the legacy of max weber - mises.at · the case of weber, a good deal of his work has been left to...

THE LEGACY OF

Max Weber

THREE ESSAYS BY

L.M. LACHMANNProfessor of Economics and Economic History

in the University of the Witwatersrand, Johannesburg

The Glendessary Press — Berkeley

Copyright© 1971 byThe Glendessary Press2512 Grove, Berkeley, California 94704All rights reserved. No part of this bookmay be reproduced in any form, bymimeograph or any other means,without permission in writingfrom the publisher.

Printed in the United States of AmericaLibrary of Congress Catalog Card Number: 76-133386Published in Great Britain byHeinemann Educational Books Ltd.

Contents

Preface vii

Introduction i

The Method of Interpretation 17On Institutions 49On Political Institutions 92

Biographical Note 143

Bibliographical Note 145

Index 149

Preface

T H E R E ARE various ways in which, in any discipline,we might look upon its history of thought. Accordingto one view the contributions of past thought aregiven to us, like sculptures in a museum, as objects ofinterest and study but as something which hasassumed a permanent form. Such a view wouldperhaps be difficult to sustain in cases where certainideas of a thinker of the past are found to be in-consistent with others of his ideas, or where, as inthe case of Weber, a good deal of his work has beenleft to us in a fragmentary form. But in principle itis a possible and consistent view of the present-dayfunction of the history of thought. It means that thepresent tasks of a discipline are one thing, and itshistory of thought is another. This will be a viewcongenial to all those who like to know for certainwhat data they confront before they start their work.

It is possible, however, to take a different view ofthe present role of the history of thought. Accordingto this other view the ideas contributed in the pastdo not at present necessarily exist in their final form.The answers which have been given are all possibleanswers to recurring questions, but there may beother answers. The problems which thinkers of thepast have attempted to solve may have solutionsother than those they proposed, but which still eludeus. A thinker may have dealt with a complex of factsand problems some of which are linked by threadsof which he may have been unaware. From time to

vii

Vlll PREFACE

time therefore at least the major contributions mustbe subjected to re-examination.

It is obvious that all those who take this latter vieware compelled to adopt a critical attitude towardstheir intellectual forebears. However great theiradmiration and reverence for a thinker of the past,they must be ever ready to re-examine his startingpoint and his final point, his assumptions and hisconclusions, the structure of his models and thesignificance of the facts from which he has abstracted.

It is in this spirit that in the present book wereassess some of Weber's contributions which haveremained relatively neglected. In the first essay anattempt is made to replace the ideal type as thecentral concept of the theory of action. The secondessay is essentially an endeavour to make explicitthose problems of institutional structure which appearto be implicit in much of Weber's work. The subjectof the third essay calls perhaps for a special word ofexplanation.

The subject-matter of Weber's political writings,largely devoted to criticism of the social and politicalstructure of Hohenzollern Germany, is no longer ofinterest to us and, as such, does not belong to thehistory of thought. But the conceptual apparatus heused in these writings is a different matter. Weber,even when writing as a politician, could not helpthinking in sociological terms. We therefore havegood reason to be interested in the concepts he usedin his political writings, in particular where theybear some relation to a theory of institutional structure.

Weber's critique of Hohenzollern Germany wasmainly based on the postulate that parliamentarydemocracy is the political form required by modernindustrial society. The postulate evidently rests on a

PREFACE IX

notion of the necessarily unitary character of institu-tional structures.

On the other hand, Weber repeatedly stressed theneed for the daily compromises of political life andinsisted that all political arrangements ultimately reston such compromises. But in his political model thereappears to be no place for those more durable com-promises between powerful social groups which, beingmore than ephemeral, often shape the course ofevents for a long period, but do not require, andoften would not be compatible with, a unitary institu-tional structure. Weber, we might say, was not athorough pluralist. It seems to us, however, thatHohenzollern Germany did offer some instructiveexamples of precisely this type. If we are right, someof his political judgments of fifty years ago may beaccounted for by certain peculiar features of hisconceptual apparatus.

The history of thought is a critical enterprise.Every idea contributed in the past stands in need offrequent re-examination and reinterpretation. Themore important we think an idea, the more often weshall have to do that. This certainly was Max Weber'sown view. It is to be hoped that this book, which hasbeen written to honour his memory, will be regardedas no less a tribute to his name for being part of thiscritical enterprise.

I have to thank Mrs Heather Karolyi who com-piled the biographical and bibliographical notes andMrs Jacqueline Schoneberg who typed variousversions of the script.

L. M. LACHMANNParis, March/April

L.M.W. 1

Introduction

I N THE English-speaking world of the social sciencesMax Weber is today no longer a stranger. For overthirty years now, ever since 1937 when ProfessorParsons devoted a part of his Structure of Social Actionto an interpretation of his ideas,1 there has been asteady stream of books and articles, translations andcommentaries, all concerned with Weber's work. Thetime has thus arrived when students of Weber findthemselves in the happy position of being able toconsider 'the next step'. But what precisely is this tobe ? There can, of course, be no general answer. Thesituation is different in each social science. Everystudent of Weber will have his own views, notnecessarily shared by others, on those problems in thesocial science with which he happens to be mostfamiliar that are most likely to yield fruitful solutionswhen treated by Weberian methods. To carry forwardWeber's ideas in the circumstances of today is there-fore a task about the precise nature of which, as wellas the directions in which it is to be pursued, there isunlikely to be any general agreement. Nor is thereany need for it.

What we shall attempt to do in this introduction isto outline a programme for rendering certain ideasof Weber fruitful in fields either as yet untilled, or notas yet fertilized by such ideas. This programmenaturally reflects a certain view on the relative

1 Talcott Parsons, The Structure of Social Action (McGraw-Hill,1937), Part III, chs. XIII-XVIII.

2 INTRODUCTION

importance of Weber's different ideas, as well as onthe requirements of progress in different socialsciences, in the circumstances which each finds itselfin today. These are of course matters of value judge-ment on which there can be little useful discussion.All we can do is to present the reasons that haveprompted us to choose this particular programme asnecessary and feasible. This we shall do in thisIntroduction, and then attempt to carry out theprogramme in the three essays which follow.

Others would choose other programmes. Today weall know the general character of the legacy whichWeber has left us. More or less every item of theinventory has been carefully recorded. The questionbefore us is how it is to be utilized. It would not besurprising if views on investment opportunities wereto differ among heirs.

It is a curious fact that, while Weber's ideas havemade their impact on almost every social science, andmany branches of historical study have been in-fluenced by him, the one discipline he was mainlyconcerned with when, in the early years of thiscentury, he began to take an interest in questions ofmethodology, viz. the science of economics, todaybears almost no trace of his influence. Weber espousedthe method of interpretation (Verstehen) for the socialsciences. In economics today the prevailing style ofthought is a neoclassical formalism which is quiteuntouched by Weber's methodology and inclined totake it for granted that the methods of the naturalsciences are the only scientific methods known toman. We shall try to show why in our view this is afield in which the dissemination of Weberian ideaspromises to yield a rich harvest.

In the three essays which follow we propose to make

INTRODUCTION 3

use of the legacy of Max Weber in three directions:In the first place, it seems to us that a re-examina-

tion of Weber's starting-point is now called for. Inthe history of human thought it is a matter of commonobservation that a thinker's actual starting-point isoften arbitrarily chosen. Once a train of thought hasbeen completed we are often able to see, looking back,that a different starting-point would have served usbetter. In the case of Weber, however, we have toremember that our legacy really consists of a mass offragments. His main work, Wirtschaft und Gesellschaft,was not completed at the time of his death. Inattempting to assess his work as a whole we are thusdeprived of the benefit enjoyed by those who are ableto assess an author's starting-point from a firm pointof arrival chosen by the author himself.

Weber's starting-point, the fundamental concept ofhis methodology, is of course the Idealtypus. He choseit in 1904, at the very outset of his career as a studentof the methods of the social sciences and before heever came to think of himself as a sociologist, even asa future sociologist. As subsequent discussion hasshown, its meaning is by no means unambiguous. It isalso not a concept the scope and relevance of whichis in any way confined to the social world. Hence itsqualification for service as the fundamental concept ofsocial science remains in doubt. We shall have toconsider whether it might not be better to substitutefor it something more closely akin to human actionand society.

The second way in which we might make use of ourlegacy is by trying to bring some order into the massof fragments left to us. In particular, in fields whereWeber himself, partly for reasons of intellectualbackground and partly perhaps for lack of inclination,

4 INTRODUCTION

was reluctant to establish generalizations of a levelhigh enough to warrant their use in a general analy-tical scheme, some attempt in this direction seems tous to be called for. Thus we find in his work a numberof generalizations about institutions of various kinds,generalizations at different levels of abstraction, butno general theory of institutions. It seems to us thatthe time has come for his heirs to endeavour to fill thisgap. Accordingly, in our second essay, we shallpresent an outline of such a general theory of institu-tions which is, in keeping with the spirit of Weber,dynamic rather than static in that our emphasis is onprocesses of change rather than on the state of affairswithin a given social system. We shall of course haveto construct our analytical scheme largely on afoundation to be found in Weber's work.

The third direction in which we might seek toexploit our legacy concerns Weber's political thought.In the years of the First World War and its aftermathWeber took an active interest in German politics. It isfairly easy to distil from his political writings certainfundamental ideas, e.g. on modern parliamentarydemocracy and the political requisites of an industrialsociety. It is less easy, as we shall see, to reconcilesome of these ideas with certain generalizations aboutthe nature of the institutional order that emerge aselements of a general theory of institutions based onhis work. In the third essay we shall deal with MaxWeber's political thought. In so far as we thereexamine the links between his ideas on politicalinstitutions and other generalizations on institutions,the third essay is a continuation of the second. Inboth, our aim is to conduct tests of coherence betweenthe various parts of our legacy, to see which of thefragments will fit together.

INTRODUCTION 5

Two developments which have of late taken placein different fields of social thought prompt us to thinkthat the present time may offer a suitable juncture foran endeavour to exploit the legacy of Weber in thethree directions indicated. Each of these two develop-ments marks a kind of crisis, or at least a serioussetback in the progress of a social science. The firstconcerns the growing disenchantment of economistswith those 'growth models' which members of theirprofession have produced in some profusion duringthe past two decades. Economic growth is of course asubject still very much in fashion. But it is graduallycoming to be recognized that growth processes areprocesses of historical change, that they are promptedby many forces, not all of them economic, and that,whatever may be the best way of studying them, it isimpossible to reduce the rich variety of forces inoperation to one simple analytical model. In particular,the notion of equilibrium, which economists havelong been used to regard as the pivot of their analyticalapparatus, can find no application here.

One result of the discussion which has led to thissceptical conclusion is of particular interest to us. Ina world of uncertainty in which the outcome of everyhuman action can only be guessed with more or lessconfidence, the various economic agents (e.g. farmers,merchants and industrial entrepreneurs) will eachpursue plans prompted by certain expectations aboutfuture events. These expectations will diverge, henceso will the plans prompted by them. But if differentmen hold different expectations about a future event,at best only one can be right, and the others must bewrong. This means that some of the plans whichimplement their expectations must miscarry, and thatsome of the capital invested in accordance with these

D INTRODUCTION

plans will turn out to have been malinvested. Hence,there can be no such thing as 'equilibrium growth',which is of course incompatible with malinvestment.1

The recognition of these facts means discomfiture tothe neoclassical formalism which has dominatedeconomic thought in recent decades. Following theexample of the natural sciences, it has sought torepresent economic life as a system of interacting'forces', i.e. formal entities which can be regarded asvariables, the complex of relationships between whichcan be denoted as a system of simultaneous equations.There is, in fact there has to be, a niche in such aformal system for human dispositions, i.e. for humanpreferences. Everybody has a scale of preferences forgoods and services. But there is not, in fact therecannot be, any place whatevei for genuine humanaction as distinct from mere reaction to events.

For precisely the same reasons for which thisdevelopment has caused dismay to the formalists,students of Weber will welcome it. In what for manydecades looked an unpromising field, 'methodologicalindividualism' as espoused by Weber now appearsvindicated. If the course of economic processes can

1 A similar development has taken place of late in monetarytheory, a part of economics which often provides pointers tofuture developments in other parts of the discipline. 'Money isnot a mechanism, it is a human institution, one of the mostremarkable of human institutions. Even the simplest forms ofmoney, even metallic coinage, even the use of metals as moneythat preceded coinage, none can function without some minimumof trust . . . At the earlier and cruder stages, mechanical theories(such as the Quantity Theory) give a reasonably good approxi-mation to the working of money, but the subtlety of the monetaryfacts have [sic] gone on increasing, and theory has had a hardjob to keep up.' John Hicks, Critical Essays in Monetary Theory(Oxford University Press, 1967), p. 59.

INTRODUCTION 7

be shown to depend on certain qualities of individualplans, e.g. the degree of their divergence or con-vergence, then these plans and the meaning theplanners attach to them are things that matter, andmust be included in every attempted explanation ofsuch processes. There are thus good reasons, in thestudy of human action, to give careful attention to theplan which guides and directs action.

The second of the two developments mentioned isthe somewhat inconclusive outcome of the long dis-cussion on the merits of 'structural functionalism' insociology. On the one hand, it is clearly impossible toregard every society presented to us by history as a'social system' in any strict sense of this word. Thesuccesses of some planned actions of individuals willalways be found to be incompatible with one another.While, if by a social system we mean no more than acomplex of social relationships, the notion becomes sowide as to be useless.

Weber was certainly opposed to all those schools ofthought which were in his own day the predecessorsof today's functionalists. He took the view that allbiological and similar analogies in the social scienceshad at best to be regarded as interesting hypotheses,heuristic devices of stronger or weaker suggestivepower, which in each case required independentverification of their modus operandi. A praxeologicaltheory of society constructed on the firm basis ofpurpose and plan, such as emerges from Weber's work,is evidently not compatible with a functionalist view oflife in society.

On the other hand, forms of complementarity doexist in social life and are conspicuous in particular asregards institutions. One of Weber's main concernswas, after all, to unravel the intricate network of

8 INTRODUCTION

relationships between institutions of various types,religious, political, economic, legal, etc., in differentsocieties, even though his concern was mostly of acomparative historical nature. It may be rash tospeak of a social system in the sense of a complex ofsocial forces functioning indefinitely in such a manneras to maintain themselves permanently in operation,or even more modestly of an institutional order in thesense of complete complementarity of the institutionsof various types. But neither do the institutions ofsociety offer a picture of chaos. The pattern ofrelationships they exhibit is not completely un-intelligible, even though it may be intricate. It ispossible for our minds to discern some intelligibleorder among institutions even where there are manyrents in the pattern of the relationships existingbetween them.

The first task that emerges from the confusion andperplexity which have thus far accompanied the dis-cussion on structural functionalism is thus to constructa theory of institutions which does not take theircomplete complementarity for granted, but whichexamines carefully those features of the institutions ofthe real world that appear to display some degree ofcomplementarity. The second task will then consistin welding together the results obtained, with a viewto the possibility that enough elements of comple-mentarity may have been assembled to warrant ouruse of the notion of an institutional order, and to con-struct a theory on this basis.

