The Science of Public Administration: Three Problems

Author(s): Robert A. Dahl

Source: Public Administration Review , Winter, 1947, Vol. 7, No. 1 (Winter, 1947), pp.
1-11
Published by: Wiley on behalf of the American Society for Public Administration

Stable URL: https://www.jstor.org/stable/972349

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/972349?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Wiley and American Society for Public Administration are collaborating with JSTOR to
digitize, preserve and extend access to Public Administration Review

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
 The Science of Public Administration:
Three Problems

By ROBERT A. DAHL
Department of Political Science
Yale University

T XHE effort to create a science of public ential one, is L. Urwick's contention that
administration has often led to the for-"there are certain principles which govern the
mulation of universal laws or, more com- association of human beings for any purpose,
monly, to the assertion that such universal just as there are certain engineering principles
laws could be formulated for public adminis- which govern the building of a bridge."3
tration.l In an attempt to make the science of Others argue merely that it is possible to dis-
public administration analogous to the natu- cover general principles of wide, although not
ral sciences, the laws or putative laws are necessarily of universal validity.4 Surely this
stripped of normative values, of the distor-more modest assessment of the role of public
tions caused by the incorrigible individual administration as a study is not, as an abstract
psyche, and of the presumably irrelevant ef- statement, open to controversy. Yet even the
fects of the cultural environment. It is often discovery of these more limited principles is
implied that "principles of public administra- handicapped by the three basic problems of
tion" have a universal validity independent values, the individual personality, and the so-
not only of moral and political ends, but cial
of framework.
the frequently nonconformist personality of
the individual, and the social and cultural set- Public Administration and Normative Values
ting as well.
THE first difficulty of constructing a science
Perhaps the best known expression of this of public administration stems from the
kind is that of W. F. Willoughby. Although he frequent impossibility of excluding normative
refused to commit himself as to the propriety considerations from the problems of public
of designating administration as a science, Wil- administration. Science as such is not con-
loughby nevertheless asserted that "in admin-
cerned with the discovery or elucidation of
istration, there are certain fundamental prin- normative values; indeed, the doctrine is gen-
ciples of general application analogous to those
erally, if not quite universally, accepted that
characterizing any science .. ."2 A more re- science cannot demonstrate moral values, that
cent statement, and evidently an equally influ-
science cannot construct a bridge across the
I See, for example, F. Merson, "Public Administra- great gap from "is" to "ought." So long as the
tion: A Science," i Public Administration 220 (1923); naturalistic fallacy is a stumbling block to
B. W. Walker Watson, "The Elements of Public Ad-
ministration, A Dogmatic Introduction," lo Public philosophers, it must likewise impede the
Ad-
progress of social scientists.
ministration 397 (1932); L. Gulick, "Science, Values and
Public Administration," Papers on the Science of Ad-Much could be gained if the clandestine
ministration, ed. by Gulick & Urwick, (Institute of Pub-
lic Administration, 1937); Cyril Renwick, "Public Ad- See fn. 12, infra, for the full quotation and citation.
ministration: Towards a Science," The Australian 4This I take to be Professor Leonard D. White's posi-
Quarterly (March 1944), p. 73. tion. See his "The Meaning of Principles in Public Ad-
2Principles of Public Administration (The Brook- ministration," in The Frontiers of Public Administra-
ings Institution, 1927), Preface, p. ix. tion (University of Chicago Press, 1936), pp. 13-25.

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
2 PUBLIC ADMINISTRATION REVIEW

