CHAPTER 1

MANAGEMENT AND SCIENTIFIC KNOWLEDGE

Every professional is concerned with the use of knowledge in the achievement ofobjectives: the engineer as he designs equipment, the medical practitioner as hediagnoses and prescribes for the ills of his patients, the lawyer or thearchitect as he serves his clients. The professional draws upon the knowledge ofscience and of his colleagues, and upon knowledge gained through personalexperience. The degree to which he relies upon the first two of these ratherthan the third is one of the ways in which the professional may be distinguishedfrom the layman.

It is beginning to be possible for the industrial manager to be a professionalin this respect. He can draw upon a reasonable and growing body of knowledge inthe social sciences as an aid to achieving his managerial objectives. He neednot rely exclusively on personal experience and observation.

Progress in any profession is associated with the ability to predict andcontrol, and this is true also of industrial management. One of the major tasksof management is to organize human effort in the service of the economicobjectives of the enterprise. Every managerial decision has behavioralconsequences. Successful management depends—not alone, butsignificantly—upon the ability to predict and control human behavior.

Our ability along these lines today is spotty. It is remarkably good in somerespects. Consider such everyday acts as making an appointment, signing apurchase agreement, placing a long-distance call, asking a subordinate toprepare a report, making a hotel reservation, mailing a letter. In literallythousands of ways we predict with a high degree of accuracy what others will do,and we control their behavior in the sense that our actions lead to the desiredconsequences.

At the same time, it is true that other attempts at prediction and control arequite inadequate. Many of the important social problems of our time reflect thisinadequacy: juvenile delinquency, crime, the high traffic fatality rate,management-labor conflict, the cold war.

The results so far achieved in the management of business and industry reflectconsiderable ability to predict and control human behavior. The fact that acompany is economically successful means, among other things, that managementhas been able to attract people into the organization and to organize and directtheir efforts toward the production and sale of goods or services at a profit.Nevertheless, few managers are satisfied with their ability to predict andcontrol the behavior of the members of their organizations. The interestexpressed in new developments in this field is an indication of management'srecognition of the opportunity for improvement. The frequent success of theoutright charlatan in peddling managerial patent medicines also reflects theconsciousness of inadequacy. Many managers would agree that the effectiveness oftheir organizations would be at least doubled if they could discover how to tapthe unrealized potential present in their human resources.

I share with some of my colleagues the conviction that the social sciences couldcontribute more effectively than they have to managerial progress with respectto the human side of enterprise. There are, of course, many reasons whyimprovement has been slow. Some have to do with the social sciences themselves:they are still in their adolescence in comparison with the physical sciences;their findings are piecemeal and scattered; they lack precision; many criticalissues are still in controversy. These are relative matters, however. One needonly contrast the situation today with that thirty years ago to recognize thatmuch has been accomplished. The social sciences are a rich resource today formanagement even though they have not reached full maturity.

I am not particularly impressed with arguments that social scientists do notpublish their findings in language intelligible to the layman. Neither dophysicists! Also, while it is lamentable that some social scientists jumpincautiously from relatively precarious theory to practical applications, andothers refuse to concern themselves at all with applications, there is nothingunique about social science in these respects. Today most managers are forced torely on "middlemen" in the form of social scientist consultants or staff, or onliterature intermediate between scientific journals and the Sunday supplementsto interpret theory and research or to help them judge the scientific adequacyof claims or proposals. The time is not far off when the competentmanager—like any other professional practitioner—will find it anecessity to be well enough versed in the scientific disciplines relevant tothis work to be able to read the literature and judge the adequacy of scientificfindings and claims.

This is not to say that we social scientists can ignore our responsibilities. Itis to say that the position of the manager vis-à-vis the social sciences willone day be no different than that of the engineer vis-à-vis the physicalsciences or the doctor vis-à-vis chemistry or biology. The professional need notbe a scientist, but he must be sophisticated enough to make competent use ofscientific knowledge.

EVERY MANAGERIAL ACT RESTS ON THEORY

There are some other reasons why management has been relatively slow to utilizesocial science knowledge. Two of these are especially important. The first isthat every manager quite naturally considers himself his own social scientist.His personal experience with people from childhood on has been so rich that hefeels little real need to turn elsewhere for knowledge of human behavior. Thesocial scientist's knowledge often appears to him to be theoretical andunrelated to the realities with which he must deal, whereas his own experience-basedknowledge is practical and useful.

This frequent, invidious comparison of the practical and the theoretical withrespect to the management of human resources has been a severe handicap toprogress in this field. It has led to premature and misguided attempts totranslate scientific findings into action; it has permitted the quack and thecharlatan to peddle worthless gimmicks and programs.

