing an impressive sample of past and

contemporary writings on diverse aspects of bureaucracy.
These are sure signs that the serious study of the subject
has come of age.
Professor Blau's Short Study is the first systematic socio-
logical textbook on bureaucracy as such. In contrast with
some pioneering textual ventures, however, Bureaucracy in
Modern Society reveals, on the one hand, the author's
close familiarity with the numerous and frequently fugitive
contributions to the field and, on the WEor hand, keen
insights derived from his own investigations. These qualities
help to make this study a text of many merits' theoretical
sophistication and conceptual precision, skillful and illumi-
nating utilization of concrete materials--note especially
the rewarding exploitation of case studies in the treatment
of "Bureaucracy in Process", clarity of exposition, free of
unnecessary jargon and designed to hold the reader to the
march of the analysis, economy of presentation, encour-
aging the student to read more widely in the Held, a pur-
suit now abetted by the availability of 3.11 excellent source
book.
Bureaucracy in Modern Society possesses a further vir-
tue. For the author brings out sharply positive as well as
negative functional interrelations between bureaucracy and
democracy, some of which are by no means apparent. To
be sure, Professor Bleu aligns himself with democratic
values. But his study, I believe, is a stimulating and valu-
able book for bureaucrats M mselves, for bureaucracy's
severest critics, and for both students and readers in gen-
eral, more anti more o w h o r n today must man bureauc-
racy's posts.
CHARLES H. PAGE
* Reader in Bureaucracy, Glencoe, Ill.: Free Press, 1952.

Merton's coeditors are Ailsa P. Gray, Barbara Hockey, and
Hanan C. Selvin.
CONTENTS
FOREWORD by Charles H. Page 5
I. Why Study Bureaucracy? 13
The Rationalization of Modem Life 14

The Value of Studying Bureaucracy 20
Today 20
In a Democracy 21
For Sociologists 23
II. Theory cmd Development of Bureaucracy 27
The Concept of Bureaucracy 28

Implications of the Idea1~Type Construct 34
Conditions That Give Rise to Bureaucratization 36
Historical Conditions 36
Structural Conditions 40
Ill. Bureaucracy in Process 45
Bureaucracy's Other Face 46

In the Navy 46
In a Factory 48
In a Federal Law~Enforcement Agency 50
9
10 BURBAUCRACY :n momznn SOCIETY
Organization of Work Groups 53

Bureaucracy's New Face 5'7
Irrationality of Rationalistic Administration 58
Conditions of Adjustive Development 61
The Task of the Administrator 66
IV. Bureaucratic Authority 69
Strategic Leniency and Authority 70

Power of Sanction 74
Inequality in Hierarchical Organizations 80
v. Bureaucracy and Social Change 85
Who Are the Ritualists? 86

Bureaucracy as Instrument of Innovation 91
Conservative Pressures in Two Social Contexts 96
VI. Bureaucracy and Democracy
The Accusation of "Red Tape" 102

Contrasting Principles of Internal Control 105
Efficiency versus Dissent 106
A Union with Two Parties 110
A Challenge for Democracy 114
FOOTNOTES 119
SELECTED READINGS 125

BUREAUCRACY
IN MODERN SOCIETY
I
1
Why Study Bureaucracy?
"THAT STUPID BU1ZNAUCRA*r!" Who has not felt this way

at one time or another? When we are sent from one official
to the next without getting the information we want; when
lengthy forms we had to E11 out in sextuplicate are re-
turned to ~us because we forgot to cross a "t" or dot an
1 when our applications are refused on some technicality
-that is when we think of bureaucracy. colloquially, HI
terns "bureaucracy" an become an epithet which refers
to inefficiency and red tape in the government, but this
was not its original meaning, and in not the way the
term will be used in this book.
If you alone had the job of collecting the dues in a
small fraternity, you could proceed at your own discretion.
But if five persons had this job in a large club, they would
Iind it necessary to organize their work lest some members
were asked for dues repeatedly and others never. If hun-
13
14 BURBAUCRACY IN MODERN SOCIETY
dreds of persons have the assignment of collecting taxes

from millions of citizens, their work must be very system-
atically organized; otherwise chaos would reign and the
assignment could not be fulfilled. The type of organization
designed to accomplish large~scale administrative tasks by
systematically coordinating the work of many individuals
is called a bureaucracy. This concept, then, applies to
organizing principles that are intended to improve admin-
istrative eliiciency and that generally do so, although
bureaucratization occasionally has the opposite effect of
producing inefficiency. Since complex administrative prob-
lems confront most large organizations, bureaucracy is not
confined to the military and civilian branches of the gov-
ernment but is also found in business, unions, churches,
universities, and even in baseball.
While the popular notion that bureaucracies are typi-
cally ineilicient is not valid, this does not mean that the
social scientist can simply dismiss it. The prevalence of this
false belief in our society is a social fact that should be
explained. In this study, after bureaucratic operations have
been analyzed and clarified, such an explanation will be
suggested in the last chapter. There we shall see that
bureaucratization has implications in a democratic society
that engender antagonism toward it. Whereas this antago-
nism usually results from the ruthless efficiency of bureauc-
racies, and not from their inefficiency, people often feel
constrained to give vent to it by accusing bureaucracies of
inefficiency, just as you might call a fellow who made you
angry "stupid" even though it was not his lack of intelli-
gence that aroused your anger.
The Rationulizotion of Modern Like

Much of the magic and mystery that used to pervade
human life and lend it enchantment has disappeared
Why Study Eureaucracy? 15
from the modem world.* This is largely the price of ra-

tionalization. In olden times, nature was full of mysteries,
and roan's most serious intellectual endeavors were di-
rected toward discovering the ultimas meaning of his
existence. Today, nature holds fewer secrets for us. Scien-
tihc advances, however, have not only made it possible
.. .
to
. . . ... . . . . .
explain many natural phenomena but have also channeled

human thinking. Modern man is less concerned than, say,
medieval man was with ultimate values and symbolic
meanings, with those aspects of mental life that are not
subject to scientific inquiry, such as religious truth and
artistic creation. This is an age of great scientists and
engineers, not of great philosophers or prophets.
The secularization of the world that spells its disen-
l
chantrnent is indicated by the large amount of time we
spend in making a living and getting ahead, and the little
time we spend in contemplation and religious activities.
Compare the low prestige of moneylenders and the high
prestige of priests in former eras with the very different
positions of bankers and preachers today. reoccupied
with perfecting efficient means for achieving objectives,
we tend to forget why we want to reach those goals. Since
we neglect to clarify the basic values that determine why
some objectives e preferable to others, objectives lose
significance, and their pursuit becomes an end in
itself. This tendency is portrayed in Budd Shulberg's novel
What Makes Sammy Run? The answer to the question in
the title is that only running makes him run, because he is
so busy trying to get ahead that he has no time to find
out where he is going. Continuous striving for success is
* The disench;;;1;;;;1§_n; of the world is a main theme running
through the writings M the German sociologist Max Weber,
whose classical analysis of bureaucratic structure be dis-
Classed presently.
not Sammy's means for the attainment of certain ends but

the very goal of his life.
These consequences of rationalization have often been
deplored, and some observers have even suggested that it
is not worth the price.1 There is no conclusive evidence,
however, that alienation from profound values is the in-
evitable and permanent by-product of rationalization, and
not merely an expression of its growing pains. The bene-
Hcial results of rationalization-notably the higher stand-
ard of living and the greater amount of leisure it makes
possible, and the raising of the level of popular education
it makes necessary-*permit an increasing proportion of
the population, not just a privileged elite, to participate
actively in the cultural life of the society. This could
ultimately lead to a flowering of the arts and other cultural
pursuits OD. a wider scale than that in any earlier period.
Our high standard of living is usually attributed to the
spectacular technological developments that have occurred
since the Industrial Revolution, but this explanation ig-
nores two related facts. First, the living conditions of most
people during the early stages of industrialization, after
they had moved from the land into the cities with their
sweatshops, were probably much worse than they had
been before. Dickens depicts these terrible conditions in
certain novels, and Marx describes them in his biting
critique of the capitalistic economy? Second, major im-
provements in the standard of living did not take place
until administrative procedures as well as the material
technology had been revolutionized. Modern machines
could not be utilized without the complex administrative
machinery needed for running factories employing thou-
sands of workers. It was not so much the invention of new
machines as the introduction of mass-production methods
that enabled Henry Ford to increase wages and yet pro-
duce a car so cheaply that it ceased to be a luxury. When
Why Study Bureaucracy? 17
Ford later refused to make further administrative innova-

tions, in the manner of his competitors, the position of his
company suffered, but after his grandson instituted such
changes the company manifested IICW competitive strength.
Rationalization in administration is a prerequisite for the
full exploitation of technological knowledge in mass pro-
duction, and thus for a high standard of living*
Let us examine some of the administrative principles
on which the productive efficiency of the modem factory
depends. If every worker manufactured a complete car,
each would have to be a graduate of an engineering col-
lege, and even then he could not do a very good job, since
it would be impossible for him to be at once an expert
mechanical engineer, electrical engineer, industrial
IN
designer. Besides, ere would not be enough people with
engineering degrees in the country to fill all the positions.
Specialization numits the employrnnlullllql
trained workers, which lowers production costs. Moreover,
whereas the jack-of-all-trades is necessarily master of none,
each employee can become a highly skilled expert in his
particular field of specialization.
What has been taken apart must be put together again.
A high degree of specialization creates a need for a com-
plex system of coordination. 111 such need exists in the
small shop, where the work is less specialized, all workers
have direct contact with one another, and the boss can
supervise the performance of all of them. Un. president
of a large company cannot possibly discharge his man-
* To be sure, activities of trade unions have greatly con-
tributed to the raising of our standard of living by forcing
employers to distribute a larger proportion of their income
to workers. Without administrative efficiency in the production
and distribution of goods, however, there would be less income
to distribute, and fewer goods could be bought with a given
amount of income. Moreover, the strength of unions also
depends on an efficient administrative machinery.
18 BUREAUCRAJCY IN MODERN SOCIETY
agerial responsibility for coordination through direct con-

sultation with each one of several thousand workers.
Managerial responsibility, therefore, is exercised through
a hierarchy of authority, which furnishes lines of com-
munication between top management and every employee
for obtaining information OI1 operations and transmitting
operating directives- (Sometimes, these lines of communi-
cation become blocked, and this is a major source of in-
eliiciency in administration.)
Effective coordination requires disciplined performance,
which cannot be achieved by supervision alone but must
pervade the work process itself. This is the function of
rules and regulations that govern operations whether they
specify the dimensions of nuts and bolts or the criteria to
be used in promoting subordinates. Even in the ideal case
where every employee is a highly intelligent and skilled
expert, there is a need for disciplined adherence to regula-
tions. Say one worker had discovered that he could pro-
duce bolts of superior quality by making them one-eighth
of an inch larger, and another worker had found that he
could increase efficiency by making nuts one-eighth of an
inch smaller. Although each one made the most rational
decision in terms of his own operations, the nuts and bolts
would Of course be useless because they would not match.
How one's own work tits together with that of others is
usually far less obvious than in this illustration. For the
operations of hundreds of employees to be coordinated,
each individual must conform to prescribed standards even
in situations where a different course of action appears to
him to be most rational. This is a requirement of all team-
work, although in genuine teamwork the rules are not
imposed from above but are based on common agreement.
Efficiency also suffers when emotions or personal con-
siderations influence administrative decisions. If the owner
of a small grocery expands his business and opens a
second store he may put his SOI1 in charge even though

another employee is better qualified for the job. He acts
on the basis of his personal attachment rather than in the
interest of business etliciency. Similarly, an official in a
large company might not promote the best-qualified worker
to foreman if one of the candidates were his brother. In-
deed, his personal feelings could prevent him from recog-
nizing that the qualifications of his brother were inferior.
Since the subtle eiiects of strong emotions cannot easily
'interference
be suppressed, the best way fo check their
with efficiency is to exclude from the administrative hier-
archy those interpersonal relationships that are character~
ized by emotional attachments. While relatives sometimes
work for the same company, typically they are not put
in charge of one another. Impersonal relationships assure
the detachment necessary if efficiency alone is to govern
administrative decisions. However, relationships between
employees who have frequent social contacts do not re-
main purely impersonal, as we shall see.
These four factors-specialization, a hierarchy of au-
Mority, a system of rules, and impersonality-are the
basic characteristics of bureaucratic organization. Fac-
tories are bureaucratically organized, as are government
agencies, and if this were not the case they could not
operate efficiently on a large scale. Chapter Two is devoted
to a more detailed analysis of bureaucratic structure and
the conditions that give rise to bureaucratization. But
actual operations do not exactly follow the formal blue-
print. To understand how bureaucracies function, we must
observe them in action. This is the task of Chapters Three
and Four, which are concerned, respectively, with bureau-
cratic work groups and relationships of authority. After
discussing the internal structure and functioning of bu-
reaucracies, we shall turn in the final two chapters to
their implications for the society of which they are a part.
20 BUREAUCRACY IN NIODERN SOCIETY
SpecMcally, we shall examine Me consequences of bu-

reaucratization for social change and for democracy. First,
however, the question raised in the title of this introduc-
tory chapter should be answered: why study bureaucracy?
The Value of Studying Bureaucracy

Learning to understand bureaucracies is more important
today than it ever was. It is, besides, of special significance
in a democracy. Finally, the study of bureaucratic organi-
zation makes a particular contribution to the advancement
of sociological knowledge.
Today Bureaucracy is not a new phenomenon. It

existed in rudimentary forms thousands of years ago in
Egypt and Rome. But the trend toward bureaucratization
has greatly accelerated during the last century. In contem-
porary society bureaucracy has become a dominant institu-
tion, indeed, the institution that epitomizes the modem
era. Unless we understand this institutional form, we can-
not understand the social life of today.
The enormous size of modern nations and the organiza-
tions within them is one reason for the spread of bureauc-
racy. in earlier periods, mm countrify were small, even
large ones had only a loose central administration, no
there were few formal organizations except the govern-
ment. Modern countries have many millions of citizens,
vast armies, giant corporations, huge unions, and numer-
ous large voluntary associations? To be sure, large size is
not synonymous with bureaucratic organization. However,
the problems posed by administration on a large scale tend
to lead to bureaucratization. As a matter of fact, the large
organizations that persisted longest in antiquity and even
survived this period, the Roman Empire and the Catholic
Church, were thoroughly bureaucratized.
In the United States, employment statistics illustrate

the trend toward large, bureaucratic organizations. The
federal government employed 8000 civil servants in 1820,
a quarter of a million fifty years ago, and ten times that
number today. If the men in military service are added,
nearly 10 per cent of the American labor force, six mil-
lion people, are in the employ of the federal government.
Still larger is the number who work for large-scale private
concerns, the extreme example being the American Tele-
phone and Telegraph Company with three-quarters of a
million employees. More than three-quarters of the em-
ployees in manufacturing work for Firms with one hundred
or more employees, and even in the retail trades, the bul-
wark of small business, one-sixth of all employees work
for Nuns of the same size.
A large and increasing proportion of the American
people spend their working lives as small cogs in the com-
plex mechanisms of bureaucratic organizations. And this
is not all, for bureaucracies also affect much of the rest
of our lives. The employment agency we approach to get
a job, and the union we join to protect it; the supermarket
and the chain store where we shop, the school our children
attend, and the political parties for whose candidates we
vote, the fraternal organization re we play, and the
church where we worship-all these more often than not
are large organizations of the kind that tends to be bu-
reaucratically organized.
In a Democracy Bureaucracy, as the foremost theo-

retician on the subject points out, "is a power instrument
of the first order for the one who controls the bureau-
cratic apparatus." 4
Under normal conditions, the power position of a fully

developed bureaucracy is always overpowering. The
"political master" Ends himself in the position of the

"dilettante" who stands opposite the "expert," facing
the trained otlicial who stands within the management
of administration. This holds whether the "master"
whom the bureaucracy serves is a "people," equipped
with the weapons of "legislative initiative," the "referen-
dum," and the right to remove officials, or a parliament,
elected on a .. . "democratic" basis and equipped with
the right to vote a lack of confidence, or with the actual
authority to vote it.5
Totalitarianism is the polar case of such bureaucratic

concentration of power that destroys democratic processes,
but not the only one. The same tendency can be observed
in political machines that transfer the power that legally
belongs to voters to political bosses, in business corpora-
tions that vest the power that rightfully belongs to stock-
holders in corporation officials, and in those unions that
bestow the power that rightfully belongs to rank-and-lilo
members upon union leaders. These cases lead some writers
to contend that the present trend toward bureaucratization
spells the doom of democratic institutions. This may well
be too fatalistic a viewpoint, but there can be 110 doubt
that this trend constitutes a challenge. To protect ourselves
agalmst this threat, while continuing to utilize these etlicient
administrative mechanisms, we must first learn fully to
understand how bureaucracies function. Knowledge alone
is not power, but ignorance surely facilitates subjugation.
This is the reason why the study of bureaucratic organiza-
tion has such great significance in a democracy.
The problem of efficiency versus democracy, which will
occupy us at length later, can initially be clarified by dis-
tinguishing three types of association. If an association
among men is established for the explicit purpose of
producing jointly certain end-products, whether it is manu-
facturing cars or winning wars, considerations of efficiency
are of primary importance, hence bureaucratization will

further the achievement of this objective. However, if an
association is established for the purpose of finding intrinsic
satisfaction in common activities, say in religious worship,
considerations of efficiency are less relevant. When such
an association, for instance a religious body, grows so
large that administrative problems engender bureaucratiza-
tion, the pursuit of the original objectives may, indeed, be
hamperedf* Finally, if an association is established for the
purpose of deciding upon common goals and courses of
action to implement them, which is the function of demo-
cratic government (but not that of government agencies),
the free expression of opinion must be safeguarded against
other considerations, including those of efficiency. Since
bureaucratization prevents the attainment of this objective,
it must be avoided at all cost. Ideally, organizations of the
first type would always be bureaucratized, and those al
the last type, never. But one of the difficulties is that many
organizations, such as unionE are of a mixed type.
For Sociologists The study of bureaucratic organiza-

nu is of special significance for sociologists because m
helps them in their task of finding an order in the complex
interdependencies of social phenomena. The sociologist is
concerned with explaining patterY. of human behavior
in terms of relationships between people and shared norma-
tive beliefs of people. For example, to explain why some
students get poorer grades than others who are DO more
intelligent, this sociological hypothesis could be advanced:
the former have fewer friends and the discomfort of their
social isolation interferes with their work. Let us assume we
would actually ind that the grades of isolated students are
lower than those of the rest. Would that prove the hy-
pothesis? By no means, since the difference could be due
to the fact that students who appear stupid in class become
less popular, or that radicals (or any other group) are

discriminated against by teachers and are also disliked by
fellow students.
This problem can be solved in the controlled experiment,
which makes it possible to demonstrate that a specific
factor has certain effects because all other factors are held
constant. If two test tubes have exactly the same content
and a-re kept under the same conditions except that one is
heated, the changes that occur in one liquid but not in the
other must be the result of heat. Many social conditions,
however, in contrast to most physical conditions, cannot
be duplicated in the laboratory. Although we can make
human subjects feel isolated in an experimental session,
this is not the same experience as having no friends in col-
lege, and other social conditions, such as international
warfare, cannot be reproduced in the laboratory at all.
This is a dilemma of social research: controlled conditions
are required for the testing of hypotheses, but the artificial
situation in laboratory experiments is usually not suitable
for this purpose. Not that this is an insurmountable dif-
ficulty, techniques have been devised to approidmate the
analytical model of the controlled experiment outside the
laboratory. still, the larger the number of varying factors
in the social situation, the smaller is the chance that ex-
planatory hypotheses can be confirmed.
Bureaucracy provides, as it were, a natural laboratory
for social research. The formal organization, with its ex-
plicit regulations and official positions, constitutes COD-
trolled conditions, and these controls have not been artifi-
cially introduced by the scientist but are an inherent part
of the bureaucratic structure. To be sure, the daily activi-
ties and interactions of the members of a bureaucracy
cannot be entirely accounted for by the oliicial blueprint.
If they could, there would be no need for conducting ern-
pirical studies in bureaucracies, since everything about
them could be learned by examining organizational charts

and procedure manuals. Several factors in addition to
oliicial requirements influence daily operations, which
means, of course, that conditions are not as fully COD.-
trolled as in a laboratory experiment. Nevertheless, the
explicit formal organization, the characteristics of which
can be easily ascertained, reduces the number of variable
conditions in the bureaucratic situation and thereby facili-
tates the search for and the testing of explanatory hypoth-
eses. n o
In summary, the prevalence. o_t. bureaucracies in our

society furnishes a practical reason for studying them;
fact that they endanger democratic institutions supplies an
ideological reason; and the contribution their study can
make to sociological knowledge provides ala scientific
reason for undertaking this task.
2
Theory and Development

of Bureaucracy
ADVANCEMENT in any science tiepeniis OH developments in

both theory and empirical research and on a close con-
nection between them. sobjectiyes of science are to
improve the accuracy and scope of explanations of phe-
nomena as a basis for better predictability and control. A
system of interrelated explanatory propositions is a scien-
`msight,
tihc theory. Not every however, is a scientific

proposition, this term refers only to those that have been
confirmed in systematic research or can at least be con-
firmed in future research, which is not the case for all
explanations. Toynbee's interpretation of history in terms
of challenge and response, time instance, although it may
provide new insights into the course of history, cannot be
empirically tested, since there is no conceivable factual
27
28 BUREAUCRACY :n MODERN SOCIETY
evidence that would clearly disprove it. An important

methodological principle of science holds that only those
propositions can be empirically conlirrned that indicate
precisely the evidence necessary for disproving them.
If undisciplined speculating does not further the ad-
vancement of science, neither does random data-collecting.
A large number of miscellaneous facts contribute as little
to the building of systematic theory as a large number of
odd stones contribute to the building of a house. To he
sure, unsophisticated fact-finding has its uses, and so does
undisciplined imagination, but for empirical research and
theoretical insights to serve science, they must be in-
tegrated. This requires that theory be precise enough to
direct research, and that research be oriented toward
establishing theoretical generalizations.
The lesson to be learned from these considerations is
that the study of bureaucracy should be governed by a
theoretical orientation and should teens upon the investiga-
tion of empirical cases. These case studies of bureaucra-
cies, in turn, will help to clarify and refine our theoretical
understanding of this social structure and its functioning.
Following this procedure, we shall start with Max Weber's
famous theory of bureaucracy.
The Concept of Bureaucracy