Finally, and this will constitute a third task, we haveto see how such an order, if it can be shown to exist,stands up to processes of change; how it can combinecoherence with flexibility in a world of uncertainty.

It seems to us that here again, as in the case of the

INTRODUCTION 9

discomfiture of equilibrium economics mentionedabove, students of Weber hold the key which mayunlock the door that has barred progress. Institutionsare very important, but their modus operandi must beexamined from a firmly founded praxeological basisrather than taken for granted on the strength of somedubious biological analogy.

We have mentioned these two developments indifferent fields of social thought in order to show thatfrom the first we are able to gain a new starting-point,based on the method of interpretation, for a theory ofaction, namely the plan, while the second provides uswith an indication of the direction in which a generaltheory of institutions erected on Weberian foundationswill have to be sought.

The first essay, 'The Method of Interpretation', is anattempt to find a new starting-point for a theory ofaction inspired by the Weberian notion that actionderives its meaning from the mind of the actor. Wehave to start by dispelling a misunderstanding aboutthe intellectual ancestry of the method of Verstehen.Professor Parsons has linked it to the tendency ofidealistic philosophy to see in events observable intime and space concrete 'emanations' of permanentforces which are regarded as the 'real' agents of allchange.1 However strong such a connection may havebeen in the case of some German nineteenth-centurythinkers, we doubt whether it is useful to see Weberand his method of interpretation against this back-ground. Firstly, Weber was strongly opposed to allforms of 'emanationism' as methods of social science.Secondly, the method of interpretation (Verstehen) isone the origins of which have nothing whatever to do

1 Talcott Parsons, The Structure of Social Action (1937), Part III ,ch. XIII , 'The Idealistic Tradition'.

IO INTRODUCTION

with any philosophy. It is nothing less than thetraditional method of scholarship which scholars haveused throughout the ages whenever they were con-cerned with the interpretation of texts. Whenever oneis in doubt about the meaning of a passage one tries toestablish what the author {meant by it5, i.e. to whatideas he attempted to give expression when he wroteit. This, and not an axiom of the philosophy ofidealism, is the true origin of the method of inter-pretation. It is evidently possible to extend this classicalmethod of scholarship to human acts other thanwritings. This is what all historians, whether philo-sophically minded or not, have always done. It is this'positive5 method of the German Historical Schoolthat Weber took over and adapted to his purpose.

The question now arises whether it would be wiseto discard this method of cognition in favour of themethod of modern natural sciences. We learn fromSir Karl Popper that scientific explanation of eventsruns in terms of universal laws, and the impact of theforces described by the laws on an 'initial situation'A which is transformed by them into a subsequentsituation B. Scientific activity consists in the attemptsto formulate hypotheses of laws which are then to betested by making predictions about the transformationof initial situations.

Our choice of method must depend on whether thephenomena in which we are interested, those ofhuman conduct towards other human beings, lendthemselves readily to treatment in terms of thesecategories.

As regards universal laws, nobody doubts thathuman beings are subject to them. The question weface is not whether such laws exist, but whether thosewhich do ('All men must eat in order to live5) are of

INTRODUCTION II

much help in enabling us to understand how socialsituations change. As regards initial situations, ahuman situation can never be defined exclusively inobservable terms because all human action is alsoconcerned with an unknown and unknowable future.Two business-men, say, partners in a firm, may welllook at the same 'objective' situation with differenteyes, and may desire to take different actions. Humanaction cannot be regarded as mere reaction tostimulus. To understand it we have to understand whatimage of the future the actors are bearing in theirminds. Hence, any initial situation, however muchcare we may take to define it precisely and objectively,may suddenly, without any impact of external forcessuch as may operate under a universal law, turn intoanother different situation merely because the in-dividuals acting have 'changed their minds'.1 Theapplicability of the vaunted one-and-only method ofscience to human action must therefore remain indoubt. It seems wiser to leave other avenues open.

We mentioned above that the status of Weber'sideal type as the fundamental concept of a theory or

action which seeks to interpret meaningful action israther dubious. Discussion devoted to the task ofclarifying the ambiguity surrounding this term hasbeen rather inconclusive, except that everybodyagrees that the term is much too wide to be useful.The ideal type was really meant by Weber as ameasuring-rod, a device to bring order into a mass of

1 It does not help at all to speak here of 'dispositions' and toinclude them among the data of the initial situation. One of thetwo business-men may, of course, be an optimist and the othera pessimist. But they need not be. They may at first take identicalviews and subsequently change their minds, or at least one ofthem may do so.

12 INTRODUCTION

facts. As such it has nothing to do with the intelligibilityof human action.

We propose to replace it by the notion of plan. Allhuman action, if it is to be successful, requires a planto guide it. To understand an action means to under-stand the plan which is being carried out here andnow. A phenomenon of human action is an observableevent; so, in principle, is the making of plans. Hencethe application of the method of interpretation toaction does not mean, as some have thought, that wehave to regard action as the unfolding of a meaningthe source of which remains hidden from us: theexternal manifestation of an otherwise unknowableentity. Plans, strategic, economic or otherwise, areobservable events. At least wherever a number ofindividuals have to agree on a common plan, thereshould be no difficulty in establishing its character.

The correspondence between plan and action whichthus provides an almost 'natural5 conceptual basis forthe study of human action has no counterpart innature. A claim for the methodological autonomy ofthe social sciences, such as Weber endeavoured toestablish, is thus most strongly substantiated when webase it on the existence of this correspondence.

The frequent failure of plans we observe in realitysays nothing against this correspondence. It remainstrue that all action derives its meaning from the planwhich guides it. On the contrary, we should be unableeven to speak of success or failure of an action withoutthe notion of a plan in our minds.

Our second essay is concerned with institutions andtheir order. Its theme follows from that of the firstessay. Human action in society is interaction. Eachplan must take account of, among many other facts,favourable and unfavourable, the plans of other

INTRODUCTION 13

actors. But these cannot all be known to the planner.Institutions serve as orientation maps concerningfuture actions of the anonymous mass of other actors.They help the planner by making the social world alittle less uncertain than it would be otherwise. Butin reality institutions always change, some faster thanothers.

The second essay is thus a continuation of the first,and it also draws upon the legacy of Weber. In hiswork we find a number of generalizations aboutinstitutions at various levels of abstraction, but nogeneral analytical scheme. Various possible explana-tions of this fact are discussed. The outline of thescheme we present in this essay goes back to an earlierscheme by Menger which Weber, in our interpreta-tion, endorsed in his Sociology of Law. An institu-tional order must always exist, though in a muchlooser form than the exponents of the 'social systems'theories currently in fashion would be ready to grant.In our view the central problem of the institutionalorder hinges on the contrast between coherence andflexibility, between the necessarily durable nature ofthe institutional order as a whole and the requisiteflexibility of the individual institution. In other words,this central problem does not become apparent untilwe come to view the institutional order in the per-spective of time, which was also Weber's perspective.In order to elucidate various aspects of this problemwe introduce in this essay various distinctions, e.g.that between fundamental and secondary institutions. Wealso find that Menger's distinction between designedand undesigned institutions still has a deserved place insuch an analytical scheme, though often in history weobserve that institutions originally undesigned werelater on given a design by their being cast in a legal

14 INTRODUCTION

form. Our main conclusion is that it is impossible forall institutions to change at the same rate, and thatthe relative immutability of some institutions is alwaysa necessary prerequisite for the relative flexibility ofthe rest.

Our third essay deals with the role of politicalinstitutions within the institutional order. Its themeis a continuation of one of the themes explored in thesecond essay, in that we make a closer study of certainfundamental institutions. We again draw uponWeber's legacy by relating some of our conclusions tohis political thought. Weber was a vigorous politicalthinker who took a close interest in the politics ofWilhelminian Germany and its immediate aftermath.As was mentioned above, while it is fairly easy todistil from his political writings certain fundamentalideas, empirical generalizations about the politicalinstitutions of modern industrial society, it is notalways obvious how well these would fit into a generalanalytical scheme, a theory of institutions based onhis work. In fact, it is sometimes possible to criticizethe former in terms of the latter.

Much of what Weber said in his political writingsnow belongs to past history. But he saw very clearlyat least one problem of modern democracy which hasby no means lost its significance: how to co-ordinategroup interests with the wider interests of the nationas a whole. As we shall see in this essay, Weber'sanswer was that the democratic process, the struggleof parties for political power, will give rise to leaderswho as party leaders learn how to co-ordinate groupinterests and as national leaders bring about co-ordination on a wider scale. It is a conception whichdoes not lack subtlety in that one kind of pluralism,that of interest groups, is here offset by another kind

INTRODUCTION 15

of pluralism, that of political parties. In fact a partyis here regarded as an intermediate co-ordinatingmechanism for group interests. The national leader,product of the democratic process which is a con-tinuous struggle for power, reigns supreme as theco-ordinator-in-chief.

His discussion of these matters raises three questionswhich we regard as central to the legacy of Weber as apolitical sociologist.

In the first place, what happens if no leader of therequisite qualities and status emerges from thedemocratic process, but only men of mediocre talentsand without the gift to subjugate group interests tothe wider aims they pursue? Or perhaps even menwho simply do not pursue such aims? In the light ofmuch that has happened in the fifty years since Weberdied such a question cannot be ignored.

Secondly, may not the struggle for power amongpolitical leaders, even irrespective of their personalqualities, gradually lead to an erosion of those funda-mental institutions which circumscribe and limit theexercise of political power? Will not the leaders inthe course of the political struggle have to makepromises to the electorate which cannot be redeemedwithout whittling away some of the very institutionson which the democratic process rests? This is aquestion to which Weber did not address himself. Inthe Germany of his last years such institutions wereonly just coming into existence. Actually he helped tocreate them. But it is a question which arises at oncefrom the scheme of institutional change we presentin our second essay and which owes much to thelegacy he left us.

We may raise a final question which flows from acombination of the first two. Every political system,

l6 INTRODUCTION

whether democratic or not, ultimately rests for itsstability on the broad consensus of certain majorsocial groups in the society which supports it. It maywell be that for the protection of our fundamentalinstitutions against such erosion as we referred to inour second question we have to look to the actual, butof course unwritten, terms of the compromise ofmajor social forces on which our political order rests.If so, a good deal will depend, for the stability of thepolitical structure, on whether such a compromise isfreely accepted by the parties to the bargain as anintegral part of their way of life from now onwards,or regarded by them as a merely temporary arrange-ment which leaves the future in a dark void. We shallexemplify this distinction, which we regard as vital,with respect to Hohenzollern Germany and theWeimar Republic. It is not a distinction we shouldexpect to find in Weber's work. In his politicalthought struggle always occupies a more prominentplace than compromise, which to him in any case isalways temporary. We hope to show, however, thatit is a distinction not unconnected with a sub-jectivist interpretation of political form, and onewhich appears to flow naturally from a view ofinstitutions which, in the spirit of Weber, looks atthem as instruments of intelligible human action.

The Method of Interpretation

I T HAS often been said that Max Weber, in pro-pounding the method of Verstehen as the method ofstudy appropriate to human action, was essentiallydefending the heritage of German idealism against theonslaught of positivism.1 This view, though there issome truth in it, fails to do justice to the complexnature of the situation in which Weber found himselfno less than to the subtlety of his mind. It is quite truethat in some important respects he remained very muchthe heir of the German Historical School all his life.But, though a disciple, he was by no means an un-critical admirer of this school of thought, and did nothesitate to criticize its protagonists, as well as some ofits major articles of faith, where he thought itnecessary. Moreover, the beginnings of his interest inquestions of methodology fall into a period when,after a long illness, he came to feel the need to re-examine his own position as well as that of the schoolin which he had grown up.

But, whatever Weber's own attitude towards it, themethod of Verstehen, the interpretation of humanutterance in order to make it intelligible, is mucholder than German idealism and the HistoricalSchool which, partly, sprang from it. We might evenperhaps say that it is the 'natural5 method of renderingan intelligible account of the manifestations of the

1 See, for example, Talcott Parsons, The Structure of SocialAction (McGraw-Hill, 1937), ch. XIII, 'The Idealistic Tradi-tion'.

17

l8 THE METHOD OF INTERPRETATION

human mind. It is nothing less than the traditionalmethod of classical scholarship.

Whenever we wish to 'understand' a text, be it of areligious, literary, legal or other nature, we have toemploy a number of procedures all of which aim atour greatest possible certainty as to what the author'wanted to say' in the first place. Where the textstudied contains a generalization, e.g. a religious pre-cept or a legal norm, we also have to decide to whatkind of concrete situation our text may apply. Textualinterpretation is therefore the prototype of Verstehen.Until the rise of modern natural science this was thecommonly accepted method of all scholars, whetherthey studied the Bible or the Corpus Juris Civilis, readHomer or translated Avicenna or Averroes from theArabic. It will be readily appreciated how little allthis has to do with 'intuition'. The procedure is arational procedure of discursive study.

In interpreting a text, what essentially we are tryingto do is to identify a 'meaning', an idea, to which thetext in question is designed to give expression. Inother words, interpretation is a method of comparativestudy by means of which we are attempting to establisha relation between an observable event (a readabletext) and an idea which existed in a human mindprior to the writing of the text, and to which thelatter is designed to lend expression. The object of ourstudy is therefore to establish a degree of correspon-dence between a phenomenon and an idea.

The method of study to be employed for such apurpose must largely take the form of coherencetests. In interpreting a text of uncertain meaning wehave to ask for each possible meaning, whether andhow far it would be consistent with what the sameauthor says in other passages of the same work, or in

20 THE METHOD OF INTERPRETATION

reason. He must ask how far the variety of purposespursued by the individual whose action he studies (asby any other individual) 'fitted together'. He has toascertain 'The Plan5, the coherent design behind theobservable action in which the various purposes aswell as the means employed are bound together. Hethus has to conduct coherence tests on two levels.In each case he has to ascertain:

(1) whether the purposes he ascribes to the individualacting are in fact consistent with one another andfit into the framework of a general plan, theexecution of which would account for the knownfacts;

(2) whether the design and execution of such a planare in fact consistent with whatever else is knownabout the intentions, circumstances, etc. of theindividual whose action is the subject under study.

Once we have realized that the historical method isreally nothing more or less than the classical methodof interpretation applied to overt action instead of totexts, a method aiming at identifying a human design,a 'meaning5 behind observable events, we shall haveno difficulty in accepting that it can be just as wellapplied to human interaction as to individual actors.From this point of view all history is interaction,which has to be interpreted in terms of the rival plansof various actors. All historiography has in fact pro-ceeded in this manner.

The question we now have to face is whether groupaction lends itself to treatment in the same terms. Is itamenable to study in terms of purpose and plan inthe same way as individual action? Or does ourmethod of interpretation here encounter an insur-mountable obstacle ?

