‘aul Appleby in any ofits fields of operations —raw ma- 1s, marketing, production, labor relations, ‘ce, and management, Yetin relation othe ‘ed States government even the very largest ‘oration is small and simple. And the more corporations we have, the more complex ¢ the government become, Govemment, ing in one way or another with almost ‘yihing, requires in its highest officials a Jal competence in handling relationships ing all the varied and powerful forces, ac- jes, and elements in the country. At the top job is that of managing relationships be- “nthe complex pats of the entire nation, of ng both form and leadership t the lite of ‘whole people, At that level itis an art—the of politics. Only a politician ean be Presi . The President needs economic under- ving, but he should not function as an rnomist; he needs legal understanding, bot should not function as lawyer, he needs iness understanding, but he should not ion as a businessman; he needs social un standing, but he should not function asso: logist; he needs understanding of research, he should not function as a scientist; he ‘ds understanding of agriculture, labor, tnce, but he should not function asa farmer, corer, or banker. He needs to understand ‘= broadly in orde to understand polities: success or fare as President depend on wv he functions as a politician. AC its best, ites is statesmanship. Staterafi—government—is diffrent from other professions because itis bro than sting else inthe field of ation. Purely speo- tive thought and emotion may range wider Id, yot even this may be doubted, fr govern sat must be concersed with intellectual and totional outreachings 190. Government is erent because it must take account of all the sires, needs, actions, thoughts, and senti- ants of 140,000,000 people. Government is ‘erent because government is politics. The Proverbs of Administration - Herbert A Simon 7 The Proverbs of Administration Herbert A, Simon A fact about proverbs that greatly enhances their quotability is that they almost always ‘occur in mutually contradictory pairs. “Look before you leap!”—but “He who hesitates is fost” ‘This is both a convenience and a serious efoct—depending on the use to which one wishes to put the proverbs in question. If itis a ‘matter of rationalizing behavior that has al- ready taken place or justifying action that has already been decided upon, proverbs are ideal. Since one is never ata loss to find one that will rove his point—or the precisely contradictory point, for that matter—they are a great help in persuasion, politcal debate, and all forms of thetorie. [But when one seeks fo use proverbs as the basis ofa scientific theory, the situation is less happy. It isnot thatthe propositions expressed by the proverbs are insufficient itis rather that they prove too much. A scientific theory should tell what is true but also what is false. If New- ton had announced tothe world that particles of ‘matter exert either an attraction ora repulsion ‘on each other, he would not have added much to scientific knowledge. His contibution con sisted in showing that an attraction was ex ecised and in announcing the precise law governing its operation “Most of the propositions that make up the body of administrative theary today share, un- fortunately, this defect of proverbs. For almost cevery principle one ean find an equally plau sible and acceptable contradictory principle. Although the two principles of the pair will “See Red with penn on Pale Abaeraion Rei © 1946by te Aresen Sef Pi Adie tenon ASPA) 106 Soee NW Sut 70, Wats, Dic 20018, Al eho seed 16 lead to exactly opposite organizational recom- ‘mendations, there is nothing in the theory toi dicate whichis the proper one to apply" Ikis the purpose ofthis paper to substantiate the sweeping criticism of administrative theory, and to present some suggestions —perhaps less ‘concrete than they should be—as to how the ‘existing dilemma can be solved. Some Accepted Administrative Principles ‘Among the more common principles that ‘occur in the literature of administration are these: 1, Administrative efficiency is increased by a specialization of the task among the group. 2. Administrative efficiency is increased by ar- ranging the members of the group in de terminate hierarchy of authority. 3. Administrative efficiency is increased by limiting the span of control at any point in the hierarchy fo a small number. 