The main characteristics of a bureaucratic structure (in
the "ideal-typical" case*), according to Weber, are the
following:
1. "The regular activities required for the purposes of
the organization are distributed in a fixed way as oiiiciai
duties." 1 The clear-out division of labor makes it possible
to employ only specialized experts in each particular posi-
* The "ideal type" is discussed later in this chapter.
Theory and Development of Bureaucracy 29
son and to make every one of them responsible for the

effective p_erformance of his duties. This high degree of
specialization has become so much part of our socio-eco-
nomic life that we tend to forget that it did not prevail in
former eras but is a relatively recent bureaucratic innova-
:nm
w ,
2. "The organization of offices follows the principle of

hierarchy; that is, each lower office is under the contre-l
and supervision of a higher one."2 Every official in this
administrative hierarchy is accountable to his superior for
his subordinates' decisions and actions as well as his own.
To be able to discharge his responsibility for the work of
subordinates, he bas authority over them, which means
that he has the right to issue directives and they have the
duty to obey them. This authority is strictly circumscribed
and confined to those directives that are relevant for of~
facial operations. The use of status prerogatives to extend
the power of control over subordinates beyond these limits
does not constitute the legitimate exercise of bureaucratic
authority. "r»n..
. Operations are governed Wu consistent system of

.
abstract rules I us [and] consist an
the application of
these rules to particular cases.'w»a J system of Stalldtlfds
is designed to assure uniformity in the performance of
every task, regardless of the number of persons engaged
in it, and the coordination of different tasks. Hence explicit
rules and regulations define the responsibility of each
member of the organization and the relationships between
them. This does not imply that bureaucratic duties are
necessarily simple and routine. It must be remembered
that strict adherence to general standards in deciding
specific cases characterizes not only the job of the file
clerk but also that Of the Supreme -cOurt justice. For the
former, it may involve merely filing alphabetically; for
the latter, it involves interpreting the law of the land in
30 BUREAUCRAT-CY IN MODERN SOCIETY
order to settle the most complicated legal issues. Bureau-

cratic duties range in complexity from one of these ex-
tremes to the other.
4. "The ideal official conducts his office . ..[in] a
spirit of formalistic impersonality, 'Sine fro of studio,'
without hatred or passion, and hence without affection or
enthusiasm." 4 For rational standards to govern operations
without interference from personal considerations, a de-
tached approach must prevail within the organization and
especially toward clients. If an oii';icia1 develops strong
feelings about some subordinates or clients, he can hardly
help letting those feelings influence his official decisions.
As a result, and often without being aware of it himself,
he might be particularly lenient in evaluating the work of
one of his subordinates or might discriminate against
some clients and in favor of others. The exclusion of per-
sonal considerations from official business is a prerequisite
for impartiality as well as for efficiency. The very factors
that make a government bureaucrat unpopular with his
clients, 3.11 aloof attitude and a lack of genuine concern
with their problems, ctually benefit these clients. Disin-
terestedness and lack of personal interest go together. The
official who does not maintain social distance and becomes
personally interested Io the cases of his clients tends to be
partial in his treatment of them, favoring those he likes
over others. Impersonal detachment engenders equitable
treatment of all pens BS and thus fosters democracy in
administration.
5. Employment in the bureaucratic organization is based
on technical qualifications and is protected against arbi-
tr my dismiss I. "It c sit tes career. There is sys in
of 'promotionS sccording to seniority or to achievement,
or both." 5 These personnel policies, which are found not
only in civil service but also in many private companies,
Theory and Development Go Bureaucracy 31
encourage the development of loyalty to the organization

and esprit de corps among its members. The consequent
identification of employees with the organization motivates
them to exert greater efforts in advancing its interests. It
\
may also give rise to a tendency to think of themselves as
a class apart from and superior to the rest of the society.
Among civil servants, this tendency has been more pro-
nounced in Europe, notably in Germany, than in the
United States, but among military oiiicers, it may be
found here, llmllliia
"Experience tends universally to show that the purely
bureaucratic type M dministrative organization .. . is,
from a purely technical point of view, capable of attaining
the highest degree. of efiicie "The fully developed
bureaucratic mechanism compares with other organizations
exactly as does the machine with nor mechanical modes
of production." 7 Bureaucracy solves the distinctive organi-
zational problem of maximizing organizational efficiency,
not merely that of individuals.
The superior administrative efficiency of bureaucracy is
the expected result of its various characteristics as out-
lined by Weber. For an individual to work efficiently, he
must have the necessary skills and apply them rationally
and energetically, but for an organization to operate effi-
ciently, more is required. Every one of its members must
have the expert skills needed for the performance of his
tasks. This is the purpose of specialization and of employ~
meet on the basis of technical qual'cations, often ascer-
tained by objective tests. Even experts, however, may be
prevented by personal bias from making rational decisions.
The emphasis on impersonal detachment is intended to
eliminate this source of irrational action. But individual
rationality is not enough. If the members of the organiza-
tion were to make rational decisions independently, their
work would not be coordinated and the efficiency of the

organization would suffer. Hence there is need for disci-
pline to limit the scope of rational discretion, which is met
by the system of rules and regulations and the hierarchy
of supervision. Moreover, personnel policies that permit
employees to feel secure in their jobs and to anticipate ad-
vancements for faithful performance of duties discourage
attempts to impress superiors by introducing clever in~
novations, which may endanger coordination. Lest this
stress on disciplined obedience to rules and rulings under-
mine the employees motivation to devote his energies to
his job, incentives for exerting effort must be furnished.
Personnel policies that cultivate organizational loyalty and
that provide for promotion on the basis of merit serve this
function. In other words, the combined effect of bureauc-
racy's characteristics is to create social conditions which
constrain each member of the organization to act in ways
that, whether they appear rational or otherwise from his
individual standpoint, further the rational pursuit of
organizational objectives.
Without explicitly stating so, Weber supplies a functional
analysis of bureaucracy. In this type of analysis, a social
structure is explained by showing how each of its elements
Contributes to its persistence and elective operations. Con-
oem with discovering all these contributions, however,
entails the danger thatthe scientist may neglect to inves-
tigate the disturbances that various elements produce in
the structure. As a result, his presentation may make the
social structure appear to function more smoothly than it
actually does, since he neglects the disruptions that do in
fact exist. To protect ourselves against this danger, it is
essential to extend the analysis beyond the mere considera-
tion of functions, as Robert K. Merton points out.8 Of
particular importance for avoiding false implications of
stability and for explaining social change is the study of
dysfunctions, those consequences that interfere with adjust-

ment and create problems in the structure."
A re-examination of the foregoing discussion of bureau-
cratic features in the light of the concept of dysfunction
reveals inconsistencies and conflicting tendencies. If re-
served detachment characterizes the attitudes of the mem-
bers of the organization toward one another, it Is unlikely
that high esprit de corps will develop among them. The
strict exercise of authority in the interest of discipline
induces subordinates, anxious to be highly thought of by
their superiors, b conceal defects in operations from
superiors, and this obstruction of the flow of information
upward in the hierarchy impedes effective management.
insistence on conformity also tends to engender rigidities
in official conduct and to inhibit the rational exercise of
judgment needed for efficient performance of tasks. If
promotions are based on merit, many employees will not
experience advancements in their careers, if they are based
primarily OD seniority SO as to give employees this experi-
ence and thereby to encourage them to become identified
with the organization, i promotion system will not
furnish strong incentives for exerting efforts and excelling
in one's job. These illustrations surplice to indicate that the
same factor that enhances efficiency in one respect often
threatens it in another; it may have both functional and
dysfunctional consequences.
Weber was well aware of such contradicatory tendencies
in the bureaucratic structure. But since' he treats dysfunc-
tions only incidentally, his discussion leaves the impression
that administrative efficiency in bureaucracies is more
stable and less problematical than it actually is. In part, it
was his intention to present an idealized image of bureau-
cratic structure, and he used u conceptual tool appro-
priate for this purpose. Let us critically examine this
conceptual tool. I
34 BUREAUCRACY IN MODERN socznnr
Implications of the Ideal-Type Construct

Weber dealt with bureaucracy as what he termed an
"ideal type." This methodological concept does not repre-
sent 3.11 average of the attributes of all existing bureaucra-
cies (or other social structures) , but a pure type, derived by
abstracting the most characteristic bureaucratic aspects of
all known organizations. Since perfect bureaucratization
is never fully realized, no empirical organization cor-
responds exactly to this scientific construct.
The criticism has been made that Weber's analysis of an
imaginary ideal type does not provide understanding of
concrete bureaucratic structures. But this criticism obscures
the fact that the ideal-type construct is intended as a guide
in empirical research, not as a substitute for it. By indicat-
ing the characteristics of bureaucracy in its pure form, it
directs the researcher to those aspects of organizations
that he must examine in order to determine the extent of
their bureaucratization. This is the function of all concep-
tual schemes: to specify the factors that must be taken
into consideration in investigations and to define them
clearly.
The ideal type, however, is not simply a conceptual
scheme. It includes not only definitions of concepts but
also generalizations about the relationships between them,
specifically the hypothesis that the diverse bureaucratic
characteristics increase administrative efficiency. Whereas
conceptual definitions are presupposed in research and
not subject to verification by research findings, hypotheses
concerning relationships between factors are subject to
such verification. Whether strict hierarchical authority, for
example, in fact furthers efficiency is a question of em-
pirical fact and not one of definition. But as a scientific
Theory and Development OJ' Bureaucracy 35
construct, the ideal type cannot be refuted by empirical

evidence. If a study of several organizations were to find
that strict hierarchical authority is not related to efficiency,
this would not prove that no such relationship exists in the
ideal-type bureaucracy, it would show only that these
organizations are not fully bureaucratized. Since generali-
zations about idealized states defy' testing in systematic
research, they have no place in science. On Ere other hand,
if empirical evidence is taken into consideration and gen-
eralizations are modified accordingly, we deal with prevail-
ing tendencies in bureaucratic structures and no longer
with a pure type.
Two misleading implications of the ideal-type concep-
tion of bureaucracy deserve special mention. The student
of social organization is concerned with the patterns of
activities and interactions that reveal how social conduct
is organized, and not with exceptional deviations from
these patterns. The fact that one official becomes excited
and shouts at his colleague, or that another arrives late
at the othce, is unimportant in understanding the organiza-
tion, except that the rare occurrence of such events in-
dicates that they are idiosyncratic, differing from the
prevailing patterns. Weber's decision to treat only the
purely formal organization of bureaucracy implies that all
deviations from these formal requirements are idiosyn-
cratic and of 110 interest for the student of organization.
Recent empirical studies have shown this approach to be
misleading. Informal relations and unofficial practices
develop among the members of bureaucracies and assume
an organized form without being otiicially sanctioned.
Chester I. Barnard, one of the first to call attention to this
phenomenon, held that these "informal organizations are
necessary to the operations of formal organizations." 10
These informal patterns, in contrast to exceptional occur-
fences, as we shall see in Chapter Three, are a regular part

of bureaucratic organizations and therefore must be taken
into account in their analysis.
Weber's approach also implies that any deviation from
the formal structure is detrimental to administrative ef-
ficiency. Since the ideal type is conceived as the perfectly
efficient organization, all differences from it must DBCES-
sarily interfere with efficiency. There is considerable
evidence that suggests the opposite conclusion; informal
relations and unofficial practices often contribute to elii-
cient operations. In any case, the significance of these
unofficial patterns for operations cannot be determined in
advance OD theoretical grounds but only on the basis of
factual investigations. Before examining such case studies
of bureaucracies it is useful to explore the conditions
that give rise to bureaucratization.
Conditions that Give Rise

to Bureauczratizution
To say that there is a historical trend toward bLlreauc~
racy is to state that many organizations change from less
to more bureaucratic forms of administration. Yet the
historical trend itself and the changes in any specific
organization are different phenomena. Both are expres-
sions of the process of bureaucratization, but since different
conditions account for them, they will be discussed sep-
arately.
Historical Conditions One of the historical conditions

that favors the development of bureaucracy is a money
economy. This is not an absolute prerequisite. Bureaucra-
cies based on compensation in kind existed, for example,
in Egypt, Rome, and China. Generally, however, a money
economy permits the payment of regular salaries, which,
In turn, creates the combination of dependence and indy

pendence that is most conducive to the faithful perfonn-
ance of bureaucratic duties- Unpaid volunteers are too
independent of the organization to submit unfailingly to its
discipline. Slaves, on the other hand, are too dependent on
their masters to have the incentive to assume responsibili-
ties and carry them out on their own initiative. The eco-
nomic dependence of the salaried employee on his job and
his freedom to advance himself in his career engender the
orientation toward work required for disciplined and
responsible conduct. Consequently, there were few bu-
reaucracies prior to the development of a monetary system
and the abolition of slavery.
It has already been mentioned that sheer size encourages
the development of bureaucracies, since they are mechan-
isms for executing large~scale administrative tasks. The
la_rge modern nation, business, or union is more likely to
be bureaucratized than was its smaller counterpart in the
WW»=W'M important than size as such, however, is the
emergence of special administrative problems. Thus in
ancient Egypt the complex job of constructing and regulat-
ing waterways throughout the country gave rise to the
first known large-scale bureaucracy in history. In other
countries, notably those with long frontiers requiring
defense, bureaucratic methods were introduced to solve
the problem of organizing an effective army and the related
one of raising taxes for this purpose. England, without
land frontiers, maintained only a small army in earlier
centuries, which may in part account for the fact that the
trend toward bureaucratization was less pronounced there
than in continental nations, which had to support large
armies. Weber cites the victory of the Puritans under the
leadership of Cromwell over the Cavaliers, who fought
more heroically but with less discipline, as an illustration
of the superior effectiveness of a bureaucratized army."
38 BUREAUCRASY IN MODERN SOCIETY
The capitalistic system also has furthered the advance

of bureaucracy. The rational esteem ation of economic risks,
which is presupposed in capitalism, requires that the
regular processes of the competitive market not be inter-
rupted by external forces in unpredictable ways. Arbitrary
actions of political tyrants interfere with the rational cal-
culation of gain or loss, and so do banditry, piracy, and
social upheavals. The interest of capitalism demands, there-
fore, not only the overthrow of tyrannical rulers but also
the establishment of governments strong enough to main-
tain order and stability. Note that after the American
Revolution such representatives of the capitalists as Alex-
ander Hamilton advocated a strong federal government,
while representatives of farmers, in the manner of Jeffer-
son, favored a weak central government.
Capitalism then promotes effective and extensive opera-
tions of the government. It also leads to bureaucratization
in other spheres. The expansion of business firms and the
consequent removal of most employees from activities
directly governed by the profit principle make it increas-
ingly necessary to introduce bureaucratic methods of
administration for the sake of efficiency. These giant cor-
porations, in turn, constrain workers, who no longer C311
bargain individually with an employer they know per-
sonally, to organize into large unions with complex
administrative machineries. Strange as it may seem, the
free-enterprise system fosters the development of bureauc-
racy in the government, in private companies, and in
unions.
These historical conditions were not causes of bureauc-
racy in the usual sense of the term. Evidently, a large and
effective army did not cause bureaucracy, on the contrary,
bureaucratic methods of operation produced an effective
large army. The need for these methods, however, arose
in the course of trying to build such an army without them
and helped bring about a bureaucratic form of organiza-

tion. The qualifying word "helped" is essential. If needs
inevitably created ways of meeting them, human society
would be paradise. In this world, wishes are not horses,
and beggars do not ride. Social needs, just as individual
I
ones, often persist without being met. knowledge of the
conditions that end .endered.a-need§gr. .bureaucracy does
not answer the question' what made its development ac-
tually possible under some circumstances and not under
others? The Cavaliers were in need of a better fighting
force, as their defeat demonstrates. Why was it not they
but the Puritans who organized a disciplined army?
In The Protestant Ethic and the Spirit of Capitalism,
Weber indirectly answers this question. He shows that the
Reformation-especially Calvinism, the religious doctrine
of the Puritans-apart from its spiritual significance, had
the social consequence of giving rise to this-worldly
asceticism, a disciplined devotion to hard work in the
pursuit of one's vocation. The Protestant has no Pope or
priest to furnish spiritual guidance and absolve him for
his sins, but must ultimately rely on his own conscience
and faith, this encourages the emergence of self-imposed
discipline. The strong condemnation of pleasure and emo-
tions, exemplified by the Puritan "blue laws," generates
the sobriety and detachment conducive to rational conduct.
Moreover, in contrast to Catholicism and even Lutheran~
ism, Calvinism does not emphasize that the existing order
is God's creation but that it has been corrupted by nan's
sinfulness. Man's. religious duty is not to adapt to this
wicked world, nor to withdraw from it into a monastery,
but to help transform it pro gloriam Dei through methodi-
cal efforts in his everyday life and regular work. The
anxieties aroused by the doctrine of double predestination,
according to which man cannot ajfect his predestined fate
or CVBII. how whether he will be saved or damned, rein-
forced the Calvinist's tendency to adopt a rigorous discif-