THE METHOD OF INTERPRETATION 21

In answering this question we have to distinguishbetween organized and unorganized group action.Where a group is organized in such a way that thetask of acting for it, of planning action and carryingout such plans, is entrusted to certain individualsdesignated for such purposes, there is of course noproblem for us. Everything said so far about historicalexplanation of individual action here applies to theaction of these officials. This is, in fact, how mostpolitical and diplomatic history has been written.Historians have regarded it as their main task toexplain the action of political leaders in terms of their'policies', i.e. in terms of consistent plans. Even whenhistorians speak of 'the foreign policy' of a countryover a period exceeding any individual's span of life,what is meant is of course the continuous execution ofa coherent plan over a long period.

Economic and social history can hardly be writtenin these terms except perhaps in such cases where, forinstance, the growth of a business enterprise isdescribed. But our method of interpretation need notfail us even in this field. The task of historical explana-tion consists here in accounting for a recurrent patternof action, and such an account, if it is to be anintelligible account, again requires interpretation interms of the typical elements of plans to be found in'anonymous mass action'. The only difference con-sists in the fact that here the plan elements whichinterest us are not the millions of individual purposespursued, but the common elements of norms, institu-tions, and of the general environment in which allthese plans have to be carried out. It is these commonelements, which millions of plans have to contain andto which all individual action in a given society hasto be oriented, which it is here the task of the historian


to explain. His task, no longer primarily concernedwith purposes, which are here taken for grantedrather than ignored, is still the interpretation of actionin terms of plans and their elements. The method ofinterpretation that aims at linking observable eventsto types of design existing in the minds of peopleacting still holds good, and should enable us topresent an intelligible account of what is happening.

On the other hand, we have to admit that in caseswhere we have to deal with group action which iscompletely unorganized, neither organized and di-rected by leaders nor oriented to common norms,rules or institutions, our method will fail. A com-pletely spontaneous riot (if there is such a thing) isbest treated as a 'natural event5. We should be unableto give an intelligible account of it; there can beno question here of coherence of plans.

Finally we come to the question whether the methodof interpretation may be employed beyond the bordersof history, namely in the analytical social sciences.

The answer to this question is in the affirmative.There seems to be no reason why a method which isuseful in the explanation of individual action shouldbe less so in the explanation of classes of such actions.The case is exactly parallel to the one discussed above,when we were concerned with economic and socialhistory. It is true that in explaining recurrent patternsof action, the essential subject-matter of all socialsciences, we cannot provide such explanation in termsof purposes, as elements of plans, because the purposespursued by millions of people are of course numberedin millions. But often we are none the less able toprovide explanations in terms of the elements commonto all these plans, such as norms, institutions, andsometimes institutionalized behaviour, the maximiza-


tion of profits, or the avoidance of the risk of insolvency.As long as we are able to account for the recurrence ofpatterns of action in terms of such elements of plans,we are successfully employing the classical method ofinterpretation. We are still explaining subsequentevents in terms of ideas. Moreover, the line thatdivides concrete historical phenomenona from per-manent social structures is notoriously thin. To whichof these two classes should we assign, for instance,the medieval town economy, the political system ofthe Republic of Venice, or the public finances ofFrederician Prussia? The plain fact is that everyrecurrent pattern of events, anything we should feelat all entitled to call a 'structure', requires explanationin terms of permanent forces as well as in terms ofconcrete historical circumstances. Interpretation isneeded in the former as well as the latter type ofexplanation.

Weber found merit in this method, which theprocedure adopted by the natural sciences lacks. Hefelt that in the study of human action we should notdispense with a method which enables us to ascertain'the meaning5 of action, individual or collective, whilethe natural sciences are, in any case, unable to domore than bring a large number of observablephenomena within the bounds of an analyticalscheme. Natural phenomena can have no 'meaning'.In espousing the method of Verstehen Weber went farbeyond defending the heritage of the GermanHistorical School.

We must now give a brief account of the particularsituation in which Weber found himself compelled toconfront this crucial issue in the methodology of theSocial Sciences.

When in 1902 he gradually recovered from hisL.M.W.—2


illness, the Methodenstreit, that is to say the controversyon the merits of abstract analytical schemes for thestudy of social, and in particular economic, eventswhich had divided Austrian and German economistsinto two hostile camps, had lasted twenty years, andsigns of weariness were becoming apparent on bothsides. This controversy started in 1883 when CarlMenger, Professor of Economics in the University ofVienna, published a book on the methods of the socialsciences in which he defended classical economictheory and criticized the Historical School, thendominant in Germany.1 Schmoller, the head of thatschool, wrote an acrimonious review of the book towhich Menger replied the following year, 1884, witha tract on 'The Errors of the Historical School' which,as Schumpeter has said, 'fairly steamed with wrath'and contained a number of personal attacks onSchmoller. After that friendly relations betweeneconomists of the two Empires became rather strained.

It is noteworthy that in this controversy Mengerand the Austrians were throughout on the defensive.They did not deny the justification for historicalstudies in the economic and social field, but strove touphold the right of, indeed the need for, abstractanalysis of economic phenomena.

Weber, who at the start of his career had been adisciple of Schmoller and whose first economicstudies had been of a historical nature, was eager tobring the controversy to an end. In the years of hisillness he had moved far away from his early moorings.

1 Carl Menger, Untersuchungen iiber die Methode der Sozial-wissenschqften und der Politischen Oekonomie insbesondere (Leipzig1883), translated as Problems of Economics and Sociology, edited withan introduction by Louis Schneider (University of Illinois Press1963)-


He was quite willing to agree with the Austrians thatall historical explanation requires causal schemeswhich are of a general nature. He recognized thatthere is a need for economic theory. But there werecertain aspects of Menger's methodology which hewas unable to accept.

Menger regarded it as the main task of all thesciences to find and formulate 'exact laws'. But henever stressed the distinction between empirical re-gularity and logical necessity, between what Leibnizcalled verites de fait and verites de raison. He seems tohave regarded the £law' of diminishing marginalutility as an empirical law of nature based on psycho-logical 'drives'.

Weber denied Menger's contention that the 'laws'governing economic conduct (among which wasMenger's own creation, the law of marginal utility)are 'exact laws' in the same sense as those found innature. He regarded this as a 'naturalistic fallacy'. Heinsisted that the observable uniformity of humanconduct in economics, profit maximization in businessetc. is essentially of a 'pragmatic' nature and hasnothing whatever to do with 'psychology' of any kind.Once a man decided to conduct his business with theaim to maximize his profits, certain necessary con-sequences followed, but such necessity was of a strictlyconditional nature, and its source was 'pragmatic' inthe sense that it lay in the 'logic of the situation' thebusiness-man confronted.1 In fact, abstract economictheory consisted essentially of rational schemes inwhich the conditions of successful action were definedin such a way as to require certain kinds of action.This is something very different from the way in which

1 Gesammelte Aufsdtze zur Wissenschaftslehre (Second Edition,> P- 396.


natural events are 'determined5 by their causes. Thenaturalistic fallacy consists in confusing the two.

Weber remained very much the heir of the HistoricalSchool also in other respects. Adherents of thisschool had always objected to what they regarded asthe 'artificial separation3 of economic from othersocial activity, of which they held the classicaleconomists guilty. Weber saw no reason why theabstract schemes, the need for which he acknowledged,should have to be confined to schemes of rationaleconomic conduct. To this end he devised the famousnotion of the Idealtypus, which has given rise to somuch criticism.

We shall choose a different starting-point. But wefirst have to explain why the 'Ideal Type5 does notappear to us to offer an ideal starting-ground for ajourney into the theory of social action.1

For Weber the ideal type was the chief instrumentof causal analysis in society, the fundamental conceptof all social sciences. Like the concepts of allgeneralizing sciences, it is obtained by a process ofabstraction and must therefore be relatively empty ofcontent when compared to reality. But in our caseabstraction is not completely arbitrary. What we getin exchange for relative emptiness of content is a'higher degree of unambiguity5 (gesteigerte Eindeutigkeit)of our concepts. This enables us to go beyond therational schemes of classical economics and to graspthe meaning of irrational action, for instance that ofmystics, or of the action of a crowd in a state of massemotion.

The ideal type is essentially a measuring rod. When1 Weber discusses his notion at length in The Methodology of the

Social Sciences, translated by Edward A. Shils and Henry A.Finch (Free Press of Glencoe, New York, 1949), pp. 89-111.


we use an ideal type we stand at a distance fromreality, but for precisely this reason are able to gainknowledge of it: 'By indicating the magnitude ofapproximation of an historical phenomenon to one orseveral of our concepts we can order these pheno-mena.' In other words, the ideal type serves thepurpose of ordering concrete phenomena in terms oftheir distance from it. It is readily seen how differentis Weber's ideal type from the model, a methodo-logical device currently in fashion in many of thesocial sciences. Both are 'mental constructs', both aregained by abstraction. But the virtue of models restsin their being 'testable'. They must serve the purposeof predicting concrete events. In choosing betweendifferent models we must choose the one whichenables us to make predictions which come nearest toevents actually observed. With ideal types this is notso. With models, however remote from reality theirelements may be, any distance between the pre-dictions derivable from them and reality is a seriousdefect. With our ideal types, by contrast, suchdistance is a positive virtue since it offers us 'space' inwhich to display the ordering of our observed events.

As already mentioned, what Weber wanted as chiefconceptual tool of the social sciences was a conceptsufficiently wide to comprehend both rational schemesand all kinds of historical generalizations. How widein fact was the scheme he envisaged is seen from the'sample card' of ideal types he presented when hefirst set forth the notion in his 'Essay on Methodology'in 1904.

Class or generic concepts (Gattungsbegrife)—ideal types—ideal—typical generic concepts—ideas in the sense ofthought—patterns which actually exist in the minds ofhuman beings—ideal types of such ideas—ideals which


govern human beings—ideal types of such ideals—idealswith which the historian approaches historical facts—theoretical constructs using empirical data illustratively—historical investigations which utilize theoretical conceptsas ideal limiting cases.1

The Ideal Type has become the centre of a fiercecontroversy, which it is neither possible nor necessaryto concern ourselves with here. To deal with it at alladequately would by itself require a long essay, andit is doubtful whether the result would justify theeffort. It is noteworthy, however, that in this contro-versy some of Weber's most ardent admirers turnedinto his most severe critics. For example, Sombart,his comrade-in-arms in the early days of the Werturteildiscussion, severely criticized Weber2 for having con-fused the true ideal type, which does not permit ofmore than a modicum of abstraction, with the purelyformal generic concepts of all sciences. All the critics,however different their points of view, agreed thatWeber's concept was much too wide to be useful,while each wanted to see it narrowed down in onedirection or another. On the scope and direction ofthis narrowing-down process, and in particular on thequestion what segment of reality it should cover, noagreement was reached. The conclusion emergedpretty clearly in the end, however, that differentconceptual tools are required to deal with rationalschemes of action on the one hand, and historicalgeneralizations on the other.3

1 ibid., p. 103.2 Werner Sombart Die drei Nationalukonomien (Munich 1930),

pp. 245-6.8 For an interesting survey of the controversy see Fritz Machlup

'Idealtypus, Wirklichkeit und Konstruktion', in Ordo, vol. XII(1960-61). See also Walter Eucken, The Foundations of Economics

» P- 348, n. 66.

THE METHOD OF INTERPRETATION 2O,

But whatever view we may adopt on these contro-versies, there is one (to us overwhelming) reason whywe are unable to accept the ideal type as our funda-mental concept. This reason lies in the simple factthat Weber's ideal type lacks any specific reference tohuman action and seems to be as readily applicableto the animal kingdom or the plant world as to thehuman sphere.1 It seems better to start our journey onmore promising ground and adopt as our fundamentalconcept a notion germane to human action, a notion,that is, in which the meaning of action is pre-conceived even before the very moment at which thecourse of action begins to unfold.

We propose therefore to proceed by an altogetherdifferent path. We shall start from a notion at oncesimpler and more comprehensive than the contro-versial Idealtypus and its many different variants. Weshall attempt to show that this notion constitutes thenatural centre of the method of interpretation andthat most of the other concepts we need in order togive an account of human action and its results can bederived from it. This notion is the plan. Phenomena ofhuman action are without doubt observable events,and may be treated as such. We may, precisely as inthe study of nature, put forward various hypothesesabout the way in which these events are related toeach other, and then proceed to judge their rivalmerits by the criterion of falsifiability of prediction.Nothing said in what follows must be taken to implydenial of this possibility. But in the field of culture itis not at all a rewarding one.

1 In 1913 Weber admitted as much. 'Logically it makes nodifference whether an ideal type is formed from meaningful andintelligible or from specifically meaningless relationships.'Gesammette Aufsdtze zur Wissenschqftslehre (our translation), p. 438.


One trait distinguishes all cultural phenomena fromnatural ones. When men act they carry in their mindsan image of what they want to achieve. All humanaction can be regarded as the carrying out of projectsthat are designed to give effect to imagined ends. Butevery man pursues a multiplicity of ends, the achieve-ment of at least some of which precludes that of otherends. Moreover, the scarcity of the means at thedisposal of each actor imposes further restraints uponhis choice. In other words, men have to choose thepurposes they wish to achieve, and they have to makesuch choice within the constraints of a given 'situation'.To act at all, men have to make plans, comprehensivesurveys of the means at their disposal and the ways inwhich they might be used, and let their actions beguided by them.

Nature offers no parallel to this possibility. Naturalphenomena exist in time and space only, and ob-servability is the only criterion of their existence. Thefact, on the other hand, that human action exists inthe form of plans, i.e. mental design, before it iscarried out in time and space, permits us to studythe relationships between human action and theplans which guide it. The method of interpretation inthe social sciences ultimately rests on the possibilityof, and the need for, such comparative study. In thissense, then, we may say that we are able to give an'intelligible account' of human action by revealingthe plans which guide it, a task beyond the grasp ofthe natural sciences. The mere fact that this possibilityexists is the foundation of the method of interpretationand thus offers a vindication of the plea for themethodological autonomy of the social sciences.1

1 Cf. Alfred Schutz, 'Choosing among projects of action',Collected Papers, vol. I (The Hague, 1962), pp. 67-96.


This method of comparative study, as we saw above,is applicable in the historical as well as the theoretical,field. Historical explanation was seen to be theexplanation of human action, individual or group, interms of plans. As a matter of fact, those parts ofhistory most closely concerned with res gestae, diplo-matic, military, and naval history, have never beenwritten otherwise. It is perhaps only a little lessobvious that at the other end of the scale, in thehistory of thought and ideas (Geistesgeschichte), thechief matters to be unravelled are the changes in shapeand design which successive thinkers gave to thevarying content of similar ideas.

In social theory our main task is to explainobservable social phenomena by reducing them to theindividual plans (their elements, their shape anddesign) that typically give rise to them. This is whatWeber meant by the explanation of action 'in termsof the meaning attached to it by the actor'. It is notto be denied that social phenomena may be studiedby other methods, for instance by correlating seriesof events observed in time. We shall try to show whysuch efforts are unlikely to be rewarding.

It may be held against us that few men achieve theiraims, or even a gratifying part of them, and that inreality no action ever goes entirely according to plan.Our provisional answer to this charge has to be that itmerely serves to emphasize the complexity of therelationships between plan and action, and that the factthat no course of action is ever a full replica of theplan which guides it does not enable us to dispensewith the plan as our prime tool. In fact we mightourselves take the charge a step farther and say thatthe unintended consequences of action are probablymore important than those intended, and that they

L.M.W.—2*


indeed constitute the most interesting problem of theanalytical social sciences. All this merely goes to showwhat an interesting field the comparative study ofplan and action is, and how wide a range of impor-tant problems presents itself to us once we haveentered it.