4, Administrative efficiency is increased by srouping the workers, for purposes of con- trol, according to (a) purpose, (b) process, (© clientele, (@) place. (This is really an elaboration of the firs princple but de- serves separate discussion) Since these principles appear relatively simple and clea, it would seem that their ap- plication to conerete problems of administra- tive organization would be unambiguous and that their validity would be easily submitted to empirical test. Such, however, seems not to be the case. To show why its not each ofthe four principles just listed will be considered in turn. Specialization, Administrative efficiency {is supposed to increase with an increase in spe- cialization, But is this intended to mean that ‘any increase in specialization will increase efficiency? If so, which ofthe following ater natives is the correct application of the prin- ciple in a particular case?128 “The Proves of Adminstration » Herbert A. Simon 1. A plan of nursing should be put into effect by which nurses will be assigned to districts and do all nursing within that district, inctuding. school examinations, visits 10 homes of school children, and tuberculosis nursing 2. A functional plan of nursing should be put into effect by which diferent nurses will be assigned to school examinations, visits to homes of school children, and tuberculosis nursing. The present method of generalized nursing by districts impedes the develop- ‘ment of specialized skis in the three very diverse programs. Both of these administrative arrangements satisfy the requirement of specialization —the first provides spocialization by place; the sec ‘ond, specialization by function. The principles ‘of specialization is of no help at all in choosing between the two altematives. appears that the simplicity ofthe principle ‘of specialization is a deceptive simplicity—a simplicity which conceals fundamental ambi ites. For “specialization” is nota condition of efficent administration; itis an inevitable characteristic of all group effort; however cfficent or inefficient that effort may be. Spe- cialization merely means that different persons sre doing different things—and since it is physically impossible for two persons tobe do- ing the same thing in.the same place at the same time, two persons are always doing dif- ferent things. ‘The real problem of administration, then is not to “specialize,” bt to specialize in that par- ticular manner and along those particular lines ‘which will lead to administrative efficiency ‘Bot, in thus rephrasing this “principle” of ad- ‘ministration, there has been brought clearly tothe open its fundamental ambiguity: “Ad- ministrative efficiency is increased by a spe cialization of the task among the group in the ol, and staffed by men of similarly appro- te background. What reason is there (0 be- > that if attached toa Ministry of Education * would teach old-fashioned farming by -fashioned methods, while if attached to a istry of Agriculture they would teach new tioned farming by old-fashioned methods? administrative problem of such a bureau id be to teach new-fashioned farming by fashioned methods, and itis a litle dif- Ito see how the departmental location of ‘uit would affct this esult. “The question ‘vers itself" only if one has 2 rather mystical 1 in the potency of bureau-shuffing as a ins for rediecting the activities of an ey, These contradictions and competitions have sived increasing attention from students of sinistration during the past few years. For ‘mple, Gulick, Wallace, and Benson have ed certain advantages and disadvantages of several modes of specialization, and have sidered the conditions under which one or ‘other mode might best be adopted. All, analysis has been ata theoretical leve!—in sense that data have not been employed to onstrate the superior effectiveness claimed the different modes. But though theoretical, analysis has lacked a theory. Since no prehensive framework has been con- tcted within which the discussion could take 2e, the analysis ha tended either tothe logi- ‘one-sidedness which characterizes. the _mples quoted above orto inconelusivenes, The impasse of administrative theory. The + “principles of administration” that were “The Proves of Administration » Herbert. Simon 135 set forth at the beginning of this paper have now been subjected to critical analysis. None of the four survived in very good shape, for in each case there was found, instead of fn unequivocal principle, a set of two or ‘mote mutually incompatible principles appar- cently equally applicable to the administrative situation. Moreover, the reader will sce thatthe very same objections can be urged against the cus- tomary discussions of “centralization” versus “ecentralzation,” which usually conclude in cffect, that “on the one hand, centralization of ‘decision-making functions is desirable; on the other hand, there are definite advantages in decentralization.” ‘Can anything be salvaged which will be ‘wef inthe construction of an administrative theory? As a matte of fact, almost everything canbe salvaged. The difficulty as arisen from treating as principles of administration” what are really only criteria for describing and diag nosing administrative situations. Closet space is certainly an important item in the design of socessful house; yet house designed entirely ‘with a view to securing a maximum of closet space—all other considerations