pline and immerse himself in his work as a way of relieving
his anxieties.
Protestantism, therefore, has transplanted the ascetic
devotion to disciplined hard work (which must be dis-
tinguished from the exertion of effort as a means for
reaching specific ends) from monastic life, to which it
was largely confined earlier, to the mundane affairs of
economic life. Although the explicit purposes of the Re-
formation were other-worldly and not this»worldly, the
psychological orientation it created had the unanticipated
consequence of helping to revolutionize the secular world.
For without this orientation toward ceaseless effort and
rational conduct as intrinsic moral values, Weber argues
convincingly, capitalism could not have come into exist-
ence, and neither, it should be added, could full-blown
bureaucracy have developed, because it too depends on
rational discipline."
Structural Conditions The historical conditions that

led to the pervasiveness of bureaucracy today do not, of
course, explain why some organizations in contemporary
society are highly bureaucratized and others are not. These
variations raise the problem of the conditions within a
given social structure that give rise to its bureaucratization.
A recent empirical study is concerned with this problem.
Alvin W. Gouldner investigated the process of bureauc-
ratization in a gypsum plant." After the death of the old
manager, the company that owned the plant appointed a
man who had been in charge of one of its smaller factories
as his successor. The new manager, anxious to prove hirin-
self worthy of the promotion by improving productivity,
was faced with special ditliculties. He was not familiar
with the ways of working that had become customary in
this plant, had not established informal relations with
Theory and Deveiopmenf of Bureaucracy 41
his subordinates, and did not command the allegiance of

workers, who still felt loyal to his predecessor. To marshal
the willing support of workers and induce them to identify
with his managerial objectives, he attempted to cultivate
informal relations with them; but this cannot be done
overnight. In the meantime, he found it necessary to dis-
charge his managerial responsibilities by resorting to for-
mal procedures. In the absence of of
___ to keep him informed about the rk
situation, the new manager instituted a system of regular
operational reports for this purpose. Since he did not know
the workers well enough to trust them, he closely checked
O11 their operations ordered bis lieutenants to estab-
lish strict discipline. When some of these lieutenants, used
to the more lenient of the former manager, failed
to adopt rigorous methods of close an
envision, he re-
placed them by outsiders who were more sympathetic with
his disciplinarian approach. These innovations alienated
workers and deepened the gulf between them and the
manager, with the result that he had to rely increasingly
on formal bureaucratic methods of administration.
The role of the successor . . . confronted Peele with
distinctive problems. He had to solve these problems if
he wished to hold his job as manager. In the process of
solving them, the successor was compelled to use
bureaucratic methods. Peele intensified bureaucracy not
merely because he wanted to, not necessarily because he
liked bureaucracy, nor because he valued it above other
techniques, but also because he was constrained to do
so by the tensions of his succession."
In the interest of his objective of gaining control over
the operations in the plant, it was necessary for the suc-
cessor to introduce bureaucratic procedures. At the same
time, for workers to realize their objective of maintaining
some independent control over their own work, it was
4-2 BUREAUCRACY IN. MODERN SOCIETY
necessary for them to oppose the introduction of discipli-

narian measures. As noted above, the existence of a need
does not explain why it is met. In this case, two conflict-
ing needs existed side by side, with the "victor" determined
by the power structure in the organization. The powerful
position of the manager was responsible for his ability to
meet his need by bureaucratizing operations, as indicated
by the following comparison with a situation where he was
not similarly successful.
This plant consisted of a gypsum mine and a wallboard
factory, but the process of bureaucratic formalization was
confined to the factory. Stronger informal ties and more
pronounced group solidarity prevailed among miners than
among factory workers, partly as a consequence of the
common danger to which they were exposed in the mine.
Miners were highly motivated to work hard, and they had
developed their own unofficial system of assigning tasks
among themselves; for instance, new miners had to do the
dirty jobs. Hence there was less need in the mine for
formal discipline and rules prescribing exact duties. Never-
theless, Peele attempted to formalize operating procedures
there, too. The strength of their. informal organization,
however, made it possible for miners, in contrast to fac-
tory workers, effectively to resist these attempts. The
process of bureaucratic for-rnalization generated by suc-
cession in management is not inevitable; collective resist-
ance can arrest it.
The miners, so to speak, had evolved an unofficial bu-
reaucratic apparatus of their OWI1. Their effective informal
organization, by regulating their work, took the place of
a more formal system of control and simultaneously gave
them sufficient power to defeat endeavors to impose a
formal system of discipline upon them against their will.
Did efficiency suffer? Gouldner implies it did not, although
Theory and Devefopmenr of Bureaucracy 43
he does not speciiicay deal with this question. In any

case, the conduct of the miners calls attention, once more,
to the importance of informal relations gnd unofiiciai
practical in bureaucratic structures, which is the topic
of the next chapter.
3
Bureaucracy in Process
A BIIREAUCRACY in operation appears quite different from

the abstract portrayal of its formal structure. Many otlicial
rules are honored in the breach, the members of the or-
ganization act as human beings--often friendly and some-
times annoyed--rather than like dehuxnanized impersonal
machines.
But this contradiction between official requirements and
actual conduct in bureaucracies may be more apparent
than real. Perhaps the violation of some rules is incon~
sequential for the organization and the essential regulations
are regularly obeyed. It is also possible that a detached
attitude is required only in those relationships flat are
involved in the transaction of otlicial business, such as
employee-client or subordinate-superior, and congenial in-
formality is ed to relationships between employees
who work next to one another but not with one another,
such as the members of a stenographic pool. However,
45
even if such clear-cut divisions between formal and in-

formal spheres were always to exist, and this is by HO
means the case, it would still be relevant to inquire whether
informal relations and unofficial practices have any signifi-
cant effects on operations and the achievement of organi-
zational objectives. This is the task of the present chapter.
Bureaucracy's Other Face

Cases from three different kinds of organization have
been selected for presentation. They deal, respectively, with
a military, an industrial, and a civil-service bureaucracy.
All of them reveal "bureaucracy's other face," as Charles
H. Page calls the unofficial activities and interactions that
are so prominent in the daily operations of formal organi-
zations. These concrete cases furnish the basis for a re-
examination of the concept of bureaucratic organization
and its relation to administrative efficiency.
In the IJavy
The existence and importance of the informal struc-
ture of the Navy would hardly be denied by any experi-
enced participant . . . Like the formal, it consists of
rules, groupings, and sanctioned systems of procedure.
They are informal because they are never recorded in
the codes or official blueprints and because they are
generated and maintained with a degree of spontaneity
always lacking in the activities which make up the
formal structure. These rules, groupings, and proce-
dures do, nevertheless, form a structure, for, though not
officially recognized, they are clearly and semiperma-
nendy established, they are just as "real" and just as
compelling on the membership as the elements of the
official structure, and they maintain their eidstence and
social significance throughout many changes of person-
nel. . . .
Bureaucracy in Process 47
[The newcomer] has two large segments d HH - vy

organization to learn. The high-pre§"ure-Tn-§tiil5iion of
the indoctrination school or boot an-In
teacher and his own study of the documents can reveal
vym
the intricacies of the Navy's formal structure. ..
. But
knowledge of the informal structure, which is at least
as necessary for successful participation, must be gained
through experience in the group itself. . ..
Many pressing problems develop within the Navy,
eyflieient solutions for which are not possible within the
framework of the official institutional structure. ...
Such a problem is the constant and, to the initiated,
conspicuous one of oilicial communication between
officers. Official communications in most cases must,
according to regulations, be routed through; II chain
of command" for whatever endorsements thI! Hicers in
the chain judge appropriate. . .
. Yet very frequently
the circumvention of this regulation appears as pre-
cisely the solution of a pressing problem. When such a
development occurs the individuals involved, if they are
sophisticated in the ways of their organization, will
operate on the level of the informal structure wherein
a solution is usually possible, and will thereby avoid
that bureaucratic frustration so frequently felt by those
umullll are strict followers of "the book." ....
_I extreme example, an island air base whose posi-
tion and absence of native popul ation guaranteed almost
no contact with extra-Navy persons, had experienced a
major structural change from the time it had been based
in the United States. In this case the informal structure
had almost altogether lost its private sanctification and
stood, in large measure, as the officially recognized pat-
tem of this group of temporat'y.island residents. One
visiting oflieer described this as a "breakdown" of the
organization. This was clearly not the case, as shown
by the high morale and the effective accomplishment of
missions. What had "broken down" was a large part of
the formal structure, or rather it had been submerged
48 BUREAUCRMCY IN MODERN SOCIETY
as the informal structure rose into overt recognition and

use. Fortunately the "skipper" as well as several other
officers and petty officers were "natural leaders": their
status and role definitions were somewhat parallel in
the two structures. However, unmistakable indications
of the superordination of the informal included the re-
placement of the social isolation of the commanding
oliicer by his very keen participation in all activities of
the unit, the submergence of the rejected types whatever
their rank or rate to the informally deNned roles, the
emergence of the natural leaders to what amounted to
official recognition, the abandonment of most of the
officially governing protocol (except in the treatment of
visitors), and accomplishment of the day-to-day and
long-run tasks with efficiency, zeal, and spontaneous
initiative not characteristic of official bureaucratic ma-
chinery.1
In a Factory
Of the fourteen men, or operators, as they were called
in the Western Electric Company, who were regularly
in the Observation Room, nine were wiremen, ...
. .
three were soldernten, . and two were inspectors.
... The men were engaged in making parts of switches
for central office telephone equipment. Specically, they
were connecting wires tO banks of terminals. . . A .
Wireman took the necessary number of banks for an
equipment and placed them in a holder or mixture on a
workbench. Then he connected the terminals of the
banks together in a certain order with wire. ..
.A
wireman worked on two equipments at a time. Having
finished a level on one equipment, he moved to the
second equipment. In the meantime, a solderrnan fixed
in place the Finished connections of the first equipment,
and an inspector tested and scrutinized the work of both
men. ...
Let us now turn to some of the activities, over and
above each man's special job, that were observed in the
room. One of the commonest was helping another man

O t by doing some of his wiring for him when he had
fallen behind. Although no formal rule of the company
said that one man should not help another, helping was
in practice forbidden, OH the theory that the jobs were
e-man j bs a d that each man could do his own best.
Nevertheless a good deal of help was given. The wire
men said it made them feel good to be helped. ..
Everyone took part in helping. Unlike some other ac
.
ii ties, it was not confined to one social group. . .
In the lunch hour an-El-fiom time to time during the
work, the men in the room took part in all sorts of
games. Almost anything was an excuse for a bet'
matching coins, lagging coins, shooting craps, cards,
combinations of digits in the serial numbers of weekly
pay checks. Pools were organized on horse racing, base-
. .
Ii and quality records. . Participation in games
.
occurred for the most part within two groups . . , a
group at the front of the room . . . [and] a group at
the back. . .
» The material collected by the observer
could also be interpreted to show that friendships or
antagonisms existed between certain men in the room
. . . Except for a friendship between [two men], all
friendships occurred within one or the other of the two
groups already mapped out on the basis of participation
in games. ...
Roethlisberger and Dickson sum up all this evidence
by saying that, although the members of the Bank Wir-
ing Observation Room were pulled together in some
ways, for instance, in mutual help and in restriction of
output, in others they were divided. In particular, there
were two cliques in the room, whose membership was
approximately that revealed by participation in games
..
. [But three men] were in no sense members of
either clique, [two of them] attracting much antago~
nism. Each clique had its own games and activities
noticeably different from those of the other group. .
If, as we have seen, in output rates of the wiremen
could not be correlated with their intelligence or dex-

terity, they could clearly be correlated with clique mem-
bership. . . . [The members of one clique] had the
lowest 0utPI_1t.2
In a Federal Law-Enforcement Agency
The principal duties of agents were carried out in the
field. Cases of firms to be investigated were assigned to
them indivi dually by the supervisor. Processing a case
involved an audit of the books and records of the Brin,
interviews with the employer (or his representative)
and a sample of employees, the determination of the
existence of legal violation and the appropriate action
to be taken, and negotiations with employers. . If . .
an agent encountered a problem he could not solve, he
was expected to consult his supervisor, who, if he could
not furnish the requested advice himself, gave the agent
permission to consult a staff attorney. Agents were not
allowed to consult anyone else directly, not even their
colleagues. . ..
Agents, however, were reluctant to reveal to their
supervisor their inability to solve a problem for fear
that their ratings would be adversely affected. ...
Their need for getting advice without exposing their
difficulties to the supervisor constrained agents to c011-
sult one another, in violation of the oliicial rule. ...
A consultation can be considered an exchange of
values; both participants gain something, and both have
to pay a price. The questioning agent is enabled to per-
form better than he could otherwise have done, without
exposing his difficulties to the supervisor. By asking for
advice, he implicitly pays his respect to the superior
proficiency of his colleague. This acknowledgement of
inferiority is the cost of receiving assistance. The con-
sultant gains prestige, in return for which he is willing
to devote some time to the consultation and permit it to
disrupt his own work. The following remark of an agent
illustrates this: "I like giving advice. It's iiattering, I
Rureaucracy in Process 51
suppose, if you feel that the others come to you for

advice."
The expert whose advice was often sought by col-
leagues obtained social evidence of his superior abilities.
This increased his confidence in his own decisions, and
thus improved his performance as an investigator. . ..
The role of the agent who frequently solicited advice
was less enviable, even though he benefited most directly
from this unofficial practice. Asking a colleague for
guidance was less threatening than asking the super-
visor, but the repeated admission of his inability to
solve his own problems also undermined the se1f-conli-
dence of an agent and his standing in the group. The
a m , of advice became prohibitive if the consultant,
after the questioner had subordinated himself by asking
for help, was Nam least discouraging--by postponing
a discussion or by revealing his impatience during one.
W
mm. avoid such rejections, agents usually consulted a
colleague with whom they were friendly, even if he was
not an expert. an in go,
An agent who worked on an interesting case and
encountered strange problems often told his fellow
agents about it. . .. These presentations of complex
cases assisted the speaker in solving his problems. They
were consultations in disguise. . . . The agent who at-
tempted to arrive at decisions while sitting alone at his
desk defined the situation as preparing the case for sub-
mission to the supervisor. His anxiety, engendered by
an supervisor's iv aluation of his decisions, interfered
most with clear thinking in this situation. Instead of
trying to make important official decisions, an agent
could discuss the interesting aspects of his case with
one of his colleagues. This situation, defined as a dis-
cussion among friends, did not evoke anxiety. On the
contrary, it destroyed the anidety which pervaded the
decision-making process. __._
The listener was not merely a friend it a fellow
specialist in solving the problems which occurred in
investigations. This created the possibility of intenup-

tion, if the suggested interpretation required correction.
A listener might remind the speaker that he forgot to
take some factor into account, or that the data lend
themselves to alternative conclusions. The assent implicit
in the absence of interruptions and in attentive listening
destroyed the doubts that continuously arose in the
process of making many minor decisions in order to
arrive at a conclusion. The admiration for the clever
solution of the problem advanced, expressed by inter-
ested questions and appreciative continents, increased
the speaker's confidence in his partial solutions while
groping for the final one. By reducing his anxiety,
"thinking out loud" enabled an official to associate
relevant pieces of information and pertinent regulations,
and thus to arrive at decisions of which he might not
have thought while alone- . . .
[This pattern of explicit and disguised consultations]
transformed an aggregate of individuals who happened
to have the same supervisor into a cohesive group. The
recurrent experience of being dependent O11 the group,
whose members furnished needed help, and of being
appreciated bY the others in the group, as indicated by
their solicitations for assistance, created strong mutual
bonds. ... Second, this practice contributed to operat-
ing eliiciency, because it improved the quality of the
decisions of agents. Every agent knew that he could
obtain help with solving problems whenever he needed
it. This knowledge, reinforced by the feeling of being
an integrated member of a cohesive group, decreased
anxiety about making decisions. Simultaneously, being
often approached for advice raised the self-confidence
of an investigator. The very existence of this practice
enhanced the ability of all agents, experts as well as
others, to make decisions iudependendy.3
Organization of Work Groups

When we examine sufficiently small segments of bu-
reaucracies to observe their operations in detail, we dis-
cover patterns of activities and interactions that cannot be
accounted for by the oliicial structure. Whether the work
group is part of the armed forces, a factory, or civilian
government, it is characterized by a network of informal
relations and a set of unofficial practices which have been
called its "informal organization." This concept calls atten-
tion to the fact that deviations from the formal blueprint
are socially organized patterns and not merely the con-
sequence of fortuitous personality differences. Helping
others or playing games was the established practice in
the Bank Wiring Observation Room, not a manifestation
of the rebellious personality of one or the other individual.
Variations in productivity were not due to the fact that
the mechanical ability of some workers happened to be
superior to that of others but to the social organization
of the group, as indicated by the finding that productivity
was related to clique membership and not to manual dex-
terity or intelligence.
Regularities do not occur accidentally. That oilicial rules
bring them about is expected, but what is the source of
those regularities in social conduct that do not reflect
official standards? They are also the result of normative
standards, but standards that have emerged in the work
group itself rather than having been officially instituted by
superiors or formal blueprints. In the course of social inter-
action at work, there arise patterned expectations and
norms, which find expression in a network of social rela-
tionships and in prevailing practices. As each worker in the
Bank Wiring Observation Room grew accustomed to play-
ing games with some coworkers and not with others, his
role became socially defined as part of one of the two
cliques, or of neither, and the group became structured ac-
cordingly. Simultaneously, there developed normative be-
liefs shared by all members of the group: "Don't be a rate»
buster by working too fast" "Don't be a chiseler by work-
ing too slowly" "Don't act bossy" "Dorl't be a squealer"
These unofficial standards governed the behavior of the
workers. One inspector was excluded from both cliques,
because he acted officiously and even reported other
workers to superiors. Despite a complicated wage incentive
system, there existed some restriction of output. Since too
fast as well as too slow work was condemned, wirernen
did not try to produce as much as they could, although
doing so would have increased their pay, but slacked
their pace after they had completed what they considered
to be "a fair day's work." Differences in output within
this group would probably have been smaller if it had not
been divided into two cliques. One of these emphasized
that the worker should not produce too much, and the
other, that he should not produce too little, so that the
members of each clique enjoyed social support for work-
ing slower or faster, respectively, than those of the other.
To be effective, social norms must be enforceable. Un-
less a member of a formal organization conforms with its
official regulations to a certain minimum degree, he will be
expelled. The reverse of this statement is also true: unless
expulsion is a serious threat, the prevalence of conformity
cannot be assured. The individual's motivation to remain
part of the organization makes him subject to its control.
Salaried employees are more dependable than unpaid vol-
unteers in large part because economic dependence is a
reliable mechanism for interesting the members of the
organization in keeping their positions. The same principle
holds for the enforcement of unofficial norms. Whereas
the work group does not have the power to remove one
of its members from his job and deprive him of his in-
come, it can. ostracize him and thereby exclude him from
genuine group membership. But for such exclusion to be
a threat that discourages deviant tendencies, the individual
must first wish to be included in the group. If a person
did not care about maintaining congenial relations with
his coworkers, being cold-shouldered by them would
neither disconcert him nor deter him from 'disregarding
their social norms, and for him it would be "their" norms
rather than "ours."
This is the reason why the existence of social cohesion
is SO significant for work groups. Strong mutual ties be~
tween the members of a group make each interested in
maintaining his position in the group. In, In: situation,.__
unofficial norms can readily be enforced, and it is rarely

necessary to resort to the extreme penalty of ostracism,
since lesser sanctions suffice to sustain conformity. arm
individual violates a norm highly valued by the other mem-
bers of the group, they will become less friendly toward
him, this is virtually an automatic reaction when some-
body's behavior displeases us. Such a change in interpersonal
relationships endangers the individual's standing in the
group and induces him, if he is identified with the group,
to refrain from similar violations in the future in order
to regain the favor of his colleagues or, at least, to prevent
his relations with them from deteriorating further. An-
other type of informal sanction can be termed "ostraciSII1
in miniature." When several members of a group together
no
ridic colleague or Xpress aggression against him in
sorrel other form because he has violated an unofiicid
noxrq, they furnish him with a brief but concentrated
demoNstration of the lature of ostracism by standing
united in opposition to him alone. The extremely disagree-
able experience of feeling isolated while witnessin_g the
$5 BUREAUCRACY IN MODERN SOCIETY
solidarity of others constitutes a powerful incentive to