There can be no doubt that the whole problem ofthe need for flexibility of plans, the expectations ofthe future and interpretations of past experienceembodied in a plan, and the question of how successor failure of one plan will affect the drawing up ofanother, will have to be discussed in detail. Beforedoing this, however, we have to justify our proposalto use the concept of 'plan' as the fundamentalconcept of the method of interpretation, by attemptingto show that in this way Weber's chief aim can beachieved. We shall also have to show why the methodscommonly employed in the natural sciences will nothelp us in our endeavour.

In substantiating our claim that by making theconcept of 'plan' the cornerstone of our analysis ofhuman action we are drawing legitimate usufructfrom Weber's legacy, we have to show of course thatthe main elements of our approach are already to befound in his work. In what has to be regarded as hiscentral thesis on the methodology of cultural studiesand what distinguishes them from the naturalsciences, in the middle of his first methodologicalessay, we find three passages which, when takentogether, seem to us to bear out such an interpretation.He first points out that causal explanation is just asnecessary in culture as in nature.1 But in the formercase 'its specific significance rests only in that we areable, and want, not merely to state but to understand

1 Gesammelte Aufsdtze zur Wissenschqftslehre, p. 182.


human action5 (his italics)1. The possibility of suchunderstanding is warranted by the purposive characterof human action. But 'purpose5, he says, 'is for us animagined end which becomes the cause of an action;we take account of it in the same way as we have totake account of any other cause which does, or may,contribute to a significant effect5.2

We hold these passages to mean that the dis-tinguishing characteristic of the causal explanation ofhuman action lies in the fact that the 'effect5 of actionin its imagined form, i.e. as 'purpose5, precedes theactual course of action, and thus has to be regardedas a cause. In human action, as in nature, causeprecedes effect. But while the effect of human actionhas many causes other than human purpose, and whileall these must of course be given a place in an in-telligible account of action, purpose as a cause ofaction must not be neglected. The essence of thematter is that the end sought, in its mental form,must precede the end achieved as an observable event.

It is readily seen (with the benefit of hindsight) thatthis conception of the nature of casual explanation ofhuman action in terms of purpose would have pro-vided a firmer and more convenient starting point forthe methodology of the social sciences than thecontroversial notion of the Ideal Type. It is alsoeasy to see how it is naturally linked to our conceptof Plan. In fact, 'plan5 is but a generalization ofpurpose. In reality actors, individuals as well asgroups, pursue many purposes simultaneously andhave to establish an order of priority among them.Moreover, the manifold constraints imposed upon thepursuit of our ends by the scarcity of means as well asby the ubiquitous presence of obstacles, actual or

1 ibid., p. 183 (our translation). 2 ibid.


potential (negative means), compels all of us to bringall our means and ends within the framework of acomprehensive computation before we set out on ourcourse of action. In this mental framework theactions of other actors, as we shall learn, play a mostimportant part. They may be either means or obstaclesto the pursuit of our own ends. But in any case weare unable to judge whether a particular purposecan be pursued with the means at our disposal, oreven whether it is worth pursuing at all, untilwe have established a comprehensive framework, aplan.

In this way we hope to have established our claimthat in grouping the phenomena of action around thecentral concept of plan we are making legitimate useof the legacy of Weber.

Our next task consists in having to show why the'scientific method' of the natural sciences can dolittle to help us in our endeavour. Since Weber wrote,natural science has had many triumphs. Almostinevitably, as one of the consequences of thesetriumphs, there have been attempts in the socialsciences to borrow these methods the application ofwhich, in whatever new fields, appeared to vouchsafesuccess. As a result we see the rise of new socialsciences whose methods are borrowed from theirsisters in the field of nature, such as econometrics andsociometry.

On the other hand, a good deal has also beenlearnt, in the last sixty years or so, on the logic andmethodology, and this means on the limitations, ofscience. When Weber wrote, in the early years of thecentury, Mach and Poincare dominated the field withtheir ideas. It is today possible to look at someaspects of these problems from a point of view which


was beyond Weber's reach, and to vindicate theautonomy of the social sciences by the use of weaponswhich the successors of Mach and Poincare have madeavailable to us.

Natural science aims at establishing universal lawswhich will 'explain', i.e. predict, a maximum numberof observable events. On the question whether any'explanation' beyond successful prediction of observ-able events is at all called for, opinions differ. The'instrumentalists' deny it. But even those who do notshare this view would, in general, be satisfied if theobserved event, which must be a change of anidentifiable object in time and space, can be shown tobe 'determined' within a closed 'system', defined interms of a universal law and the 'initial situation', i.e.the observable events obtaining at a particular pointof time and space.

About the instrumentalist view of scientific method1

we have nothing to add to what Weber said. No'explanation' which has nothing to offer beyondsuccessful prediction of observable events can satisfythe student of action who wishes to understand it. Thenon-instrumentalist view on the other hand we may(provisionally) accept, precisely because it enables usto specify with some degree of precision why we areunable to make use of it for our purposes.

The sequence of events which is to confirm orrefute a hypothesis can be regarded as determinateonly if the 'initial situation', in which the train ofevents is set in motion, is known and can be describedin detail. The scientist who proposes an experiment totest his hypothesis must pay close attention to speci-

1 For a succinct account of this view as applied to a socialscience see Milton Friedman, Essays in Positive Economics (Chicago,

> P a r t l>


fying the conditions in which the experiment is totake place. But in the case of human action, even werewe to grant the existence of 'universal laws', it isimpossible to specify such an initial situation for thesimple reason that it is impossible to specify knowledge.Evidently the knowledge of the actor is an importantelement of his action. Were we to test hypothesesconcerning action, the canon of scientific methodwould require us to describe in detail all the know-ledge possessed by the actors—an evident impossibility.We see thus that while 'description of the initialsituation5 is a fairly innocuous requirement in nature,where all we have to do is enumerate objects in timeand space, for human action this requirement cannotbe met because we should have to include somethingunspecifiable—knowledge! A human situation withoutspecific knowledge makes no sense. It follows that the'scientific method5 of the natural sciences will be oflittle use to the student of action because he is unableto use the testing procedure this method prescribes.It is impossible to account for human action withouttaking account of the state of knowledge of theindividuals acting. Any such endeavour would beopen to all the familiar arguments against behaviourism.Human action is not 'determinate5 in any sense akinto the one in which natural science has to strive forthe 'determinacy5 of the events it studies. A mechanis-tic interpretation of action, couched, say, in terms of'response to stimulus5, would have to explain awaysuch simple facts as that different men in identicalsituations may act differently because of their differentexpectations of the future. We all know that men whoshare a common experience may give it widelydiffering interpretations. The human mind, a re-ceptacle of the past as well as a screen on which our


imagination projects images of the future, defies allthose generalizations on which the methodologicalcanon of the natural sciences must rest.

In these circumstances, then, it is clear that thecultural studies, concerned as they are with humanaction, require a different method of approach totheir objects. We shall now try to give a brief outlineof the scope and nature of this method which in whatfollows we shall call the praxeological method. Humanaction is not determinate, but neither is it arbitrary.It is bounded, firstly, by the scarcity of means at thedisposal of actors. This circumstance imposes aconstraint on the freedom of action. It is bounded,secondly, by the circumstance that, while men arefree to choose ends to pursue, once they have madetheir choice they must adhere to it if consistentaction with a chance of success is to be possible at all.In other words, human action is free within an areabounded by constraints. Obstacles of various kindsfurther limit the area of freedom.

The praxeological method has to take these cir-cumstances into account. Causal explanation in thefield of action cannot hope to attain determinateness,but this does not mean that we must give up all hopeof explanation. What we may hope to accomplishhere is to be able to show to what ends, means, andobstacles human action is oriented. Orientation thusemerges as a concept as fundamental to praxeologicalstudy as determinatenes; is to natural science. As thelatter requires a 'closed' analytical system, consistingof functions like independent and dependent variablesas well as constants, to warrant the determinatecharacter of its results, so praxeology requires a moreflexible form of thought, an 'open5 analytical frame-work which will nevertheless permit us to ascertain


the boundaries of action. Orientation is the pivotalconcept within this framework.

In praxeological theory we are concerned with thetypical points of orientation of typical courses ofaction. In concrete historical study we endeavour toascertain the actual ends, means, and obstacles towhich a concrete course of action of an individualor a group was oriented. Orientation was of coursealso the fundamental concept of Weber's theory ofaction. Travelling by a somewhat different route, wehave reached the same conclusion as he did. It cannow also be seen, moreover, that orientation entailsplan. A plan has to contain a comprehensive accountof ends, means, and obstacles to which a course ofaction is oriented. It provides the systematic frame-work of all points of orientation relevant to a givencourse of action. But of course this 'orientation map5

is only one necessary ingredient of a plan. Beyond thisthe plan must inter alia contain directives for actionin space and time.

Something has now to be said about the elementof knowledge in the making of plans. In makingplans, tying means to ends and prescribing action intime and space, men evidently bring their existingfund of knowledge to bear upon a present situation.But how precisely this is done is difficult to describewith any precision. Evidently only part of a man'stotal knowledge will be relevant to a given plan.Which part? This will depend on his subjectiveinterpretation of the past and his equally subjectiveexpectation of the future. In other words, each plancontains subjective elements of more than one kind.Not merely do the purposes sought in it reflect thesubjective choice of ends, but what purposes areregarded as attainable in a given situation depend on


subjective expectations of an uncertain future as wellas on subjective judgement of the relevance of pastexperience, subjectively interpreted, to this future.

The several layers of subjectivity just encounteredin our attempt to lay bare the element of knowledgein plan-making not merely defy any behaviouristic ormechanistic interpretation of human action. Theyprovide not merely a reason, as we already saw earlieron, why an 'initial situation5 as required by thenaturalistic methodology cannot even be defined inhuman action. They also provide some useful hintsfor our later work.

In the first place, they strongly suggest that weshould be liable to make great mistakes were we toregard men as equal. For our purposes, in praxeo-logical analysis, we had better treat them as unequal.It is only too clear that different men with the sameknowledge, acquired perhaps in schools to which allmen had equal access, in a society dedicated to theideal of 'equal opportunity for all', will neverthelessapply different parts of their common knowledge toa given situation, because their judgement on what isrelevant to it will differ. Different men's action is inreality oriented to different knowledge drawn fromdifferent sources of experience, but different know-ledge may flow even from the same experience.

These circumstances make a case against what wemay call 'methodological egalitarianisnV. Secondly,they suggest the existence of certain problems whicharise in the case of the simultaneous pursuit of plansby different actors, where these plans have somemeans or ends in common. In such cases each planbecomes a point of orientation for the other plans, andproblems of interaction (friction) or co-operation mayarise. The raison d'etre of the praxeological method


rests in the fact that human action exists in mentalform, as plan, before it takes place in space and time.Hence we may regard action as the unfolding of amental scheme, and make a comparative study ofaction and scheme. In such a study crucial significanceattaches to the degree of correspondence between theconditions of action met in reality by the actor andthe points of orientation (means, obstacles) which, inadvance, reflected them within the plan.

Every plan of course has to be flexible to someextent if it is to succeed. The need for flexibility partlystems from the fact that some of the knowledgerelevant to the action will only be acquired in agendo,i.e. after the plan has been drawn up and the courseof action started. To this extent the planner will haveto leave certain blank spaces in his scheme, details tobe filled in later as new knowledge accrues in action.It is of course impossible to plan everything inadvance, down to the most minute detail.

But the new knowledge acquired during the courseof action will not only be of this kind, i.e. merelyadditional to the knowledge the actor possessed at theoutset. In the more important cases such new know-ledge will correct and replace prior knowledge. Tothis extent, then, the plan will have to be partlyrevised since the points of orientation it containsare affected by the new knowledge acquired in action.In an extreme case such new knowledge may suggestthat the purpose of the plan is altogether unattainable,and then the whole plan will have to be discontinued.

We thus see that the relationship between plan andaction is not the simple one of cause and effect, but thecomplex one of interaction between mental acts andobservable events. Even to think here of independentand dependent variables would be misleading. When


new knowledge is acquired, what happens is that fromthe broad stream of experience certain elements areselected by our mind, which then transforms them intoa new structure of knowledge, modifying and partlyreplacing a prior structure. In this as in other respectsmen are unequal, and no two minds will perform thistask in identical fashion. The mental acts by which wetransform experience into knowledge, and by whichour 'world image' is constituted, are coloured by allthe traits of our personality. It is impossible to 'predict5

what knowledge an actor will derive from a given ex-perience, since he must interpret it in terms of his exist-ing 'situation image' before it can become knowledge.

It is quite instructive to turn for a moment to thenaturalistic method and see how it attempts to copewith these problems. It will hardly surprise us to learnthat those whose behaviouristic premises compel themto regard all human action as 'response to stimulus'should be hard put to it to explain the interpretationof experience and its transformation into knowledge.The easy way out of this dilemma is to assume, ineffect, that nothing new ever happens under the sun,and that a comprehensive set of alternative plans,sufficiently comprehensive to cover every possiblecontingency, is in existence from the beginning. The'response' to changing circumstances thus consists innothing more arduous than pulling out of a drawerthat new plan which corresponds to the new cir-cumstances and which, if our set is comprehensive,must already exist I The question as to how this com-prehensive set of alternative plans, which only a truesupermind could possibly have devised, came intoexistence, is then politely ignored.

We can distinguish an older and a more recentversion of this device. The older version we find in the


economic system of Pareto1 in the form of the so-calledindifference curve analysis. Here it is assumed thatconsumers and producers each have a comprehensiveset of alternative plans (graphically depicted in theform of two-dimensional indifference curves for anypair of goods) which enables them always to find the'optimal response' to any change of market price orother market conditions. By employing the 'staticmethod5, which in effect postulates that actors reactto present circumstances only and that nobody evergives a thought to a future which may be differentfrom the present, this solution is then given theappearance of determinateness.

The more recent version of the device takes theform of looking at human action by analogy to afeedback system. The deeds of man, acting in a worldin which relevant facts became known to him onlygradually in the course of his action, are here regardedas analogous to the 'action' of, for example, thoseinterplanetary vehicles which are supposed to 'steerthemselves' by a number of technical devices describedas 'translating information into appropriate action'.Similar instances are to be found in biology wheneverorganisms are capable of resisting forces hostile totheir survival.

The example shows how the mechanistic approachis confined, in the range of analogy on which it isable to draw, as in the choice of other conceptualtools at its disposal, to the forms of thought appropriateto problems encountered in nature and technology.Biology long ago omitted 'purpose' from its vocabularyand confined itself to the description and ordering ofits objects. Technology, to be sure, is concerned withpurposes, but with the purposes of the men who

1 V. Pareto, Manuel d'Economie Politique, Second Edition (1927).


devise the feedback systems. But in any case, whetherorganism or artifact, a feedback system can 'deal5 witha finite number of occurrences because it is equippedto do so. The question why it is so equipped does notarise. Moreover, the 'information' it uses requires noact of interpretation. Wherever, on the other hand,experience requires interpretation, that is, acts of themind which no two minds perform in identicalfashion and which transform it into knowledge thatonly men can have, the mechanistic analogy cannotbe applied.