being forgot- ten—would be considered, to say the least, somewhat unbalanced. Similarly, nity of com- mand, specialization by purpose, decentraliza- tion areal items to be considered inthe design of an efficient administrative organization. No single one ofthese items is of sulicent impor- tance to suffice as a guiding principle for the administrative analyst. In the design of admin- istrative organization, as in their operation, overall efficiency must be the guiding orite- rion, Mutually incompatible advantages must be balanced against each other, just s an archi- tect weighs the advantages of additional closet space against the advantages of a larger living ‘This position, fit isa valid on, constitutes ‘an indictment of much current writing about administrative matters. As the examples cited in this chapter amply demonstrate, much admin istrative analysis procoods by selecting a single criterion and applying it 0 an administrative situation to reach a recommendation; while the fact that equally valid but contradictory, erite- rinexist which could be applied with equal rea- son, but wit a different result, is conveniently jgnored. A. valid approach to the study of ‘eiministration requires that all the relevant d- ‘agnostic criteria be identified; that each admin istrative situation be analyzed in terms of the entire set of criteria; and that research be insti- tuted to determine how weights can be as- signed to the several riteria when they are, as ‘they usually will be, mutually incompatible. An Approach to Administrative Theory ‘This program needs fo be considered step by step. First, what is included inthe descrip- tion of administrative situations for purposes of soch an analysis? Second, how can weights be assigned to the Various criteria to give them their proper pace in the total pictare? The description of administrative situa- tions, ‘Before science can develop principles, itmust possess concepts. Before a law of gra itaion could be formulated, it was necessary to have the notions of “acceleration” and “yeight”” The first task of administrative the- cory isto develop a set of concepts that will per- tit the description in terms relevant to the ‘theory, of administrative situations. These con- cepts, tobe scientifically useful, must be opera- tional; that is, their meanings must correspond to empirically observable facts or situations. ‘The definition of “authority” given earlier in this paper is an example of an operational definition, ‘What is scientifically relevant description cof an organization? It isa description tha, so far as possible, designates for each person in the organization what decisions that person makes and the influences to which he is subject in making each ofthese decisions. Curent de- scriptions of administrative organizations fall far short of this standard. For the most part, they confine themselves to the allocation of functions and the formal structure of authority. ‘They give little attention tothe oth types of organizational influence or to the system of communication. * ‘What does it mean, for example, to say: “The department is made up of three bureaus.136 ‘The Provebs of Administration - Herbert A. Simon ‘The ist hasthe function of _ the second the function of andthe third the func tion of ‘What can be learned from such a description about the workability ofthe ‘organizational arrangement? Very litle, in- deed. For from the description there is obigined no idea of the degree t0 which decisions are ‘centralized a the bureau level or atthe depart: ‘mental level. No notion is given as to the extent to which the (presumably unlimited) authority of the department over the bureau is actually ‘exercised or by what mechanisms. There is no indication of the extent to which systems of ‘communication assist the coordination of the throe bureaus or, for that mater, to what extent coordination is required by the nature of their work: There is no description of the kinds of ‘waning the members ofthe bureau have undet- gone o of the extent to which this training per- mits decentralization at the bureau level. in sum, a description of administrative organiza- tions in teams almost exclusively of functions and lines of authority is completely inadequate {or purposes of administrative analysis, Consider the term “centralization.” How is it determined whether the operations of a par ticular organization are “centralized” or “de- centralized”? Does the fact that field offices ‘exist prove anything about decentralization? ‘Might not the same decentralization tke place in the bureaus of a centrally located office? A realistic analysis of centralization must include study of te allocation of decisions inthe or- ‘ganization and the methods of influence that fare employed by the higher levels to affect the decisions atthe lower levels. Such an analysis ‘would reveal a much more complex picture of the decision-making process than