abandon deviant practices lest Uris temporary state become
a permanent one.
The effective enforcement of unotiicial standards of
conduct in cohesive work groups has important- implica-
sons for official operations. Many studies have found
that the existence of cohesive bonds between coworkers is
a prerequisite for high morale and optimum performance
of duties,<1 but this does not mean that all norms that arise
in cohesive work groups contribute to the accomplishment
of official tasks. The group's own standards in the Bank
Wiring Observation Room, for example, since they dis-
couraged the fastest workers from producing at~a maxi-
mum rate, lowered productivity (although these standards
simultaneously encouraged the slowest workers to increase
their output). On the other hand, the fact that an unoiiicial
practice directly oonfticts with official regulations does not
necessarily signify that it is detrimental to operating ef-
ticiency. The practice of consulting colleagues in violation
of an otlicial rule in the government agency apparently
improved e\ciency in operations, and so did the informal
patterns on the island air base that defied the Navy's for-
mal codes.
Paradoxically, unofficial practices that are explicitly
prohibited by official regulations sometimes further the
achievement of organizational objectives. This crucial End-
ing raises questions about the concept of "informal organi-
zation" and about bureaucratic efficiency. Social scientists
often set up a dichotomy between the informal and the
formal organization and attempt to place every observe
son into one of these pigeonholes. This procedure can
only be misleading, since the distinction is an analytical
one; there is only one actual organization. When govern-
ment agents make official decisions in the course of in-
formal discussions, their conduct cannot meaningfully be
classified as belonging to either the formal or the informal

organization. Even when a factory employee worked more
slowly than he otherwise might have ill conformity with
unolicial norms, his behavior was also influenced by the
formal requirements to manufacture telephone equipment
and to use certain production methods for this purpose.
Official as well as unoilicial stander , formal as well as
informal social relations, affect the ways in which the daily
operations in work groups become organized, but the
result is one social organization in each work group, not
two.
Other problems posed by the finding that operating
efficiency is sometimes increased by violating the very
rules designed to mazdmize efficiency call for a more
detailed discussion.
Bureclucrucy's New Face
Bureaucracies are not such rigid structures as is popu-

larly assumed. Their organization does not remain fixed
according to the formal blueprint, but always evolves into
I1€W sorts. Conditions change, problems arise, and, in
the course of coping with them, the members of the organic
zation establish new procedures and often transform their
social relationships, thereby modifying the structure. The
organized patterns of activities and interactions that have
not-perhaps, not yet-been officially institutionalized
reveal bureaucracy in the process of change.
Some of the practices that emerge in the course of
operations further the attainment of organizational objec-
tives, while others hinder it. The official interest of the
bureaucratic organization demands that the latter develop-
ment be discouraged and the former encouraged. The ad-
ministrative problem is how to bring this about.
Irrationality of Ra tionalfstic Administration "Scientific

management" has attempted to rationalize industrial pro-
duction and administration by discovering and applying
the most efficient methods of operationsi' Time-and-motion
studies are a well-known illustration of this approach: the
motions required by the most skilled workers for perform-
ing a given task in the shortest possible time are deter-
mined, and these exact motions are taught to other work-
ers. But, as one writer points out, "managerial technologists
have been far more successful in demonstrating efficient
procedures for maximum productivity than they have been
in getting such procedures accepted by workers."6 This
failure of scientific management was the inevitable result
of its assumption, most evident in "scientific" wage incen-
tive systems, that rational economic interests alone govern
the conduct of employees and of its neglect of social fac-
tors. To administer a social organization according to
purely technical criteria of rationality is irrational, because
it ignores the notational aspects of social conduct.
From an abstract standpoint, the most rational method
of effecting uniformity and coordination in a large organi-
zation would appear to be to devise efficient procedures
for every task and insist that they be strictly followed. In
practice, however, such a system would not function ef-
fectively for several reasons. One is that it implicitly
assumes that management is omniscient. No system of
rules and supervision can be so finely spun that it antici-
pates all exigencies that may arise. Changes in external
conditions create HGW administrative problems, and the
very innovations introduced to solve them often have
unanticipated consequences that produce further problems.
For example, the interviewers in a public employment
agency were evaluated on the basis of the number of
applicants for jobs they interviewed per month. As jobs
became scarce after World War II, interviewers, induced
by this method of evaluation to work fast, tended to dis-

miss clients for whom jobs could not be located quickly.
In the interest of effective employment service, it was
necessary to discourage such tendencies. For this purpose,
e new method of evaluation, based primarily OI] the num-
of applicants placed in jobs, was instituted. This
innovation M motivate . erviewers to exert greater
edorts to find jobs for clients, but it also gave rise to com-
petition for the slips on which job openings were recorded,
which interviewers sometimes even hid from one another,
and these competitive practices were a new obstacle to
efficient operations. In response to this emergent problem,
the most cohesive group of interviewers developed co-
operative norms and successfully suppressed competitive
tendencies, with the result that productive eliiciency in-
creased." Unless the members of the organization have
the freedom and initiative to deal with operating problems
as they come up, efficiency will suffer.
Moreover, some impediments to operating efficiency
cannot be eradicated by oliicial decree. This is the case
with respect to the anxieties and feelings of anomie (a state
of feeling isolated and disoriented) that often arise among
the lower echelons of bureaucratic hierarchies. Informal
relations in cohesive work groups reduce such disruptive
tensions. But once cohesive groups exist in the bureauc-
racy, as we have seen, they will develop their own stand-
ards of conduct and enforce them among their members.
Administrative efficiency cannot be served by ignoring the
fact that the performance of individuals is affected by their
relations with colleagues, but only by taking cognizance
of this fact and attempting to create those conditions in
the organization that lead to unofficial practices which
further rather than hinder the achievement of its objec-
tives.
Finally, in a democratic culture, where independence of
action and equality of status are highly valued, detailed

rules and close supervision are resented, and resentful
employees are poorly motivated to perform their duties
faithfully and energetically. A striking contrast exists
between the rigorous discipline employees willingly impose
upon themselves, because they realize that their work
requires strict operating standards, and the constant an-
noyance at being hamstrung by picayune rules that they
experience as arbitrarily imposed upon them. The members
of the federal agency, for instance, often objected to
having to till out forms precisely, as we all do, and to
other minor internal rules, but they freely accepted the
much more stringent discipline of adhering strictly to legal
regulations in their investigations, which was necessitated
by law enforcement itself. To repress the ability for self-
imposed discipline and to underrnine the motivation to
exert efforts by prescribing in detail how every task is to
be performed is wasteful, to say the least. A more efficient
method of bureaucratic administration is to channel this
ability and motivation to serve the ends of the organiza-
tion.
These considerations suggest a revision of the concept
of bureaucratic structure. Rather than considering it an
administrative system with particular characteristics, it
may be preferable to follow another lead of Weber's and
to conceive of bureaucracy in terms of its purpose. Bu-
reaucracy, then, can be defined as organization that maxi-
mizes efficiency in administration, whatever its formal
characteristics, or as an institutionalized method of or-
ganizing social conduct in the interest of? administrative
etiiciency. On the basis of this definition, the problem of
central concern is the expeditious removal of the obstacles
to efficient operations which recurrently arise. This cannot
be accomplished by a preconceived system of rigid pro-
cedures, as the preceding discussion suggests, but only by
creating conditions favorable to continuous adjustive devel-

opment in the organization. To establish such a pattern
of self-adlustment in a bureaucracy, conditions must pre-
vail that encourage its members to Q e with emergent
problems "Io and the best method for producing
specified results on their own initiative, and that obviate
the need for unofficial practices which thwart the objec-
tives of the organization, such as restriction of output.
What are these conditions? We do not have sutcient
empirical evidence to give a conclusive answer to this
question. But some tentative hypotheses can be advanced,
although these must be qualified by the recognition that
the same conditions may not be required for adjustive
development in other cultures or in other historical periods.
Conditions of Adjustive Development

1. Employment security For the members of an organ-
ization to assume responsibility for finding 116W ways of
solving problems, they must have some employment secur-
ity. It is often held that only the danger of losing one's
job stimulates initiative and that security kills it, but this
view seems to be fallacious. Say that your job would not be
affected OI1€ way or the other, would YOU, personally, rather
follow simple routines or assume challenging responsibilities
in your work? Most people prefer the latter and will exer-
cise initiative if they are given a chance. The insecurity en-
gendered by the knowledge that his job hangs in the bal-
ance, however, constrains the employee to adhere closely
to familiar and officially sanctioned procedures and to
avoid taking risks. It therefore destroys his initiative, since
taking the initiative always involves the risk of possible
failure.
The situation in the employment agency referred to
above illustrates this principle. When a new evaluation
system was introduced, it will be remembered, competitive
62 BITREAUCRACY :n MODERN SOCIETY
coniiicts were effectively eliminated in one group but

persisted in another. A main reason for this difference was
that the interviewers in the first group held secure civil
service positions, whereas most interviewers in the second
group did not; they were on probation pending permanent
appointment at that time, because only temporary appoint-
ments had been made during the war. The insecure mem-
bers of the latter group were so anxious to comply with
oihcial demands that they dared not initiate cooperative
practices on their own. The secure members of the former
group, in contrast, felt free to institute cooperative prac-
tices, which resulted in higher productivity. Insecurity
generates rigidity and resistance to change, as will be
shown at greater length in Chapter Five.
2. Internalized standards of workmanship Employees
must feel free to exercise initiative, but they must also feel
constrained by strict operating principles in doing so, lest
their spontaneous actions interfere with the attainment of
organizational objectives. External restraints, such as de-
tailed rules of operations, are not well suited for this
purpose, since they tend to eliminate discretion com-
pletely. If a person, on the other hand, has fully internal-
ized rigorous standards of workmanship, he can exercise his
ingenuity while remaining guided by them; indeed, he finds
gratification in doing precisely that. This is true of the sur-
geon, who applies his talent in new ways in virtually every
operation without deviating from exacting Medical stand-
ards, of the scientist, whose search for new explanations is
governed by disciplined scientific methods, of the artist,
the expert mechanic, and, as a matter of fact, of every
skilled craftsman who is interested in competent perform-
ance of tasks in accordance with abstract principles com-
rnonly agreed upon among the members of the occupa-
tional group. Identification with abstract standards of
performance permits disciplined discretion in finding new
solutions to problems that conform with these standards.

Furthermore, it makes this course of action intrinsically
satisfying and thus supplies strong incentives for exerting
efforts in one's work.
The prevalence of such a workmanlike or professional
orientation among the members of an organization de-
pends in part on certain employment and working Condi
sons. Personnel policies must insure, as they usually do in
bureaucracies, that only employees with adequate training
and technical qualifications are hired, and that their
careers are relatively secure. In addition,. however, employ-
ees must be made responsible for the performance of
challenging tasks, not obligated to follow rigid routines
and they, collectivel3l, must have a voice in the determine
dion of the standards of workmanship that govern their
work. It can hardly be an accident that the emergence of
such standards has been most pronounced among inde-
pendent professionals and free artisans. The occupational
group that develops its own discipline can most readily
enforce it. Finally, it is unlikely that this type of orienta-
tion will be characteristic of most workers in industrial
concerns and white~collar offices until technological prog
less has eliminated the most routine tasks, which cannot
be expected to arouse a pride of workmanship.
3. Cohesive work groups. The dilettante, who has no
definite responsibilities, and the human robot, whose duties
are rigidly fixed, have, for opposite reasons, no cause for
anxiety. But responsibility for the effective performance
of complex tasks tends to engender anxiety. In the federal
agency of law enforcement, for example, officials were
held responsible for making accurate decisions in diilicult
investigations, which required ingenuity, since the diversity
of situations prevented strict adherence to exact rules-
Their freedom to exercise initiative in the process of arrive
ing at correct conclusions made their job interesting, but
64 BUREAUCRACY IN MODERN SoCIETY
it also evoked anxieties over decision-making, which inter-

fered with clear thinking and effective operations. As in
this case, integrative interaction in cohesive work groups
generally relieves disruptive anxieties and tensions and
often leads to common new practices that contribute to
operating efficiency.
Although social cohesion, another prerequisite of recur-
rent self~adjusfment in the organization, cannot be oNi-
cially created, conditions favorable for its development
can be. Low turn-over of personnel and infrequent trans-
fers within the organization establish the stability of mem-
bership conducive to the formation of mutual tics in work
groups. In the absence of explicit personnel policies, how-
ever, employees may expect to be dismissed or promoted
at any time, and hence they are apt to worry about their
careers and to feel the need to impress their superiors at
all costs. This need is likely to lead to competitive rivalry,
which destroys cohesiveness. When promotions and neces-
sary dismissals depend on explicit and openly announced
staNdards, on the other hand, employees are able to pre-
dict their career chances with relative accuracy. Most of
them, if not all, know that they will not be dismissed ex-
cept for specified misconduct, and whether there is any
likelihood of their being promoted in the near future, as
well as the technical criteria that govern advancement.
Employees secure in this knowledge have little inclination
to endanger their informal relations with colleagues
through competitive practices. Explicit personnel policies,
which are being adopted by an increasing number of
private as well as public bureaucracies, therefore promote
social cohesion. .
4. Split in managerial authority Cohesiveness empowers
work groups to institute COl'I1I'I10l1 adjustive practices, hut
these will not advance the organization's objectives if em-
ployees feel that their interest conflicts with that of man-

agement. Restriction of output among factory workers, for
example, is designed to protect their economic interest
against management by reducing the chances of being
laid oil. (The study of the Western Electric Company was
conducted during the depression of the 1930s, when the
danger of layoffs due to lack of work was very great in-
deed.) If employees fear that optimum performance will
put some of them out of their jobs, their collective en-
deavors will hinder rather than further it, since the domi-
nant concern with earning a livelihood will in all likeli-
hood override pride of workmanship. The basic source of
this fear is the conflict between the employer's interest in
reducing cost and that of employees in their income. This
is a real conflict of interest, which cannot be talked out
of existence by good labor-managem q relations, ,Maui
which would probably persist even in a socialist economy,
but its detrimental eiiects on operation; c a n e avoided
this writer believes, by a split in managerial authority.
Such a split in authority exists in civil service. Manage-
ment in government agencies, in contrast to management
in private industry, controls only operations
.
and not em-
ployment conditions. Salaries the procedures that
govern promotion and discharge are determined by civil~
to
service commissio h accord-ance with legal statutes. In-
dividual government officials, just as private employees,
occasionally come into conflict with management in the
course of operations. Man conflict Between the collective
economic interest of operating oilicials and the budgetary
considerations of their employer, however, finds expression
in their opposition to the civil-service commission and the
legislature, which set the conditions of their employment.
This conflict does not harm their relationship with the
agency's administration. On the contrary, operating offi-
cials and administrators are united by a common interest

in legislation that benefits civil servants. Employees who
have no reason to protect their economic welfare against
the management of their own organization are more apt
to maintain a professional or workmanlike concern with
perfecting methods of operations and thus to contribute to
continuous adjustive development in the organization.
5. Evaluation on the basis of clearly specilted results
The standardization required for uniform and coordinated
performance would seem to preclude the exercise of initia-
tive which has been so much emphasized in the preceding
discussion. To be sure, this is the case when operating
procedures are standardized in minute detail, but there is
no need for such close regulations unless the members of
the organization are not qualified to perform responsible
tasks. If they are qualified, it is probably enough to stand-
ardize the end-products of their operations and not the
precise ways they arrive at them. Evaluation on the basis
of clearly specified results, which employees are expected
to accomplish in their work, encourages ingenuity and
simultaneously assures the standardization necessary for
effective bureaucratic operation.
The Task of the Administrator These five conditions

and, at least, one other, which will be discussed in the
following chapter, characterize bureaucracy's new face.
Once they are met, needed adjustments occur quite spon-
taneously, as it were, within the organization. This self-
adjustment largely relieves the administrator of the duty
of coping with emergent operating problems himself, giv-
ing him more time for discharging other responsibilities.
Bureaucratic processes continually endanger the conditions
for optimum performance, necessary reductions in staff
give rise to feelings of insecurity despite explicit personnel
Bureaucracy in Process 6'7
policies; the anxieties engendered by evaluation 011 the basis

of results may not be relieved by social cohesion; In d so
forth. The main task of the new administrator is to keep
vigilant watch over these conditions of adjustive develop-
ment, which are perpetually threatened, but without w`l':'iI&l1,
if the hypotheses advanced here are correct, efficiency in
the bureaucracy suffers.
Bureaucratic Authority
THE IHERARCHY of authority in a bureaucracy, essential

for coordination, often produces among its lower echelons
profound feeling; of. ineclua-ITty and apathy that impede
identification with the organization's objectives. The initia-
tion of needed adjustments by the operating members of
the organization presupposes, in addition to the give con-
ditions already discussed, a method of hierarchical coor-
dination that minimizes these harmful consequences for
work motivation. After analyzing the ways in which
bureaucratic authority is exercised, we shall return to this
problem at the end of this chapter.
To start with, let us consider another paradox between
official requirements and actual practica. In theory,
reaucratic superiors are expected to exert strict and imper-
sonal control over subordinates. But in fact, immediate
supervisors and foremen frequently "play bail" with their
69
70 BUREAUCRACY IN. MODERN SOCIETY
subordinates and let them "get away with" infractions of

many rules. What accounts for this leniency?
Strategic Leniency and Authority

A psychological explanation of the failure to enforce
strict discipline among subordinates might attribute it to
poor leadership. Some supervisors are overly lenient, it
could be held, because inborn or acquired personality
traits prevent them from asserting their authority over
others and maintaining effective leadership. Note that this
explanation assumes as a matter of course that the bureau-
cratic superior who appears lenient merely indulges his
subordinates and is less effective than the disciplinarian in
discharging his supervisory responsibilities. Empirical evi-
dence, however, indicates that the very opposite is the
case.
A study of twenty-four clerical sections in an insurance
company analyzed the relationship between method of su-
pervision and productive etiiciencyl In closely supervised
1 sections, whose heads gave clerks detailed instructions and
frequently checked up on them, productivity was usually
lower than in sections where employees were given more
freedom to do the work in their OWI1 way. Moreover, su-
pervisors who were primarily concerned with maintaining
a high level of production, interestingly enough, were less
successful in meeting this goal than those supervisors who
were more interested in the welfare of their subordinates
than in sheer production; in the latter case, productivity
was generally higher. Finally, groups who worked under
more authoritarian supervisors were, on the whole, less
productive than those supervised in a relatively democratic
fashion. Other studies have also found that disciplinarian
supervisors are less effective than more liberal ones."
Such findings are often misinterpreted as signifying that
Bureaucratic Authority 71
democratic ways are superior to authoritarian ones. But

this is a rather loose use of the term "democratic," the
exact meaning of which is worth preserving. Since "de-
mocracy" denotes rule from below (literally, "people's
rule") and not from above, one person's supervision of
others can, by definition, not be democratic. This is not
the place for a discussion of the relation between de-
mocracy and bureaucracy, the final chapter is reserved
for this purpose. But here it should be noted that tolerant
supervisory practices, in contrast to disciplinarian ones, are
neither democratic nor an indication that controlling
power over subordinates has been surrendered. On the
contrary, leniency in supervision is a potent strategy, con-
sciously or unconsciously employed, for establishing au-
thority over subordinates, and this is why the liberal
supervisor is particularly effective.
Let us clarify the concept of authority. First, it refers
to a relationship between persons and not to an attribute
of one individual. Second, authority involves exercise of
social control which rests on the willing compliance of
subordinates with certain directives of the superior. He
need not coerce or persuade subordinates in order IO ill-°
iiuence them, because they have accepted as legitimate the
principle that some of their actions should be governed
by his decisions. Third, authority is an observable pattern
of interaction and not an official definition of a social
_
relationship. If a mutinous crew refuses to obey the cap-
tain's orders, he does not in fact have authority over his
men. Whatever the superio as official rights to command
obedience and the subordinates' official duties ."'"
him, his authority over them extends only to conduct that