smuggling of moral values intobethe erected. They serve

social sci-to condition and to
ences could be converted into open and complicate, but not to change the single ulti-
honest commerce. Writers on public adminis- mate test of value in administration."8
tration often assume that they are snugly insu- It is far from clear what Gulick means to
lated from the storms of clashing values; usu- imply in saying that "interferences with effi-
ally, however, they are most concerned with ciency" caused by ultimate political values
ends at the very moment that they profess to may "condition" and "complicate" but do not
be least concerned with them. The doctrine of "change" the "single ultimate test" of effi-
efficiency is a case in point; it runs like a half-ciency as the goal of administration. Is effi-
visible thread through the fabric of public ad-ciency the supreme goal not only of private
ministration literature as a dominant goal of administration, but also of public administra-
administration. Harvey Walker has stated thattion, as Gulick contends? If so, how can one
"the objective of administration is to secure say, as Gulick does, that "there are...
the maximum beneficial result contemplated highly inefficient arrangements like citizen
by the law with the minimum expenditure of boards and small local governments which may
the social resources."5 The term "social re- be necessary in a democracy as educational de-
sources" is sufficiently ambiguous to allow vices"?
for Why speak of efficiency as the "single
almost any interpretation, but it suggests ultimate
that test of value in administration" if it
the general concept involved is one of maxi- is not ultimate at all-if, that is to say, in a
mizing "output" and minimizing "cost." conflict Like- between efficiency and "the demo-
wise, many of the promised benefits of admin- cratic dogma" (to use Gulick's expression) the
istrative reorganization in state governments latter must prevail? Must this dogma prevail
are presumed to follow from proposed only
im-because it has greater political and social
provements in "efficiency in operation."force Andbehind it than the dogma of efficiency;
yet, as Charles Hyneman has so trenchantly or ought it to prevail because it has, in some
observed, there are in a democratic society sense, greater value? How can administrators
other criteria than simple efficiency in opera- and students of public administration dis-
tion.6 criminate between those parts of the demo-
Luther Gulick concedes that the goal of cratic effi- dogma that are so strategic they ought
ciency is limited by other values. to prevail in any conflict with efficiency and
those that are essentially subordinate, irrele-
In the science of administration, whether public vant,oror even false intrusions into the demo-
private, the basic "good" is efficiency. The fundamental
cratic hypothesis? What is efficiency? Belsen
objective of the science of administration is the ac-
and Dachau were "efficient" by one scale of
complishment of the work in hand with the least ex-
values.
penditure of man-power and materials. Efficiency is And in any case, why is efficiency the
thus axiom number one in the value scale of admin- ultimate test? According to what and whose
istration. This brings administration into apparent con- scale of values is efficiency placed on the high-
flict with certain elements of the value scale of politics,
whether we use that term in its scientific or in its pop-
est pedestal? Is not the worship of efficiency it-
ular sense. But both public administration and politics self a particular expression of a special value
are branches of political science, so that we are in thejudgment? Does it not stem from a mode of
end compelled to mitigate the pure concept of efficiency thinking and a special moral hypothesis rest-
in the light of the value scale of politics and the social
order."
ing on a sharp distinction between means and
ends?
He concludes, nevertheless, "that these inter- The basic problems of public administra-
ferences with efficiency [do not] in any way tion as a discipline and as a potential science
eliminate efficiency as the fundamental valueare much wider than the problems of mere
upon which the science of administration mayadministration. The necessarily wider preoc-
6Public Administration (Farrar & Rinehart, 1937), cupation of a study of public administration,
p. 8. as contrasted with private administration, in-
6 "Administrative Reorganization," i The Journal of evitably enmeshes the problems of public
Politics 62-65 (1939).
7 Op. cit., pp. 192-93. 8 p. cit., p. 193.

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
 SCIENCE OF PUBLIC ADMINISTRATION 3

administration in the toils of ethical considera- sistency opposed the growth of delegated legis-
tions. Thus the tangled question of the right lation and the expansion of the powers of
of public employees to strike can scarcely be administrative tribunals-no doubt from a con-
answered without a tacit normative assump- viction that previously existing economic
tion of some kind. A pragmatic answer is satis- rights and privileges are safer in the courts
factory only so long as no one raises the ques- than in administrative tribunals; whereas
tion of the "rights" involved. And to resolve those who support this expansion of adminis-
the question of rights merely by reciting legal trative power and techniques generally also
norms is to beg the whole issue; it is to confess favor a larger measure of economic regulation
that an answer to this vital problem of public and control. Much of the debate that has been
personnel must be sought elsewhere than with phrased in terms of means ought more prop-
students of public administration. Moreover, erly to be evaluated as a conflict over general
if one were content to rest one's case on legal social goals.
rights, it would be impossible to reconcile in a One might justifiably contend that it is the
single "science of public administration" the function of a science of public administration,
diverse legal and institutional aspects of the not to determine ends, but to devise the best
right to strike in France, Great Britain, and means to the ends established by those agen-
the United States. cies entrusted with the setting of social policy.
The great question of responsibility, cer- The science of public administration, it might
tainly a central one to the study of public ad-be argued, would be totally nonnormative,
ministration once it is raised above the level of and its doctrines would apply with equal va-
academic disquisitions on office management,lidity to any regime, democratic or totalitar-
hinges ultimately on some definition of ends,ian, once the ends were made clear. "Tell me
purposes, and values in society. The sharpwhat you wish to achieve," the public admin-
conflict of views on responsibility expressedistration scientist might say, "and I will tell
several years ago by Carl Friedrich and Her- you what administrative means are best de-
man Finer resulted from basically different signed for your purposes." Yet even this view
interpretations of the nature and purposes of has difficulties, for in most societies, and par-
democratic government. Friedrich tacitly as-ticularly in democratic ones, ends are often in
sumed certain values in his discussion of the dispute; rarely are they clearly and unequivo-
importance of the bureaucrat's "inner check" cally determined. Nor can ends and means ever
as an instrument of control. Finer brought be sharply distinguished, since ends determine
Friedrich's unexpressed values into sharp focus means and often means ultimately determine
and in a warm criticism challenged their com- ends.10
patibility with the democratic faith.9 The student of public administration can-
It is difficult, moreover, to escape the con- not avoid a concern with ends. What he ought
clusion that much of the debate over delegated to avoid is the failure to make explicit the
legislation and administrative adjudication,
ends or values that form the groundwork of his
both in this country and in England, actually doctrine. If purposes and normative consid-
erations were consistently made plain, a net
arises from a concealed conflict in objectives.
Those to whom economic regulation and con- gain to the science of public administration
trol are anathema have with considerable con- would result. But to refuse to recognize that
the study of public administration must be
9C. J. Friedrich, "Public Policy and the Nature of founded on some clarification of ends is to
Administrative Responsibility," in Public Policy (Har-
vard University Press, 1940); Herman Finer, "Adminis- perpetuate the gobbledygook of science in the
trative Responsibility in Democratic Government," 1 area of moral purposes.
Public Administration Review 335 (1940-41). See also A science of public administration might
Friedrich's earlier formulation, which touched off the proceed, then, along these lines:
dispute, "Responsible Government Service under the
i. Establishing a basic hypothesis. A nonnormative
American Constitution," in Problems of the American
Public Service (McGraw-Hill Book Co., 1935); and 0 See Aldous Huxley's discussion in Ends and Means
Finer's answer to Friedrich in 51 Political Science Quar- (Harper & Bros., 1937), and Arthur Koestler, The Yogi
terly 582 (1936). and the Commissar (Macmillan Co., 1945).