any enumer- ation ofthe geographical locations of organiza tional units atthe different levels. Administrative description sufers currently from superficiality, oversimpliication, lack of realism, Itas confined itself too closely tothe ‘mechanism of authority and has filed to bring within its orbit the other, equally important, ‘modes of inuence on organizational behavior It has refused to undertake the tiresome take of studying the sctual allocation of decision- ‘making functions. Ithas been satisfied to speak of “authority,” “centralization,” “span of con- “function,” without secking operational efinitions ofthese terms. Uatil administra, description reaches a higher level of sophiy. ticaton, there is litle reason to hope tht rap progress will be made toward the ident cation and verification of valid administrative principles Does this mean that a purely formal de- scription of an administrative organization is impossible—ahat a relevant description must include an account ofthe content ofthe organ zation's decisions? This isa question thats a. ‘mast impossible to answer in the present site (of knowledge of administative theory. One ‘thing seems certain: content plays a greater role in the application of administrative principles than is allowed fori the formal administrative theory ofthe present time. This isa fact that is bbepinning to be recognized in the literature of administration. If one examines the chain of publications extending from Mooney and Reil- Jey, through Gulick and the President's Com miltee controversy, to Schuyler Wallace and Benson, he sees a steady shift of emphasis fom the "principles of administration” themselves toa study of the conditions under which com- peting principles are respectively applicable Recent publications seldom say that “organi zation should be by purpose,” but rather that “under such and such conditions purpose orga- nization is desirable.” ILis to ese conditions ‘Which underlie the application of the proverbs of administration that administrative theory and analysis must turn in their search for really valid principles to replace the proverbs ‘The diagnosis of administrative situations. Before any positive suggestions car be made it is mecessary 10 digress a bit and to consider ‘more closely the exact nature ofthe propesition of administrative theory. The theory of admin- istration is concerned with how an organization should be constructed and operated in order to accomplish its work efficenly. A fundamental principle of administration, which follows al- ‘most imumediately from the rational character cof “good” administration, i that among several altematives involving the same expenditure that one should always be selected which leads to the greatest accomplishment of administra- tive objectives; and among several alternatives that lead tothe same accomplishment that one1 + Herbert. Simon on pepe eed eae ee eee coor eee Se Sect pra ae es Varies aeete ed The diagnosis of administrative situations. sfore any postive suggestions can be made, it necessary to digress a bit and to consider ‘ore close the exact nature of the proposition administrative theory. The theory of admin- -ration is concered with how an organization ‘nkd be constructed and operated in order to compli ts work efficiently. A. fundamental inciple of administration, which follows al- ost immediately from the rational character “good” administration, is that among several cematives involving the same expenditure atone should always be selected which leads the greatest accomplishment of administra € objectives; and among several alternatives itlead tothe same accomplishment that one ‘The Proverbs of Administration - Herbert A. Simon 137 seus be seed which involves the east fe. Since is "principle of fiieney"s vensnsic of any actvy Oa tps 1 her fo maximize the aiomentof certain tote wih the se of scarce means, its as Started of econ toory as itis fa aerne theory. Te aise ma” thes his lace alongside the clasical“eo- some man" ctl, the principle” feficieny shold te conseed a deinton rather than a pin Sista deinton of what i meant by eo" or comet” adnate behav. fee oe tl how accomplishment are to be eect, but merely sates Hat this mas tain i the sm of administrate actin, fn at adminisrte teary mast disclose hee what condos he maximization takes ace. : Now what are the factors that determine the eve ofelceney which achive by an a- pinisratv organization? Ie is not posible to mk an exons st of ese bu he prin pal ecegrics ca be ener raps the ples method of ppc ito consider the Sage ember ofthe adnate organiza Son ad ack what he its are to the quantity tnd goat of his cot, Tese its ncude {2 Limits on his bit to peor and @) in {on his ably to make Comet deisions. To te exten hat these inte are removed te ad- instateonanzton apace is gal of High eflency, Two persons, even the same Sil teste objectives and ales, Be same Iowieige and infra, ean