The Proverbs of Administration

Author(s): Herbert A. Simon

Source: Public Administration Review , Winter, 1946, Vol. 6, No. 1 (Winter, 1946), pp.
53-67
Published by: Wiley on behalf of the American Society for Public Administration

Stable URL: http://www.jstor.com/stable/973030

REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.com/stable/973030?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

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

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
The Proverbs of Administration
By HERBERT A. SIMON
Associate Professor of Political Science
Illinois Institute
of Technology

A FACT about proverbs that greatly opposite organizational recommendations,

enhances their quotability is that there is nothing in the theory to indicate
they almost always occur in mutu- which is the proper one to apply.1
ally contradictory pairs. "Look before you It is the purpose of this paper to substan-
leapl"-but "He who hesitates is lost." tiate this sweeping criticism of administra-
This is both a great convenience and a tive theory, and to present some suggestions
serious defect-depending on the use to -perhaps less concrete than they should be
which one wishes to put the proverbs in -as to how the existing dilemma can be
question. If it is a matter of rationalizing solved.
behavior that has already taken place or
justifying action that has already been de- Some Accepted Administrative
cided upon, proverbs are ideal. Since one Principles
is never at a loss to find one that will prove
AMONG the more common "principles"
his point-or the precisely contradictory that occur in the literature of admin-
point, for that matter-they are a great help istration are these:
in persuasion, political debate, and all 1. Administrative efficiency is increased
forms of rhetoric.
by a specialization of the task among the
But when one seeks to use proverbs as group.
the basis of a scientific theory, the situation
2. Administrative efficiency is increased
is less happy. It is not that the propositions
by arranging the members of the group in
expressed by the proverbs are insufficient; a determinate hierarchy of authority.
it is rather that they prove too much. A 3. Administrative efficiency is increased
scientific theory should tell what is true but
by limiting the span of control at any point
also what is false. If Newton had announced
in the hierarchy to a small number.
to the world that particles of matter exert
4. Administrative efficiency is increased
either an attraction or a repulsion on each
other, he would not have added much to
by grouping the workers, for purposes of
control, according to (a) purpose, (b) proc-
scientific knowledge. His contribution con-
ess, (c) clientele, or (d) place. (This is really
sisted in showing that an attraction was ex-
an elaboration of the first principle but de-
ercised and in announcing the precise law
serves separate discussion).
governing its operation.
Since these principles appear relatively
Most of the propositions that make up
simple and clear, it would seem that their
the body of administrative theory today
application to concrete problems of admin-
share, unfortunately, this defect of prov-
erbs. For almost every principle one can1 Lest it be thought that this deficiency is peculiar to
find an equally plausible and acceptable
the science-or "art"-of administration, it should be
pointed out that the same trouble is shared by most
contradictory principle. Although the two Freudian psychological theories, as well as by some
principles of the pair will lead to exactly
sociological theories.

53

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
54 PUBLIC ADMINISTRATION REVIEW

istrative organization would be

trative unambigu-
efficiency. But, in thus rephrasing
ous and that their validity would
this be
"principle" easily
of administration, there has
beenSuch,
submitted to empirical test. brought however,
clearly into the open its fun-
seems not to be the case. damental
To show why
ambiguity: it
"Administrative effi-
ciency is increased
is not, each of the four principles by a specialization of the
just listed
will be considered in turn. task among the group in the direction
Specialization. Administrative efficiency which will lead to greater efficiency."
is supposed to increase with an increase inFurther discussion of the choice between
specialization. But is this intended to mean competing bases of specialization will be
that any increase in specialization will in- undertaken after two other principles of
crease efficiency? If so, which of the follow- administration have been examined.
ing alternatives is the correct application Unity of Command. Administrative effi-
of the principle in a particular case? ciency is supposed to be enhanced by ar-
1. A plan of nursing should be put into effect ranging the members of the organization in
by which nurses will be assigned to districts and a determinate hierarchy of authority in or-
der to preserve "unity of command."
do all nursing within that district, including school
examinations, visits to homes or school children, and
Analysis of this "principle" requires a
tuberculosis nursing.
clear understanding of what is meant by
2. A functional plan of nursing should be put
the term "authority." A subordinate may
into effect by which different nurses will be as-
be said to accept authority whenever he
signed to school examinations, visits to homes of
permits his behavior to be guided by a
school children, and tuberculosis nursing. The
present method of generalized nursing by districtsdecision reached by another, irrespective
impedes the development of specialized skills in the
of his own judgment as to the merits of that
three very diverse programs. decision.
Both of these administrative arrange- In one sense the principle of unity of
ments satisfy the requirement of specializa-command, like the principle of specializa-
tion-the first provides specialization by tion, cannot be violated; for it is physically
place; the second, specialization by func- impossible for a man to obey two contra-
tion. The principle of specialization is of no dictory commands-that is what is meant by
help at all in choosing between the two "contradictory commands." Presumably, if
alternatives. unity of command is a principle of admin-
It appears that the simplicity of the prin- istration, it must assert something more
ciple of specialization is a deceptive sim-than this physical impossibility. Perhaps it
plicity-a simplicity which conceals funda- asserts this: that it is undesirable to place
mental ambiguities. For "specialization" is a member of an organization in a position
not a condition of efficient administration; where he receives orders from more than
it is an inevitable characteristic of all group one superior. This is evidently the meaning
effort, however efficient or inefficient that that Gulick attaches to the principle when
effort may be. Specialization merely meanshe says,
that different persons are doing different The significance of this principle in the process
things-and since it is physically impossibleof co-ordination and organization must not be lost
for two persons to be doing the same thingsight of. In building a structure of co-ordination,
in the same place at the same time, twoit is often tempting to set up more than one boss
for a man who is doing work which has more
persons are always doing different things.
than one relationship. Even as great a philosopher
The real problem of administration, of management as Taylor fell into this error in
then, is not to "specialize," but to specializesetting up separate foremen to deal with machinery,
in that particular manner and along thosewith materials, with speed, etc., each with the power
particular lines which will lead to adminis-of giving orders directly to the individual work-