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
4 PUBLIC ADMINISTRATION REVIEW

science of public administration might

This rest on with
concern a basic
human behavior greatly
hypothesis that removed ethical problems from the area
limits the immediate potentialities of a science
covered by the science. The science of public adminis-
tration would begin where the basicof hypothesis
public administration.
leaves First, it diminishes
the possibility
off. One could quarrel with the moral of using experimental proce-
or metaphysical
dures;
assumptions in the basic hypothesis; but and experiment, though perhaps not
all normative
argument would have to be carried on at that level,
indispensable and
to the scientific method, is of
not at the level of the science. The science, as such,
enormous aid. Second, concern with human
would have no ethical content.
behavior seriously limits the uniformity of
Can such a basic hypothesis be created? To this
writer the problem appears loaded with enormous data,
and since the datum is the discrete and
highly
perhaps insuperable difficulties; yet it is unlikely that a variable man or woman. Third, be-
science of public administration will ever be possible
cause the data concerning human behavior
until this initial step is taken.
constitute an incredibly vast and complex
2. Stating ends honestly. Some problems of the public
services, like that of responsibility, evidently cannot be the part played by the preferences of the
mass,
observer is exaggerated, and possibilities of in-
divorced from certain ends implied in the society
dependent
served by the public services. If this is true, there can verification are diminished. Fourth,
never be a universal science of public administration
concern with human action weakens the relia-
so long as societies and states vary in their objectives.
bility of all "laws of public administration,"
In all cases where problems of public administration
are inherently related to specific social ends andsince
pur- too little is known of the mainsprings of
human action to insure certitude, or even high
poses, the most that can be done is to force all norma-
probability,
tive assumptions into the open, and not let them lie half in predictions about man's con-
concealed in the jungle of fact and inference to duct.
slaughter the unwary.
All these weaknesses have been pointed out
so often in discussing the problems of the so-
Public Administration and Human Behavior
cial sciences that it should be unnecessary to
SECOND major problem stems from the in- repeat them here. And yet many of the sup-
escapable fact that a science of public ad-
posed laws of public administration and much
ministration must be a study of certain aspects
of the claim to a science of public administra-
of human behavior. To be sure, there are parts
tion derive from assumptions about the na-
of public administration in which man's be- ture of man that are scarcely tenable at this
havior can safely be ignored; perhaps it is pos-
late date.
sible to discuss the question of governmentalThe field of organizational theory serves as
accounting and auditing without much con- an extreme example, for it is there particu-
sideration of the behavior patterns of govern-
larly that the nature of man is often lost sight
mental accountants and auditors. But most of in the interminable discussions over ideal-
problems of public administration revolve
ized and abstract organizational forms. In this
around human beings; and the study of public
development, writers on public administra-
administration is therefore essentially a study
tion have been heavily influenced by the ra-
of human beings as they have behaved, and as character that capitalism has imposed
tional
they may be expected or predicted to behave,
on the organization of production, and have
under certain special circumstances. What ignored the irrational qualities of man him-
marks off the field of public administration self.
from psychology or sociology or political Capitalism,
in- especially in its industrial form,
stitutions is its concern with human behavior
was essentially an attempt to organize produc-
in the area of services performed by govern- tion along rational lines. In the organization
mental agencies.1l of the productive process, the capitalistic en-
u See Ernest Barker's excellent and useful distinctions trepreneur sought to destroy the old restrictive
between state, government, and administration, in The
Development of Public Services in Western Europe, duly 'served.' Every right and duty implies a corre-
sponding 'service'; and the more the State multiplies
1660-i930 (Oxford University Press, 1944), p. 3. Admin-
istration "is the sum of persons and bodies who arerights
en- and duties, the more it multiplies the necessary
services of its ministering officials." See also Leon
gaged, under the direction of government, in discharg-
Duguit, Law in the Modern State (B. W. Huebsch,
ing the ordinary public services which must be rendered
daily if the system of law and duties and rights is 1919),
to beCh. II.