anally dei only upon te same couse of ection. Hence ainstatve tory st be i eed fn he factors tht wil determine with tht sis, vals, and knowledge th ozani- Jaton metiber undertakes his work. These ae the "Limits oration wih wish be pee plesof aniston must deal ‘On one side, the dvi United by those als, habis, and reflexes which ae no Tong in he rei of he conscous His per formanes for example, maybe Kime by his manual dexterity or Bis ration Gime of his Strength. His decison making prooeses may te imitate sped of is mental process, his sil in elementary arid, an 9 fot In this area, dhe principles of sdisistaion _must be concerned with the physiology of the human body and withthe laws ofskll- training and of habit. Tiss the fel that has been most successfully cultivated by the followers of Tay Tor and in which has been developed ime-and- ‘mation study, ‘On a second side, the individual is limited by his values and those conceptions of purpose ‘which influence him in making his decisions. It his loyalty to the organization is high, his deci- sions may evidence sincere acceptance of the “objectives set for the organization; if that loy- alty is lacking, personal motives may interfere with his administrative efficiency If his loyal ties are attached to the bureau by which he is ‘employed, he may sometimes make decisions that are inimical fo the larger unit of whic the ‘bureau isa part. In this area the principles of administration must be concerned with the determinants of loyalty and morale, with lead- ership and initiative, and withthe influences that determine where the individual's organiza- tional loyalties wil be attached. ‘Ona tind side, the individual is timited by the extent of his knowledge of things relevant to his ob, This applies both tothe basic knowl- edge required in decision making—a bridge designer must know the fundamentals of me- cchanics—and © the information that #8 re ‘quired to make his decisions appropriate to the given situation In this area administrative the- ‘ory is concerned with such fundamental ques- tions as these: What ae the mits on the mass of knowledge that human minds can accum- Tate and apply? How rapidly can knowledge be assimilated? How is specialization in the ad- ministrative organization to be related to the specializations of knowledge that are prevalent inthe community's occupational structure? How is the system of communication to canoe! knowledge and information to the appropriate decision-points? What typesof knowledge can, ‘and what types cannot, be easily transmitted? How isthe need for intercommnication of in- formation affected by the modes of spcislza- tion in the organization? This is perhaps the terra incognita of administrative theory, and undoubtedly its careful exploration will cat great light on the proper application of the proverbs of administration. Perhaps this triangle of limits does not138 ‘The Proverbs of Administration - Herbert A. Simon ‘completely bound the area of rationality, and other sides need to be add tothe figure. In any case, this enumeration wil serve to indi- cate the kinds of considerations that must g0 into the construction of valid and non-conta- dctry principles of administration. ‘An important fact tobe kept in ming is that the limits of rationality are variable limits, Most important of all, consciousness of the limits may in tseif alter them. Suppose it were discovered ina particular organization, for ex- ample, that organizational loyalties attached 10 small units had frequently led to harmful de- ‘7c of intraorganizational competition. Then, ‘ program which trained memibers ofthe orga- nization to be conscious of their loyalties, and to subordinate loyalties to the smaller group to those of the large, might lead 10 a very ‘considerable alteration of the limits in that corganization.* ‘A related points that the term “rational be- hhaviog” as employed here, refers to rationality ‘when that behaviors evaluated in terms ofthe objectives ofthe lager organization; fr, as just pointed out, the difirence in direction ofthe individual's aims from those ofthe larger orpa- nization is just one of those elements of non- ‘tionality with which the theory must deal. ‘A inal observation is that, sine administa- tive theory is concerned with the nonrational Timitsof the rational, i follows thatthe larger the aeain which rationality hs been achieved the less important is the exact form of the ‘sdministrtive organization, For example, the function of plan preparation, or design, if ite sults in a writen plan that can be communi- cated interpessonally without dificaity, can be located almost anywhere in the organization without affecting results. All thats needed i a procedure whereby the plan can be given au- thoritatve status, and this can be provided in ‘a number of ways. A discussion, then, of the proper locaton fora planning or designing unit is apt to be highly inconclusive and is