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
 PROVERBS OF ADMINISTRATION 55

man. The rigid adherence to theissue

principle of unity
direct orders to him regarding the
of command may have its absurdities; these are, how-
technical, accounting aspects of his work.
ever, unimportant in comparison with the certainty
Similarly, the director of motor vehicles
of confusion, inefficiency and irresponsibility which
arise from the violation of the in the public works department will be un-
principle.'
able to issue direct orders on care of motor
Certainly the principle of unity
equipment offire-truck
to the com- driver.2
mand, thus interpreted, cannot be criticized
Gulick, in the statement quoted above,
for any lack of clarity orclearly
anyindicates
ambiguity.
the difficulties to be faced
The definition of authority given
if unity above
of command is not observed. A
should provide a clear test certain whether,amount ofin any
irresponsibility and con-
concrete situation, the principle is almost
fusion are observed.
certain to ensue. But per-
The real fault that must be found with this
haps this is not too great a price to pay for
principle is that it is incompatible with the
the increased expertise that can be applied
principle of specialization. One of the mostto decisions. What is needed to decide the
important uses to which authority is put in issue is a principle of administration that
organization is to bring about specializationwould enable one to weigh the relative ad-
in the work of making decisions, so that vantages of the two courses of action. But
each decision is made at a point in the or-neither the principle of unity of command
ganization where it can be made most ex-nor the principle of specialization is helpful
pertly. As a result, the use of authorityin adjudicating the controversy. They
permits a greater degree of expertness tomerely contradict each other without indi-
be achieved in decision-making than wouldcating any procedure for resolving the con-
be possible if each operative employee had
tradiction.
himself to make all the decisions upon If this were merely an academic con-
which his activity is predicated. The in-troversy-if it were generally agreed and
dividual fireman does not decide whether
had been generally demonstrated that unity
to use a two-inch hose or a fire extinguisher;
of command must be preserved in all cases,
that is decided for him by his officers, and even with a loss in expertise-one could
the decision is communicated to him in the assert that in case of conflict between the
form of a command.
two principles, unity of command should
However, if unity of command, in prevail. But the issue is far from clear, and
Gulick's sense, is observed, the decisions experts can be ranged on both sides of the
of a person at any point in the administra-controversy. On the side of unity of com-
tive hierarchy are subject to influence mand there may be cited the dictums of
through only one channel of authority; and Gulick and others.3 On the side of speciali-
if his decisions are of a kind that requirezation there are Taylor's theory of func-
expertise in more than one field of knowl- tional supervision, Macmahon and Millett's
edge, then advisory and informational serv-
idea of "dual supervision," and the practice
ices must be relied upon to supply those of technical supervision in military organi-
premises which lie in a field not recognized
zation.4
by the mode of specialization in the or-
ganization. For example, if an accountant 2 This point is discussed in Herbert A. Simon "De-
in a school department is subordinate to an
cision-Making and Administrative Organization," 4
educator, and if unity of command is ob- Public Administration Review 20-21 (Winter, 1944).
3 Gulick, "Notes on the Theory of Organization," p.
served, then the finance department cannot 9; L. D. White, Introduction to the Study of Public
Administration (Macmillan Co., 1939), p. 45.
1 Luther Gulick, "Notes on the Theory of Organiza- *Frederick W. Taylor, Shop Management (Harper &
tion," in Luther Gulick and L. Urwick (eds.), Papers on
Bros., 1911). p. 99; Macmahon, Millett, and Ogden
the Science of Administration (Institute of Public Ad-The Administration of Federal Work Relief (Public
ministration, Columbia University, 1937), p. 9. Administration Service, 1941), pp. 265-68; and L. Ur-

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
56 PUBLIC ADMINISTRATION REVIEW

It may be, as Gulick asserts, form,that

it also the
solves no-
fewer problems. In the
tion of Taylor and these firstothers is an
place, it no longer requires, except for
"error." If so, the evidence that
settling it isofan
conflicts authority, a single hier-
error has never been marshalled or pub-
archy of authority. Consequently, it leaves
lished-apart from loose unsettled heuristic the very important question of
argu-
ments like that quoted above. how authority
One isshould leftbe zoned in a par-
with a choice between equally ticular organization
eminent (i.e., the modes of spe-
theorists of administration and cialization)
withoutand through
any what channels it
evidential basis for making should
that be exercised.
choice.Finally, even this nar-
What evidence there is of actual adminis- rower concept of unity of command con-
trative practice would seem to indicate that flicts with the principle of specialization,
the need for specialization is to a very large for whenever disagreement does occur and
degree given priority over the need for the organization members revert to the for-
unity of command. As a matter of fact, it mal lines of authority, then only those types
does not go too far to say that unity of of specialization which are represented in
command, in Gulick's sense, never has ex- the hierarchy of authority can impress them-
isted in any administrative organization. If selves on decision. If the training officer of
a line officer accepts the regulations of an a city exercises only functional supervision
accounting department with regard to the over the police training officer, then in case
procedure for making requisitions, can it of disagreement with the police chief, spe-
be said that, in this sphere, he is not sub- cialized knowledge of police problems will
ject to the authority of the accounting de- determine the outcome while specialized
partment? In any actual administrative knowledge of training problems will be sub-
situation authority is zoned, and to main- ordinated or ignored. That this actually oc-
tain that this zoning does not contradict the curs is shown by the frustration so com-
principle of unity of command requires a monly expressed by functional supervisors
very different definition of authority from at their lack of authority to apply sanctions.
that used here. This subjection of the line Span of Control. Administrative effi-
officer to the accounting department is no ciency is supposed to be enhanced by limit-
different, in principle, from Taylor's recom- ing the number of subordinates who report
mendation that in the matter of work pro- directly to any one administrator to a small
gramming a workman be subject to one number-say six. This notion that the
foreman, in the matter of machine opera- "span of control" should be narrow is con-
tion to another. fidently asserted as a third incontrovertible
The principle of unity of comand is per- principle of administration. The usual com-
haps more detensible if narrowed down to mon-sense arguments for restricting the
the following: In case two authoritative span of control are familiar and need not be
repeated here. What is not so generally
commands conflict, there should be a single
determinate person whom the subordinate recognized is that a contradictory proverb
of administration can be stated which,
is expected to obey; and the sanctions of
authority should be applied against the sub-though it is not so familiar as the principle
ordinate only to enforce his obedience to of span of control, can be supported by
that one person. arguments of equal plausibility. The prov-
If the principle of unity of command is erb in question is the following: Adminis-
more defensible when stated in this limitedtrative efficiency is enhanced by keeping
at a minimum the number of organizational
wick, who describes British army practice in "Organiza- levels through which a matter must pass
tion as a Technical Problem," Gulick and Urwick (eds.),
op. cit., pp. 67-69. before it is acted upon.