they voluntarily permit to be governed by his directives.
Actual authority, consequently, is mu granted by
formal organizational chart, but must be established in
the course of social interaction, although the official bu-
reaucratic structure, as we shall see presently, facilitates its

establishment.
What are some of the practices of a lenient foreman
or supervisor? Above all, he allows subordinates to violate
minor rules, to smoke or talk, for example, despite the
fact that it is prohibited by management. This permissive-
ness often increases his power over them by furnishing
him with legitimate sanctions that he can use as he sees lit.
If an action of his subordinates displeases him, the super-
viser can punish them by commanding: "Cut out the
smoking! Can't you read the sign'?" Had he always en-
forced the rule, this penalty would not have been available
to him. Indeed, so crude a use of sanctions is rarely neces-
sary. The mere knowledge that the rule exists and, possi-
bly, that it is enforced elsewhere, instills a sense of obliga-
tion to liberal superiors and induces subordinates more
readily to comply with their requests.
Whereas the disciplinarian supervisor generally asserts
his official prerogatives, the lenient and relaxed one does
not. The latter attempts to take the wishes of his subordi-
nates into account in arranging their work schedule, al-
though he has the right to assign their work at his own
discretion. Sometimes he goes to special trouble to ac-
commodate a subordinate. Instead of issuing curt com-
snands, he usually explains the reasons for his directives.
He calls his subordinates by their first names and en-
courages their use of his first name (especially in demo-
cratically minded American organizations). When one of
his subordinates gets into difficulties with management,
he is apt to speak up for him and to defend him. These
different actions have two things in common: the superior
is not required to do them, and his subordinates greatly
welcome his doing them. Such conduct therefore creates
social obligations. To repay the supervisor for past favors,
and not to risk the cessation of similar favors in the fu-
tube, subordinates voluntarily comply with many Of Es

requests, including some they are not officially-required
to obey. By refraining from exercising his power of COD-
trol whenever it is legitimate to do so, the bureaucratic
superior establishes effective authority over subordinates,
which enables him to control them much more effectively
than otherwise would be possible.
Complementary role expectations arise in the course of
mferaction
' between superior and subordinates and become
crystallized in the course of interaction among subordi-
nates. As the superior permits subordinates to violate
some rules and to make certain decisions themselves, and
as they grow accustomed to condo with many of his
directives, they learn to- expect to exercise discretion in
some areas and to follow supervisory directives in others,
and he learns to expect this pattern of conduct from them.
The members of the work group, by watching one another
at work and talking among themselves about the manner
in which they perform their duties, develop social con-
sensus about these role expectations and thereby reinforce
them. The newcomer to the group, who must be taught
"how things are done around here" as distinguished from
"what's in the book," provides an opportunity for further
affirming this consensus by making it explicit.
The resulting common role expectations are often so
fully internalized that employees are hardly aware of being
governed by them. The members of one department might
find it natural for their supervisor to interrupt their work
and tell them to start on a new task. The members of an-
other department in the same organization might consider
such a supervisory order as gross interference with their
work, since they had become accustomed to using their
discretion about the sequence of their tasks, yet readily
comply with other directives of the supervision. These role
expectations of independence from the supervisor in some
areas and unquestioning obedience in others define the

limits of his authority over subordinates.
Power of Sanction
The preceding comments apply to informal leadership
as well as to burea socratic authority. The informal leader,
like the prudent bureaucratic superior, establishes his au-
thority over his followers by creating social obligations?
Once a relationship of authority exists, both bureaucratic
superior and informal leader can alford to word their
orders as mere suggestions, because even these are readily
followed by the rest of the group. Neither of them usually
needs sanctions to command obedience, though sanctions
are available to both of them in case they wish to use
special inducements, since praise or blame of the person
in the superordinate position itself exerts a powerful injiu-
ence.
Nevertheless, there is a fundamental distinction between
informal leadership and bureaucratic authority. Informal
leadership freely emerges among a group of peers. It is
initially the result of personality differences that have be-
come socially magnified. Some members of the group ex-
cel in activities that are highly valued by all, whether
these are street fighting or solving complex problems;
these few will be more respected, and their opinions will
carry greater weight. The person in the extreme position,
if he also finds ways to obligate the others to him, is ex-
pected to be the group's leader.
Bureaucratic authority, on the other hand, prevents the
group itself from conferring the position of leadership
upon the member of their choice. The voluntary obedience
of subordinates must converge upon the individual offli-
cially placed in the position of supervisor, irrespective of
his personal characteristics. The bureaucratic mechanism
Bureaucrazie Authority 75
that makes this state of affairs a predictable occurrence is

the superior's power to impose sanctions, typically in the
form of periodic ratings of the performance of his sub-
ordinates, which iniiuence their chances of advancement
and of keeping their jobs.
The dependency of bureaucratic subordinates upon their
immediate superior produced by his rating power engen-
ders frustrations and anxieties for adults. It forces employ-
ees to worry about their supervisor's reaction at every step
they take. An effective way to weaken or avoid such feel-
ings is to identify with the bureaucratic system of norma-
tive standards and objectives. By making this system a part
of their own thinking, employees transform conformity
with its principles from submission to the superior's de-
mands into voluntary action. Guided by internalized stand-
ards, they are less likely to experience external restraints
in performing their duties. Moreover, once the hierarchical
division of responsibility has been accepted as a basic
principle of the organization, it becomes less threatening
to a person's self-esteem to obey the supervisor's directives,
since he is known to be duty-bound to issue them, just as
it is not degrading to obey the tragic directions of a police-
man. Dependence on the superior's rating encourages the
adoption of a bureaucratic orientation, for the disadvan-
tages of dependence can thereby be evaded.
It is of crucial importance that this process of identifica-
tion with bureaucratic standards does not occur in isola-
tion but in a social situation. All members of the work
group find themselves in the same position of dependence
on their supervisor. (In fact, all members of the bureau-
cratic organization are, in varying degrees, dependent on
their immediate superiors.) Together, they can obtain
concessions from the supervisor, because he is anxious to
obligate them by granting some of their demands. In ex-
change. they feel constrained to comply with many of his
1
directives. Typically, a strict definition is given to the

limits of this effective authority. Subordinates can often
be heard to remark: "That's the supervisor's responsibility.
He gets paid for making those decisions." This does not
mean that operating employees shirk responsibilities, as
indicated by their willingness to shoulder those they de-
fine as their own. But the social agreement among the
members of the work group that making certain decisions
and issuing certain directives is the duty of the supervisor,
not merely his privilege, serves to emphasize that follow-
ing them does not constitute submission to his arbitrary
will but conformity with commonly accepted operating
principles. In such a situation, which prevails in some
organizations though by no means in all, subordinates do
not experience the supervisor's exercise of authority over
them as domination, neither are they necessarily envious
of' his responsibilities, since they frequently consider their
own more challenging than his.
The effective establishment of authority obviates the
need for sanctions in daily operations. If a supervisor com-
rnands the voluntary obedience of subordinates, he need
not induce them to obey him by promising them rewards
or threatening them with punishment. In fact, the use of
sanctions undermines authority. A supervisor who is in the
habit of invoking sanctions to back his orders-"You
won't get a good rating unless you do this"--shows that
he does not expect unqualified compliance. As subordi-
nates learn that he does not expect it, they will no longer
feel obligated unconditionally to accept his directives.
Moreover, employees resent being continually reminded
of their dependence on the supervisor by his promises and
threats, and such resentment makes them less inclined to
carry out his orders.
This is the dilemma of bureaucratic authority: it rests
on the power of sanction but is weakened by frequent
resort to sanctions in operations. A basic difference, how-

ever, should be noted between the periodic rating of the
performance of subordinates, which can be called a disuse
sanction, and specific sanctions recurrently employed to
enforce particular commands. Since all employees know
that their immediate superior is officially required to
evaluate their operations at periodic intervals, this evalu~
ation is neither a sign that he does not expect un-
qualified compliance with his directives nor a reason for
annoyance with him. This diffuse sanction, imposed only
annually or every few months, though creating the depend-
ence of subordinates upon their supervisor, does so with-
out constantly endangering their willingness to be guided
by his requests, as the habitual use of specific sanctions
(including promises of good ratings and threats of poor
ones) would.
While the mere fact that the supervisor administers rat-
ings is not resented by his subordinates, low ratings might
well antagonize some of them. But bureaucratic mechan-
isms exist that enable the supervisor to shift the blame for
negative sanctions. For example, statistical records of per-
formance, which are kept in many white~collar offices as
well as factories, furnish the supervisor with objective
evidence with which he can justify low ratings by showing
the recipients that the poor quality of their work left
him no other choice. Instead of blaming the supervisor
for giving them a poor rating, these employees are forced
to blame themselves or to attribute the rating to the "sta-
tistics," which are often accused, rightly or wrongly, of
failing to measure the qualitative aspects of performance*
His intermediate position in the hierarchy provides the
supervisor with another justification mechanism. He can
place the responsibility for giving low ratings or instituting
* Of course, quantitative records also facilitate the super~
visor's task of evaluating operations.
'78 BUREAUCRACY IN MODERN SOCIETY
unpopular requirements on his superiors, to whom he is

accountable. Oftentimes a supervisor or foreman will tell
his subordinates that he does not like certain standards
any better than they do but "those brass-hats in the front
office" insist on them. In most organizations, one or a few
superintendents or assistant managers (or deans) become
the scapegoats who are blamed for all negative sanctions
and unpopular requirements. Since the attitudes of em-
ployees toward these administrators in removed positions
is much less relevant for effective operations than their
attitudes toward their immediate superior, the displace-
ment of aggression from him to them is in the interest of
the organization. Clients or customers can also serve as
scapegoats of aggression-the supervisor can blame their
demands for instituting procedures that inconvenience ern-
ployees. And if he joins subordinates in ridiculing clients
or customers, a frequent practice in service occupations,
the supervisor further reduces antagonism against himself
by standing united with the employees against outsiders.
Periodic ratings, then, increase the dependency of the
members of a bureaucracy on their superiors but at the
same time allow them to escape from disturbing feelings
of dependency by internalizing the principles that govern
operations. Although the responsibilities the supervisor is
required to discharge occasionally arouse the animosity of
some subordinates, various mechanisms divert such antag-
onism from the supervisor to other objects. These two
elements of the bureaucratic structure conspire to provide
a fertile soil for the establishment of supervisory authority.
Together, they permit supervisors to obligate subordinates
willingly to follow directives.
Various circumstances, however, can prevent such fa-
vorable conditions in the bureaucratic organization. The
disciplinarian supervisor may antagonize subordinates,
Bureaucratic Authority '79
through recurrent use of sanctions and in other ways, and

thereby undermine his effective authority over them as
well as their motivation to put effort into their work. The
lenient supervisor may be so reluctant to displease sub-
ordinates that he refrains from evaluating their perform-
ance in accordance with rigorous standards, giving all of
them high ratings. This practice invalidates the incentive
system, which. enhances the interest of employees in ac-
complishing specified results in their operations. The ma-
nipulative supervisor may employ devious techniques to
conceal from subordinate* _ul attempts- 'l impose his
arbitrary will upon them, -u ~n mple, by frequent and
unwarranted utilization of scapegoats. While manipulative
_
techniques he a fair chance " bf being successful in
temporary pIf relationsh as. as between -A omer and
salesman, their chances of success in relatively permanent
relationships within a group are very slim. For sooner or
later, some member is apt to see through them, and he is
not likely to keep this a secret. Once they are discovered,
mani techniques have a boomerang effect. Em-
ployees who realize that their superior tries to manipulate
them are prone to suspect all of his statements and gen-
erally to resist his efforts to influence their performance.
These and other disruptive tendencies can be observed in
hierarchical organizations, but methods of supervision that
encourage operating efficiency are also evident. In the ab-
sence of a much larger body of information about bureauc-
racies than we now possess, it is impossible to know which of
these opposite conditions is more frequent. Nevertheless,
the fact that authority is sometimes effectively exercised
without domineering subordinates or lowering their mo-
rale, rare as this may be, demonstrates that such a state
of affairs is actually possible and not merely a utopian
ideal type.
Inequality in Hierarchical Organizations

If we assume that hierarchical authority is a prerequisite
for effective coordination in a large organization (and
though this is not a conclusively established fact over-
whelming evidence points in its direction), its members
cannot be fully equal in status and power. In a democracy,
however, where status prerogatives are frowned upon, in-
tense feelings of inequality among the lower echelons of
a bureaucracy have several effects that are detrimental for
operations. They inhibit identification with the organiza-
tion and its objectives, lessen interest in performing tasks
to the best of one's abilities, kill initiative, and reduce the
chances that emergent operating problems will be readily
met. Unless employees consider themselves partners in a
common enterprise rather than tools in the hands of man-
agement, they are not prone willNigly to assume responsi-
bilities of their own.
Whereas a basic conflict exists between coordination and
work motivation in bureaucratic organizations, this does
not mean that it cannot possibly be resolved. Social in-
equality is not an all-or-none proposition: there are varia-
tions in kind as well as in degree. The high value a person
places upon equality of status does not prevent him from
obeying the orders of his physician when he is ill. Sub-
mitting to the authority of his physician--or lawyer or any
expert-~does not violate his integrity. This is the case be-
cause he is convinced that the doctor does not arbitrarily
impose his will upon him, but that the doctor's orders are
governed by rational principles that serve a common in-
terest, namely, curing his illness? When it is socially ac-
cepted that the person in authority is guided by rational
standards in the pursuit of common ends, unquestioning
obedience to his directives, although an undeniable sign
of i n e a l i t i in the relationship, is not experienced as

subjugation nor does it engender profound feelings of in-
equality. I
Let us DOW examine, in contrast, three bureaucratic con-
ditions that definitely create pronounced inequalities. 1.
Close supervision forces employees continually to submit
to the demands of their superior. To be told what to do
and to be checked throughout the day are quite a different
experience from that involved in being governed by the ex-
pectafion that one's completed work must meet rigorous
standards. The latter situation does not entail being con-
stantly dominated by another person, the former does.
2. Employment of sanctions in daily operations recurrently
asserts the supervisor's power oveN members If the
work group and contradicts the assumption that je is
merely the first among peers. Promises of rewards have
this effect no less than threats of punishment. auform-
ance ratings are another major source of inequality. 3. Ar-
bitrafjy power produces even more extensive differences in
social status. If their supervisor regularly makes specified
decisions, employees can adjust to the situation by accept-
ing these limits to their discretion. If, O11 the other hand,
he expects them to assume responsibility for deciding cert
rain matters, but from time to time imposes his own de-
cisions in these very matters, no similar adjustment is pos~
sible. Through this type of action, die supervisor subjects
his subordinates to his arbitrary will, probably he is moti-
vated, though he seldom admits this even to himself,
primarily by a desire to impress others with his power over
them. For were it necessary for him to make these de-
cisions, he could not permit them to be made most of the
time by the subordinates themselves. Arbitrary exercise of
power arouses the most intense feelings of inequality.
But bureaucratic authority neither depends on these
three conditions nor is inevitably accompanied by them,
as can be briefly shown. In the first place, evaluation on

the basis of standards that specify the results to be accom-
plished constrains employees to discipline themselves and
renders close supervision as well as detailed rules super-
iiuous. Statistical records of pertormance are particularly
elective as an impersonal mechanism of control.5 Sec-
ondly, the supervisor who commands the voluntary obedi-
ence of his subordinates has little need for sanctions to
enforce his directives. To be sure, his effective authority,
which enables him to discharge his supervisory responsi-
bilities without frequent resort to specific sanctions, ulti-
mately rests O11 a diffuse sanction, the periodic rating.
However, the tendency to put the blame for low ratings OI1
objective standards or higher administrative oliicials is not
simply a manipulative technique. Although the supervisor
exercises judgment in evaluating subordinates, he, just as
they, is presumably guided by exacting standards in arriv-
ing at his decisions, and he is held accountable by his
superiors for making correct judgments. If the supervisor
conforms with these principles and strives to give fair rat-
ings, his subordinates, while not becoming his equals in
the bureaucracy, do not find themselves dominated by an
unpredictable power that demolishes their self-respect.
Finally, the exercise of bureaucratic authority is a duty,
not a privilege that can be abrogated at will. In this re-
spect, formal requirements and infomial expectations
among work groups are in agreement. Only by overstep-
ping both the official and the unofficial limits of his au-
thority can a superior dominate his subordinates in an
arbitrary manner.
These considerations suggest that the pronounced in-
equalities that are often found in bureaucratic organiza-
tions are not essential for coordination or uniformity. If this
conclusion is correct, one of the main tasks of the adminis-
trator is to minimize such disruptive inequalities without,
of course, endangering the hierarchical authority needed

"E coordination. That this is not an impossible task is
indicated the fact that some bureaucratic superiors
main accomplished it. That it is not an easy task is the
result of inherent tendencies in hierarchical organizations.

Superiors, whose responsibilities require that they have
some control over subordinates, are under perennial temp-
tation to utilize their power not only in the interest of the
organization but also in their own interest, for instance, to
facilitate their own work or to satisfy their need of dom-
inating others. If they yield to this temptation, minor
inequalities expand into major ones. To suppress such
tendencies is far from simple, and even after they have
been suppressed, continued watchfulness is required lest
they emerge again. The goal, however, may, well be worth
the effort, since lesser inequalities promise to find expres-
sion in a more highly motivated working force.
5
Bureaucracy and Social Change
In' n1sc1p1.1ntz does not suffice for effective bureaucratic

operation, Flexibility also being necessary, it follows that
rigidity is disadvantageous for orgaIlizaiiol2L Whereas
this principle has been stressed throughout the preceding
discussion, we have not yet examined carefully the socio-
psychological processes involved. Why do some members
w
of large organizations resist any change mi rocedures
while others accept innovations with ease? What are the
organizational conditions in bureaucracies that give rise
to these opposite tendencies? -
This question of internal change is distinct from, though
not unrelated to, the problem of external change, Qhat is,
bureaucracy's role in changing the society of which it is a
part. Bureaucratization has been held to be a revolutionary
force, on the one hand, and a potent instrument of reac-
tion that makes it virtually impossible to alter the existing
85
institutional structure, on the other. Again, there appear to

be contradictory strains Mat require exploration.
Who Are the Ritualisms?