In the light of this brief and inevitably ratherinadequate outline of the praxeological method wemust now attempt to answer three questions which willprobably have arisen in the mind of the reader.

1. Plans often fail. Will not such failure invalidatethat correspondence between plan and action, betweenmental scheme and observable events, on which ourmethod rests? Does this not mean that only successfulaction lends itself to our method of treatment ?

2. Can we say anything about relations betweenplans of different actors? In reality, it seems, suchrelations may be of very different character. Some-times we find co-operation, where the actions ofdifferent actors all contribute towards achieving thesame end. But sometimes we find rivalry, wheredifferent men act at cross purposes. What have weto say about these possibilities?

3. Every plan is geared to certain concrete purposesthe actor sets out to achieve. But it has often been saidthat the unintended consequences of action are among themost important problems of the social sciences. If so,how can an analytical scheme such as ours, in whichpurpose and plan are the fundamental concepts, copewith phenomena which were not intended or planned

44 T H E METHOD OF INTERPRETATION

by any actor and which thus appear to transcend thecategories of our scheme of thought ?

In answering the first question we need do no morethan refer to what was said above about the elementof knowledge in planning and the role of points oforientation. Failure of a plan must be due to in-adequate knowledge of the circumstances in whichaction has to be taken. We pointed out above thatnew knowledge acquired during the course of actionmay invalidate and replace the knowledge on whichthe plan was based. We mentioned the possibility thatas a result the whole plan may have to be abandoned.

We may even go a step further. Mere observationof external events can tell us nothing about successand failure. It is only in terms of the degree of corres-pondence between plan and outcome of action thatwe can meaningfully speak of success and failure at all.Correspondence between plan and action, so far frombeing invalidated by failure, thus, on the contrary,proves itself an indispensable conceptual tool for thestudy of failure.

The second question raises issues of fundamentalimportance that we shall have to deal with throughoutthe rest of this book.

Certainly the analysis of the individual plan canbe no more than the first step in constructing a theoryof action. But it is also an indispensable step. Havingtaken it, we shall not find it altogether too difficult toaccommodate the plans and actions of other actorswithin the framework of our analytical scheme. Forour actor they are simply points of orientation in noway different from other circumstances of action. Theother actors, be they allies or rivals, widen or restrictour own freedom of action. In the former case their


co-operation provides us with means; in the latter casetheir rivalry offers an obstacle. In either case theirprojected action constitutes points of orientaton forus.

The real difficulty lies here in a circumstance towhich we have already drawn attention. At themoment of planning the future actions of others, likeso many other future conditions of success, are un-certain and unknown. We can only form expectationsabout them and use them for our orientation. But ofcourse expectations may be disappointed with con-sequences we already know. Since human action ismore volatile than the conditions of nature, we havehere a source of danger to successful action, theimportance of which grows as society grows morecomplex. At the same time all societies have evolvedinstitutions which are calculated to reduce this un-certainty. To them most of our second essay will bedevoted.

The subject of the third question, the unintendedconsequences of action, will also occupy us on laterpages. For the moment we shall confine ourselves todrawing a distinction for the purpose of which we areable to make use of our reply to the first question.Here we have to distinguish between cases in whichthe unplanned consequences flow from the success ofindividual plans, and those where they flow fromfailure. The best-known example of the former caseis a market economy in which the pursuit of their ownpurposes by consumers and producers, individualwant-satisfaction and the maximization of profits,leads to the the optimal allocation of resources and thehighest possible satisfaction of wants. An 'equilibriumposition for the economic system as a whole' is thusreached and maintained owing to the repeated


success of all individual plans. The success of theseplans here entails the stability of the economic system.

But where the unintended consequence flow fromfailure, such stability will obviously not exist. Whetherthe plan revisions thus made necessary and thesubsequent concatenation of events are ever likely togenerate stability in a new system, or whether theywill lead to chronic instability, possibly even pro-gressively weakening the forces tending to integratethe social system, is an interesting question which weshall turn to in a moment.

The ordinary effect of the coexistence of a number ofdivergent plans which concern, partly at least, thesame means and ends will of course be that some plansfail and have to be revised. Unsuccessful planningthus prompts the need for more, and possibly better,planning. It in no way invalidates the need for ouranalysis of action in terms of plans.

It is impossible to show that, as a result of repeatedfailures and revisions, the various divergent plans willtend to grow closer together and in the end converge.This would be so only in a stationary world in whichit might be legitimate to expect that actors, like menshooting at a fixed target, will as a result of a processof trial and error gradually come to learn more andmore about the circumstances in which they have toact and thus be able progressively to correct theirmistakes. But the real world is a world of continuousunexpected change in which targets are moving ratherthan fixed. This means that even while men are gainingadditional knowledge by learning from earlier mis-takes, at the very same time some of their existingknowledge is continuously becoming obsolete. Asituation is even possible, which we might call 'thetragedy of the premature pioneer5, in which an actor's


sole mistake consists in anticipating a future event attoo early a date, so that, were he here to 'learn' fromhis mistake, he would actually nullify valid knowledgewhich, if retained to a later date, would probablyprove to be useful. We have to conclude that in aworld in motion forces reducing the divergence ofplans and other forces tending to widen such di-vergence will both be in operation, and that it isimpossible to say which set of forces will prevail in anyconcrete situation.

Finally, there is one aspect of the complex ofproblems caused by the unintended consequences ofaction to which we wish to call attention here, thoughseveral of its manifestations will occupy us in sub-sequent parts of this book. The problem is one wemight call the problem of the inter generational successionof plans. The world in which we are planning todaywith its houses, streets, parks, means of communica-tions, etc. is largely the cumulative result of plansmade by our ancestors. Some of these permanentresources were of course planned to be used by futuregenerations, but some were not. In old towns, forinstance, we find many buildings which throughoutthe centuries have served a succession of uses thatwere never dreamt of by their original architects,mansions which are now hotels, stables which havebeen turned into garages, assembly halls which havebecome post offices, and so on. Two interpretationsof these facts in terms of our analytical scheme arepossible of which one is unfavourable, the other favour-able, to the analysis of action in terms of plans such as weare advocating. On the one hand we might say that thesefacts go to show the ubiquitous nature of the unintended,and hence unplanned, consequences of action, and thatthe lesson we should learn from them is that the


sphere of planned action is after all only a small partof the whole realm of human action. This is the viewunfavourable to the type of analysis for which we arepleading.

But another interpretation of these facts appears tous to be more subtle and more compelling. We mightcall such plans as leave us, when completed, withpermanent artifacts such as those mentioned abovefor use by future generations, 'open-ended' plans, anddistinguish them from all other plans. This inter-pretation raises the difficult question as to whenexactly we are entitled to speak of the completion of aplan.

In an important sense acting man is at everymoment of time engaged in carrying out some in-complete plan. No man ever lives to see the day whenhe has carried out all his plans. These plans formsomething like an echelon, one starting and one endingevery day, so that the later are still incomplete whenthe earlier are completed. To speak of an inter-generational succession of plans might be actuallymisleading if by doing so we were to convey thenotion that one plan must be completed beforeanother can start. This of course is quite wrong, andit might be better to speak of the intertemporalnetwork of plans to give expression to the intricatenature of the forms of integration by means of whichplanning and action are welded into a whole. Withinthis network the artifacts mentioned above, relics ofopen-ended plans of the past, but at the same timepresent resources available for planning the future,occupy nodal points, the continuous existence ofwhich underlines the permanent significance of theunintended consequences of intentional human action.

On Institutions

To UNDERSTAND human action means to understandthe plan which guides the observable acts to which itgives rise. The praxeological method, which aims atenabling us to understand action, rests on the paral-lelism between action and plan, a fact which has nocounterpart in nature. The plan which gradually un-folds in space and time contains, we saw, an orientationscheme which must comprehend purpose, means, andobstacles. Action is thus oriented to them.

We must now return to the second question whichwe raised at the end of our first essay, that is, theinterrelationship between the actions of variousactors. We said there that formally for the actor thereis no difference between the action of others and anyother circumstances affecting the constraints boundinghis freedom of action. But we also pointed out thatmaterially a significant difference lies in the fact that,since human action is more volatile than the condi-tions of nature, it is far less easy to predict. In a com-plex society such as our own, in which the success ofour plans indirectly depends on the actions of millionsof other people, how can our orientation scheme pro-vide us with firm guidance? The answer has to besought in the existence, nature, and functions ofinstitutions.

An institution provides means of orientation to a largenumber of actors. It enables them to co-ordinate their

49

50 ON INSTITUTIONS

actions by means of orientation to a common signpost.If the plan is a mental scheme in which the conditionsof action are co-ordinated, we may regard institutions,as it were, as orientation schemes of the second order,to which planners orientate their plans as actorsorientate their actions to a plan. To investigate thenature, functions, and structural relationships betweeninstitutions is the main task of this essay.

Whether we post a letter, wait for a train, or draw acheque, our action is in each case orientated towardsa complex network of human action of which we knowenough to make it serve our ends, though we mayknow next to nothing about the internal working-order of these institutions. We know of course thatsuch an internal working-order exists, but in oureveryday life take no interest whatever in its details.We know very well that the Post Office worksaccording to a general plan, but such knowledge aswe have about it is usually quite irrelevant to theachievement of our purpose in posting a letter. Onlya few aspects of this general plan, perhaps the times ofcollection and delivery of mail, need be ofconcern to us.

The existence of such institutions is fundamental tocivilized society. They enable each of us to rely on theactions of thousands of anonymous others about whoseindividual purposes and plans we can know nothing.They are nodal points of society, co-ordinating theactions of millions whom they relieve of the need toacquire and digest detailed knowledge about othersand form detailed expectations about their futureaction. But even what knowledge of society they doprovide in highly condensed form may not all berelevant to the achievement of our immediate purposes.Economy of effort may induce us to ignore most ofthe time a good deal of the knowledge available to us.

ON INSTITUTIONS 51

Most banks proudly display their balance sheets intheir branch offices, but a normal customer hardlyever looks at them.

The existence of institutions raises a large numberof problems, only a few of which we are able toconsider here. But three of them appear to occupysuch a prominent place that we shall have to examinethem in detail.

There is, in the first place, the problem of institu-tional change. If institutions are to serve us as firmpoints of orientation their position in the socialfirmament must be fixed. Signposts must not beshifted. On the other hand, it is hardly possible toimagine that banks, railways, and other institutionsare totally exempt from change. It appears that suchchange need not interfere with the plans of users ofinstitutions provided it is known in advance. But somechanges will not comply with this condition. Whathappens then? Are situations possible in whichinstitutions mislead rather than guide planned action ?

There is, secondly, the problem of the institutionalorder and its unity. If institutions are to serve asinstruments of co-ordination, do they not themselveshave to be co-ordinated ? If so, that is, if each institu-tion forms part of a comprehensive structure, what isthe nature of the forces which integrate it? And whatwould be the character of circumstances in whichthese forces ceased to work ? In other words, what arethe conditions of integration and disintegration ?

From the confluence of these two problems therearises, thirdly, the question whether the forces ofintegration, supposing they do operate, would operatein all conditions of change. It goes without saying thatthe rise of new institutions, partly to replace olderones, but partly to fill 'gaps' in the institutional

52 ON INSTITUTIONS

structure, raises questions which belong to this thirdcategory.

What is the general nature of the conditions inwhich such new institutions would 'fit' into theexisting structure ? And where these conditions do notexist, is it impossible for new institutions to arise? Ifnot, does it mean that the existing institutionalstructure would have to change in such a way as toaccommodate the new accretions, or that it will beundermined by them?

To enumerate these questions is only to give avery rough outline of the tasks confronting us. Butbefore coming to grips with them we shall first haveto turn aside and examine what Weber thought aboutthem. In scrutinizing Weber's legacy, however, weshall soon have to learn that the construction of atheory of institutions designed to answer our questionson the basis of this legacy is anything but easy.

II

No general theory of institutions is to be foundanywhere in Weber's work. To be sure, he has muchto say about institutions and their modes of change.Even today his work is one of our richest mines ofinformation on institutions and their changes through-out history. Certainly we are entitled to say that thewhole range of institutions, religious, political, eco-nomic, legal, and educational which his powerfulmind encompassed, and their modes of change underthe impact of various social forces, were always inthe forefront of his interests. For all this it remainstrue that he never formulated a General Theory ofInstitutions. Fragments of such a theory can be foundand we shall of course have to examine them carefully.

ON INSTITUTIONS 53

But a coherent general framework within which thesefragments would find their places is not part ofWeber's legacy. It is possible to find reasons for thisabsence of a general framework which look super-ficially plausible but provide no real explanation.Three such reasons suggest themselves readily to thestudent of Weber's work.

The first is a linguistic one. Modern German hasno word which corresponds exactly to the meaning ofthe English word 'institution'. The German wordInstitution has a narrower meaning, confined toorganized institutions. In modern German usage thefamily is, but language is not, an Institution. Weberusually avoids the term altogether and speaks ofAnstalt, a legal term denoting an organized association.Modern German sociologists, on the other hand, haveadopted the term Gebilde precisely in order to renderthe meaning of the wider term, and Weber knew theword. Moreover, Menger in his Untersuchungen usedthe word Institution in exactly the same sense which ithas in current English. The suggestion therefore thatWeber, even had he wished to formulate a generaltheory of institutions, would have lacked the linguisticmould in which to cast his thought, fails to carry muchconviction.

A second, and stronger, reason we might find inWeber's repeatedly expressed view that theory, whilea necessary tool in the kit-bag of the historian, mustnever be allowed to become an end in itself. Hecertainly deprecated all theory for its own sake. Ingeneral he saw no reason for a higher level of abstrac-tion than the nature of the concrete object of enquirywarranted. Thus he may have thought a generaltheory of institutions unnecessary.

But are we really to believe that a mind as powerful

54 ON INSTITUTIONS

as his, having mastered a well-nigh incredible numberof detailed facts about institutions of the most diversekind, from ancient Judaism to Tsarist Russia, fromChina to modern America, never felt the need for aframework of generalizations to be drawn from thesefacts? How, indeed, is it possible even to order thisvast store of facts without establishing a certainnumber of generalizations at some level ?

Weber was not opposed to theory as such, but onlyto unnecessary theory or, what is the same thing,theory at a higher level of abstraction than the objectof enquiry warrants. We have attempted to show inthe first section of this essay why a general theory ofaction such as Weber envisaged not merely warrants,but actually requires, a general theory of institutions.Moreover, the facts show that on occasion, especiallywhen the (usually polemical) context of the discourseappeared to require it, Weber was by no meansaverse to establishing generalizations of a fairly highorder of abstraction. We shall have to devote attentionto some of these. What remains a puzzle is thus not theabsence of generalizations in Weber's work, but hisfailure to integrate what generalizations there areinto a coherent framework.