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
 PROVERBS OF ADMINISTRATION 57

This latter proverb is one theof

position
the of funda-
chief inspector and head nurse be
mental criteria that guideabolished and that the employees now filling these
administrative
positions perform regular inspectional and nursing
analysts in procedures simplification
duties. The details ofwork.
work scheduling now handled
Yet in many situations theby results to which
these two employees can be taken care of more
this principle leads are in direct
economicallycontradic-
by the secretary to the health officer,
tion to the requirements of and, since
thebroader matters of policy have, in any
principle
event, always
of span of control, the principle of unity required the personal attention of
the health officer, the abolition of these two posi-
of command, and the principle of specializa-
tions will eliminate a wholly unnecessary step in
tion. The present discussion iswill
review, concerned
allow an expansion of inspectional and
with the first of these conflicts. To illus- nursing services, and will permit at least a begin-
trate the difficulty, two alternative pro- ning to be made in the recommended program of
posals for the organization of a small healthhealth education. The number of persons report-
ing directly to the health officer will be increased
department will be presented-one based on
to nine, but since there are few matters requiring
the restriction of span of control, the otherthe coordination of these employees, other than the
on the limitation of number of organization
work schedules and policy questions referred to
levels: above, this change will not materially increase his
work load.
1. The present organization of the department
places an administrative overload on the health The dilemma is this: in a large organiza-
officer by reason of the fact that all eleven em- tion with complex interrelations between
ployees of the department report directly to him
and the further fact that some of the staff lack members, a restricted span of control in-
evitably produces excessive red tape, for
adequate technical training. Consequently, venereal
each contact between organization members
disease clinic treatments and other details require
must be carried upward until a common
an undue amount of the health officer's personal
attention.
superior is found. If the organization is at
It has previously been recommended that the
proposed medical officer be placed in charge of the
all large, this will involve carrying all such
venereal disease and chest clinics and all child matters upward through several levels of
hygiene work. It is further recommended thatofficials
one for decision and then downward
againand
of the inspectors be designated chief inspector in the form of orders and instructions
placed in charge of all the department's inspection-
-a cumbersome and time-consuming proc-
al activities and that one of the nurses be ess.
desig-
nated as head nurse. This will relieve the health
commissioner of considerable detail and will leave The alternative is to increase the number
of persons who are under the command of
him greater freedom to plan and supervise the
health program as a whole, to conduct health edu-
each officer, so that the pyramid will come
cation, and to coordinate the work of the depart-
more rapidly to a peak, with fewer inter-
ment with that of other community agencies. If the
vening levels. But this, too, leads to diffi-
department were thus organized, the effectiveness
of all employees could be substantially increased. culty, for if an officer is required to super-
2. The present organization of the departmentvise too many employees, his control over
them is weakened.
leads to inefficiency and excessive red tape by rea-
son of the fact that an unnecessary supervisory If it is granted, then, that both the in-
level intervenes between the health officer and the
operative employees, and that those four of the
crease and the decrease in span of control
twelve employees who are best trained technicallyhas some undesirable consequences, what is
are engaged largely in "overhead" administrativethe optimum point? Proponents of a re-
duties. Consequently, unnecessary delays occur instricted span of control have suggested
securing the approval of the health officer on mat-
three, five, even eleven, as suitable num-
ters requiring his attention, and too many matters
require review and re-review.
bers, but nowhere have they explained the
The medical officer should be left in charge ofreasoning which led them to the particular
the venereal disease and chest clinics and child number they selected. The principle as
hygiene work. It is recommended, however, that
stated casts no light on this very crucial

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
58 PUBLIC ADMINISTRATION REVIEW

clientele
question. One is reminded of currentin the organization
argu-
department.
ments about the proper size of the national
debt. 1. Public health administration consists of the
Organization by Purpose, following
Process, Clien-
activities for the prevention of disease
and the maintenance
tele, Place. Administrative efficiency of healthful conditions: (i)
is sup-
vital statistics; (2) child hygiene-prenatal, mater-
posed to be increased by grouping workers
nity, postnatal, infant, preschool, and school health
according to (a) purpose,programs;
(b) process, (c)
(3) communicable disease control; (4)
clientele, or (d) place. But from
inspection the discus-
of milk, foods, and drugs; (5) sanitary
sion of specialization it isinspection;
clear (6)that this
laboratory service; (7) health educa-
tion.
principle is internally inconsistent; for pur-
pose, process, clientele, and One of the are
place handicaps
com-under which the health de-
partment labors is the fact that the department has
peting bases of organization, and
no control at any
over school health, that being an ac-
given point of division the advantages
tivity ofof education, and there
of the county board
three must be sacrificed to secure the
is little or advan- between that highly
no coordination
tages of the fourth. If the major
important partdepart-
of the community health program
ments of a city, for example, and the balance
are of the program which is conducted
organized
by the city-county health unit. It is recommended
on the basis of major purpose, then it fol-
that the city and county open negotiations with the
lows that all the physicians, board
all ofthe lawyers,
education for the transfer of all school
all the engineers, all the statisticians will
health work and the appropriation therefor to the
joint health unit.... ex-
not be located in a single department
clusively composed of members 2. To of
the modem
their school
pro-department is entrusted
the care of children during almost the entire period
fession but will be distributed among the
that they are absent from the parental home. It
various city departments needing their responsibilities
has three principal serv- toward them:
ices. The advantages of organization by
(1) to provide for their education in useful skills
and knowledge
process will thereby be partly lost. and in character; (2) to provide
Some of these advantages them
canwith bewholesome
regained play activities outside school
hours; (3) to care for their health and to assure
by organizing on the basis of
the process
attainment of within
minimum standards of nutri-
the major departments. Thus
tion. there may be
an engineering bureau within the
One of the public
handicaps under which the school
works department, or theboard
boardlabors of
is theeduca-
fact that, except for school
lunches, the board has no control over child health
tion may have a school health service as a
and nutrition, and there is little or no coordination
major division of its work. Similarly, within
between that highly important part of the child
smaller units there may bedevelopment
division by and
program area
the balance of the pro-
or by clientele: e.g., a fire gram
department
which is conductedwill
by the board of educa-
tion. It is recommended
have separate companies located through- that the city and county
out the city, while a welfare open negotiations
department for the transfer of all health
work for children of school age to the board of
may have intake and case work education. agencies in
various locations. Again, however, these
Here
major types of specialization again is be
cannot posed
si-the dilemma of
multaneously achieved, forchoosing
at any between
pointalternative,
in equally
plausible, administrative
the organization it must be decided whether principles. But
this is not
specialization at the next level the only
will be difficulty
ac- in the present
complished by distinction ofcase, for a purpose,
major closer study of the situation
major process, clientele, orshows
area.there are fundamental ambiguities in
the meanings
The conflict may be illustrated byofshow-
the key terms-"purpose,"
"process," "clientele,"
ing how the principle of specialization ac-and "place."
cording to purpose would lead"Purpose"
to amay be roughly defined as the
differ-
objective
ent result from specialization or end for which
according to an activity is