ape to hhinge on the personalities in the organization and their relative enthusiasm, or lack oft 0- ward the planning function rathec than upon any abstract principles of good administration. ‘On the other hand, when factors of commu- ication or faiths or loyalty are crucial t0 the making of a decision, the location ofthe dag son inthe organization i of great importa, ‘The method of allocating decisions in the amy for instance, automatically provides (at lege jn the period prior to the actual bate) thy each decision will be made where the kno ‘edge is available for coordinating it With oth decisions. Assigning weights tothe criteria. A fing step, then, in the overhauling ofthe proverbs cf ‘administration is to develop a vocabulay, ‘along the lines just suggested, for the descrip tion of administrative organization. A seco step, which has also been outlined, is to study the limits of rationality in order t develop a complete and comprehensive enumeration of the criteria that must be weighed in evaluating fan administrative organization. The curest proverbs represent only a fragmentary and un- systematized portion ofthese criteria When these two tasks have been carried ‘ut, it remains to assign weights to the criteria Since the criteria, or “proverbs” are often mu- tually competitive or contradictory, itis not saficient merely to identify them. Merely to ‘know, for example, that a specified change in ‘organization will reduce the span of conta is not enough to justify the change. This gain snus be balanced against the possible resulting Joss of contact between the higher and lower ranks ofthe hierar Hence, administrative theory must also be concerned with the question ofthe weights tha ‘are tobe applied to these eriteria—to the prob- Jems of ther relative importance in any con- ‘rete situation. This question is net one that can be solved in a vacuum. Armehar philosophiz- ing about administration—of which the pres- cnt paper is an example—has gone about as far a8 it ean profitably go inthis particular di rection, What is needed now is empirical re search and experimentation to determine the relative desirability of altemative administra tive arrangements “The methodological framework for this re- search is aleady at hand in the principle of efficiency. If an administrative organization whose activities are susceptible to objective evaluation be subjected to study, then the1+ Herbert A. Simon luking of a decision, the location ofthe dec ‘om in the organization is of great importance, he method of allocating decisions in the army, v instance, automatically provides (at leat 1 the period prior to the actual battle) that sch decision willbe made where the knowi- Age is available for coordinating it with other Assigning weights tothe criteria. A fis, ep, then, in th overhauling of the proverbs of ministration is to develop 2 vocabulary ‘ong the lines just suggested, for the descrip on of administrative organization, A. second =p, which has also been outlined, isto study «limits of rationality in order to develop a rmplete and comprehensive enumeration of e criteria that must be weighed in evaluating \ administrative organization, The current ‘verbs represent only a fragmentary and un- stematized portion ofthese criteria, When these two tasks have been carried 1 itemains to assign weights to the criteria ‘ce the criteria, or “proverbs.” are often mu- ally competitive or contradictory, it is not ficient merely wo identify them. Merely to ow, for example, that a specified change in ganization will reduce the span of contol is 4 enough to justify the change. This gain sste balanced against the possible resulting 's of contact between the higher and lower 1ks ofthe hierarchy. Hence, administrative theory must also be cemed with the question of the weights that tobe applied to these criteria —to the prob- 1s of their relative importance in any eon- ‘Ae stuation. This question is not one that can solved ina vacuum, Armchair philosophiz § about administration—of which the pres- paper is an example—has gone about as as it can profitably go inthis particular di tion. What is needed now is empirical re atch and experimentation to determine the ative desirability of alternative administra > arrangements ‘The methodological framework for ths re sch is already at hand in the principle of ciency. If an administrative organization se activities are susceptible to objective ‘uation be subjected to study, then the ‘The Proves of Administration - Herbert A. Simon 139 sual change in accomplishment that results fom modifying administrative arrangements fn these organizations can be observed and analyzed “There are two indispensable conditions to ccessfl esearch along these lines. Firs itis necessary thatthe objectives of dhe aiministra~ te organization under study be defined in con- fue terms so that results, expressed in terms Of these objectives, can be accurately