"The clerks of departments find themselves sooner or
later in the condition of a wheel screwed on to a machine;
the only variation of their lot is to be more or less oiled."
In these words, Balzac describes the lot of the bureaucrat
in the novel The Civil Service. Students of administration
similarly have often called attention to the ritualistic con-
cem with the minutiae of routine and the resulting ineli-
ciency that one often encounters in bureaucracies. An ex-
ceipt from a study of the civil service in France illustrates
these conditions:
Every large-scale organization controlled from a sin-
gle center sooner or later finds it advisable to elaborate
systematic routine procedures in the interest of fiscal
regularity and operational consistency. Private business
corporations are no more immune to this process than
are government departments. Nor do routine procedures
necessarily slow up staff deCiSiODS. On the contrary, if
they are properly adapted to the daily problems of the
enterprise, they expedite action.
An organization conforming closely to the hierar-
chical principle, however, faces the constant danger that
these routine operations will become sterilizing ends in
themselves rather than effective means to desirable ends.
When this happens the usual result is an entanglement
of "red tape," or as the French are wont to call it, La
paperasserie, mere routine thereby becoming bad row
tine. Formal instructions issued at the center Overwhelm
those who have to handle out on the circumference con-
crete situations unforeseen in their variety. Almost in-
evitably an adequate delegation of discretion to subor-
dinate officials is missing in such a system and the field
Bureaucracy and Social Change 87
agent stationed OH the administrative tiring line stands

helpless before demands for prompt decision and im-
mediate action. The fact that every case must be "re-
ferred" somewhere means a postponement of any de-
cision about it, the more circuitous the course of refer-
ence, the greater the delay....
Mr. Ford radon Ford relates his adventures in try-
ing to trace a postal money order gone astray. When
this occurs, the usual _course is to take the matter up
through official channels, give the postman a big tip, or
put the case into the hands of "an adviser of public
companies." On this occasion, however, Mr. Ford de-
cided to go directly to the Direction de la Seine des
P. T. T. on the Boulevard Montparnasse. At two o'clock
he was ushered into the Director's office by a smiling
charwoman. After a half hour the Director returned
from lunch and scrutinized the documents with great
. care. Following further consultation with an otlicial in
a blue uniform, the Director announced that Ford
should betake himself to the "Chief Sub-oiiice for the
Recovery of Money Orders" on the other side of Paris.
There he was directed to Room V on the sixth floor.
While he conversed with an attractive young woman
for an hour about face powders and the like, her chief
examin
. ed the papers and asked questions about Ford's
.. . . . .
war record and family, finally instructing him to return

to the Boulevard Montparnasse, this time to Room XVI
on the third floor. From there he was sent back to
Room XI in the Chief Sub-office; thence to Room IV,
Boulevard Montparnasse, next to Room III, Chief Sub-
odice, and finally to the "open sesame e"-Room XIII,
on Montparnasse. Although assured there that he would
receive his money by the first delivery the following day,
it actually arrived seven weeks later, only after a gener-
ous tip had been showered upon the postman.1
Inefficiency of this sort occurs when the members of an
organization become so preoccupied with meticulous appli-
cation of detailed rules that they lose sight of the very pur-
88 BUREAUCRALCY IN MODERN SOCIETY
pose of their action. Certain conditions in bureaucratic

structures encourage the development of this ritualistic
orientation, as Merton notes:
Discipline can be effective only if the ideal patterns are
buttressed by strong sentiments which entail devotion to
one's duties, a keen sense of the limitation of one's au-
thority and competence, and methodical performance
of routine activities. The ef'cacy of social structure de-
pends ultimately upon infusing group participants with
appropriate attitudes and sentiments. . . . These senti-
ments are often more intense than is technically neces-
sary. There is a margin of safety, so to speak, in the
pressure exerted by these sentiments upon the hnraml-
crat to conform to his patterned obligations, in much
the same sense that added allowances (precautionary
overestimations) are made by the engineer in designing
the supports for a bridge. But this very emphasis leads
to a transference of the sentiments from the aims of
the organization onto the particular details of behavior
required by the rules. Adherence to the rules, originally
conceived as a means, becomes transformed into an end-
in-itself, there occurs the familiar process o f displace-
_
ment of goals whereby "an instrumental value becomes
a terminal value." Discinline, readily interpreted as con-
formate with regulations, whatever the situation, is
seen not as a measure designed for specific purposes but
becomes an immediate value in the life-organization of
the bureaucrat.-2
The prevention of arbitrary decisions requires that a
high respect for disciplined performance of duties be fos-
tered among the members of a bureaucracy. This emphasis
sometimes becomes overpowering, with the result that
punctilious adherence to formalized procedures is elevated
into the primary objective of bureaucratic activities and
displaces their original objectives in the thinking of offi-
cials. Compelled by this orientation to find the right rule
before making the least commitment, a bureaucrat will

refuse to take any action if there is no clearcut precedent
or if there is the slightest doubt about whether it is entirely
within his official sphere jurisdiction. The well-known
phenornenon "passing the buck" and other practices
that obstruct operations are often expressions of this ten-
dency. In one case, an official in the employment agency
previously mentioned postponed deciding on the color of a
new set of index cards until he could determine what color
they were "supposed" to have, completely ignoring that
the only purpose of assigning a color to them was to dis-
tinguish them from other sets of cards. Olhcials who 'rind
their security in strict adherence to familiar routines,
moreover, strongly resist changes in the organization and
are incapacitated by new problems that confront them.
Rigidities are dysfunctional for operating eiliciency even
under stable conditions and particularly when emergent
problems call for a reorganization of working procedures.
Ritualistic displacement of goals, however, is not char-
acteristic of all members of bureaucratic organizations.
Many of them, far from deriving satisfaction from con-
stantly following the same routine, find doing so extremely
boring. They often express a desire for more variety in
their work and for changes that would relieve its mo-
notony. Since even complex tasks become less interesting
once they are fully mastered, many employees welcome
frequent changes in procedures because these create new
problems which recurrently make their work challenging.
As one civil servant in the federal agency previously dis-
cussed put it: "Lots of us gripe about the fact that they
change things all the time. But if I should be completely
honest with you, although; .\ l~ so gripe about having to
n
keep on learning new things, really like it. Tllat's what
keeps the job interesting."
Some officials rigidly oppose innovations in the organi-
zation, while others favor them. What are the stntctural

constraints in bureaucracies that account for these diiter-
ences? One of them is the nature of the incentive system.
When strict conformity with specific operating rules is the
basis for evaluation, employees are motivated, as a way of
adapting to this situation, to think of bureaucratic proce-
dures as if they were a sacred ritual, and strong resistance
to change in these procedures must be expected. When
employees, on the other hand, are evaluated in terms of
the results accomplished in their operations, they are en-
couraged to exercise ingenuity and employ diverse methods
in the interest of maximizing specified accomplishments.
But even in the situation where the evaluation system it-
self does not foster ritualism other conditions in the
bureaucratic structure may do so.
An analysis of instances of extreme rigidity in hierarchi»
cal organizations reveals that they are usually associated
with fear of superiors. For example, a group of otiicials
was once reprimanded by a high administrative oilicial for
having made an incorrect decision in one of their cases.
Thereupon, they applied the rules literally in similar cases
and refrained from exercising any discretion even when it
was clearly called tor, afraid of further reprimand, they
attempted to protect themselves against this danger with
overconformity. Bureaucratic superiors cannot generally
C€I1Sl.I1'€ a subordinate for following official regulations
exactly, regardless of how inetlicient or ridiculous such
action may be in a particular case. Hence, feelings of de-
pendency on superiors and anxiety over their reactions
engender ritualistic tendencies.
Rigid adherence to the established routine is a defense
mechanism against feelings of insecurity. In the study of the
federal agency, the attitudes of a group of officials toward
changes in regulations, which occurred frequently, were
ascertained and related to their competence as investiga-
tors. Not one of the more competent half of this group,

but most of the less competent half, voiced objections to
these recurrent innovations. From a purely rational stand-
point, the opposite finding might have been expected: the
. .. . ... . .
agent most familiar with existing regulations and most

adept in applying them presumably should have been most
disturbed when they were superseded by new ones. This
reasoning, however, fa q to take mm Consideration the
»
l
emotional factors that influence conduct. The anxieties

generated by the experience that one's knowledge is not
always adequate for one's tasks can be calmed by making
a ritual of conformity with those procedures with which
one has become familiar. Changes in procedures constitute
a threat to this method of coping with anxieties and, con-
sequently, must be strenuously resisted. Only in the ab-
sence of predominant feelings of insecurity can the desire
to escape monotony emerge as a motivating force. Officials
who feel secure in their ability to handle their responsibil-
ities and do not continually worry about the reactions of
superiors conceive of new problems as stimulating chal-
lenges and welcome frequent changes which prevent their
jobs from becoming monotonous*
Bureaucracy as Instrument of Innovation

In the large and complex societies of today, the imple- i
mentation of 136W social policies requires bureaucratic
machinery. Consider the case mm
invetitions, which are
sometimes viewed as spontaneous sources of social change,
for example, the atomic bomb. To be sure, had Enrico
Fermi and other scientists not had some brilliant ideas,
there would be no atomic bomb, but these ideas alone did
not bring it into evdstence. A complicated bureaucratic
* This situation probably also holds for teachers and other
professions.
organization had to be set up both to produce atomic

bombs and to furnish scientists with laboratories where
they could work together on improvements and new de-
velopments. Not that all social change in modern society
is bureaucratically instituted. New customs constantly arise
without the intervention of bureaucracies. But the deliber-
ate introduction of a social innovation on a large scale,
whether it involves the production of a new weapon or the
enforcement of a new law, depends on bureaucratic meth-
ods of administration.
Trade unions illustrate this point and some of its impli-
cations. For workers to realize their collective goal of im-
proving their standard of living, they organized. To estab-
lish a strong labor union against the opposition of employ-
ers was, and still is, a very difficult task. It could not have
been accomplished unless many workers, at least tempo-
rarily, had set aside their economic interests, often sacri-
ficing their jobs and sometimes their very lives, because
they were idealists whose primary objective was the crea-
tion of an effective labor organization. The need for such
idealism in the establishment of trade unions was pointed
out a century ago by Karl Marx, the man who is often
assumed to interpret social conduct as determined pri-
marily by economic interests and not by ideals. He wrote:
If the first aim of the general resistance was merely the

maintenance of wages, combinations [of workers], at
first isolated, constitute themselves into groups as the
capitalists in their turn unite in the idea of repression,
and in the face of always united capital, the mainte-
nance of the association [union] becomes more neces-
sary to them than that of wages. This is so true that
the English economists are amazed to see the workers
sacrifice a good part of their wages in favor of associa-
tions, which in the eyes of the economists, are estab-
lished solely in favor of wages?
Without using the term, Marx described in this passage

the process of displacement of goals from high wages to
maintenance of the organization. Observing the early
struggles of the labor movement, he assumed this process
to be highly beneficial for it. The Endings of more recent
studies of trade unions suggest that he was too optimistic,
in this respect as in many others. Displacement of goals
frequently results in a preoccupation with keeping the
bureaucratic apparatus going at the expense 'its basic
objectives.
Robert Michels's famous study of labor unions and
democratic parties in Germany at the beginning of the
present century is concerned with this problem. Even a
socialist party or a progressive union, regardless of how
egalitarian its principles, must establish a hierarchical be
reaucracy to put its reform program into effect. (The
issues raised by this so-called "iron law of oligarchy" will
be discussed in the next chapter.) The major interest of
party or union olNcials is to strengthen the organization,
not only because their jobs depend on its survival, but also
because a powerful machine is needed in the might for the
intended reforms. In this respect, the self-interest Of the
leadership and the collective interest of the membership
coincide. Officials, consequently, are willing to make great
sacrifices for the sake of fortifying the organization. To
s, they will abandon unpopular points
of the program. To prevent the possibility of a crushing
defeat, they will fail to enforce union demands by calling
a strike. "Thus, from a means, organization becomes an
end." 4 Step by step, tbe original objectives are surrendered
in the interest of increased organizational strength. The
resulting organization may be extremely strong, but it is
no longs; an instrument for effecting the social reforms
that were initially planned. What was once a socialist
party (or a radical union) has turned into a rather con-
94 BUREAUCRACY IN MODERN soczan'
sedative one. The inevitable fate of all reformist move»

merits, according to Michels, is to grow conservative in the
course of becoming organized.
This conclusion has important implications that extend
far beyond the question of the future of socialism. The
radical ideas of the Reformation spread more or less spou-
taneously without the aid of a bureaucratic apparatus, pro-
ducing profound changes in the institutional structure of
European society. Michcls suggested that this cannot hap-
pen in today's bureaucratized societies. For new ideas to
find expression in institutional change, they must be bu-
reaucratically implemented. In the process of creating an
effective bureaucratic apparatus, radical new ideas are al-
ways renounced in favor of more conservative ones. In
other words, people cannot possibly control their common
destiny by instituting desirable social reforms. For unless
they establish a bureaucratic organization for this purpose,
they will not be successful in realizing their new ideals,
and if they do, they will abandon them. There is reason
to assume, however, that this impasse is not entirely in-
surmountable. Although Michels analyzed a doubtlessly
prevalent feature of organizational l i f e , he ignored another
trend that points in the Opposite direction.
An examination of unions and parties that began with a
very radical program reveals, indeed, that most of those
that survived replaced their earlier radical goals by more
conservative ones and greater concern with administrative
matters in the course of establishing an effective organiza-
tion. But study of unions that initially had more limited.
plans for change discloses different tendencies. The Ameri-
can labor movement provides a good illustration. After
the decline seventy years ago of the Knights of Labor,
who had advocated a radical political program, most un-
ions confined their efforts to the pursuit of two objectives :
establishing the right of collective bargaining and raising
wages. To be sure_, concern with building a strong union

sometimes pushed these two goals into the background, as
jurisdictional strikes indicate, and some union leaders be-
came increasingly conservative.-5 'Mi main development,
however, was not the one stressed by Michels: unions did"
not relinquish their original objectives. Quite the contrary,
they achieved them in large part and strove for new,
further-reaching reforms. Thus, the right of collective bar-
gaining supplied a basis for the might for workers' pensions,
a social innovation far surpassing the aspirations of union
members a few decades ago. This process, the reverse of
displacement of goals, can be called "succession of goals",
as earlier objectives are attained, they become stepping
stones fOr new ones.
The succession of goals, of course, is not primarily the
result of the superior idealism of American as compared
with German labor leaders, but the consequence of struc-
tural constraints H organization. lm union has
achieved its major ogectives, the enthusiasm of its mem-
bers tends to wane. Many withdraw their support, inan-
cially and otherwisis and thereby threaten the persistence
of the organization. The very fact that union officials are
interested in maintaining their job and power constrains
them to seek new ways of stimulating membership sup-
port. An effective method for recreating vigorous interest
in union affairs is to establish new objectives for which
workers are willing to fight. Hence, new goals often
emerge in organizations as old ones have been reached.
This is the case not only in unions but also in other
organizations marked by bureaucracy's stamp.6
What determines whether displacement of goals or suc-
cession of goals predominates in an organization? This
crucial question can be only partially answered. When the
original objectives of a social movement arouse intense
hostility and violent attacks, the insecurity of its members
and their preoccupation with creating an organization and

preserving it are likely to constrain them to compromise
their ideals in order to avoid annihilation. When the com-
munity permits an organization, if only by default, to be~
come established and attain at least some of its first ob-
jectives in a relatively short period, it will probably find
new fields to conquer in the course of its development.
How radical can social movements be without provoking
hostilities that destroy them? How long is the period of
grace before the struggle is given up as hopeless and the
initial objectives are abandoned to maintain the organiza-
tion? We do not yet know the answers to these questions,
although a few recent researches provide suggestive leads
to the study of interconnections between organizational
factors and social change.
Conservative Pressures in
Two Social Contexts
A brief review of two empirical investigations of pro-
gressive programs and their fate indicates the complex re-
lationship between bureaucratic structures and changes in
social policies. Whereas both studies found strains toward
conservativism, the social forces responsible for them
were quite different.
Philip Selznick's study of the Tennessee Valley Authority
shows that the grass-roots policy adopted by this New
Deal agency had unanticipated consequences that brought
about fundamental changes in its progressive program?
The principle of grass-roots democracy emphasizes that
the central government should not simply impose its au-
thority upon the people in a region, but should give them
a voice in the management of the federal agencies that
affect their lives. Since all people in the Tennessee Valley
could not directly participate in administrative decisions,
this principle was implemented in actual practice by ap»

pointing representatives of powerful local institutions,
notably the land-grant colleges, to positions on the policy-
making body of the TVA. Many of these influential per-
sons and organizations had been strongly opposed to the
TVA, and their opposition might well have put serious
obstacles 'iiiTt§ Way; The cooptation of-representatives of
these powerful conservative groups by the TVA, mi is,
HI absorption into its leadership structure, averted this
threat. E grass-roots method, as it was interpreted,
constituted a mechanism that permitted a New Deal
agency to function in a region dominated by conservative
forces.
m I o work through locally estab-
lished institutions brought about unforeseen effects that,
paradoxically, contradicted the democratic spirit of the
grass-roots doctrine that had been the reason for making
the commitment. As men with coNservative views who
represented vested interests and not the majority of people
in the area were appointed to its board of directors, TVA's
policies became increasingly conservative and removed
from New Deal principles. Thus, the TVA discriminated
against Negroes, it came into conflict with other New Deal
agencies, such as the Farm Security Administration, and
various policies that had been designed to protect the
public interest against special private interests were re-
versed. The last point is exemplified by the changes that
occurred in the purchase of land for reservoirs. Building
a reservoir improves the soil around it. To permit the
public, whose funds paid for the reservoir, to benefit from
this increment in land value, the TVA established the
policy that the purchase of land for each reservoir include
a surrounding protective strip 300 to 1,000 feet wide.
Many landowners, anxious to reap these benefits them-
selves, were opposed to this program of public ownership.
Since their interest was represented on TVA's board, their

pressure was successful. In 1942, the policy was reversed,
the board of directors deciding "to limit the purchase of
land for reservoir purposes to the minimum abprebriate
for the particular project," 8 which usually did not include
any protective strip.
Selznick does not deny that the TVA produced pro-
found changes in the Tennessee Valley and greatly con-
tributed to the welfare of its economically deprived people.
But he shows that these changes were not as far-reaching
and not as unequivocally in the interest of the larger
population in the area as had been originally planned. In
the course of its development, the bureaucracy became
more conservative.
S. M. Lipset, in his study of a socialist government in a
Canadian province, also observes that the bureaucratic
implementation of a progressive program occasioned its
modification." When the C.C.F. (Cooperative Common-
wealth Federation) came into power in Saskatchewan in
1944 and its members took over all cabinet posts, they
retained the former administrators of government bureauc-
racies as their deputies. Although it was known that most
of the high government officials were middle-class persons
not at all sympathetic to the socialist program of the
C.C.F., the leaders held that they needed these administra-
tive experts to operate the bureaucratic machinery in the
various governmental departments. Moreover, ministers
assumed that they would determine policy and that their
deputies would only carry it out. In the process of being
administered by conservative bureaucrats, however, social-
ist policies were often basically altered. Here are a few
illustrations from Lipset's book: a
A number of civil servants were able to convince their

ministers that certain changes were not administratively
feasible or that they would incur too much opposition.
Some deputy ininisters exchanged information with

other deputies lm their technique of controlling their
ministers. g]Some key officials boasted of "running
my department completely," and of "stopping hare-
brained radical schemes."
One cabinet minister, who has since discharged a large

part of his field staff, found as a result of complaints
from local members of the C.C.F. that members-of TO-is
staff continued to grant leases and farming privileges to
well-to-do persons who had secured them under previous
governments, 'though it was now government policy to
give them to poorer farmers and landless veterans.
One cabinet minister decided that certain government

work that had previously been contracted out to private
concerns should be done by government employees
whenever possible. His deputy minister, however, con-
tinued sending the work out to private concems.1°
An important similarity marks the Endings of Lipset
and Selznick, but there is also a difference that, though
less obvious, is no less important. If we focus our atten-
tion upon the political program, we notice the similarity.
In both instances, a progressive program was modified in
the course of being implemented by bureaucratic methods.
If we focus our attention upon the bureaucratic organiza-
tion, on the other hand, we can see that it played quite a
different role in the two cases. In Saskatchewan, bureauc-
racies obstructed plans for reform that had originated
elsewhere, whereas in Tennessee, external forces obstructed,
partly through infiltration, the bureaucracy's original
plans for reform. The fact that government policy was
modified in a conservative direction in both cases reveals
the power of conservative forces in Canadian and Ameri-
can society. In one case, however, conservative pressure
was exercised by bureaucracies and its success indicates
100 BUREAUCRACY IN MODERN SOCIE1IY
their strength, while in the other case, conservative pres-

sures were exercised upon the bureaucracy and their sue-
cess indicates its weakness.
It should be noted that neither study suggests that the
bureaucratic structure itself generated the conservative
trend. Selznick shows that in the case of the TVA it
originated outside the bureaucracy, and Lipset stresses
that in Saskatchewan it was due to the conservative
orientation of the particular oiiicials and not to Inherent
tendencies in the bureaucratic form of organization. More-
over, many parts of the progressive programs of the TVA
and the C.C.F. were actually realized. Michels's conclusion
that the bureaucratic machinery necessary for implement-
ing new social ideals invariably destroys them is perhaps
too pessimistic.
Bureaucracy and Democracy
To Assess bureaucracy's impact on democratic values, "H-

internal and external consequences must be distinguished.
Either we are concerned with the particular structures that
are bureaucratically organized and raise the questions of
whether this organizing principle is compatible with in-
ternal democracy and, if not, whether the bureaucratiza-
tion of some organizations is nevertheless justified in a
democracy; or we are concerned with the society within
which numerous bureaucracies exist and raise the questions
of whether they threaten its democratic institutions and, if
so, how to protect democracy against this threat. Answers
to the first two questions will furnish clues for answering
the second two.
Bureaucracy's power of control has implications for its
own members, on the one hand, and for society at large,
on the other. In addition, it has implications for its clients,
who constitute a border~1ine group, being neither fully
101
102 BURBAUCRACY IN MODERN socnz'ry
part of the organization nor completely external to it. At

the outset of this study, it was mentioned that the accu»
sation of bureaucratic inefficiency, even when it is not
factually correct, reveals attitudes of clients toward .bu-
reaucracies that must be explained. An analysis of the
sources of such antagonistic attitudes can serve as an
introduction for the discussion of the other problems of
bureaucratic power.
The Accusation of "Red Tape"