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
 PROVERBS OF ADMINISTRATION 59

carried on; "process" as a means for ac- be considered a single function depends
complishing a purpose. Processes, then, are entirely on language and techniques.2 If
carried on in order to achieve purposes. the English language has a comprehensive
But purposes themselves may generally be term which covers both of two subpurposes
arranged in some sort of hierarchy. A typist it is natural to think of the two together as
moves her fingers in order to type; types in a single purpose. If such a term is lacking,
order to reproduce a letter; reproduces a the two subpurposes become purposes in
letter in order that an inquiry may be an- their own right. On the other hand, a single
swered. Writing a letter is then the purpose activity may contribute to several objec-
for which the typing is performed; while tives, but since they are technically (pro-
writing a letter is also the process whereby cedurally) inseparable, the activity is con-
the purpose of replying to an inquiry is sidered a single function or purpose.
achieved. It follows that the same activity The fact, mentioned previously, that pur-
may be described as purpose or as process. poses form a hierarchy, each subpurpose
This ambiguity is easily illustrated for contributing to some more final and com-
the case of an administrative organization. prehensive end, helps to make clear the
A health department conceived as a unit relation between purpose and process. "Or-
whose task it is to care for the health of the ganization by major process," says Gulick,
community is a purpose organization; the ". .. tends to bring together in a single
same department conceived as a unit which department all of those who are at work
makes use of the medical arts to carry on its making use of a given special skill or tech-
work is a process organization. In the same nology, or are members of a given profes-
way, an education department may be sion."3 Consider a simple skill of this kind
viewed as a purpose (to educate) organiza- -typing. Typing is a skill which brings
tion, or a clientele (children) organization; about a means-end coordination of muscu-
the forest service as a purpose (forest con- lar movements, but at a very low level in
servation), process (forest management), the means-end hierarchy. The content of
clientele (lumbermen and cattlemen utiliz- the typewritten letter is indifferent to the
ing public forests), or area (publicly owned skill that produces it. The skill consists
forest lands) organization. When concrete merely in the ability to hit the letter "t"
illustrations of this sort are selected, the quickly whenever the letter "t" is required
lines of demarcation between these cate-
by the content and to hit the letter "a"
gories become very hazy and unclearwhenever
in- the letter "a" is required by the
deed. content.

"Organization by major purpose," says There is, then, no essential difference

Gulick, ". . . serves to bring together in abetween a "purpose" and a "process," but
single large department all of those who are only a distinction of degree. A "process" is
at work endeavoring to render a particular an activity whose immediate purpose is at a
service."1 But what is a particular service?low level in the hierarchy of means and
Is fire protection a single purpose, or is itends, while a "purpose" is a collection of
merely a part of the purpose of public activities whose orienting value or aim is at
safety?-or is it a combination of purposes a high level in the means-end hierarchy.
including fire prevention and fire fighting?
It must be concluded that there is no such
2 If this is correct, then any attempt to prove that
certain activities belong in a single department because
thing as a purpose, or a unifunctionalthey relate to a single purpose is doomed to fail. See,
(single-purpose) organization. What isfor
toexample, John M. Gaus and Leon Wolcott, Public
Administration and the U.S. Department of Agricul-
ture (Public Administration Service, 1940.)
1 op. cit., p 21. a Op. cit., p. 23.

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
60 PUBLIC ADMINISTRATION REVIEW