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
 SCIENCE OF PUBLIC ADMINISTRATION 5

practices and standards of feudalism and mer-

the motive underlying persistence in bad
cantilism; to rid the productive process structure it
ofisthe
always more hurtful to the
inherited cluster of methods and technics that greatest number than good structure."'3
characterized the guilds and medieval crafts- Sweeping generalizations such as these gave
men; in short, to organize production accord- promise of a set of "universal principles": i.e.,
ing to rational rather than traditional con- a science. American students of public admin-
cepts. Combined with a new acquisitive ideal, istration could not fail to be impressed.
this rational approach to production trans- Aside from the fact that Urwick ignored the
formed not only the whole economic process whole question of ends, it is clear that he also
but society itself. The rapid growth of mech- presupposed (though he nowhere stated what
anization, routine, and specialization of labor sort of human personality he did presuppose)
further increased the technically rational an essentially rational, amenable individual;
quality of capitalist production. It was per- he presupposed, that is to say, individuals who
haps inevitable that concepts should arise would accept logical organization and would
which subordinated individual vagaries and not (for irrelevant and irrational reasons) re-
differences to the ordered requirements of the bel against it or silently supersede it with an
productive process: for it was this very subor- informal organization better suited to their
dination that the replacement of feudal and personality needs. Urwick must have supposed
mercantilist institutions by capitalism had ac- this. For if there is a large measure of irration-
complished. The organization (though not the ality in human behavior, then an organiza-
control) of production became the concern of tional structure formed on "logical" lines may
the engineer; and because the restrictive prac- in practice frustrate, anger, and embitter its
tices authorized by tradition, the protective personnel. By contrast, an organization not
standards of the guilds, the benevolent regula- based on the logic of organizational principles
tions of a mercantilist monarchy, and even the
may better utilize the peculiar and varying
non-acquisitive ideals of the individual had personalities of its members. Is there any evi-
all been swept away, it was actually feasible to dence to suggest that in such a case the "logi-
organize production without much regard for cal" organization will achieve its purposes in
the varying individual personalities of those some sense "better" or more efficiently than
in the productive process. The productive the organization that adapts personality needs
process, which to the medieval craftsman was to the purposes of the organization?14 On what
both a means and an end in itself, became
wholly a means. added.) See also his "Executive Decentralisation with
Ultimately, of course, men like Taylor pro- Functional Co-ordination," 13 Public Administration
vided an imposing theoretical basis for regard- 344 (1935), in which he sets forth "some axioms of
organisation," among others that "there are certain
ing function, based on a logical distribution
principles which govern the association of human be-
and specialization of labor, as the true basis of
ings for any purpose, just as there are certain engineer-
organization. Men like Urwick modified and ing principles which govern the building of a bridge.
carried forward Taylor's work, and in the Such principles should take priority of all traditional,
process have tremendously influenced writers personal or political considerations. If they are not ob-
on public administration. Urwick, so it must served, co-operation between those concerned will be
less effective than it should be in realising the purpose
have appeared, provided a basis for a genuine for which they have decided to co-operate. There will
science of administration. "There are prin- be waste of effort." (Italics added.) See also his criticisms
ciples," he wrote, "which should govern ar- of the "practical man fallacy," p. 346.
rangements for human association of any kind. 13 Ibid., p. 85.
These principles can be studied as a technical 14 See John M. Gaus's excellent definitions: "Organ-
ization is the arrangement of personnel for facilitating
question, irrespective of the purpose of the en-
the accomplishment of some agreed purpose through
terprise, the personnel composing it, or any the allocation of functions and responsibilities. It is the
constitutional, political, or social theory un- relating of efforts and capacities of individuals and
derlying its creation."'2 And again, "Whatever groups engaged upon a common task in such a way as
to secure the desired objective with the least friction
1 L. Urwick, "Organization as a Technical Problem," and the most satisfaction to those for whom the task is
Papers on the Science of Administration, p. 49. (Italics done and those engaged in the enterprise.... Since

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
6 PUBLIC ADMINISTRATION REVIEW