A third reason, which some will regard as a variantof the second, might be found in the circumstancessurrounding Weber's early training in the HistoricalSchool. Abstract theory, one might say, Weber didnot feel to be his metier. He did not deny the need forit, but in general, except in cases of (polemical)emergency, was inclined to leave it to others betterequipped than he to supply. Not for him the longchains of deductive reasoning proceeding from a fewaptly, but always arbitrarily, chosen axioms. He felthe could do his best work tilling other fields.

ON INSTITUTIONS 55

The trouble with this explanation is that, as wealready know, Weber did take an interest in the placeof abstract theory in social thought. How can one beinterested in methodology without being interested inall the methods, however abstract some of them maybe, which might be used in a discipline? To thisquestion the reply may be that it is one thing to be acritic of methods, yet quite another thing to practisethem.

The fact remains, however, that, especially inpolemical argument, Weber did not shun levels ofabstraction of which, were this explanation valid, heshould have been wary. In any case we may feel surethat, if he had thought a General Theory of Institu-tions called for, neither the limitations of his trainingin economic theory nor anything else would haveprevented him from creating it. In proffering our ownhypothesis why he did not do so we therefore have toexplain, in the first place, why he may not havethought it called for.

In the Methodenstreit, an interest in which, as wesaw earlier, sparked off Weber's methodologicalstudies, the origin, nature, and functions of institutionshad occupied a prominent place. The GermanHistorical School had taxed classical economists withignoring the effects of the institutional environmenton human action. Pointing to the variety and diversityof economic institutions in different societies andcenturies, adherents of this school asked how oneanalytical model could possibly account for all thevarieties of economic action in circumstances sodiverse. It seemed to them that this diversity ofinstitutions by itself invalidated that universal theoryof the market economy which lies at the heart ofclassical economics.

L.M.W. 3

56 ON INSTITUTIONS

Facing this challenge, Menger decided to turn theflank of his enemy's position by a bold move.1 Headmitted the importance of institutions for economicaction but distinguished between those which are theproduct of legislation (cthe common will') and thosewhich are not. He then raised the famous question'How can it be that institutions which serve thecommon welfare and are extremely significant for itsdevelopment come into being without a common willdirected towards establishing them?', which hedescribed as 'perhaps the most noteworthy problemof the social sciences' (p. 146).

His answer was, briefly, that 'those social structureswhich are the unintended result of social development'are all, more or less, like market prices and wage-ratesin that in a long historical process they have come intoexistence as a result of men pursuing their interests.'They present themselves to us as the unintendedresult of individual efforts of members of society, i.e.of efforts in pursuit of individual interests . . . they are. . . the unintended social result of individuallyteleological factors', (p. 158) In the Marshallian idiomwe might say that, while in the short run economicphenomena are indeed shaped by existing institutions,in the long run these institutions themselves are shapedby the very forces whose ubiquity and universal powerthe Historical School had denied. In this way Mengerclaimed to have wrested a most powerful weapon fromthe hands of his opponents. For they had failed tounderstand the true nature of institutions, 'a naturewhich has up to now been characterized merely byvague analogies or by meaningless phrases' (p. 158),

1 Carl Menger, Problems of Economics and Sociology, edited withan introduction by Louis Schneider (University of IllinoisPress 1963).

ON INSTITUTIONS 57

while he had shown that this nature is identical withthat of such strictly economic phenomena as marketprices, wage-rates, etc.

We have here, then, what we may call a praxeo-logical theory of institutions, admittedly in roughoutline, in which the existence of certain institutionsis explained as the unintended result of the pursuit ofindividual plans by large numbers of actors—as a'resultant of social forces', not a product of socialdesign. In Menger's terminology, they are theinstitutions of organic, not pragmatic origin. We maynote that in this part of his book Menger says nothingabout what determines the human actions which havesuch undesigned social effects. The pursuit of indivi-dual interests is here a wide notion without anydeterministic connotation. Within the constraints ofthe given situation men are presumed free to pursuetheir ends.

Alas, this voluntaristic trait of Menger's thesis wasmarred by his Appendix VI which bears the title'The Starting Point and the Goal of all HumanEconomy are Strictly Determined' (pp. 216-19). HereMenger argues that all economic action is strictlydetermined by human needs and the resourcesavailable to satisfy them: 'Our direct need and theimmediately available goods are in respect to anypresent moment given facts that are not within ourdiscretion' (p. 217). He admits that human action assuch, 'the way which can really be taken or actuallywill be taken by human agents . . . is by no meansstrictly determined a priori . . .'. But the reasons forthis are 'Arbitry, error and other influences'. Withoutsuch influences therefore all human action would bedeterminate.

It is possible, to be sure, to see in this Appendix VI

58 ON INSTITUTIONS

a relapse into an earlier period of Menger's thought,a more rigid determinism oriented to the ideals ofnineteenth-century natural science, to which Weber,as we know, objected. But a reader must be forgivenif he is baffled by the contrast between the twopassages.

What was Weber's attitude towards this issue? Aswe see it, he disagreed with the Historical School andwas quite willing to give Menger his carefully qualifiedblessing, but one can sense that he felt uneasy never-theless about Menger's rather ambiguous position ondeterminism and found it possible to convince himselfthat Menger did not have the whole answer either. Inthese circumstances he may have thought it wise toleave the whole question open—an attitude which inany case would come naturally to a disciple of theHistorical School. To Weber, with his aversion to'unnecessary' theory, no general theory seemed to becalled for in this situation.

Weber's rejection of the Volksgeist theory of institu-tions, espoused by some, though not all, adherents ofthe Historical School, is emphatic. In 'Roscher'sHistorical Method', the first paper he wrote afterrecovering from his illness,1 he explicitly endorsesMenger's criticism that Roscher and his followers, theHistorical School of economists, had misunderstoodthe method of Savigny and the Historical Law School,by making far more of the Volksgeist than the latterintended. Weber points out that this notion, at best'an auxiliary concept for the preliminary denotationof a multitude of concrete phenomena not yet logicallyworked out' and a 'resultant of innumerable cultural

1 Gesammelte Aufsdtze zur Wissenschaftslehre (Second Edition,1951), pp. 1-42. All translations from the German text of theseessays are ours.

ON INSTITUTIONS 59

effects' had been endowed by Roscher with a 'meta-physical character' and regarded as 'the real causeof the individual cultural manifestations of a peoplewhich emanate from it.'1 Such metaphysics was distaste-ful to him.

Otherwise, however, his attitude to Menger and histheory of institutions is rather ambiguous. It is acurious fact that in his greatest paper on methodology,the Essay on the 'Objectivity of the Social Sciences'of 1904, Menger's name is not mentioned once,2

though the whole essay is clearly directed against'naturalism', i.e. the dogmatic belief that there is andcan be only one truly scientific method. Menger'sview on determinism in human action is here evidentlyrelevant. Later on, however, in his Sociology of Law,Weber took over, with some qualifications, most ofMenger's thesis on the origin of 'organic' institutionsas the unintended results of individual action in thepursuit of interests, as a 'resultant of social forces'—at least in the field of legal institutions. On the otherhand he was careful to point out that he did notregard this thesis as a complete explanation.

It seems legitimate to infer that Weber's ambiguous1 ibid., p. 10.2 There is a reference to Menger, though not by name. 'In

spite of the fundamental methodological distinction betweenhistorical knowledge and the knowledge of "laws" which thecreator of the theory drew as the first and only one, he nowclaims empirical validity, in the sense of the deducibility ofreality from "laws", for the propositions of abstract theory.'(The Methodology of the Social Sciences, p . 87).

This is rather odd. Menger, while a strong defender of theabstract method of classical economics, can hardly be regardedas its 'creator' nor as the 'first and only one' to draw this par-ticular distinction. In our view this strange passage confirms theextent to which Weber's mind was preoccupied with Menger'swork.

60 ON INSTITUTIONS

attitude towards Menger was due to Menger'sambiguous attitude on the freedom of human action.Weber, uncertain to what extent Menger's praxeo-logical theory of institutions, towards which he wasquite sympathetic, rested ultimately on a deterministicpremise, wanted to avoid a 'confrontation5 with him.But had he tried his hand at a general theory ofinstitutions, such a confrontation could not have beenavoided. In this situation he did not feel the time wasripe for generalizations on such a precarious matter.

Il l

Whether or not our explanation is accepted, the factremains that no general theory of institutions is to befound in Weber's writings. But we need such a theory,as without it the theory of action which is to giveexpression to the praxeological method would beincomplete. We thus face the arduous task of piecingtogether what generalizations on institutions, of asufficiently high order of abstraction, we are able tolay our hands on in Weber's work, in order to seewhether they can serve as a foundation for the buildingwe have to erect.

Taking Weber's utterances in order of time, thefirst is one we already know: his rejection of the'emanationist' interpretation of the origin of institu-tions in the 1903 paper on Roscher mentioned above.We may here perhaps note that when Weber describesthe Volksgeist as nothing but a 'resultant of in-numerable cultural influences' this expression issimilar to one sometimes found in Menger, forexample when institutions are described as 'resultantsof social forces'.

ON INSTITUTIONS 6l

The second statement by Weber which is of interestto us we find in the context of his criticism of the workof the legal philosopher Stammler in 1907, in the'Paradigm of the Skat game'1 (a German card game).Stammler, not given to a very careful use of terms,held that the outstanding characteristic of social lifewas its being governed by rules, and had spoken ofthe analogy of'rules of the game5. Since a game maybe regarded as an institution, what Weber says in thecontext of his polemic against Stammler throws somelight on his general view on institutions.

His main point against Stammler is that though theplayers' action is of course oriented towards the rulesof the game they are playing, and though we mighttherefore call the rules a 'presupposition' of any con-crete game, this tells us nothing about the actualhappenings in a concrete game. In our terminology,the rules of the game constitute a set of orientationpoints, limiting the range of action of each playerbut also permitting him, because his rivals' actionsare equally subject to limitation, to guess with greaterconfidence what they will do. Within these limitshuman action here as elsewhere remains free. Weber'sargument thus follows the general line of anti-determinism. Norms as such cannot determine aconcrete outcome. But nothing has as yet been saidabout the origin of the rules of the game.

In 1913 Weber published an essay in which heelucidated the meaning of some of the fundamentalconcepts to be used in his magnum opus Wirtschaft undGesellschqft, which at that time was still in its earlystages. The last part of this essay is devoted to theAnstalt, the organized institution. Here he makesthree points which are of special interest to us.

1 Ibid., pp. 337-40-

62 ON INSTITUTIONS

Firstly, on the origin of such institutions, he stressesthat the norms which govern them arise conly in therarest cases by autonomous agreement of all thoseparticipating in future action from whom . . . loyaltytowards the norms is expected5.1 Almost always somepeople proclaim such norms and the others thensubmit to them. Institutional norms, then, have theirusual origin in Oktrqyierung, in the few imposing theirwill upon the many. We find here, in rough outline,an elite theory of the origin of institutions, and Weberclose to the position of Mosca and Pareto of whom, sofar as we know, he knew nothing.

Secondly, he points out that the same institutionoften comes to mean different things to differentpeople, and why this is so. It is created by a firstgroup who impose it upon, or 'suggest5 it to, others.It is 'run5 by a second group, namely of executives,who may interpret its purpose differently from thefirst. It is then used 'for their private purposes5 by athird group for whose members it is 'a means oforientation of their (legal or illegal) acts becausecertain expectations concerning the conduct of othersattach to them5 (p. 472). A fourth group, 'and theseare the masses5, simply learns by tradition certainmodes of conduct in respect of the institution 'mostlywithout any knowledge of purpose and meaning, oreven awareness of the existence of the norms5 (p. 473).He shows that the same principle applies to money,which is not an 'organized5 institution, an Anstalt.'How this has actually acquired its peculiar qualitiesthe money-user does not know—since even theexperts quarrel about it so violently.5

At the end of the essay he stresses once more thesignificant function of institutions, which lies in the

1 Gesamelte Aufsdtze zur Wissenschqftslehre, p . 468.

ON INSTITUTIONS 63

fact that they enable us to orientate our actiontowards 'unambiguous expectations to which theygive rise. And here rests the specific interest of rationalcapitalistic "enterprise" in "rational" norms whosepractical functioning, in terms of chances, can bejust as well calculated as that of a machine' (p. 474).We shall have to return later on to this significantpoint.

The most important generalizations, from our pointof view, are to be found in the Sociology of Law.1 Tobe sure, what is said here applies, strictly speaking,only to legal institutions. But the generalizations weencounter here are of such a fundamental nature thatthey are readily extended beyond the legal sphere.

It is hardly surprising that in this part of his workWeber was at his very best, if we remember that hisoriginal training was in law and legal history, andthat it embodies a lifetime's experience. Weberhimself must have felt this when he told his wife thatit was the most 'complete' part of his work.

In the Sociology of Law we find certain themes, bynow familiar, with a number of interesting variationsadded. Weber asks 'How do new legal normsoriginate?' and answers that, while today this largelyhappens by legislation, it has not always been so andneed not be so. He again rejects the metaphysicalexplanation of institutions: 'Scientifically, however,this conception leads nowhere' (p. 67). He also rejectsthe view that changes in 'external conditions ofexistence' by themselves are causes of legal change.'The really decisive element has always been a newline of conduct which then results either in a change

1 Max Weber on Law in Economy and Society, edited by MaxRheinstein (Harvard University Press, 1954). Here referred toas: Sociology of Law.

L.M.W.—3*

64 ON INSTITUTIONS

of the meaning of the existing rules of law or in thecreation of new rules of law' (p. 68).

He then cautiously adopts what is in essenceMenger's praxeological theory of the origin of un-designed institutions. New institutional forms aremost frequently created by individuals through'invention' and then disseminated by imitation andselection: 'Not merely in modern times has this lattersituation been of significance as a source of economicreorientation, but in all systems in which the mode of lifehas reached at least a measure of rationalization' (p. 68).

On the other hand, the systematic character of law,the postulate that all legal norms constitute a co-herent system, the legal system, is a late product ofour civilization. Weber, who in this whole chapter is,perhaps inevitably, more the legal historian than thesociologist, ascribes its evolution primarily to themental habits of the academically trained continentallawyers who were naturally inclined to interpret thelegal order as a 'closed' system (just as modernscience does). He is aware that: 'Among the conditionsfor the development of a market economy, thecalculability of the functioning of the coercivemachinery constitutes the technical prerequisite aswell as one of the incentives for the inventive geniusof the cautelary jurists' (p. 72). But it does not seemto have occurred to him to link this development of'calculability' to the necessarily formal character of acoherent legal system. On the contrary, he stressesseveral times the conflicts which are apt to arisebetween the 'formalism' of the logical thought of thelawyers and the needs of their clients and the publicat large:

It is by no means the peculiar foolishness of modernjurisprudence which leads to such conflicts. To a large

ON INSTITUTIONS 65

extent such conflicts rather are the inevitable consequenceof the incompatibility that exists between the intrinsicnecessities of logically consistent formal legal thinking andthe fact that the legally relevant agreements and activitiesof private parties are aimed at economic results andoriented towards economically determined expectations,(p. 308)

To us this discrepancy merely reflects a more deep-seated problem of society.