Next consider "clientele"mean that Bureau

and "place" as A incorpo
bases of organization. These
processes
categories
of aarecertain kind in
X, without
really not separate from purpose, reference to any
but a part
of it. A complete statementpurposes, subareas, or subclie
of the purpose
of a fire department would partment X. Now it is conc
have to include
particular
the area served by it: "to reduce fire unit
lossesmight incorpo
esses of a certain
on property in the city of X." Objectives kind but th
of an administrative organization
esses might are
relate to only certa
subpurposes of the departm
phrased in terms of a service to be provided
and an area for which it is In provided.
this case, which
Usu- corresponds
unit in an
ally, the term "purpose" is meant to refer education departme
above, the
only to the first element, but the second is unit would be s
both
just as legitimately an aspect of purpose
purpose.and process. Th
would
Area of service, of course, may be be the only one in the
a specified
clientele quite as well aspartment
a geographical
using the medical
and which
area. In the case of an agency concerned
workswith health (s
Even dimension
on "shifts," time will be a third when the problem i
of purpose-to provide a given proper usage
service in for
a the term
"process,"
given area (or to a given clientele) "clientele,"
during a and "ar
given time period. ciples of administration give
With this clarification of terminology,
which of these four competing
the next task is to reconsidercialization is applicable in an
the problem
situation.
of specializing the work of an organization.The British Machin
It is no longer legitimate to ernment
speak of a Committee
"pur- had no d
the matter.
pose" organization, a "process" It considered purpo
organiza-
tion, a "clientele" organization,
tele as the or
twoan
possible bases of
and put
"area" organization. The same unitits faith entirely in
might
Others have
fall into any one of these four categories, had equal assura
depending on the nature of ing between
the larger purpose
or- and pro
ganizational unit of whichsoning
it was awhich
part. A leads to these
unit providing public healthconclusions
and medicalleaves something
The Machinery
services for school-age children in Mult- of Governme
nomah County might be considered
gives this sole(1) argument
an for it
"area" organization if it were
Nowpart
the of a unit outcome of t
inevitable
providing the same serviceorganization
for the state[by clientele]
of is a ten
Oregon; (2) a "clientele" putian administration.
organization if It is imp
specialized service which each Dep
it were part of a unit providing similar
render to the community can be of
services for children of all ages;
ard when(3) a work
its "pur- is at the same t
pose" or a "process" organization
particular(it would
class of persons and ext
be impossible to say which) if it of
variety were part for them, as w
provision
of an education department. ment concentrates itself on the p
particular
It is incorrect to say that Bureauservice
A is aonly by whoms
and looks beyond the interest o
process bureau; the correct small
statement
classes.2 is that
Bureau A is a process bureau within De-
usually speaks of "grouping activities" rather than "di-
partment X.1 This latter viding
statement would
work," the relative character of these categories
is not always apparent in this discussion (op. cit., pp.
This distinction is implicit in15-30).
most of Gulick's
2Report of the Machinery of Government Commit-
analysis of specialization. However, since he cites as
examples single departments within (H.
tee a M. Stationery
city, Office,since
and 1918). he

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
 PROVERBS OF ADMINISTRATION 61

istry of
The faults in this analysis Agriculture
are obvious. they would teach
new-fashioned
First, there is no attempt to determine farming
howby old-fashioned
a service is to be recognized. Second,
methods? there problem of
The administrative
is a bald assumption, absolutely without
such a bureau would be to teach new-
fashioned
proof, that a child health unit, for farming
example, by new-fashioned meth
ods, and it iscould
in a department of child welfare a little difficult
not to see how the
offer services of "as high adepartmental
standard" location
as the of the unit would
same unit if it were locatedaffect
in a this
department
result. "The question answers
of health. Just how the shifting ofif the
itself" only unit
one has a rather mystical faith
in the potency
from one department to another would of bureau-shuffling
im- as
prove or damage the quality
means of its
for work is
redirecting the activities of an
not explained. Third, no agency.
basis is set forth
for adjudicating the competing claims of and competitions
These contradictions
purpose and process-thehave
two are increasing
received merged attention from stu-
in the ambiguous term "service." It is not
dents of administration during the past few
necessary here to decide whether the com-
years. For example, Gulick, Wallace, and
mittee was right or wrong in its recom-
Benson have stated certain advantages and
mendation; the importantdisadvantages
point is that of the the
several modes of spe-
recommendation represented a have
cialization, and choice,
considered the condi-
tions under
without any apparent logical or which one or the other mode
empirical
grounds, between contradictory principles
might best be adopted.2 All this analysis
of administration. has been at a theoretical level-in the sense
Even more remarkable illustrations of
that data have not been employed to dem-
illogic can be found in most discussions of
onstrate the superior effectiveness claimed
purpose vs. process. They would befor too
the different modes. But though the-
ridiculous to cite if they were not oretical,
com- the analysis has lacked a theory.
monly used in serious political and admin-
Since no comprehensive framework has
istrative debate. been constructed within which the discus-
For instance, where should agricultural educationsion could take place, the analysis has
come: in the Ministry of Education, or of Agricul- tended either to the logical one-sidedness
ture? That depends on whether we want to see the
which characterizes the examples quoted
best farming taught, though possibly by old meth- above or to inconclusiveness.
ods, or a possibly out-of-date style of farming,
taught in the most modern and compelling man- The Impasse of Administrative Theory.
ner. The question answers itself.' The four "principles of administration"
that were set forth at the beginning of this
But does the question really answer it-
paper have now been subjected to critical
self? Suppose a bureau of agricultural edu-
analysis. None of the four survived in very
cation were set up, headed, for example, by
good shape, for in each case there was
a man who had had extensive experience in
found,
agricultural research or as administrator of instead of an unequivocal principle,
a set of two or more mutually incompatible
an agricultural school, and staffed by men
principles apparently equally applicable to
of similarly appropriate background. What the administrative situation.
reason is there to believe that if attached
Moreover, the reader will see that the
to a Ministry of Education they would
very same objections can be urged against
teach old-fashioned farming by new-fash-
ioned methods, while if attached to a Min- 2Gulick, "Notes on the Theory of Organization," pp.
21-30; Schuyler Wallace, Federal Departmentalization
(Columbia University Press, 1941); George C. S. Benson,
1Sir Charles Harris, "Decentralization," 3 Journal "International Administrative Organization," i Public
of Public Administration 117-33 (April, 1925). Administration Review 473-86 (Autumn, 1941).

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
62 PUBLIC ADMINISTRATION REVIEW