kind of evidence are we compelleddence):to "The idea that organizations should

assume
be built up
that the rationality of organizational round and adjusted to individual
structure
idiosyncracies,
will prevail over the irrationality of man?rather than that individuals
Patently the contention that oneshould be adapted
system ofto the requirements of
organization is more rational sound
than another,
principles of organization, is . . .fool-
ish ... ."
and therefore better, is valid only The
(a) ifHawthorne
indi- experiment dem-
viduals are dominated by reason onstrated,
or (b) if on they
the contrary, that ". . . no
are so thoroughly dominated by study
the of technical
human situations which fails to take
process (as on the assembly line,account of the non-logical
perhaps) that social routines can
their individual preferences may hopesafely
for practical success."'7
be ig-
nored. However much the latterInassumption
1939, Leonard White seriously qualified
might apply to industry (a matterthe principle of subordinating individuals to
of consid-
structure
erable doubt), clearly it has little by adding the saving phrase of the
application
neo-classical
to public administration, where economists: "in the long run."
technical
processes are, on the whole, of"To what extent,"
quite subordi- he said, "it is desirable to
nate importance. As for the firstrearrange structure
assumption, itin preference to replacing
personnel
has been discredited by all the is a practical
findings of matter to be deter-
mined
modern psychology. The science of in the light of special cases. In the long
organiza-
run,
tion had learned too much from the demands
industry andof sound organization re-
not enough from Freud. quire the fitting of personnel to it, rather than
sacrificing
The more that writers on public normal organizational relationships
adminis-
tration have moved from the classroom to the to the needs or whims of individuals."'l In
administrator's office, the more Urwick's uni-
the same year, Macmahon and Millett went far
versal principles have receded. As early beyond
as the customary deductive principles of
public administration theory by making an
1930, in a pioneering work, Harold Lasswell
described the irrational and unconscious ele- actual biographical study of a number of fed-
ments in the successful and unsuccessful ad- eral administrators.19 In the most recent text
ministrator.1' Meanwhile, experiments in on
thepublic administration, the importance of
Hawthorne plant of Western Electric Com- personality is frankly admitted. ". . . admin-
istrative research," say the authors, "does not
pany were indicating beyond doubt that indi-
vidual personalities and social relationships
17 Urwick, op. cit., p. 85, and Henderson, et al., p. 155.
had great effects even on routinized workUrwick
in has set up a false dilemma that makes his
industry. Increased output was the result choice
of more persuasive. Actually, the choice is not be-
tween (a) wholly subordinating organizational struc-
"the organization of human relations, rather
ture to individual personalities, which obviously might
than the organization of technics."'6 Urwick
lead to chaos or (b) forcing all personalities into an ab-
had said (with little or no supporting evi-
stractly correct organizational structure which might
(and often does) lead to waste and friction. There is a
organization consists of people brought into a certain third choice, (c) employing organizational structure
relationship because of a humanly evolved purpose, andit personalities to the achievement of a purpose. By
is clear that it should be flexible rather than rigid. excluding purpose, Urwick has, in effect, set up organ-
There will be constant readjustments necessary because ization as an end in itself. An army may be organized
of personalities and other natural forces and because more efficiently (according to abstract organizational
of the unpredicted and unpredictable situations principles)
con- than the political structure of a democratic
fronted in its operations." "A Theory of Organization state, but no one except an authoritarian is likely to
in Public Administration," in The Frontiers of Public contend that it is a superior organization-except for
Administration, pp. 66-67. the purposes it is designed to achieve. Yet once one ad-
15 Psychopathology and Politics (University of Chi- mits the element of purpose, easy generalizations about
cago Press, g93o), Ch. 8 "Political Administrators." organizational principles become difficult if not impos-
"I L. J. Henderson, T. N. Whitehead, and Elton Mayo, sible; and the admission presupposes, particularly in
"The Effects of Social Environment," in Papers on the the case of public organizations, a clear statement of
Science of Administration, op. cit., p. 149. It is worth ends and purposes.
noting that this essay properly interpreted contradicts 18 Leonard White, Introduction to the Study of Public
the implicit assumptions of virtually every other essay Administration (Macmillan Co., 1939), p. 38.
in that volume; and it is, incidentally, the only wholly 19 A. W. Madmahon and J. D. Millett, Federal Admin-
empirical study in the entire volume. istrators (Columbia University Press, 1939).

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
 SCIENCE OF PUBLIC ADMINISTRATION 7

seek its goal in the formulation ofOrmechanical does it mean rather that in plumbing hu-
rules or equations, into which human man behavior
behav- the researcher must be capable
ior must be molded. Rather, it looks toward of using the investigations of the psychiatrist
the systematic ordering of functions and hu- and sociologist? The need for specialization-
man relationships so that organizational deci- a need, incidentally, which science itself seems
sions can and will be based upon the certainty to impose on human inquiry-suggests that
that each step taken will actually serve the pur- the latter alternative must be the pragmatic
pose of the organization as a whole."20 And one answer.
whole chapter of this text is devoted to infor- Development of a science of public ad
mal organizations-the shadow relationships tration implies the development of a
that frequently dominate the formal structure of man in the area of services adminis
of the organization. the public. No such development
Thus by a lengthy and circumspect route, brought about merely by the constantly
man has been led through the back door and ated assertion that public administr
readmitted to respectability. It is convenient already a science. We cannot achieve a
to exile man from the science of public ad- by creating in a mechanized "adminis
ministration; it is simpler to forget man and man" a modern descendant of the eig
write with "scientific" precision than to re- century's rational man, whose only ex
is in books on public administrat
member him and be cursed with his madden-
ing unpredictability. Yet his exclusionwhose
is only activity is strict obedience
versal laws of the science of administration."
certain to make the study of public administra-
tion sterile, unrewarding, and essentially un-
real.
Public Administration and the Social Setting
If there is ever to be a science of public ad-
ministration it must derive from an under- IF WE know precious little about "administra-
standing of man's behavior in the area marked tive man" as an individual, perhaps we
know even less about him as a social animal.
off by the boundaries of public administra-
tion. This area, to be sure, can never be Yet we cannot afford to ignore the relation-
clearly separated from man's behavior in othership between public adminstration and its so-
cial setting.
fields; all the social sciences are interdepend-
ent and all are limited by the basic lack of un- No anthropologist would suggest that a so-
derstanding of man's motivations and re- cial principle drawn from one distinct culture
sponses. Yet the ground of peculiar concern is likely to be transmitted unchanged to an-
for a prospective science of public administra- culture; Ruth Benedict's descriptions of
other
tion is that broad region of services adminis-the Pueblo Indians of Zufii, the Melanesians
tered by the government; until the manifold of Dobu, and the Kwakiutl Indians of Van-
couver Island leave little doubt that cultures
motivations and actions in this broad region
can be integrated on such distinctly different
have been explored and rendered predictable,
lines
there can be no science of public administra- as to be almost noncomparable.21 If the
tion. nation-states of western civilization by no
means possess such wholly contrasting cultures
It is easier to define this area in space than
as the natives of Zufii, Dobu, and Vancouver
in depth. One can arbitrarily restrict the pros-
pective science of public administration toIsland,
a nevertheless few political scientists
certain region of human activity; but one can-would contend that a principle of political or-
not say with certainty how deeply one must ganization drawn from one nation could be
mine this region in order to uncover its se- adopted with equal success by another; one
crets. Does concern with human behavior would scarcely argue that federalism has
mean that the researcher in public administra-
everywhere the same utility or that the uni-
tion must be a psychiatrist and a sociologist?
tary state would be equally viable in Britain
0 Fritz Morstein Marx, ed., Elements of Publicand
Ad-the United States or that the American
ministration (Prentice-Hall, 1946), p. 49. (Italics
added.). 1 Patterns of Culture (Houghton Mifflin Co., 1934).