To Weber, then, the systematic character of thelegal order is a late product of modern history, likeother manifestations of'rationalization'.

To a youthful law, it is unknown. According to presentmodes of thought it represents an integration of allanalytically derived legal propositions in such a way thatthey constitute a logically clear, internally consistent, and,at least in theory, gapless system of rules, under which, itis implied, all conceivable fact situations must be capableof being logically subsumed lest their order lack aneffective guaranty . . . In the main, the 'system' haspredominantly been an external scheme for the orderingof legal data and has been of only minor significance in theanalytical derivation of legal propositions and in theconstruction of legal relationships. The specificallymodern form of systematization, which developed out ofRoman Law, has its point of departure in the logicalanalysis of the meaning of the legal propositions as wellas of the social actions, (p. 62)

On the other hand,

The increased need for specialized legal knowledgecreated the professional lawyer. This growing demand forexperience and specialized knowledge and the consequentstimulus for increasing rationalization of the law havealmost always come from increasing significance ofcommerce and those participating in it. For the solution

66 ON INSTITUTIONS

of the new problems thus created, specialized, i.e. rational,training is an indispensable requirement.

However,

a body of law can be 'rationalized' in various ways andby no means necessarily in the direction of the develop-ment of its 'juristic' qualities. The direction in which theseformal qualities develop is, however, conditioned directlyby 'intrajuristic' conditions: the particular character ofthe individuals who are in a position to influence 'pro-fessionally' the ways in which the law is shaped, and onlyindirectly by general economic and social conditions.(P- 97)

Weber also noted by what social forces the systematiccharacter of the modern legal order is threatened:

New demands for a 'social law' to be based upon suchemotionally coloured ethical postulates as justice orhuman dignity, and thus directed against the verydominance of a mere business morality, have arisen inmodern times with the emergence of the modern classproblem, (p. 308)

IV

We must now turn to our task of constructing a theoryof institutions which fits into our conceptual scheme.In examining the legacy which Max Weber left us wehave come across a number of generalizations whichmay serve us well as building blocks, but we shall alsohave to look for other building material. If we are toconduct ourselves like wise and responsible heirs,drawing our rightful usufruct but also adding to ourlegacy by our own efforts, we cannot rest content withWeber's generalizations as they are. We have to fitthem into the edifice we are about to construct, atheory of institutions which can be linked to the

ON INSTITUTIONS 67

theory of action set forth in the first essay. We shallfind that, as is often the case, as soon as we try to fita number of hitherto isolated generalizations into acoherent framework, they begin to reveal certainproblematical features, which without this test wouldprobably have gone unnoticed.

In turning to our task we shall not have to spendmuch time in examining the needs theory of institu-tions to be found in Menger's Appendix VI. We mustreject it. To be sure, no institution can exist for longunless it satisfies some need. But not every needgenerates an institution. The weakness of this theorylies in its failure to provide us with any criterion bywhich to distinguish between those needs which willfind their satisfaction through appropriate institutionsand those which will not. Menger was, here aselsewhere, too readily inclined to draw on the analogyof the market. In a market economy of course theprice system acts as a 'centralized agency5 for thedistribution of goods and services. We have here asimple criterion by which to determine which needswill in fact be satisfied. In the market, in this sense,all needs are brought into harmony, provided weregard prices as objective indices of the needs whichthe goods bought at these prices are to satisfy. Wherethis provision does not hold, no comparison of needsis possible. Outside the sphere of the market not evensuch a unifying agency as the price system is to befound. We must therefore conclude that the needstheory of institutions fails to satisfy our need for acoherent theory of the origin and functions ofinstitutions.

On the other hand, Menger's praxeological theoryof the origin of undesigned ('organic') institutions ismuch better suited to our analytical needs. Here we

68 ON INSTITUTIONS

have a theory which explains the origin of suchinstitutions in the same way as other innovations.Some men realize that it is possible to pursue theirinterests more effectively than they have done so farand that an existing situation offers opportunities notso far exploited. In concert with others they do exploitthem. If they are successful their example will findready imitators, at first a few, later on many.

Successful plans thus gradually crystallize intoinstitutions. Within the sphere of freedom of actionnew institutions arise as additional orientation points,which may take the place of older institutions thatbecame obsolete. Imitation of the successful is, hereas elsewhere, the most important form by which theways of the elite become the property of the masses.Once an idea originally grasped by an eager mind hasbeen 'tested' and found successful, it can be safelyemployed as a means to success by minds less eagerand lacking originality. Institutions are the relics ofthe pioneering efforts of former generations fromwhich we are still drawing benefit. Drawing oncemore on the analogy of the market, we may say thatthe theory of institutions is the sociological counterpartof the theory of competition in economics. In bothcases innovation and imitation are the complementaryelements of what is virtually the same social process.

But even if we were to regard the answer justoutlined as, by and large, a satisfactory answer to thequestion about the origin and functions of undesignedinstitutions, a new host of intricate questions wouldmake its appearance on the horizon. Most of thesecluster around the problem of the nature and per-manence of the institutional order, a problem whichwill have to remain at the centre of our stage until theend of the book.

ON INSTITUTIONS 69

When different men, successfully pursuing differentinterests, shape types of action which, by multifariousimitation, gradually crystallize into institutions, howcan we know that these undesigned products ofindividual pursuit will all be compatible with oneanother ? Will they all come to form a coherent system ?What problems will arise if this is not the case ?

In trying to answer these questions we shall get aslittle help from invoking the analogy of the marketmechanism as we did in the case of the needs theoryof institutions. In a market economy a 'tendencytowards a general equilibrium5 of prices and quantitiesproduced and exchanged can be shown to exist,subject to a number of conditions, which includeabsence of unexpected change that would disruptplans. Outside the market sphere no such pre-dominant tendency towards a general equilibriumcan be meaningfully asserted to operate. In everyconceivable situation there are 'destabilizing' as wellas 'stabilizing' forces at work. Moreover, it is hard toimagine any kind of institutional change which wouldnot upset at least some existing plans. To invoke theanalogy of the market forces will not therefore help usmuch.

We also have to remember that besides the un-designed institutions so far discussed there are thoseof the designed variety, the products of legislation andother manifestations of the 'social will'. What reasonshave we to believe that all institutions, designed andundesigned, will easily fit into a coherent whole, whenalready the undesigned by themselves leave us in somedoubt ?

In these circumstances we must clearly establish, asour next step, whether the coherence of the institutionalstructure as a whole is of great importance to us in our

70 ON INSTITUTIONS

task of constructing a theory of institutions. For if thiswere not so, if, for example, this coherence were to usa feature of secondary importance to the task at hand,we might perhaps safely ignore the difficulties whichnow appear on our horizon.

Unfortunately this easy way out is not open to us.The coherence and permanence of the institutionalorder are of paramount importance to those engaged,as we are, in tracing all the major conditions ofrational action. In reducing the uncertainty of thefuture which enshrouds all human action, and helpingus overcome the limitations of our ignorance of thepresent, such coherence and permanence are indeed ofprimary importance.

That this is so is most readily seen if we at firstconsider only legal institutions. Here it is indeedobvious that any act by which somebody commitshimself for a period of significant length, if for examplehe grants a loan repayable after twenty years, involvesthe coherence and permanence of the whole legalorder. That in any agreement between creditor anddebtor coherence and permanence of the legal rules isinvolved is obvious enough, but it might be thoughtat first that this requirement applies only to the rulesconcerning loan contracts. Of course this is a fallacy.In the first place, there can be no permanence of a setof norms unless they are coherent. Secondly, it isimpossible to separate the legal provisions governingloans from the rest of the legal order. Every concretebusiness transaction involves such a large number oflegal rules that it would be impossible to enumeratethem all and, hence, separate them from the rest ofthe legal order. That in our everyday lives we remainunaware of this fact is of course due to the relativelyinfrequent occurrence of legal disputes in the lives of

ON INSTITUTIONS 71

non-lawyers, since it is as a rule only in the case oflegal dispute that these matters are called to ourattention, and even in such a dispute only a few rulesbecome the subject of litigation. Finally, in a legalsystem that lacked coherence it would be impossibleto predict the outcome of a single case, as it would beimpossible to determine the scope and nature of the' gaps' in the system as well as of all the conceivablecontradictions in it.

It is therefore wrong to see, with Weber, in thecoherence and permanence of modern legal systemsnothing but the sediment of a certain type of legaleducation, the product of lawyers whose minds,trained to logic and order, demanded an orderlyarrangement of their tool-box. These features of ourlegal order are typically undesigned features of ourtype of civilization.

This is not to deny that there is solid merit inWeber's way of looking at this development. Intracing the history of an institution there is always agood deal to be said for stressing the intellectualpropensity, the 'spirit', of the elite which has createdit. Successful institutions often bear the unmistakableimprint of the spirit of their creators even aftercenturies of change. But the tendency to stress suchspiritual origins is a virtue which, like other virtues,can be practised to excess. In this case the lawyersclearly also had to take their orientation from theneeds of their clients for a coherent legal order.

The discrepancy, stressed by Weber and mentionedabove (p. 64), between the need of business-men forsimple rules and the complex characteristics of legallogic, does of course exist and often leads to conflictbetween the lawyers and the public. But the realconflict exists here rather between the short-term need

72 ON INSTITUTIONS

of the individual businessman for simplicity and thelong-term need of the business community as a wholefor a coherent legal order, which entails complexlogical rules. In this situation the lawyers merely actas intermediaries.

The question now arises as to how the coherenceand permanence of the legal order can be reconciledwith the facts of annual legislation. How can we speakof the uniformity and continuous existence of a bodyof norms if every one of these norms can be changedevery year ? The answer has to be that this is preciselyhow the lawyer has to regard the legal system, verymuch as the merchant looks upon his stock as a whole,as it appears in his balance sheet, as consisting entirelyof easily replaceable parts. It is a legal fiction necessaryto lend coherence to the framework of legal thought.

But we have to look at the matter from the praxeo-logical, as distinct from the legal, point of view, andmust disregard legal fictions. In the light of what hasbeen said above about the social function of institu-tions as signposts, it must be clear that the more oftenthe legal order is subjected to change, by legislationor judicial interpretation, the more it loses its capacityto serve as a means of orientation in relation to theaction of others. This fact has some bearing on thewhole question of the status of designed institutionswithin our conceptual edifice. There must clearly be alimit to the amount of annual designing and re-designing of institutions which society can stand. Thelegal order can absorb some changes, but not toomany of them, and they must not be of a fundamentalkind.

If we are now to extend our perspective from thelegal sphere to the institutions of society as a wholewe have to establish, first of all, the existence of

ON INSTITUTIONS 73

coherence and permanence in this wider sphere.We are no doubt entitled to speak of a legal order,but with what right may we claim to speak of aninstitutional order in general ? Even if we succeededin establishing the existence of such an order, it islikely that it would have to be a much looser and lesscoherent order than that of the legal sphere.

In our situation it might be tempting to invoke thesupport of one of the many 'social system' theorieswhich now abound in the field of the social sciences.It would seem that if the network of social relation-ships is to lend itself to description in terms of a'system5 at all, institutions will largely have to provideits structure and thus have an important part to playin it. And since institutions have an importantfunction in guiding social action, do they not thuslend themselves readily to treatment in terms of the'structural-functional' variety of social-system theories?

There are a number of reasons why we should notrely on such support, and why we are compelled toseek to establish the existence of an institutional orderby our own efforts. In the first place, there aresubstantial differences between these various theories,in particular as regards their level of abstraction.Some authors do not seem to mean by 'system'anything more than the existence of a set of socialrelationships. Others rely largely on functionalspecialization. The status of institutions within thecontext of these theories would require a considerableeffort at clarification. It is clear that a good deal of'structure' rests on them. Unfortunately, however,

74 ON INSTITUTIONS

most of these theories proceed on such a high levelof abstraction that one never knows when the institu-tions referred to are meant to be concrete institutions,and when they are elements of an abstract system.

There is another reason, even more important tous, why we should not invoke the support of thesocial-system theories currently in fashion to help usin our endeavour. We are concerned with the legacyof Max Weber. As we pointed out in the Introduction,Weber's approach to social action is something verydifferent from that of the structural-functionaltheories.1 Weber was concerned with the meaning theactor attributes to his action. Most social-systemtheories ignore this aspect of action. As regardsinstitutions in particular, when we speak of the'function' of institutions in guiding and co-ordinatingthe actions of millions of individuals we are followingWeber in using this word in a sense very different fromthe one it has in the words 'structural-functional'. Thetheory we are attempting to establish aims at thereduction of certain social phenomena to humanmental acts as manifested in plans. Most of thetheories mentioned, by contrast, aim at establishingtheir 'systems' in terms of recurrent patterns of actionwithout reference to the meaning such action has tothe individuals acting. We believe we are makinglegitimate usufruct of Weber's legacy. It follows thatwe can hardly hope to draw benefit from social-systemtheories of the type characterized.

We also have to remember that Weber, as he1 For a detailed critique of 'functionalism' in social theory

from a point of view very close to Weber's see John Rex, KeyProblems of Sociological Theory (Routledge and Kegan Paul, 1961),ch. IV. See also W. G. Runciman, Social Science and PoliticalTheory (1963), GUP, pp. 110-22.

ON INSTITUTIONS 75

explained at length in his famous critique of Stammler,1

attributed great importance to the distinction betweenlegal norms and human conduct oriented to suchnorms, and emphasized that the former in no way'determine3 the latter. We thus have good reason todistinguish carefully between legal norms and thoserecurrent patterns of conduct which we call institu-tions. Are we, then, entitled to speak of an institutionalorder ?

First of all, the mere fact that each institutiondenotes a recurrent pattern of conduct does not byitself entail the existence of an over-all institutionalorder. As was the case with the legal order, thecriterion of existence of the wider institutional order,if such can be shown to exist, would have to be soughtin its capacity to outlast its individual elements. Theforms and character of its existence have therefore tobe established separately from that of its componentelements. In comparing the legal system with thewider institutional order we have to remember thatthe unity of the former lies in its character as a systemof norms, while the unity of the latter will have to besought elsewhere.

We shall now compare the legal system and theinstitutional order with respect to their degrees ofcoherence, and we shall do so by comparing them withregard to four of their characteristics: permanence,consistency, unity, and over-all complementarity('gaplessness').

The first characteristic is evidently shared by boththe legal system and the institutional order as a whole.We simply cannot speak of an aggregate as a 'whole'unless it outlasts its component elements. The im-portance of the permanence of the institutional order

1 Gesammelte Aufsdtze zur Wissenschaftslehre (1951), pp. 291-359.

76 ON INSTITUTIONS

in general, as well as for a theory of action such asours, requires no comment.

As regards consistency, our second characteristic,the matter is already more complicated. In the caseof legal institutions the range of required com-patibility comprehends them all. No two legalinstitutions can be incompatible. With other institu-tions the requirement of compatibility is less strictsince not all institutions are used by the same actorsor figure in the same plans. The existence of militaryinstitutions based on absolute obedience to superiorauthority does not preclude the existence of otherinstitutions in which orders may well and are expectedto be questioned, even though the two principles areincompatible. But there must indeed always be somefundamental institutions with which all others arecompatible.