the customary discussions tified; that each administrative

of "centraliza-
tion" vs. "decentralization," which usually
analyzed in terms of the entir
teria;
conclude, in effect, that "on the and that research be ins
one hand,
determine how weights can be
centralization of decision-making functions
is desirable; on the other hand, there criteria
the several are when they
usually will be, mutually inco
definite advantages in decentralization."
Can anything be salvaged which will be
useful in the construction ofAn
an Approach
administra- to Administra
tive theory? As a matter of fact,
THIS almost
program needs to be
everything can be salvaged. The stepdifficulty
by step. First, what is i
has arisen from treating as "principles
the description ofof administrati
administration" what are really only cri-
for purposes of such an analy
how can weights
teria for describing and diagnosing admin- be assigned t
criteria
istrative situations. Closet space to give them their pro
is certainly
an important item in the design ofpicture?
the total a suc-
The Description
cessful house; yet a house designed entirely of Administr
tions. Before
with a view to securing a maximum a science can d
of closet
ciples,
space-all other considerations it must
being for-possess concept
law of gravitation
gotten-would be considered, to say the could be fo
least, somewhat unbalanced.
wasSimilarly,
necessary to have the not
celeration"
unity of command, specialization and "weight." The f
by pur-
pose, decentralization are all items to be theory is to d
administrative
considered in the design ofof anconcepts
efficient ad-will permit the
that
ministrative organization. Noin terms
singlerelevant
one of to the theory
these items is of sufficienttrative situations.
importance to These con
scientifically
suffice as a guiding principle for the ad- useful, must be
ministrative analyst. In the that is, their
design of ad-meanings must c
ministrative organizations,empirically
as in their observable
op- facts o
The definition
eration, over-all efficiency must be the guid-of "authority"
in this paper
ing criterion. Mutually incompatible is an example o
ad-
vantages must be balancedtional against
definition. each
What is
other, just as an architect weighs a scientifically
the advan- relevant descrip-
tion of an
tages of additional closet space organization?
against the It is a description
advantages of a larger livingthat, so far as possible, designates for each
room.
personone,
This position, if it is a valid in theconsti-
organization what decisions
tutes an indictment of much thatcurrent
person makes writ-
and the influences to
which he
ing about administrative matters. As theis subject in making each of these
decisions.
examples cited in this chapter amply dem-Current descriptions of adminis-
onstrate, much administrativetrativeanalysis
organizations fall far short of this
pro-
ceeds by selecting a single standard. For the most
criterion and part, they confine
themselves
applying it to an administrative situation to the allocation of functions
to reach a recommendation; andwhile
the formal thestructure
fact of authority. They
that equally valid, but contradictory, cri-
give little attention to the other types of or-
teria exist which could be applied
ganizational with
influence or to the system of
communication.1
equal reason, but with a different result, is
conveniently ignored. A valid Whatapproach
does it mean, tofor example to say:
the study of administration requires that
1The monograph by Macmahon, Millett, and Ogden,
all the relevant diagnostic criteria be iden-

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
 PROVERBS OF ADMINISTRATION 63

"The department is made tions

up ofof threeunits
organizational bu-
at the different
reaus. The first has the function of levels.
the second the function of , and the Administrative description suffers cur-
third the function of ?" What can be rently from superficiality, oversimplifica-
learned from such a description about tion, the
lack of realism. It has confined itself
workability of the organizational arrange-
too closely to the mechanism of authority
ment? Very little, indeed. For from thefailed to bring within its orbit the
and has
description there is obtained no ideaother,
of the equally important, modes of influ-
degree to which decisions are centralized
ence on organizational behavior. It has re-
fused to undertake the tiresome task of
at the bureau level or at the departmental
level. No notion is given as to the extent
studyingto the actual allocation of decision-
which the (presumably unlimited) author-making functions. It has been satisfied to
ity of the department over the bureau speakisof "authority," "centralization,"
actually exercised or by what mechanisms.
"span of control," "function," without seek-
There is no indication of the extent to ing operational definitions of these terms.
which systems of communication assist Until
the administrative description reaches a
coordination of the three bureaus or, for
higher level of sophistication, there is little
that matter, to what extent coordination reason
is to hope that rapid progress will be
made toward the identification and verifica-
required by the nature of their work. There
is no description of the kinds of trainingtion
the of valid administrative principles.
members of the bureau have undergone or
Does this mean that a purely formal de-
of the extent to which this training permits
scription of an administrative organization
decentralization at the bureau level. In
is impossible-that a relevant description
must include an account of the content of
sum, a description of administrative organi-
zations in terms almost exclusively of func-
the organization's decisions? This is a ques-
tions and lines of authority is completely tion that is almost impossible to answer in
inadequate for purposes of administrative the present state of knowledge of adminis-
analysis. trative theory. One thing seems certain:
Consider the term "centralization." How
content plays a greater role in the applica-
is it determined whether the operations tion of of administrative principles than is
a particular organization are "centralized" allowed for in the formal administrative
or "decentralized"? Does the fact that field theory of the present time. This is a fact
offices exist prove anything about decen- that is beginning to be recognized in the
tralization? Might not the same decentrali- literature of administration. If one ex-
zation take place in the bureaus of a cen- amines the chain of publications extending
trally located office? A realistic analysis of from Mooney and Reilley, through Gulick
centralization must include a study of theand the President's Committee controversy,
allocation of decisions in the organization to Schuyler Wallace and Benson, he sees a
and the methods of influence that are em-
steady shift of emphasis from the "princi-
ployed by the higher levels to affect the ples of administration" themselves to a
decisions at the lower levels. Such an analy- study of the conditions under which com-
sis would reveal a much more complex pic- peting principles are respectively applica-
ture of the decision-making process than ble. Recent publications seldom say that
any enumeration of the geographical loca- "organization should be by purpose," but
rather that "under such and such condi-
op. cit., perhaps approaches nearer than any other pub-
lished administrative study to the sophistication re- tions purpose organization is desirable." It
quired in administrative description. See, for example,
the discussion on pO. 233-36 of headquarters-field rela- is to these conditions which underlie the ap-
tionships. plication of the proverbs of administration