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
8 PUBLIC ADMINISTRATION REVIEW

not rest
presidential system would operate on a thorough examination of de-
unchanged
in France or Germany. velopmental and environmental differences.
The manifest
There should be no reason for benefits and merits of the Brit-
supposing,
then, that a principle of public ish administrative class have sometimes led
administration
has equal validity in every nation-state,
American students or of public administration t
that successful public administration
suggest practices
the development of an administrative
in one country will necessarilyclass
prove success-
in the American civil service; but propo
sals of
ful in a different social, economic, and this kind have rarely depended on
politi-
cal environment. A particular nation-state em-
thorough comparison of the historical factor
that made episodes,
bodies the results of many historical the administrative class a successful
traumas, failures, and successes achievement
which have in
in Britain, and may or may not
turn created peculiar habits, mores, institu-
be duplicated here. Thus Wilmerding has vir-
tionalized patterns of behaviour, Weltan-
tually proposed the transfer to the United
States psycholo-
schauungen, and even "national of all the detailed elements in the Brit-
gies."22 One cannot assume thatish civil service;
public admin- although he does not explic-
istration can escape the effectsitly ofbase
this hiscondi-
proposals on British experience
tioning; or that it is somehow independent except in a few instances,
of they follow British
and isolated from the culture or practices
social withsetting almost complete fidelity.23
in which it develops. At the White same hastime,
likewiseas argued for the creation of
value can be gained by a comparative an "administrative
study of corps" along the lines of
government based upon a due respect the British for administrative
dif- class. He has sug-
ferences in the political, social,gested and that reform of the civil service in Brit-
economic
environment of nation-states, so aintoo
and the
creationcom-of an administrative class
parative study of public administration were accomplished
ought in little more than tw
to be rewarding. Yet the comparative generations; profiting by British experience
aspects
of public administration have largely he argues, been we ought
ig- to be able to accomplish
nored; and as long as the studysuch ofa reform
public in even
ad-shorter time.24 Since the
ministration is not comparative, claims
question of anfor "a
administrative class is perhaps
science of public administration" sound the outstanding case where American writers
rather hollow. Conceivably there might be a on public administration have employed the
science of American public administration comparative method to the extent of borrow-
and a science of British public administration ing from foreign experience, it is worthy of a
and a science of French public administration; brief analysis to uncover some of the problems
but can there be "a science of public adminis- of a comparative "science of public adminis-
tration" in the sense of a body of generalized tration." For it throws into stark perspective
principles independent of their peculiar na- the fundamental difficulties of drawing uni-
tional setting? versal conclusions from the institutions of any
Today we stand in almost total ignorance of
the relationship between "principles of public 23 Lucius Wilmerding, Jr., Government by Merit (Mc-
Graw-Hill Book Co., 1935).
administration" and their general setting. Can 24 "The British civil service, which the whole world
it be safely affirmed, on the basis of existing now admires, went through nearly twenty years of
knowledge of comparative public administra- transition before its foundations even were properly
tion, that there are any principles independ- laid. It went through another twenty years of gradual
ent of their special environment? adjustment before the modern service as we know it to-
The discussion over an administrative class day was fully in operation. ... In the light of British
experience, and by taking advantage of modern knowl-
in the civil service furnishes a useful example edge about large-scale organization, we can easily save
of the difficulties of any approach that does the twenty years in which the British were experiment-
ing to find the proper basis for their splendid service.
2 See the fragmentary but revealing discussion on We na-shall, however, need ten years of steady growth,
tional differences in Human Nature and Enduring consciously guided and planned, to put a new admin-
Peace (Third Yearbook of the Society for the Psycholog- istrative corps into operation, and probably another
ical Study of Social Issues) Gardner Murphy, ed. ten years before it is completely installed." Government
(Houghton Mifflin, 1945). Career Service (University of Chicago Press, 1935), p. 8.

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
 SCIENCE OF PUBLIC ADMINISTRATION 9