There is 110 doubt that bureaucracies sometimes operate
ineihciently. When this occurs, however, clients rarely
have an opportunity to observe it. Conversely, many bu-
reaucratic practices condemned by clients are not in fact
inetiicient. For example, being required to fill out lengthy
.forms in minute detail, including entries that are clearly
not pertinent to the particular case, is inconvenient from
the standpoint of the client but may be expedient for the
bureaucracy. This requirement is more efficient than per-
mitting clients to decide which entries are relevant, since
even occasional omissions of pertinent information would
interfere with operations. Think of the last time you
accused some otlicials of being SO entangled in red tape
that they could not work effectively. Was it after you had
made a careful investigation and obtained evidence that
given operating methods were disadvantageous for the
bureaucracy? More likely, it was when you felt disadvan-
taged by a bureaucratic decision, and you gave vent to
your powerless anger by leveling the accusation without
knowing whether inefficiency was involved or not. We all
do this it makes us feel better.
The individual client stands helpless before the power-
ful bureaucracy, awaiting decisions that often vitally affect
his interests. Greatly concerned with his case, he sees in
Bureaucracy and Democracy 103
it a number of exceptional circumstances that deserve

special consideration H the impersonal bureaucratic
machinery disre_gards these and handles the case simply as
one of a general category. Raging against adverse decisions
or into able delays worse than futile, since it does
not sway H impersonal organization and merely ern-
phaizes 0 impotence. Frustrated clients can relieve
their pent-up aggression, 15wever, in discussions of bu-
reaucratic s t u p i y and red tape. Vifhereas the organiza-
tion's ruthlessness, not- Ts inefficiency, is the source of
their antagonism, by expressing it in the form of an ap-
parently disinterested criticism of performance, clients
derive a feeling of superiority over the "blundering bu-
reaucrats" that serves as a psychological compensation for
being under their power. To be sure, we are incapable of
direct retaliation when the actions of powerful bureauc-
racies hurt our interests, hue retaliate indirectly by
contributing through our opinion and ridicule to the low
public esteem of bureaucrats in our society.
Findings of a survey on attitudes toward bureaucratic
red tape support this interpretation! People who placed a
high value on social equality were found to be more critical
of red tape than those who did not. If this criticism were
based entirely on factual observation, such a difference
would probably not exist, since persons without an ega1i-
tarian orientation are as likely to have encountered bu-
'inefficiency
reaucratic as those with one. If severe censure

of red tape, on the other hand, is motivated by resentment
against bureaucratic power, the reason for the difference
becomes apparent: the more a person values equality, the
more objectionable is the experience of being subjected
to the controlling power of oiiicials. The same principle
C811 account for the Ending that criticism of red tape was
most pronounced among individuals who were particularly
sensitive about their powerless position.
In the same study, "conservatives" were found to attach

more importance to the problem of red tape than "radicals."
(Respondents were divided into these two political camps
on the basis of their attitudes toward labor unions.) This
ending may seem surprising, inasmuch as radicals might
be expected to be most eager to condemn the operations
of the government and of private bureaucracies. This very
fact, however, may explain their lesser inclination to worry
about red tape. When a radical comes into convict with
power structures, this confirms his political conviction that
existing institutions are unjust and should be changed. His
radical ideology supplies a channel of aggression against
the existing social system, obviating the need for express-
ing his aggression in other forms. But when a conservative
comes into conflict with power structures, he is in a more
difficult position. Since his ideology does not allow him to
denounce the government and private enterprise, eyen
when his interests have been injured by their actions, he
often seeks to relieve his feelings of frustration through
attacks OH the administrative machinery .and its red tape.
implicit in the prevalent condemnation of red tape is a
significant social consequence of this practice, which can
be most clearly seen in totalitarian countries. In Soviet
Russia, for instance, criticism of the government and its
institutions is strictly prohibited with one notable excep-
tion. Sharp criticism of bureaucratic mismanagement and
red tape is permitted and, indeed, encouraged, as indicated
by its frequent appearance in the government-owned press.
The dictatorship, through ruthless suppression and dis-
regard for the public's interests, engenders considerable
hostility, which might lead to attempts to overthrow it.
Although the nature of government policies rather than
lack of cliicicncy in their administration is the basis of
this hostility, by providing an opportunity for releasing
aggression in complaints about administrative red tape,

the dictatorship reduces the danger that the people will
rebel. But if this scapegoating serves an important function
for a totalitarian regime., H dysfunctional for a demo-
cratic society. Fuming against red tape and bureaucratic
methods serves as a psychological substitute for opposition
to bureaucratic policies that violate the interests of in-
dividuals. It therefore interferes with democratic processes,
which presuppose that people are able and willing freely
to express their opposition to existing power structures and
their policies.
Contrasting Principles of Infernal Control

Whatever the specific objectives of a bureaucratic or-
ganization, its formal purpose can be defined as the effective
accomplishment of these objectives. This is the case for
an army, for ..e;;am.ple, which is
for a fact
. . .. . . .
or ected to win battles;
which is expected to produce goods that
can be sold for a profit, for an employment agency, which
is expected to End jobs for applicants and workers for
employers. There is a second type of organization, how-
ever, which has no specific objectives but the purpose of
furnishing mechanisms for establishing consensus on com-
mon objectives, the machinery of a democratic govern-
ment is of this type. A third type, cited in Chapter One,
consists of organizations designed to supply their members
with intrinsic satisfactions either of a spiritual sort, as in a
church, or of a secular nature, as in a social club. The
following discussion is primarily concerned with the
contrast between internal social control in the first two
kinds of organization. Internal conditions in the third type,
which will not be analyzed here, are more similar to those
of the second than those of the first type.
106 BUREAUCQRACY IN MODERN SOCIETY
Efficiency versus Dissent If an organization is estab-,

fished for the explicit purpose of realizing specified objec-
tives, it is expected to be governed by the criterion of
efficiency. An organization so governed has been defined
as a bureaucracy. Defined in this way, the bureaucratic
form of organization is fundamentally different from both
the democratic and the autocratic forms. Neither the will
of the majority nor the personal choice of a ruler or a
ruling clique reigns supreme, but the rational judgment
of experts. Although both authoritarian elements and con-
cessions to democratic values are found in bureaucratic
structures,-efficiency is the ultimate basis for evaluating
whether such elements are appropriate. Disciplined obedi-
ence in the hierarchy of authority, ideally, is not valued
for its own sake, as it is in an autocracy, but is encouraged
to the extent to which it contributes to effective coordina-
tion and uniform operations. Similarly, while pronounced
inequalities are inherently opposed in a democracy, the
principal reason for minimizing them in a bureaucracy
is that# they inhibit optimum performance. Bureaucratiza-
tion implies that considerations of efficiency outweigh all
others in the formation and development of the organiza-
tion.
But if men organize in order to ascertain the ideas that
prevail among them and then to agree on common objec-
tives, their purpose requires that the basic principle which
governs their action is freedom of dissent. In this type of
democratic organization, considerations of efficiency are
expected to be subordinated to the central aim of stimulat-
ing the free expression of conflicting opinions. Of course,
democratic processes are not the most expeditious way of
arriving at decisions either for total societies or for limited
associations, such as trade unions. But the fact that it
would be more efficient if the leader were to decide on the
objectives to be pursued is irrelevant, since this policy
Rureaucracy and Democracy 107
could not possibly accomplish the purpose of determining

those objectives that are commonly agreed upon or express
the view of the majority. To assure that the majority view-
point remains supreme, a limitation has to be imposed on
the majority itself. It must not stifle the opposition of any
minority, however small its numbers or extreme its views,
for unless dissenting voices can be heard today, tomor-
row's decisions will not be democratic ones.
Bureaucratic and democratic structures can be distil
gushed, then, on the basis of the dominant organizing
principle: efficiency or freedom of dissent. Each of these
principles is suited for one purpose and not for another.
When people set themselves the task of determining the so~
c a l objectives that represent the interests of most of them,
the crucial problem is to provide an opportunity for all
conflicting viewpoints to be heard. In contrast, when the
task is the achievement of given social objectives, the es-
sential problem to be solved is to discover the efficient, not
the popular, means for doing so. Democratic values re-
quire not only that social goals be determined by majority
decision, but also that they be implemented through the
most effective methods available, that is, by establishing
organizations that are bureaucratically rather than demo-
cratically governed. The existence, therefore, of such bu-
reaucracies does not violate democratic values. But these
values are threatened by the encroachment of concern
with bureaucratic efficiency upon those institutions where
freedom of dissent is essential, where the guiding goal is
to enable men to arrive at democratic decisions. Thus
bureaucratic efficiency is expected to prevail in specialized
government agencies, but not in the political arena. Various
attempts to suppress radical political opposition in the in~
terest of national security 'Frustrate Trow illiciency CDD-
siderations intrude upon freedom of dissent.
Democratic processes are in particular danger of being

undermined by bureaucratization in organizations which
have the double purpose of deciding on common objec-
tives, on the one hand, and of carrying the decisions out,
OD the other. Political parties are a case in point. Large-
scale democracy depends on the existence of opposing
parties. They have the function of giving expression to
the political beliefs of people and to serve as channels
through which they can influence the government. To
fulfill this responsibility, a party must be democratically
organized, which means that primaries and other devices
are used to assure that the party program reflects the
wishes of its adherents. Democratic parties, however, also
seek to win elections, which requires an efficient organiza-
tion. Hence, they tend to be governed by political machines
and national committees, primaries being relegated to a
relatively inconsequential role. In order to make parties
more effective instruments for winning victories, their
function of permitting the voters to decide the political
platforms between which they will choose at election time
is sacrificed.
Incorporated business concerns furnish another illustra-
tion of this tendency. A clear distinction exists between
the management of a business and the organization of its
legal owners as stockholders in a public corporation. Ac-
cording to our laws and the principles of capitalism, it is
presumed that the business is managed OH the basis of
efficiency, but that the board of directors of the corpora-
tion is democratically elected by the stockholders. The
procedure employed for this purpose actually defeats its
intent. Proxies are provided every stockholder, which give
him about as much choice in elections as the citizen of a
totalitarian nation. He has almost no way of opposing the
existing leadership, since in most cases he can only en-
dorse this group by signing the proxy or not vote at all.
(Any OTC stockholder can attend the annual meetings and

voice his opposition, but all stockholders could not pos-
sibly do so. There is no stadium large enough to hold the
one million people who own shares of the American Tele-
phone and Telegraph Company or the hundreds of thou-
sands who share the ownership of other large corporations.)
In fact, therefore, most public corporations are dominated
by a few officials, and the large majority of stockholders,
sometimes owning more than 90 per cent of the shares,
have no influence on rnanagement.2 Only when a faction
arises, as occurred among the stockholders of the New
York Central Railroad in 1954, are there alternative slates
of candidates for directorships from which to choose, and
democratic processes zillwus temporarily revived?-
This case suggests that the perpetuation of democratic
processes depends on permanent factions, Lhat is, opposi-
tion parties within the organization, The situation in labor
unions, in this respect, is little different from that in busi-
ness corporations. As long as factional strife prevails in a
union, its members can throw their sup_p.ort to either sigil
and therefore influence the policies ormu organization,
once it ceases, they no longer have this power. Factions
rarely survive for more than brief periods, however, since
the one that palms control typically fortifies its leadership
position by suppressing the opposition. Such undemocratic
action is often justified by the need of remaining united
in the struggle with employers-internal disagreements
weaken the union and trust be avoided. Hmplicit in this
argument is the assumption that je mm function of a
union is to be an of(-icient instrument for attaining given
objectives, such as higher wages, an assumption represent-
ing a very narrow conception of unionism. Labor unions
are more broadly conceived as organizations that enable
workers to have a voice in determining their employment
and working conditions. There are various objectives that
a union can pursue; unless its members decide which of

these to seek at any given time the union does not represent
their interests. For this purpose, the union must establish
a democratic machinery and must protect it against the
threat of being destroyed for the sake of increased eff:-
ciency or strength. Let us examine how one union success-
fully maintained internal democracy.
A Union with Two Parties If internal democracy were

to incapacitate a union from being an effective bargaining
agent, workers could hardly be expected to engage in
such luxury. A recent study by Lipset and his colleagues
of the International Typographical Union shows, however,
that this development is not inevitably the case This
union, which is democratically governed, is a very strong
and elective representative of the interests of its members.
The democratic character of the I.T.U. manifests itself in
many ways. Most important is the fact that the officials of
the international union and of the larger locals are bi-
annually chosen in elections which are not purely symbolic
affairs where the membership merely endorses the existing
leadership. In many elections in the past-half of those in
the New York local, for instance-the incumbent oliicials
have been defeated. Changes in administration were usu-
ally accompanied by sharp reversals in union policy. Thus,
one administration favored arbitration with employers,
whereas the opposition advocated that union demands
should be enforced through strikes and adopted a more
militant strategy when it was elected. Since the member-
ship of the union was able to choose between such contrast-
ing alternatives, it actually determined policy. Basic
changes in union regulations must be endorsed be referen-
dum, and more often than not the proposals of the leader-
ship have been defeated, a further indication of the
independent spirit and democratic power of the rank and

File.
Probably the main source of internal democracy in the
I.T.U. is its two-party system, which has been in existence
for half a century. Institutionalized parties assure pu.-
there is always an organized opposition to the._.I€5ders'i;p
and that voters are presented with distinct alternatives,
without which democratic decisions in large groups are
impossible. Even the referendum is not an effective demo-
cratic mechanism in the absence of an opposition party,
which is interested in discussing the adlninistration's pro-
posals with the membership and in pointing out their
faults. In one-party unions, where such critical discussions
are rarely initiated, most members have no basis for op-
posing intricate proposals of their leadership, and a large
majority usually endorses them. In sharp contrast to the
situation in the I.T.U., the rank and tile in one-party
unions is not able to use the referendum to curb the power
of its leaders.
Opposing parties create the organizational conditions
necessary for democratic processes to prevail. Indeed, they
do more. It is often held that the apathy of most union
members (and of members of other voluntary associations )
is an insurmountable obstacle to democratic self~govern-
Ment. When they even fail to attend union meetings, how
can they assume responsibility for managing the affairs
of the organization? They can not, of course, in this event.
But the important question concerns the reasons for their
apathy. In the majority of unions where the leadership
makes all significant decisions, meetings are dull, since
only routine business is conducted, and there is no incen-
tive for attendance. The existence of 8.11 opposition party
greatly alters the nature of such gatherings. It forces the
leadership to present crucial issues for discussion, and,
furthermore, the controversies between opposition and

incumbent officials transform even discussions of less im-
portant topics into interesting contests. The union member
is not merely a spectator, he can actively participate in
the discussion and has a voice in deciding the contest
through his ballot. Parties are anxious to win converts, and
once a union rnernber becomes a partisan, he has an
additional inducement for attending union meetings. (Lip-
set shows that I.T.U. members who belonged to a party
attended union meetings more regularly than those who did
not.) In these various ways, institutionalized parties lessen
apathy and stimulate interest in union politics, thereby
providing a firm foundation for participation in democratic
self~government.
What were the conditions that enabled the I.T.U., vir-
tually alone among American unions, to establish and
maintain a democratic two-party system? One condition
was its strong position in the printing industry. The men
bets of a weak union usually must devote all their efforts
to the struggle with employers and have little time or
energy left for Creating a democratic machinery in their
organization. A second condition that contributed to
internal democracy was that independent local unions of
printers had existed for many years before they joined
together and formed the International Typographical Un-
ion in 1850. Several locals maintained considerable autor1~
orny, and it was the opposition of certain strong locals to
the international administration that provided the original
impetus for organizing a formal opposition party at the
beginning of this century. Thirdly, printers enjoy a fairly
high income, not much less than that of their union of-
Iicials, and they generally like their work. While union
leaders whose status is far superior to that of the workers
under their jurisdiction often view defeat in election and
the necessity to return to work in a shop as an unbearable
calamity, I.T.U. oiTicials do not. The latter, therefore, are

less prone to try to prevent defeat at all costs, including
the cost of sabotaging democratic methods. Attempts to
undermine the _twofparty system are also discouraged by
a fourth condition, namely, that the democratic tradition
has become part of the value orientation of I.T.U. mem-
bers. GfNcials are deterred from disregarding democratic
processes, since such conduct would be strongly resented
by the membership and invite defeat at the elections.
Finally, the members of the I.T.U. have created a large
number of voluntary associations within their union, such
as athletic organizations, lodges, social clubs, and the like.
Printers spend much of their recreational life with other
printers, partly because they often work at night and their
leisure hours do not coincide with those of most people.
These nonpolitical associations of. printers promote interest
i un p litics d participati i d in Cr tic s lf-g -
ernment on a wide scale. Printers who are unconcerned
with the political affairs of the union usually belong to
clubs where they c me into contact with fellow craftsmen,
some of whom are sufficiently interested in union politics
to talk about such matters on every occasion, even when
bowling or playing cards. In other words, the recreational
associations of printers expose apathetic union members to
discussions that as likely to stimulate increased concern
with the union and the way it is managed. Moreover, the
large number of v luntary organizations furnishes an op-
portunity for many union members to acquire the political
skills involved in the administration of a democratic group.
These men constitute a pool of experienced democratic
administrators, from which the officials of the two parties
and the leaders of the union can be drawn. The wide-
spread active participation in both the political and recrea-
tional affairs of the 1.T.U. sustains the internal democracy
that the two-party system makes possible.
A Challenge for Democracy