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
64 PUBLIC ADMINISTRATION REVIEW

that administrative theory and

possible analysis
to make an exhaustive list of these,
must turn in their search but for the really valid
principal categories can be enumer-
ated. Perhaps the simplest method of ap-
principles to replace the proverbs.
The Diagnosis of Administrative Situa-the single member of
proach is to consider
tions. Before any positive the administrativecan
suggestions organization and ask
be made, it is necessary to what
digressthe limits
a bit areand
to the quantity and
to consider more closely thequality
exactofnature
his output.ofThese limits include
(a) limits on his
the propositions of administrative ability to perform and (b)
theory.
The theory of administrationlimitsis on concerned
his ability to make correct deci-
with how an organization sions.
shouldTo the be
extent
con-that these limits are
removed,
structed and operated in order tothe administrative organization
accom-
plish its work efficiently.approaches
A fundamentalits goal of high efficiency. Two
principle of administration,persons,
which givenfollows
the same skills, the same ob-
almost immediately from the jectives and values,
rational char- the same knowledge and
acter of "good" administration,
information,is canthat
rationally decide only up-
among several alternatives on the same course ofthe
involving action. Hence, ad-
same expenditure that one ministrative
should always theory mustbe be interested in
selected which leads to thethe factors that will
greatest determine with what
accom-
plishment of administrative skills, values, and knowledge
objectives; and the organiza-
among several alternatives tion
thatmember
lead undertakes
to the his work. These
same accomplishment that are one
the "limits"
should to rationality
be with which
selected which involves the theleast
principles of administration must deal.
expendi-
ture. Since this "principle of Onefficiency"
one side, the individual
is is limited by
those skills,
characteristic of any activity that attempts habits, and reflexes which are
no longer
rationally to maximize the attainment of in the realm of the conscious. His
certain ends with the use of performance,
scarce means, it
for example, may be limited
by his
is as characteristic of economic manual dexterity
theory as it is or his reaction time
of administrative theory. The or his "administra-
strength. His decision-making proc-
tive man" takes his place esses may be limited
alongside the by the speed of his
classical "economic man."' mental processes, his skill in elementary
Actually, the "principle" of efficiency arithmetic, and so forth. In this area, the
should be considered a definition rather principles of administration must be con-
cerned with the physiology of the human
than a principle: it is a definition of what is
body and with the laws of skill-training
meant by "good" or "correct" administra-
tive behavior. It does not tell how accom- and of habit. This is the field that has been
plishments are to be maximized, but merely most successfully cultivated by the follow-
states that this maximization is the aim of ers of Taylor and in which has been
administrative activity, and that adminis- developed time-and-motion study and the
trative theory must disclose under what con- therblig.
ditions the maximization takes place. On a second side, the individual is lim-
Now what are the factors that determine ited by his values and those conceptions of
the level of efficiency which is achieved by purpose which influence him in making his
an administrative organization? It is not decisions. If his loyalty to the organization
is high, his decisions may evidence sincere
For an elaboration of the principle of efficiency and
its place in administrative theory see Clarence E. Rid-
acceptance of the objectives set for the or-
ley and Herbert A. Simon, Measuring Municipal Ac- ganization; if that loyalty is lacking, per-
tivities (International City Managers' Association, 2nd sonal motives may interfere with his ad-
ed., 1943), particularly Chapter I and the preface to
the second edition. ministrative efficiency. If his loyalties are

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
 PROVERBS OF ADMINISTRATION 65

attached to the bureau bythat the limits

which heofisrationality
em- are variable
ployed, he may sometimes make
limits. decisions
Most important of all, consciousness
of theunit
that are inimical to the larger limits of
may which
in itself alter them. Sup-
pose area
the bureau is a part. In this it werethediscovered
prin- in a particular or-
ciples of administration must be for
ganization, concerned
example, that organizational
with the determinants of loyalties
loyalty attached
and to small
mo-units had fre-
rale, with leadership and initiative,
quently and
led to a harmful degree of intra-
with the influences that determine where organizational competition. Then, a pro-
gram which trained members of the or-
the individual's organizational loyalties will
be attached. ganization to be conscious of their loyalties,
On a third side, the individual is limited and to subordinate loyalties to the smaller
by the extent of his knowledge of things group to those of the large, might lead to a
relevant to his job. This applies both to thevery considerable alteration of the limits in
basic knowledge required in decision-mak-
that organization.'
ing-a bridge designer must know the fun- A related point is that the term "rational
damentals of mechanics-and to the in- behavior," as employed here, refers to ra-
formation that is required to make his when that behavior is evaluated in
tionality
decisions appropriate to the given situation.
terms of the objectives of the larger organ-
In this area, administrative theory isization;
con- for, as just pointed out, the differ-
cerned with such fundamental questions ence in
as direction of the individual's aims
these: What are the limits on the mass of from those of the larger organization is just
knowledge that human minds can accumu- one of those elements of nonrationality with
late and apply? How rapidly can knowledge which the theory must deal.
be assimilated? How is specialization in theA final observation is that, since adminis-
administrative organization to be relatedtrative
to theory is concerned with the non-
the specializations of knowledge that are rational limits of the rational, it follows
prevalent in the community's occupational that the larger the area in which rationality
structure? How is the system of communica-has been achieved the less important is the
tion to channel knowledge and information exact form of the administrative organiza-
to the appropriate decision-points? What tion. For example, the function of plan
types of knowledge can, and what types can- preparation, or design, if it results in a
not, be easily transmitted? How is the need written plan that can be communicated
for intercommunication of information af- interpersonally without difficulty, can be
fected by the modes of specialization in the located almost anywhere in the organization
organization? This is perhaps the terra in- without affecting results. All that is needed
cognita of administrative theory, and un- is a procedure whereby the plan can be
doubtedly its careful exploration will cast given authoritative status, and this can be
great light on the proper application of the provided in a number of ways. A discus-
proverbs of administration. sion, then, of the proper location for a plan-
Perhaps this triangle of limits does not ning or designing unit is apt to be highly
completely bound the area of rationality, inconclusive and is apt to hinge on the per-
and other sides need to be added to the sonalities in the organization and their rel-
figure. In any case, this enumeration will ative enthusiasm, or lack of it, toward the
serve to indicate the kinds of considerations planning function rather than upon any
that must go into the construction of valid
and noncontradictory principles of adminis- For an example of the use of such training, see
tration. Herbert A. Simon and William Divine, "Controlling
Human Factors in an Administrative Experiment," 1
An important fact to be kept in mind is Public Administration Review 487-92 (Autumn, 1941).