Every managerial act rests on assumptions, generalizations, andhypotheses—that is to say, on theory. Our assumptions are frequentlyimplicit, sometimes quite unconscious, often conflicting; nevertheless, theydetermine our predictions that if we do a, b will occur. Theory and practice areinseparable.

Next time you attend a management staff meeting at which a policy problem isunder discussion or some action is being considered, try a variant on thepastime of doodling. Jot down the assumptions (beliefs, opinions, convictions,generalizations) about human behavior made during the discussion by theparticipants. Some of these will be explicitly stated ("A manager must himselfbe technically competent in a given field in order to manage professionalswithin it."). Most will be implicit, but fairly easily inferred ("We shouldrequire the office force to punch time clocks as they do in the factory."). Itwill not make too much difference whether the problem under discussion is ahuman problem, a financial or a technical one. Tune your ear to listen forassumptions about human behavior, whether they relate to an individual, aparticular group, or people in general. The length and variety of your list willsurprise you.

It is possible to have more or less adequate theoretical assumptions; it is notpossible to reach a managerial decision or take a managerial action uninfluencedby assumptions, whether adequate or not. The insistence on being practicalreally means, "Let's accept my theoretical assumptions without argument ortest." The common practice of proceeding without explicit examination oftheoretical assumptions leads, at times, to remarkable inconsistencies inmanagerial behavior.

A manager, for example, states that he delegates to his subordinates. Whenasked, he expresses assumptions such as, "People need to learn to takeresponsibility," or, "Those closer to the situation can make the bestdecisions." However, he has arranged to obtain a constant flow of detailedinformation about the behavior of his subordinates, and he uses this informationto police their behavior and to "second-guess" their decisions. He says, "I amheld responsible, so I need to know what is going on." He sees no inconsistencyin his behavior, nor does he recognize some other assumptions which areimplicit: "People can't be trusted," or, "They can't really make as gooddecisions as I can."

With one hand, and in accord with certain assumptions, he delegates; with theother, and in line with other assumptions, he takes actions which have theeffect of nullifying his delegation. Not only does he fail to recognize theinconsistencies involved, but if faced with them he is likely to deny them.

Another common way of denying the importance of theory to managerial behavior isto insist that management is an art. This also precludes critical examination ofthe theoretical assumptions underlying managerial actions by placing reliance onintuitions and feelings, which are by definition not subject to question. Theissue is not whether management is a science. It is not. Its purposes aredifferent. Science is concerned with the advancement of knowledge; management,like any profession, is concerned with the achievement of practical objectives.The issue is whether management can utilize scientific knowledge in theachievement of those objectives. To insist that management is an art isfrequently no more than a denial of the relevance of systematic, testedknowledge to practice. So long as the manager fails to question the validity ofhis personal assumptions, he is unlikely to avail himself of what is availablein science. And much is there. The knowledge in the social sciences is notsparse, but frequently it contradicts personal experience and threatens somecherished illusions. The easy way out is rejection, since one can always findimperfections and inadequacies in scientific knowledge.

CONTROL IS SELECTIVE ADAPTATION

An equally important reason for management's failure to make effective use ofcurrent social science knowledge has to do with a misconception concerning thenature of control in the field of human behavior. In engineering, controlconsists in adjustment to natural law. It does not mean making nature do ourbidding. We do not, for example, dig channels in the expectation that water willflow uphill; we do not use kerosene to put out a fire. In designing an internalcombustion engine we recognize and adjust to the fact that gases expand whenheated; we do not attempt to make them behave otherwise. With respect tophysical phenomena, control involves the selection of means which areappropriate to the nature of the phenomena with which we are concerned.

In the human field the situation is the same, but we often dig channels to makewater flow uphill. Many of our attempts to control behavior, far fromrepresenting selective adaptations, are in direct violation of human nature.They consist in trying to make people behave as we wish without concern fornatural law. Yet we can no more expect to achieve desired results throughinappropriate action in this field than in engineering.

Individual incentive plans provide a good example of an attempt to controlbehavior which fails to take sufficient account of "natural law"—in thiscase, human behavior in the industrial setting.

The practical logic of incentives is that people want money, and that they willwork harder to get more of it. In accord with this logic, we measure jobs,establish standards for "a fair day's work," and determine a scale of incentivepay which provides a bonus for productivity above the standard.

Incentive plans do not, however, take account of several other well-demonstratedcharacteristics of behavior in the organizational setting: (1) that most peoplealso want the approval of their fellow workers and that, if necessary, they willforgo increased pay to obtain this approval; (2) that no managerial assurancescan persuade workers that incentive rates will remain inviolate regardless ofhow much they produce; (3) that the ingenuity of the average worker issufficient to outwit any system of controls devised by management.