In conclusion, let us briefly examine some implications
of the prevalence of bureaucracies for democratic institu-
tions. Strangely enough, social scientists who advance
diametrically opposed theories about the historical develop-
ment of bureaucratization agree on this point. One inter-
pretation holds that the proliferation of bureaucracies is
the result of modern capitalism. The economic advantages
of large-scale production lead to the establishment of
huge industrial enterprises and ultimately to monopoliza-
tion. These powerful private bureaucracies put pressure on
the government to safeguard their interests, tor example,
by enacting protective-tariff laws and setting up the bureau-
cratic apparatus necessary for enforcing them. Hence,
bureaucracy in government as well as in private industry
is the product of forces generated by capitalism.5 Several
liberal economists, in contrast, attribute the trend toward
bureaucratization to the wilful effort of governments to
interfere with the capitalist economy. If the government
assumes the task of regulating economic life, it must greatly
expand its bureaucratic machinery, disturbing the competi-
tive mechanism of the free market and thereby facilitating
the development of monopolies. The emergence of bureau-
cratic business monopolies as well as government bu-
reaucracies is the inevitable outcome of the political
decision to meddle with free enterprise.6 The authors who
advance these conflicting theses about the historical origins
of large-scale bureaucracy, however, are in agreement
concerning its consequences. Bureaucratization concen-
trates power in the bands of a few men and curtails the
freedom of individuals that is essential for democracy?
Bureaucracies endanger democratic freedoms, but at
the same time they serve important functions in a demo-
cratic society that must not be ignored. Thus Weber points

out that bureaucratic personnel policies-employment on
the basis of technical qualifications--reduce the handicap
of underprivileged groups in the competition for jobs.
Negroes, for example, have a better chance of being hired
when objective criteria rather than personal considerations
govern the selection of candidates. Of course, the children
of wealthier families, who can more easily aliord the
education that qualities them for the most desirable jobs,
continue to have a-distinct advantage over others. Bureau-
cratization does not produce complete equality of occupa-
tional opportunities. Nevertheless, the fact that it does
minimize the direct effects of status privileges, such as
noble birth or skin color, constitutes a democratizing in-
fluence.
"Equal justice under law" is a fundamental democratic
principle. The executive agencies which help to administer
the law as well as the courts, according to this principle,
must not discriminate against any person or group. Ad-
rninistrative officials who investigate legal violations have
investigators interpret a law .

to interpret the law as they apply it to specific cases. Their
judgment is usually not appealed to the courts and con-
sequently assumes quasi-judicial significance. Unless all
the same manner, some
persons will be treated more strictly than others. Hence,
the decisions of all enforcement agents should be gov-
erned by uniform standards and protected against being
influenced by personal considerations. This is another way
of saying that bureaucratically organized enforcement
agencies are necessary for all members of the society to be
equal under the law.
Other contributions of bureaucracy have already been
discussed. Democratic objectives would be impossible of 4
attainment in modern society without bureaucratic organi-
zations to implement them. Thus, once the decision to
prow'de free employment service to the public had been

reached through democratic processes, a complex ad-
ministrative system for this purpose had to be established.
Furthermore, the high standard of living we enjoy today
depends, in part, on the adoption of efficient bureaucratic
methods of organization in private industry. Whereas, in
theory, a low standard of living does not inhibit democ-
racy, in actual practice it does. Large parts of the popula-
tion lose interest in preserving political freedoms if they
are preoccupied with finding ways to satisfy their minimum
economic wants. Under these conditions, people are least
likely to participate in their democratic government (eco-
nomically deprived persons are, in fact, less prone to
exercise their right to vote than others) and most likely
to fall prey to demagogues who promise them some relief
from their economic misery.
Were it not for these services rendered by bureaucracies
for a democratic society, their existence would not pose a
dilemma, only a problem. In such a case, the task would
still be difficult, but the decision would be clear' to 611-
deavor by all means to abolish bureaucracies, because they
have serious dysfunctions for democracy. First of all,
bureaucracies create profound inequalities of power. They
enable a few individuals, those in control of bureaucratic
machinciy, to exercise much more influence than others
in the society in general and on the government in partic-
ular. This huge differential in political and social power
violates the democratic principles that sovereignty rests
with all and that each citizen has an equal voice.
The prevalence of bureaucracies in a society also under-
mines democracy in more subtle ways. Lipset suggests that
theexistence of smaller self-governing bodies, the vol un-»
tars associations, is of crucial significance for democracy
in the I.T.U. They provide experience in democratic par-
ticipation, stimulate concern with the political affairs of

the union, and lessen the apathy that often characterizes
the members of a large organization, who feel remote
from its administration. In a society with more than one
thousand times the membership of the I.T.U., the United
States, democratic intermediary organizations would ap-
pear to be even more important for self-government in the
larger social structure. Tocqueville made this observation
over a century ago in his classic work Democracy in
America. Since his time, however, many formerly demo-
cratic organizations have become bureaucratized. We gen-
erally no longer govern our voluntary associations: we
simply join them, pay our dues, andlét experts run tliéiii.
As a result, we have less and less opportunity for acquit»
ing experiences that are essential for e-§ective-participation
in democratic government. :_.
A person must be able to communicate his ideas
-
others if he is to influence public opinion. But in a cotn~
munity the size of the United States, the individual's voice
is lost, and only organized groups have the strength to
make themselves heard. By joining democratic organiza-
tions and helping to decide their policies, people have a
chance to exert some influence OD the larger community.
The trend toward bureaucratization in all kinds of large
organizations blocks this vital source of democratic influ-
CIICE.
The proliferation of bureaucracies, then, threatens de-.
rnocracy in various ways. But even if we could abolish
them, we would be reluctant to do so, because we do not
wish to surrender the benefits we derive from them. Some
.. . ....
authors have concluded that modern society's need for

bureaucratic methods spells the inevitable doom of democ-
racy. But why interpret a historical dilemma as a sign of
an inescapable fate? Why not consider it a challenge to
find ways to avert the impending threat? If we want

to utilize efficient bureaucracies, we must find democratic
methods of controlling them lest they enslave us.
This is not an easy undertaking. Its accomplishment may
well depend on democratic participation OI] a far wider
scale than has ever been known. Perhaps, the challenge
posed by bureaucratization can be met only if all citizens
are able and motivated to devote a considerable portion of
their time and energy to activities in the political life of
their communities. Such a suggestion would have been
unrealistic a century or two ago, when most men had to
spend most of their waling hours making a living. The
efficiency of the very bureaucracies against which we
should protect democratic institutions, however, has re-
duced enormously the working week and increased the
number of hours people have at their free disposal. There-
tore, for the first time in history, all men, not merely a
privileged Jew, are freed, if they choose, to take their
duties as democratic citizens seriously. And as the level
of popular education is raised, more people are becoming
interested in political affairs. Many problems still lie ahead,
as we all know. The full realization of democracy in
modern society is a gigantic task. But would it not be a
pity to give up in despair just when -the tools needed for
completing it seem to be in our hands?
FOOTNOTES
CHAPTER 1
1. See Pitirim Sorokin, Social and Cultural Dynamics, New
* " r i c a n Book Company, 1937-1941. The author
traces fluctuations in cultural emphasis O11 science I and
rationality, on the one hand, and faith and supernatural
phenomena, on the other, from the earliest times to the
present, and w'gorously condemns the present trend toward
rationalization.
2. Capital, Vol. I, Chaps. 26 to 31.
3. See Kenneth Boulding, The Organizazional Revolution,

New York: Harper & Brother mu
4. From Max Weber: Essays In Sociology, translated by
mu. .
Gerthl nun Wright!lHklluu Oiord
.
University Press, 1946, I 228.
5. Ibid., p. 232.
119
6. For a fuller discussion of this point, see Charles H. Page,

"Bureaucracy and the Liberal Church," The Review of
Religion 171137-50 (1952).
CHAPTER 2
1. From Max Weber- Essays in Sociology, translated by
H. H. Gerth and C. Wright Mills, New York: Oxford
University Press, 1946, p. 196. By permission.
2. Max Weber, The Theory of Social and Economic Organi-
zation, translated by A. M. Henderson and Talcott Par~
sons, New York' Oxford University Press, 1947, p. 331.
3. bid., p. 330.
4. laid., p. 340.
5. Ibid., p. 334.
6. Ibid., p. 337.
7. From Max Weber: Essays in Sociology, op. cit., p. 214.
8. Robert K. Merton, Social Theory and .Social Structure,

Glencoe, Ill-: Free Press, 1949, pp. 21-81.
9. For a general discussion of functional analysis, see Ely
ChicO}', Sociological Perspective§ Basic Concepts and Their
Application(Studies in Sociology), New York' Ran-
dom House, Inc., 1954, Chap. 5.
10. Chester I. Barnard, The Functions of the Executive, Cam-

bridge: Harvard University Press, 1948, p. 123.
.
11 From Max Weber: Essays in Sociology, op. cit., pp. 256-57.
The advanced student will have recognized the indebted-
ness of the foregoing discussion to Weber's (pp. 204-16).
It goes without saying that Weber's fund of historical
knowledge and his profound theoretical insights about
bureaucracy can be acknowledged as outstanding contribu-
tions in the field even if one rejects his use of the ideal-
type construct.
Footnotes 121
12. For a fuller discussion of the unintended effects of Protes-

tantism, see Elizabeth K. Nottingham, Re figion and Society
(Studies in Sociology), New York: Random House,
Inc., 1954, pp. 50 ff.
13. Alvin W. Gouldner, Patterns of Industrial Bureaucracy,
Glencoe, Ill.: Free Press, 1954.
14. laid., pp. 97-98.
CHAPTER 3
1. Charles H. Page, "Bureaucracy's Other Face," Social

Forces 25:89-91 (1946). By permission.
2. Condensed from The HUman Group by George C. Ho-

mans, copyright, 1950, by Harcourt, Brace & Co., pp. 54-55,
66, 68-72. The empirical study of the Bank Wiring Ob-
servation Room, which Homans summarizes in this book
and in which he participated, is fully reported in F. J.
Roethlisberger and William I. Dickson, Management and
the Worker, Cambridge: Harvard University Press, 1946,
pp. 379-548.
3. Peter M. Blau, The Dynamics of Bureaucracy, Chicago:

University of Chicago Press, 1955, pp. 99, 104-6, 108-11,
113. (Copyright, 1955, by the University of Chicago.
Reprinted by permission of the University of Chicago
Press.)
4. in ustance, thlisberger and Dickson, op. cit.,

pp. l and Elton Mayo, The Human Problems of an
Industrial Civilization, New York: Macmillan Company,
m.
5. See Frederick W. Taylor, as Principles of Scientific
Management, New i n: Harper & Brothers, 1911.
6. Wilbert E. Moore, Industrial Relations and the Social
Order, New York: Macmillan Company, 1947, p. 190.
7. For. a full discussion of this case, see Blau, op. cit., pp.
49-67.
CHAPTER 4
1. Daniel Katz, Nathan MacCoby,. .and Nancy C. Morse,
Productfviry, Supervision ~and Morale in an Office Situs
dion, Ann Arbor: Institute for Social Research, University
of Michigan, 1950, especially pp. 17, 21, 29.
25. See, for instance, F J. Rbéthlisberger and William I.
Dickson, Management and the_ Worker, Cambridge: Har-
vard University Press, 1946, p`p. 452-53.~
3. For a clear illustration of this point in a street comer
gang, see William F. Whyte, Street Corner Society, Chi-
cage' University of Chicago Press, 1943, pp. 257-62.
4. For discussions of professional authority, see Talcott Par-

sons, The Social System, Glencoe, Ill.: Free Press, 1951,
pp. 428-79, the same author's introduction to Max Weber,
The Theory OJ' Social and Economic Organization, trans-
lated by A. M. Henderson and Taicott Parsons, New
York: Oxford University Press, 1947, pp. 58-59, It. 4, and
Alvin W. Gouldner, Patterns of Industrial Bureaucracy,
Glencoe, Ill.: Free Press, 1954, pp. 21-23.
5. See Peter M. Blah, The Dynamics of Bureaucracy, Chi-
cago: University of Chicago Press, 1955, pp. 33-48.
CHAPTER 5
1. Walter R. Sharp, The French Civil Service' Bureaucracy
in Transition, New York: Macmillan Company, 1931, pp.
446-50 (by permission); reprinted in Robert K. Merton
Ailsa P. Gray, Barbara Hockey, and Hanan C. Selvin
(eds.), Reader in Bureaucracy, Glencoe, IH.: Free Press,
1952, pp. 407-9.
2. Robert K. Merton, Social Theory and Social Structure,
Glencoe, HI.: Free Press, 1949, pp. 154-55. By permission.
3. Karl Marx, The Poverty of Philosophy, New York, Inter-

national Publishers, n.d., p. 145, quoted in Reinhard Bendix
Footnotes 123
r
and Seymour M. Lipset, Class, Status and Power, Glencoe,

Ill.: Free Press, 1953, p, 31.
4. Robert Michels, Political Parties, Glencoe, Ill.: Free Press,

1949, P. 373.
5. See Alvin W. Gouldner, "Attitudes of Trogressive' Trade-
Union Leaders," American Journal of Sociology 52:389-
92 (1947).
6. See Peter M. Blau, The Dynamics of Bureaucracy, Chi-

cago: University of Chicago Press, 1955, pp. 194-96.
'7. Philip Selznick, T V A and the Grass Roots, Berkeley and

Los Angeles: University of California Press, 1949.
8. Ibid., p. 204.
9. S. M. Lipset, Agrarian Socialism, Berkeley and Los A11-

gelesz University of California Press, 1950, pp. 255-75. By
permission.
10. Ibid., PP- 263, 265, 266-67.

r.
CHAPTER 6
1. Alvin W. Gouldner, "Red Tape as a Social Problem," in
Robert K. Merton, Ailsa P. Gray, Barbara Hockey, and
Hanan C. Selvin (eds.), Reader in Bureaucracy, Glencoe,
Ill.: Free Press, 1952, pp. 410-18.
2. Adolf A- Berle, Jr., and Gardiner C. Means, The Modern
Corporation and Private Property, New York: Macmillan
Company, 1932.
3. For a recent discussion of the bureaucratization of in-

dustry and trade, see Wilbert B. Moore, Economy and
Society; (Studies In Sociology), New York: Random
I
House, 1955, E
4. S. M. Lipset, Martin Trow, and James Coleman, Union

Democracy (Glencoe, Ill.: Free Press, 1956).
5. See Franz Neumann, Behemoth, New York: Oxford Uni-

versity Press, 1942, and Robert A. Brady, Business as a
System of Power, New York: Columbia University Press,
1943.
6. See Ludwig V011 Mises, Bureaucracy, New Haven: Yale
University Press, 1946; and Friederich V011 Hayek, The
Road to Serfdom, Chicago: University o f Chicago Press,
1944.
7. For a full discussion of the points made in this paragraph
and a criticism of the two theories mentioned, see Rein-
hard Bendix, "Bureaucracy and the Problem of Power,"
Public Administration Review 51194-209 (1945).
J
$El.ECTED READINGS
Basic source book
ROBERT K. MBRTON, A1LSA P. GRAY, BARBARA HOCKBY,

and HANAN C. SELVIN (eds.), Reader in Bureaucracy,
Glencoe, Iii.: Free Press, 1952.
Selections from the classics, such as Weber and Micbels, and
many pertinent smdies of a more recent vintage. Includes a
bibliography.
Theories of bureaucracy and organization
From Max or: Essays in Sociology, translated by

mm. Gerth and C. Wright Mills, New York: Oxford
University Press, 1946.
Contains the classic essay on bureaucracy as well as essays
011 discipline, power, and authority which are pertinent to
the discussions in this study. The translators provide H
introduction to Weber's life and his writings.
125
126 BURBAUCRACY IN MODERN socmnr
MAX WEBER, The Theory of Social and Economic Organ-

ization, translated by A. M. Henderson and Talcott Par-
sons, New York: Oxford University Press, 1947.
The typology of authority in this book includes another
general discussion of bureaucracy which is not as complete
but somewhat more concise and clearer than that in the
volume above.
Ronsnr MICHBLS, Political Parties, Glencoe, Ill.: Free

Press, 1949.
There is no reason to agree with Michels's conclusion that
democracy is hardly more than a utopian dream, but there
are good reasons for reading his incisive analysis of parties
and trade unions. Unless we learn to understand why
democracy often does not work, how can we learn to make
it work better?
CHESTER I. BARNARD, The Functions of the Executive,

Cambridge: Harvard University Press, 1938.
An analysis of principles of organization in business by a
former executive of a large company, who was one of the
first to stress the importance of informal organization.
HERBERT A. SIMON, Administrative Behavior, New York .

Macmillan Company, 1945. ,
A stimulating text on principles of administration, which

views them as limits of the process of decision-making.
Case studies of bureaucracies
F. J. ROETHLISBHRGER and WILLIAM J. DicKson, Manage-

ment and the Worker, Cambridge: Harvard University
Press, 1939.
The famous study of informal organization in a variety of
small work groups at the Hawthorne Works of the Western
Electric Company.
Selected Readings 127
I
PHILIP SELZNICK, TVA and the Grass Roots, Berkeley and

Los Angeles: University of California Press, 1949.
Study of the way in which initial commitments of an or-
ganization to the existing power structure affected its opera-
tions in unintended ways.
ALVIN W. GOULDNER, Patterns of Industrial Bureaucracy,

Glencoe, • Free Press, 1954.
Empirical study of the forces that engendered bureaucratiza-
tion and its consequences in an industrial concern.
PETER M. BLAU, The Dynamics of Bureaucracy, Chicago:

University of Chicago Press, 1955.
Study of interpersonal relationships in two government agen-
cies with special emphasis on the analysis of processes of
bureaucratic change.
S. M. LIPSET, MARTIN TRow, and JAMES COLEMAN,

Union Democracy (Glencoe, Ill.: Free Press, 1956).
Historical and sociological study of both the bureaucratic
and democratic features of the International Typographical
Union. (See also Lipset's "The Political Process in Trade
Unions," Chapter 4 in Monroe Berger, Theodore Abel, and
Charles H. Page (eds.)° Freedom and Control in Modem
Society, New York: D. Van Nostrand, Inc., 1954.)

Bureaucracy in Modern Society (Peter M. Blau)

Uploaded by

Copyright:

Available Formats

Bureaucracy in Modern Society (Peter M. Blau)

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

Bureaucracy in Modern Society (Peter M. Blau)

Uploaded by

Copyright:

Available Formats

4

CHARLES H. PAGE, Provost, University of

© COPYRIGHT, 1956, BY RANDOM HOUSE, INC.

All rights reserved under International and Pan-American

LIBRARY OF CONGRESS CATALOG CARD NUMBER: 56-7690

MANUFACTURED IN THE UNITED STATES OF AMERICA

THE MOUNTING interest of social scientists in the study of

of changing social organization are inescapable for social

The moral and political implications of these facts are

or the provision of social services or the education of young

readings,* containing an impressive sample of past and

* Reader in Bureaucracy, Glencoe, Ill.: Free Press, 1952.

FOREWORD by Charles H. Page 5

I. Why Study Bureaucracy? 13

The Rationalization of Modem Life 14

II. Theory cmd Development of Bureaucracy 27

The Concept of Bureaucracy 28

Ill. Bureaucracy in Process 45

Bureaucracy's Other Face 46

Organization of Work Groups 53

IV. Bureaucratic Authority 69

Strategic Leniency and Authority 70

v. Bureaucracy and Social Change 85

Who Are the Ritualists? 86

VI. Bureaucracy and Democracy

The Accusation of "Red Tape" 102

SELECTED READINGS 125

Why Study Bureaucracy?

"THAT STUPID BU1ZNAUCRA*r!" Who has not felt this way

dreds of persons have the assignment of collecting taxes

The Rationulizotion of Modern Like

from the modem world.* This is largely the price of ra-

explain many natural phenomena but have also channeled

not Sammy's means for the attainment of certain ends but

Ford later refused to make further administrative innova-

agerial responsibility for coordination through direct con-

second store he may put his SOI1 in charge even though

SpecMcally, we shall examine Me consequences of bu-

The Value of Studying Bureaucracy

Today Bureaucracy is not a new phenomenon. It

In the United States, employment statistics illustrate

In a Democracy Bureaucracy, as the foremost theo-

Under normal conditions, the power position of a fully

"political master" Ends himself in the position of the

Totalitarianism is the polar case of such bureaucratic

are of primary importance, hence bureaucratization will

For Sociologists The study of bureaucratic organiza-

less popular, or that radicals (or any other group) are

them could be learned by examining organizational charts

In summary, the prevalence. o_t. bureaucracies in our

Theory and Development

ADVANCEMENT in any science tiepeniis OH developments in

tihc theory. Not every however, is a scientific

evidence that would clearly disprove it. An important

The Concept of Bureaucracy

son and to make every one of them responsible for the

2. "The organization of offices follows the principle of

. Operations are governed Wu consistent system of

order to settle the most complicated legal issues. Bureau-