mea- sured, Second, its necessary that sufficient ex perimental contol be exercised to. make possible the isolation ofthe particular effect tinder stay from other disturbing factors that ‘ight be operating on the organization at the same ime. "These two conditions have seldom been ven partially fulfilled in so-called “administra- tie experiments." The mere fat that legisla- ture passes a law creating an administrative agency, thatthe agency operates for five years, thatthe agency is finally abolished, and that a historical study is then made of the agency's ‘operations is not sufficient to make of that agency's history an “administrative experi: ment.” Modern American legislation is fall of, such “experiments” which furnish orators in neighboring states with abundant ammunition ‘when similar issues arise in their baiiwicks, but which provide the scientific investigator with itl or nothing in the way of objective ev- ince, one way or the other. In he literature of administration, there are only a handful of research studies that satisfy ‘hese fundamental conditions of methodol- ‘ogy-—and these are, for the most part, on the periphery of the problei of organization. There ae, frst of all, the studies ofthe Taylor group which sought to determine the tecno- logical conditions of efficiency. Perhaps none ofthese is a beter example of the painstaking ‘methods of science than Taylor's own stidies ofthe cutting of metals.” Studies dealing with the human and soci sspects of administration are even rar than the technological studies. Among the more im- portant are the whole series of studies on fa tigue, starting in Great Britain during World War T and culminating in the Westinghouse experiments. * Inthe field of public administration, almost the sole example of such experimentation isthe series of studies that have been conducted in the public welfare field to determine the proper «ase loads for social workers.” ‘Because, apart from these scattered ex amples, studies of administrative agencies have been carried out without benefit of control or of objective measurement of results, they have had to depend for their recommendations and conclusions upon a priori reasoning proceed ing from “principles of administration.” The reasons have already been stated why the “principles” derived in this way cannot be more than “proverbs: Pechaps the prograin outlined ere will ap- pear an ambitious or even a quixotic one. There should certainly be no illusions, in undertaking it as tothe Fength and deviousness ofthe path Itis hard to see, however, what allemative re ‘mains open, Certainly neither the practitioner ‘of administration nor the theoretican can be satisfied with the poor analytic tools thatthe proverbs provide him. Nori there any reason to believe that a less drastic recomversion than that outined here will rebuild those tools to usefulness. Tay be objected that administration can not aspire to be a “science; that by the nature ofits subject it cannot be more than an “art.” ‘Whether true or false, this objection is irrele- vant tothe present discussion. The question of how “exact” the principles of administration can be made is one that only experience can nswer, But as to whether they shouldbe logi- cal ollogical thee can be no debate, Even an “are cannot be founded on proves. Notes 1. Lestitbe thought that ths deicieny is poco liar to the science —or “art of minis tion, it should be pointed out thatthe same trouble is shared by most Freudian psycho- Jogical theories, s well as by some socolog- ical theories 2. Lather Gulick, “Notes on the Theory of Or- saization,” in Papers om the Science of140 10. u ‘The Provetbs of Administration Adiniseration, e. Ler Gulick and L. Ur ‘wick (nsimte of Public Administration, Co- lumbia University, 1997), p. 9. “This points discused in Herbert A. Simon, “Decision-Making and Administrative Orga nization,” Pub, Adi. Rev. 4 (wintee 194): 20-21 Golick, op.cit,p.95L.D. Whit, Introduction to the Study of Public Administration (Mac- nillan Co, 1939),p. 48. Frederick W. Taylor, Shop Management (Harper and Bros. 1911), p- 99: Macho, Millet, and Ogden, The Administration of Federal Work Relef Public Administration Service, 1941), pp. 265-268; and L. Ure wick, who describes British army pretice in “Organization as a Technical Problem,” in Papers om the Science of Administration, Lather Gulick and L. Urwick dastiute of Public Administration, Cohumbia Universit 1937), pp.67-€8. Gatick, op cit, p21 If this is corect, then any attempt to prove that certain activites belong ina single de partment because they relat toa single pur- pose is doomed to fail. Ses, for example, ‘John M, Gaus and Leon Wolet, Public Ad: ‘ministration and the US. Deparment of Asgricudtre (Public Administration Service, 1540) Gato, op. cit, p. 23, This distinction is implicit in most of Glick’ analysis of specialization, However, since he cites as examples single depar- ‘ments