one country, and at the same time may sharply out-

well question whether it would be so easy
lines the correlative problem of comparing the
to create an administrative class in any society,
institutions of several nations in order to de- like the American, where egalitarianism is so
rive general principles out of the greater range firmly rooted as a political dogma; however
of experiences. desirable such a class may be, and however
little it may actually violate the democratic
The central difficulty of universal generaliza-
tions may be indicated in this way: An admin- ideal, one is entitled to doubt that the overt
istrative class based on merit rests upon four
creation of an administrative elite is a practi-
conditions. All of these prerequisites were cal possibility in American politics.25 In any
present coincidentally in Britain in the mid-
case, the idea must be fitted into the peculiar
mores and the special ethos of the United
nineteenth century; and none of them is pres-
ent in quite the same way here. States, and cannot be lightly transferred from
First of all, an administrative class of the Britain to this country.26
British type rests upon a general political ac- Second, the administrative class idea rests
ceptance of the hierarchical idea. This accept-upon a scholastic system that creates the edu-
ance in Britain was not the product of fortycated nonspecialist, and a recruiting system
years; it was the outcome of four centuries. Itthat selects him. Too often, the proposal has
is not too much to say that it was the four cen- been made to recruit persons of general rather
turies during which the public service was thethan specialized training for an "administra-
particular prerogative of the upper classes thattive corps" without solving the prior problem
made a hierarchical civil service structure fea- of producing such "generalists" in the univer-
sible in Britain. The Tudor monarchy had sities. The British public school system and
rested upon a combination of crown power the universities have long been dominated by
administered under the King by representa- the ideal of the educated gentleman; and for
tives of the upper middle and professional centuries they have succeeded admirably in
classes in the towns and newly created mem- producing the "generalist" mind, even when
bers of the gentry in the country; Tudor au- that mind is nourished on apparently special-
thority was in effect derived from an alliance ized subjects. It is a peculiarly British paradox
of King and upper middle classes against the that persons of high general ability are re-
aristocracy. From the Revolution of 1688 un- cruited into the civil service by means of ex-
til 1832, public service was the special domain aminations that heavily weight such speciali-
of an increasingly functionless aristocracy ties as classical languages and mathematics. In
whose monopoly of public office was tacitly
supported by the upper middle classes of the 5 Significantly, the most recent study of reform of the
American civil service states, "We do not recommend
cities. Whatever the Reform Bill of 1832 ac-
the formation of a specially organized administrative
complished in terms of placing the urban oli- corps for which a special type of selection and training
garchy overtly in office, no one in Britain had is proposed." Report of President's Committee on Civil
many illusions that a change in the hierarchi- Service Improvement (Government Printing Office,
cal structure of politics and public service was 1941), p. 57. Instead, the Committee recommends that
"all positions whose duties are administrative in nature,
entailed. The upper middle classes were no in grades CAF-i1, P-4, and higher . . . be identified as
more keen than were the landed gentry of the an occupational group within the existing classification
eighteenth century to throw open the doors of structure." This is a noteworthy step in an attempt to
public service and politics to "the rabble." achieve the advantages of an administrative class within
Out of this long historic background the idea the framework of American mores and institutions. It
of an administrative class emerged. The un-is therefore a great advance over the earlier proposal in
the Report of the Commission of Inquiry on Public
spoken political premises of the dominant Service Personnel, Better Government Personnel (Mc-
groups in the nation reflected an acceptance of Graw-Hill Book Co., 1935), which recommended the
hierarchy in the social, economic, and politi- outright creation of a distinct administrative class (p.
cal structure of Britain; the contention, com- 30).
mon in the American scene, that an adminis- :8 This was the essential point, stated in more specific
terms, of Lewis Meriam's criticism of the administrative
trative class is "undemocratic" played no real corps idea. See his excellent Public Service and Special
part in mid-nineteenth century Britain. One Training (University of Chicago Press, 1936).

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
 10 PUBLIC ADMINISTRATION REVIEW

so far as this country has an educational

torate that was ideal
now too large to control.28
(a question on which this writer speaks the
Meanwhile, with
development of dissolution as
considerable trepidation), it appears
a powerto be, or
available to the Prime Minister upon
to have been, the ideal of the specialist.
his request from Muchthe Crown gave the executive
more is involved, too, than a question
a means ofof edu-
party discipline and control far
cation; at base the problem ismore oneeffective
of social
than the promise of office. Fi-
mores that give the specialist a nally,
prestige and to
the accession a power of the manufac-
social utility that no person of turing
general educa-
and trading classes by the reforms of
tion is likely to attain. That the recruiting
1832 placed a new emphasis on efficiency, both
process has been forced to adapt as itself
a means to the
of cutting down public expenses
educational specialization characteristic of
and insuring economies in government, and
American universities (indeed, one might
(especially after say
1848) of warding off the revo-
of American life) is scarcely astonishing. It might develop out of
lutionary threat that
would be more astonishing if the Civil Service
governmental incompetence.29 All these condi-
Commission were able to recruit nonexistent tions made possible, and perhaps inevitable,
"generalists" to perform unrecognized func-the substitution of merit for patronage. To
tions within a corps of practitioners where al-talk as if reform arose out of some change in
most everyone regards himself as a subject- public morality, obscure and mysterious in
matter specialist.27 origin but laudable in character, is to miss the
In the third place, the administrative classwhole significance of British reforms. In the
idea rests upon the acceptance of merit as the present-day politics of the United States, it is
criterion of selection. In Britain this accept-not so clear that the utility of patronage has
ance was no mere accident of an inexplicabledisappeared; under the American system of
twenty-year change in public standards of mo- separation of powers, patronge remains almost
rality. If patronge disappeared in Britain, itas useful as it was under the British constitu-
was partly because patronge had ceased to tion of the eighteenth century. And in any
have any real function, whereas efficiency had case, it is self-evident that the problem here
acquired a new social and political utility. lies in a distinctly different political and social
Prior to the nineteenth century, patronage setting from that of Victorian England.
had two vital functions: it provided a place Last, a successful administrative class rests
upon the condition that such a group pos-
for the sons of the aristocracy who were ex-
cluded from inheritance by primogeniture; sesses the prestige of an elite; for unless the
and it placed in the hands of the King and his
class has an elite status, it is in a poor position
to compete against any other elite for the
ministers a device for guaranteeing, under the
limited franchise of the eighteenth century,brains
a and abilities of the nation. It is one
favorable House of Commons. Both these fac- thing to offer a career in a merit service; it is
tors disappeared during the first decades of quite another to insure that such a service has
the nineteenth century. With the expansion enough prestige to acquire the best of the na-
of the electorate after 1832, the monarchy was tion's competence. The argument that the
mere creation of an administrative class would
forced to withdraw from politics, or risk the
chance of a serious loss of prestige in an elec- be sufficient to endow that group with pres-
tige in the United States may or may not be
valid; it is certainly invalid to argue that this
2 It is noteworthy that the latest U. S. Civil Service
Commission announcement for the junior professional
was the causal sequence in Britain. In assess-
assistant examination (November, 1946) follows the
ing the ability of the British civil service to
subject-matter specialist concept; junior professional
recruit the best products of the universities,
assistants will be recruited in terms of specialities un-
thinkable in the British administrative class examina- one can scarcely overlook the profound signifi-
tions for university graduates. See, by comparison,
Specimen Question Papers for the Reconstruction Com- 2 See D. Lindsay Keir, The Constitutional History of
petition for Recruitment to (i) The Administrative Modern Britain 1485-1937 (A. & C. Black, 1943), p. 405.
Class of the Home Civil Service, (2) The Senior Branch
" See J. Donald Kingsley, Representative Bureaucracy,
of the Foreign Service, (CS.C. 18) (H.M. Stationery An
Of-Interpretation of the British Civil Service (Antioch
fice 1946). Press, 1944), Ch. III.