A "good" individual incentive plan may bring about a moderate increase inproductivity (perhaps 15 per cent), but it also may bring a considerable varietyof protective behaviors—deliberate restriction of output, hidden jigs andfixtures, hidden production, fudged records, grievances over rates andstandards, etc. In addition, it generally creates attitudes which are theopposite of those desired—antagonism toward those who administer the plan,cynicism with respect to management's integrity and fairness, indifference tothe importance of collaboration with other parts of the organization (except forcollusive efforts to defeat the incentive system).

All of these results are costly, and so are the managerial countermeasures whichmust be established to combat them (staff effort, elaborate control procedures,closer supervision, concessions with respect to rates, down-time provisions,setup arrangements, etc.). If the total costs of administering the incentiveprogram—both direct and indirect—were calculated, it would oftenturn out that they add up to more than the total gains from increasedproductivity. Certainly the typical incentive plan is of limited effectivenessas a method of control if the purpose is to motivate human beings to directtheir efforts toward organizational objectives.

Another fallacy is often revealed in managerial attempts to control humanbehavior. When we fail to achieve the results we desire, we tend to seek thecause everywhere but where it usually lies: in our choice of inappropriatemethods of control. The engineer does not blame water for flowing downhillrather than up, nor gases for expanding rather than contracting when heated.However, when people respond to managerial decisions in undesired ways, thenormal response is to blame them. It is their stupidity, or theircooperativeness, or their laziness which is seized on as the explanation of whathappened, not management's failure to select appropriate means for control.

The director of operations research in a large company is concerned becausefewer than half of the solutions to operating problems developed by his researchteam have been adopted by the line organization. He is currently trying topersuade higher management to issue orders to the line regarding theimplementation of certain of his findings. "If they can't recognize what's goodfor the organization, they will have to be told what to do," is his conclusion.Not only is his assumption of the line's stupidity incorrect, but so also is hisfurther assumption that commands from higher management will solve the problem.Yet, for him, the whole problem is "out there." It does not occur to him toquestion his own methods of control.

Effective prediction and control are as central to the task of management asthey are to the task of engineering or of medicine. If we would improve ourability to organize and direct human effort toward economic ends, we must notonly recognize that this is so, we must also recognize and correct some commonfallacies with respect to these matters.

Human behavior is predictable, but, as in physical science, accurate predictionhinges on the correctness of underlying theoretical assumptions. There is, infact, no prediction without theory; all managerial decisions and actions rest onassumptions about behavior. If we adopt the posture of the ostrich with respectto our assumptions under the mistaken idea that we are thus "being practical,"or that "management is an art," our progress with respect to the human side ofenterprise will indeed be slow. Only as we examine and test our theoreticalassumptions can we hope to make them more adequate, to remove inconsistencies,and thus to improve our ability to predict.

We can improve our ability to control only if we recognize that control consistsin selective adaptation to human nature rather than in attempting to make humannature conform to our wishes. If our attempts to control are unsuccessful, thecause generally lies in our choice of inappropriate means. We will be unlikelyto improve our managerial competence by blaming people for failing to behaveaccording to our predictions.

CONTROL AND PROFESSIONAL ETHICS

Discussions of the idea of controlling human behavior raise justifiableapprehensions about possible manipulation and exploitation. These concerns arenot new, but they will be intensified as the manager becomes more professionalin his use of social science knowledge to achieve the objectives of the economicenterprise. We must pause, therefore, to consider another characteristic of theprofessional: his conscious concern with ethical values.

Scientific knowledge is indifferent with respect to its uses. In this sense (andonly in this sense) science is independent of values. Scientific knowledge canbe used for good or evil purposes; it can be used to help mankind or to destroyhim, as we have seen so dramatically in recent times with respect to certainapplications of nuclear physics. It is obvious, therefore, that the moreprofessional the manager becomes in his use of scientific knowledge, the moreprofessional he must become in his sensitivity to ethical values. He must beconcerned both with broad social values and with those involved in his attemptsto control the members of his own organization.

Management's freedom to manage has been progressively curtailed in our societyduring the past century. Legislation with respect to child labor, the employmentof women, workmen's compensation, collective bargaining, and many other mattersreflects society's concern with the ethics of management. One approach to theseproblems is to see all restrictions on management as unreasonable and to fightblindly against them. This was fairly typical of industrial management ageneration or two ago. The other approach is to become more sensitive to humanvalues and to exert self-control through a positive, conscious, ethical code. Itis this latter approach which characterizes the concept of the "socialresponsibility" of management about which we hear so much today.