within a city, and since be usually speaks of “grouping activites” rather than “dividing work,” the relative character of these categories snot vas apparent inthis Adisussion (Gulick, op. ct, pp. 15-30). Report of the Machinery of Government Commie (H. M, Stationery Ofice, 1918). Sir Charis Harts, “Dooentalization,* four- nal of Public Administration 3 (Apel 1925): 117-233. Gack, op. cit, pp. 21~30; Schuyler Wal- lace, Federal Deparmentalization (Colum bia University Press, 1941); George C. 5, ‘Benson, “Intemational Administrative Org B. 14 1s. n. 18 Herbert, Simon nization,” Pub, Adm. Rev, | (automa 1941), 413-486, "The monograph by Macmabon, Milt, nd Ogden, op. cit, perhaps approaches nearee than any other publihed administrate sendy tothe sophistication required in admin. istrative description. See, for example, the scussion on pp. 233-736 of headquarters. field relationships, For an elaboration of the principe of efi. dency and its place in administrative theory see Clarence E. Ridley and Herbert A. Si. ‘mon, Measuring Municipal Activites (ner. rational City Managers’ Association, 2 ed, 1943), paricuarly chap. 1 and the preface to the second edition, Foran example ofthe ux of such training, see Hetbert A. Simon and William Divine, “Controlling Haman Factors in an Adminis- trative Experiment.” Pub Adm Rev. 1 (ax ‘umn 1981): 487492. See, for instance, Robert A, Walker, The Planning Function in Urban Government (Cniversiy of Chicago Pres, 1941), pp. 165- 175, Walker makes out ston cas fora ‘aching the planning agency to the chief ex- ‘ecutive, But he rests his entire case on the rather slender reed that “ss long asthe plan hing agency is outside the governmental stricture. planning will tend to encounter resistance ffom public officials as an inva sion oftheir responsibilty and jurisdiction: This “resistances precisely the type of non tational loyalty which has been refered to previously nd whichis certainly a variable FW. Taylor, On the Ar of Caring Metals (American Society of Mechanical Engi- neers, 1907). Great Britain, Ministry of Munitions, Healt, cof Munitions Workers Commitee, Final Re: port (L. M. Stationery Office, 1918); F 4. Roethlisberger and William’ J. Dickson, Management and the Worker Harvard Usi- versity Press, 1939), Ellery F Reed, An Experiment in Reducing ‘the Cost of Relief (American Public Welfare ‘Administration, 1937}; Rebecca Starman, “What Is the Most Economical Case Load in Public Relief Administration?” Social WorkHerbert. Simon nization,” Pub. Adm Reo (autumn 194) 473-486, ‘The monograph by Macmahon, Milt, ang Ogden, op. cit, perhaps approaches nea, than any other published administrate study tthe sophistication required in amin istrative description. See, for example, th clscussion on pp. 283-236 of headquaten. field relationships. FFor an elaboration of the principle of ef. ciency and its place in administrative theory see Clarence E. Ridley and Hetbert A. S mon, Measuring Municipal Activites (ier. national City Managers’ Association, 24 ed, 1943), particularly chap {andthe preface the second edition. For an example ofthe use of such traning, see Herbert A. Simoa and William Divine, “Controlling Human Factors in an Adminis: tustive Experiment” Pub. Adm Rev. 1 (a0 ‘umn 1941): 487-492 See, for instance, Robert A. Wilkes, The Planning Function in Urban Government (Cniversity of Chicago Press, 1941), pp. 166 175. Walker makes ont a stong case for s- teching the planning agency to the chet ex ‘cutive. But he rests his emie case on the rather slender reed that “as long asthe plan ning agency is outside the governmental structure... planning wil tend to encounter resistance from public oficials as an inva son oftheir responsibilty and jorisdiction” This “resistance is precisely the type of non rational loyalty which has boen referred to previously and which is certainly a variable. EW, Teploc, On the Art of Cutting Metals (American Society of Mectunical Engi- neers, 1907). Great Britain, Ministry of Munitions, Health ‘of Munitions Workers Committe, Final Re- port (HM, Stationery Ofte, 1918), F. 1 Roethlisberger and William J. Dickson, Management and the Worker (Harvard Uni versity Press, 1939). Ellery F. Reed, An Experiment in Reducing the Cos of Relief (American Public Welfare Administration, 1937); Rebecca Staman, “What Is the Most Economical Case Load in Public Relief Administration?” Social Work ‘The Prowsbs of Administration Technique 4 (May-June, 1938): 117~121; © Chicago Relief Administration, Adequate ‘Sta’ Brings Economy (American Public ‘Welfare Association, 1939); Constance Has ings and Saya S. Schwartz, Size of Visior’s Caseload as a Factor in Efficient Adminis- Herbert A, Simon 141 tration of Public Assistance (Philadelphia County Board of Assistance, 1929), Simon tal, Determining Work Loads for Profes- sional Staff ina Public Welfare Agency (Bu- eau of Public Administration, University of California, 194.