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms
 SCIENCE OF PUBLIC ADMINISTRATION 11

cance of the fact that for centuries the public

2. There can be no truly universal generalizations
service was one of the few careers into which
about public aadministration without a profound study
of varying national and social characteristics impinging
member of the aristocracy could enter without
on public administration, to determine what aspects of
loss of prestige. Like the church, thepublic army,administration, if any, are truly independent of
and politics, and unlike trade and commerce,
the national and social setting. Are there discoverable
public service was a profession in which theof universal validity, or are all principles
principles
aristocracy could engage without violatingvalid only in terms of a special environment?
the
3. It follows that the study of public administration
mores of the class. Even during the eighteenth
inevitably must become a much more broadly based
century and the first half of the nineteenth,
discipline, resting not on a narrowly defined knowledge
when the burden of incompetence and of patron-
techniques and processes, but rather extending to
age in the public service was at its heaviest,
the varying historical, sociological, economic, and other
government was a field into which the conditioning
social factors that give public administration its
elite could enter without a diminution of peculiar stamp in each country.

prestige, and often enough without even a loss The relation of public administration to its
in leisure. Throughout the age of patronage, peculiar environment has not been altogether
the British public service succeeded in obtain-ignored.31 Unhappily, however, comparative
ing some of the best of Britain's abilities.30 studies are all too infrequent; and at best they
The effect of the reforms after 1853 was toprovide only the groundwork. We need many
make more attractive a profession that already more studies of comparative administration
outranked business and industry in prestige before it will be possible to argue that there
values. In Britain, as in Germany, the psychicare any universal principles of public admin-
income accruing from a career in the civilistration.
service more than compensates for the smaller
In Conclusion
economic income. Contrast this with the
WTE ARE a long way from a science of public
United States, where since the Civil War pres-
administration. No science of public ad-
tige has largely accrued to acquisitive suc-
ministration is possible unless: (1) the place
cesses. It is small wonder that in the United
States the problem of government competitionof normative values is made clear; (2) the na-
with business for the abilities of the commu- ture of man in the area of public administra-
nity should be much more acute. tion is better understood and his conduct is
If these remarks about the British adminis-more predictable; and (3) there is a body of
trative class are well founded, then these con-comparative studies from which it may be pos-
clusions suggest themselves: sible to discover principles and generalities
that transcend national boundaries and pecul-
i. Generalizations derived from the operation of
public administration in the environment of one na- iar historical experiences.
tion-state cannot be universalized and applied to public 31 See, for example, Walter Dorn, "The Prussian Bu-
administration in a different environment. A principle
reaucracy in the Eighteenth Century," 46 Political Sci-
may be applicable in a different framework. But its ap-
ence Quarterly 403-23 (1931) and 47 Ibid., 75-94, 259-73
plicability can be determined only after a study of that (1932); Fritz Morstein Marx, "Civil Service in Ger-
particular framework.
many," in Civil Service Abroad (McGraw-Hill Book
Co., 1935); John M. Gaus, "American Society and Pub-
" Hiram Stout, Public Service in Great Britain (Uni- lic Administration," The Frontiers of Public Adminis-
versity of North Carolina Press, 1938), pp. 25-26, 82-83. tration (University of Chicago Press, 1936).

This content downloaded from

112.198.101.34 on Fri, 19 Jan 2024 00:08:18 +00:00
All use subject to https://about.jstor.org/terms