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
66 PUBLIC ADMINISTRATION REVIEW

weights
abstract principles of good that are to be applied to these cri-
administration.'
On the other hand, when teria-to
factors the of com-
problems of their relative im-
munication or faiths or loyalty are crucial situation. This
portance in any concrete
question
to the making of a decision, theis location
not one thatofcan be solved in a
vacuum. Arm-chair
the decision in the organization is of greatphilosophizing about
importance. The method of administration-of
allocating which
de- the present paper
is an example-has
cisions in the army, for instance, gone about as far as it
automati-
cally provides (at least in can
theprofitably
period go in this particular direc-
prior
tion. What
to the actual battle) that each is neededwill
decision now is empirical re-
be made where the knowledge search and
isexperimentation
available to determine
the relative
for coordinating it with other desirability of alternative ad-
decisions.
Assigning Weights to the ministrative
Criteria.arrangements.
A first
The methodological
step, then, in the overhauling of the prov- framework for this
erbs of administration is research is already a
to develop at vo-
hand in the principle
of efficiency.
cabulary, along the lines just suggested,If an administrative
for organiza-
tion whose activities
the description of administrative organiza- are susceptible to ob-
tion. A second step, which jective evaluation
has also be subjected to study,
been
then the actual change
outlined, is to study the limits of rationality in accomplishment
in order to develop a complete that results
and from modifying administrative
compre-
arrangements in these organizations can be
hensive enumeration of the criteria that
observed and analyzed.
must be weighed in evaluating an adminis-
There are two indispensable conditions
trative organization. The current proverbs
represent only a fragmentary and unsys-successful research along these lines.
to
tematized portion of these criteria. First, it is necessary that the objectives of
When these two tasks have been carried the administrative organization under
out, it remains to assign weights to the cri- study be defined in concrete terms so that
teria. Since the criteria, or "proverbs," are results, expressed in terms of these objec-
often mutually competitive or contradic- tives, can be accurately measured. Second,
tory, it is not sufficient merely to identifyit is necessary that sufficient experimental
them. Merely to know, for example, that a control be exercised to make possible the
specified change in organization will reduce isolation of the particular effect under
the span of control is not enough to justify study from other disturbing factors that
the change. This gain must be balanced might be operating on the organization at
the same time.
against the possible resulting loss of con-
These two conditions have seldom been
tact between the higher and lower ranks of
the hierarchy. even partially fulfilled in so-called "admin-
Hence, administrative theory must also istrative experiments." The mere fact that
be concerned with the question of the a legislature passes a law creating an ad-
ministrative agency, that the agency oper-
1 See, for instance, Robert A. Walker, The Planning
Function in Urban Government (University of Chicago ates for five years, that the agency is finally
Press, 1941), pp. 166-75. Walker makes out a strong abolished, and that a historical study is then
case for attaching the planning agency to the chief
executive. But he rests his entire case on the rather made of the agency's operations is not suf-
slender reed that "as long as the planning agency ficient
is to make of that agency's history an
outside the governmental structure . . . planning will
tend to encounter resistance from public officials as an"administrative experiment." Modern
invasion of their responsibility and jurisdiction." ThisAmerican legislation is full of such "experi-
"resistance" is precisely the type of non-rational loyalty
ments" which furnish orators in neighbor-
which has been referred to previously, and which is
certainly a variable. ing states with abundant ammunition when

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms
 PROVERBS OF ADMINISTRATION 67

similar issues arise in their bailiwicks, but control or of objective measurements of

which provide the scientific investigator results, they have had to depend for their
with little or nothing in the way of objec- recommendations and conclusions upon a
tive evidence, one way or the other. priori reasoning proceeding from "princi-
In the literature of administration, there ples of administration." The reasons have
are only a handful of research studies that already been stated why the "principles"
satisfy these fundamental conditions of derived in this way cannot be more than
methodology-and these are, for the most "proverbs."
part, on the periphery of the problem of Perhaps the program outlined here will
organization. There are, first of all, the appear an ambitious or even a quixotic one.
studies of the Taylor group which sought to There should certainly be no illusions, in
determine the technological conditions of undertaking it, as to the length and devi-
efficiency. Perhaps none of these is a better ousness of the path. It is hard to see, how-
example of the painstaking methods of sci- ever, what alternative remains open. Cer-
ence than Taylor's own studies of the cut- tainly neither the practitioner of adminis-
ting of metals.' tration nor the theoretician can be satisfied
Studies dealing with the human and so- with the poor analytic tools that the prov-
cial aspects of administration are even rarer erbs provide him. Nor is there any reason
than the technological studies. Among the to believe that a less drastic reconversion
more important are the whole series of than that outlined here will rebuild those
studies on fatigue, starting in Great Britain tools to usefulness.
during World War I and culminating in It may be objected that administration
the Westinghouse experiments.2 cannot aspire to be a "science"; that by the
In the field of public administration, al- nature of its subject it cannot be more than
most the sole example of such experimenta- an "art." Whether true or false, this objec-
tion is the series of studies that have been
tion is irrelevant to the present discussion.
conducted in the public welfare field to The question of how "exact" the principles
determine the proper case loads for social of administration can be made is one that
workers.3
only experience can answer. But as to
Because, apart from these scattered ex-
whether they should be logical or illogical
amples, studies of administrative agencies
there can be no debate. Even an "art" can-
have been carried out without benefit of
not be founded on proverbs.
lF. W. Taylor, On the Art of Cutting Metals (Ameri-
can Society of Mechanical Engineers, 1907). 4 Social Work Technique 117-21 (May-June, 1938); Chi-
2 Great Britain, Ministry of Munitions, Health cago
of Relief Administration, Adequate Staff Brings
Munitions Workers Committee, Final Report (H.M. Economy (American Public Welfare Association, 1939);
Stationery Office, 1918); F. J. Roethlisberger and Wil- Constance Hastings and Saya S. Schwartz, Size of Visi-
liam J. Dickson, Management and the Worker (Har- tor's Caseload as a Factor in Efficient Administration
vard University Press, 1939). of Public Assistance (Philadelphia County Board of
sEllery F. Reed, An Experiment in Reducing the Assistance, 1939); Simon et al., Determining Work
Cost of Relief (American Public Welfare Administra- Loads for Professional Staff in a Public Welfare Agency
tion, 1937); Rebecca Staman, "What Is the Most Eco- (Bureau of Public Administration, University of Cali-
nomical Case Load in Public Relief Administration?" fornia, 1941).

This content downloaded from

52.172.201.146 on Sun, 16 Aug 2020 08:32:34 UTC
All use subject to https://about.jstor.org/terms