The unity of the legal system rests on the logicalcharacter of its norms. It stems from the compre-hensive nature of the range of required compatibilityjust discussed. Though we find no exact counterpartof this in the wider institutional sphere, neverthelessinstitutions here also display an 'order', they are notan aggregate of random composition. The Post Officecould hardly take over the functions of the police, theclergy scarcely act as a Fire Brigade. The basis of thisorder, the characteristic property which bestows uponit what unity it has, is here evidently functionalspecialization.

The greatest difference exists with respect to ourfourth characteristic: over-all complementarity. Thelegal system is a seamless web. It has no 'gaps'. Ajudge before whom a legal case is brought can neverrefuse to give a decision on the grounds that heknows of no legal norm to apply to it. He has to find

ON INSTITUTIONS 77

one. The legal order abhors a vacuum no less thatnature does. In the wider institutional sphere we findno parallel to this characteristic. Some institutions willbe complementary to one another in that theyrequire each other's services, like Post Office andrailways or airlines. In fact, such group comple-mentarity is the inevitable result of the functionalspecialization of individual institutions. But here nointer-group complementarity need exist. 'Gaps' areubiquitous.

As a result of our comparison we have to concludethat the legal system and the wider institutional ordershare the first characteristic of permanence, while thefourth, 'gapless' complementarity, is absent from thelatter. As regards compatibility of institutions, therange is less comprehensive in the wider sphere thanin the legal sector. Both display enough unity to claimthe character of a structured whole, but in the caseof the institutions of society this unity rests on func-tional specialization and is not of a logical nature. Allin all we have to realize that the coherence of thewider institutional order, while it certainly exists, isweaker than that of that part of it which is formed bythe legal institutions.

From the fact that the two spheres, the wider andthe narrower, share the property of permanence itfollows that they also share the noteworthy character-istic that the permanence of the whole does not entailthe permanence of each of its parts. It is as true of theinstitutional order as of its legal part that the order asa whole lasts while each individual institution maychange. We shall see that this coincidence of per-manence of the whole with flexibility of its parts givesrise to a number of intricate problems.

Institutions rise and fall, they move and change. An

78 ON INSTITUTIONS

institution may last a long time, but during this timeassume new functions or discard old ones. We shallfind later on that these facts are likely to haveparticularly far-reaching effects in the sphere ofpolitical institutions.

These institutional changes no doubt often takeplace in response to changing needs, but also often forother reasons. An institution may cease to exist, forexample, because the services required for it are nolonger available, perhaps because, owing to a changein the moral and intellectual climate of society, thequalities of will and mind needed from those re-sponsible for it have become an object of contempt orderision, or perhaps because those whose skills arerequired are now attracted into other avenues. Theimportance of factors such as these on the 'supply side'provides further illustration of the inadequacy of theneeds theory of institutions that we rejected earlier.To bring a new institution into existence requires notmerely the existence of certain needs but also thespecific 'entrepreneurial' skill of the innovator, as wellas that of his successful imitators. But even to adjustan existing institution to new uses requires specificskills.

In every society we shall therefore at any momentfind institutions belonging to different historical'strata', some of which were originally devised forpurposes very different from those for which they arepresently used, and which nevertheless together forma coherent pattern—a pattern which, however, willnot last. As the present lay-out of an old town (inwhich we find buildings erected over many centuriesand built in many different styles) owes no less to theingenuity of its present users than to the genius of theoriginal architects who had probably designed its

ON INSTITUTIONS 79

buildings for entirely different purposes, so the presentpattern of the institutional order owes no less to theingenuity of present users of these institutions than tothat of their originators. Institutions change less as aresult of 'changing circumstances' than as a result ofhuman action designed to meet change.

But how much change of individual institutions iscompatible with the permanence of our structure ? Thewhole problem of 'flexibility versus coherence5 nowappears on the horizon. There must be some flexibility,some room for manoeuvre if men are to pursue theirvarious interests. But how much of it can we concedebefore the whole institutional structure is impaired?

Confronted with this dilemma we must rememberthat it is not change as such, but unexpected change,which jeopardises planned action. The position of eachinstitution on the social firmament must be given, orbe at least knowable. It need not be fixed, to be sure,but then its orbit at least must be known. Not move-ment as such, but irregular movement disqualifiesan institution from serving as a point of orientation.The crucial requirement is for actors to be able totake their bearings by existing institutions, to be ableto csteer by them'.

A night at the theatre with the first act of Hamletfollowed by the second act of Macbeth and the thirdact of King Lear might have its attractions, providedeverybody in cast and audience knew the programmebeforehand. Only if the stage management were tointroduce it unexpectedly would chaos on the stageand bewilderment in the audience be likely to result.The reason such a programme might be feasible is ofcourse that some actors and actresses have a repertoireof roles sufficiently wide to permit it, and that theimmediate plans of most theatregoers extend over a

8O ON INSTITUTIONS

few hours only. But in our society, especially inmodern industry, many plans (buildings, plant,equipment) have to extend over a large number ofyears and are therefore particularly susceptible tounexpected change. The conclusion appears in-evitable that the more important become long-termplans which, once the planned course of action hasbeen set in motion, cannot be adjusted to subsequentchange, the more damaging institutional changebecomes. Since at any moment some such long-termplan is bound to be in course of execution, the timefor painless institutional change will never arriveunless prior notice of it is given to all interestedparties sufficiently far ahead to give any plan inoperation time for completion. It is thus almostinevitable that all institutional change will upset someplans in the course of execution.

Unfortunately we have not yet reached the end ofour difficulties. Such unexpected change is likely tohave further repercussions. In particular with respectto the relationship between designed and undesignedinstitutions we have to note that institutions can onlybe designed to meet a certain known situation, or alimited number of possible, i.e. conceivable, situations,but not an unlimited number of unknown situations.We therefore face not merely the problem of howdesigned and undesigned institutions can supplementeach other in such a way as to form together acoherent institutional structure. We now encounteran even worse possibility, namely that unexpectedchange of undesigned institutions may not merelyjeopardize the coherence of the institutional structureas a whole, but in addition may obviate the verydesign of the designed institutions.

Here we might contemplate the following way out.

ON INSTITUTIONS 8l

In a society in which it is generally known thatfrequent change of undesigned institutions is in-evitable, the designers of designed institutions maydeliberately confine their activity to designing aframework which leaves room for a good deal (inprinciple an unlimited amount) of change which,since it will take place within the framework, will notaffect the latter as such. This device would serve tosolve our second problem, even though we could notbe certain that the integration of the institutionalstructure as a coherent whole can be accomplished inthis fashion.

This idea is not a mere figment of our imagination.The legal framework of modern Western societies hasin fact achieved something similar to the model justenvisaged by leaving a wide sphere of 'freedom ofcontract5 to individuals acting in pursuit of theirrespective interests. The modern market economywould not be possible without it.

In such a society it might be said that the un-designed institutions which evolve gradually as theunintended and unforeseeable result of the pursuit ofindividual interests accumulate in the interstices of thelegal order. The interstices have been planned,though the sediments accumulating in them have notand could not have been. In a society of this type wemight then distinguish between the external institutionswhich constitute, as it were, the outer framework ofsociety, the legal order, and the internal institutionswhich gradually evolve as a result of market processesand other forms of spontaneous individual action. Itseems to us that it is within a scheme such as this thatthe praxeological theory of institutions which we areattempting to establish most readily finds its place.We also believe that Menger had a scheme similar to

82 ON INSTITUTIONS

this in mind when he set forth his ideas on institutionsin the third part of his book.

But such a model of the character of the relation-ships between external and internal, designed andundesigned, institutions is not quite satisfactory for ourpurpose. It fails to take account of the complex natureof the relationships which obtain here. The impliedcontrast between firm outer structure and shapelessinner void could actually be highly misleading.

In the first place, the model rests on the assumptionthat undesigned institutions evolve while the designedform an outer structure, that is to say, that the formeralter much more rapidly than do the latter. Butdesigned institutions also change and we have noreason to believe that their speed of change willalways be less than that of the undesigned variety.The problems of structural change of designedinstitutions will occupy us in a subsequent essay.

Secondly, the processes of change of the two classescannot be regarded as being independent of eachother. Changes in the legal order may affect the areafor manoeuvre within which individuals may moveand undesigned institutions evolve. On the other hand,the evolution of undesigned institutions also createsnew problems for the legal order. Sooner or latersome of them may have to be co-ordinated. The lawmay permit everybody to form companies withlimited liability, or trade unions, but sooner or later,simply to reduce the amount of possible litigation,some legal rules about the relationship betweendirectors and shareholders, branch secretaries andmembers, have to be promulgated.

Thirdly, it is always possible that the slow evolutionof some institutions, even though at first taking placeapparently within the interstices of an existing social

ON INSTITUTIONS 83

and legal order, will gradually lead to what we mightcall 'deformation of social space5. The coherence andpermanence of the existing social order will then bejeopardized even without any change in the legalsystem. The danger will be much greater where theinstitutions growing up are in some respects in conflictwith each other, so that only one or the other, butnot both, can be integrated into the existing institu-tional structure. Quite serious problems can arise inthis way.

The problems mentioned can be grouped underthree heads:

(1) those which arise from the multiplicity of sources(interests) —coherence;

(2) those which arise from the lapse of time and theneed to adjust existing institutions to newinstitutions—flexibility, change;

(3) those which arise from uncertainty as to whichnew institutions will exist at a future time—flexibility, change, adjustment to what?

We shall now give a topical example of destabilizinginstitutional change in which all these three categoriesare involved.

When in the years following the First World Warmost countries of the Western world adopted theBritish institutions of'collective bargaining', sometimesin their pure British form, sometimes with the additionof institutions of compulsory arbitration in industrialdisputes, not many voices of dissent were heard. Someeconomists showed themselves aware of the elementof bilateral monopoly that 'industry-wide bargaining'for wage-rates and working conditions would entail,and pointed to the dangers inherent in such a situation.

84 ON INSTITUTIONS

But as a rule they, and anybody else expressing doubtsabout the excellence of the new dispensation, weresimply regarded as 'reactionaries'. It seemed that aproblem which in the early decades of the industrialage had baffled so many men of good will had nowat last found a solution: how to fit labour relations inmodern industry into the framework of the marketwithout results which appeared to deprive theindividual worker of all influence on the outcome ofthe market process. 'Collective bargaining' after allwas bargaining, and was not bargaining of theessence of the market ?

Since in the world of 1920 the framework of themarket economy was simply taken for granted, itappeared that even collusion between the bargainerscould only occur within narrow limits. With acompetitive price system in existence no single priceand hence no single wage-rate could get very far outof line. In fact, the competitive price system by itsvery existence set fairly narrow limits to the areawithin which wage-rates could be determined bybargaining. Any attempt by trade unions to induceemployers to accept wage-rates which were 'too high'would adversely affect the volume of sales and thuslead to unemployment. The institutions of collectivebargaining as seen in the perspective of 1920 appearedto be embedded in an economic order sufficientlystrong and stable to vouchsafe their beneficialcharacter.

We are living in a different world. No economistwould deny today that the continuous inflation fromwhich the Western world has suffered for more thantwo decades has something to do with the modernmethods of determining wage-rates. We now have toask what are the precise circumstances that have

ON INSTITUTIONS 85

turned institutions which in 1920 looked quiteharmless into a source of a dangerous and de-stabilizing social processes which contemporaryWestern society, for all its wealth and vauntedefficiency, appears to be unable to stop. We have herea clear example of what we called above 'deformationof social space'.

Three kinds of change in economic institutions, aswell as in the mental climate in which they flourish,appear to us as the main causes of this development,though no doubt it would be possible to enumerate anumber of other factors contributing to the inflationaryresult.

There is, in the first place, the change in the mone-tary system from a metallic standard to a debt moneysystem. Modern money consists of claims againstbanks, central banks, or governments. It is of theessence of such a system that the total number of suchclaims that might be created is in principle unlimited,though control by a public authority may limit it atany particular point of time. While in the world of1920 it was possible to hold that the limited quantityof metallic money kept the price system withinbounds and thus also set limits to the maximumwage-rates attainable by bargaining, no such 'ultimatedeterminant' exists at the present time. Today itwould be almost more correct to say that the totalquantity of money-claims is influenced by nothing somuch as by the total amount of wage-claims that havebeen granted. This is what Sir John Hicks meant bythe 'labour standard' which has replaced the old goldstandard. In other words, the transition from ametallic to a credit standard, the adoption of amonetary system in which money can be createdvirtually at will, has removed an important external

86 ON INSTITUTIONS

restraint on the wage-setting power of the industrialbargainers.

The second change concerns the price-setting powerof industrialists, who at the same time represent oneside in the bargaining process. Today most prices ofindustrial goods are set by their producers, they aretypically list or catalogue prices. When employersgrant higher wage-rates they are virtually certain thatthey will be able to recoup such increases in the costsof production in the form of higher prices of the goodsthey sell, in particular since they know that theircompetitors will have to pay the same higher wagerates.

This situation differs in important respects fromthat which prevailed in the market economy of thenineteenth century. In that period the most importantindustrial goods (textiles, coal, furniture) were typicallyproduced by firms of fairly small size, while the marketfor them was dominated by wholesale merchantsacting as intermediaries between producers and retailsellers. These wholesale merchants, deriving theirprofits solely from their turnover of goods, wereprimarily interested in maintaining their rate ofturnover, but not at all interested in productioncosts, which did not affect them directly. They hadto match supply with demand if their profits were notto suffer. A fall in demand would induce them toreduce their selling prices, and so their buying prices.Production costs of goods had then to be adjustedaccordingly. Market prices determined wage-ratesand not the other way round.

Today the wholesale merchant as a price-setter hasall but disappeared and with him the flexible pricesystem characteristic of the nineteenth-century marketeconomy. At present prices are set by industrial

ON INSTITUTIONS 87

producers more interested in their profit margin thanin their rate of turnover. A fall in demand will leadtoday to a fall in output and employment, but hardlyever to a fall in prices.

Yet these two institutional changes by themselveswould not have sufficed to bring about the permanentinflation of our age. The most important economiccharacteristic of our age is surely that in our worldprices can, in the long run at least, only rise but neverfall. Our first two reasons serve to explain merely whyin our world certain restraints on the rise of wage-rates,which in 1920 were still universally taken for granted,have disappeared. They do not yet explain why thesewage-rates rise continuously and do not fall even intimes of depression.

The third change which explains this very fact wasnot, strictly speaking, of an institutional but of amoral nature. In our world it has come to be acceptedas an article of social faith that no money wage-ratemust ever be allowed to fall, that wage- and salary-earners have a right to expect that their moneyincomes will rise, at least in the long run, and thatthis expectation must in no circumstances be dis-appointed. This means for all practical purposes that,since wage-rates can only rise and never fall, the samemust apply to prices.

Whether this change in the social atmosphere inwhich the institutions of collective bargaining functionhas to be regarded as the true cause of the inflationaryprocess, while the other two changes mentioned shouldperhaps only be regarded as necessary conditions, is aquestion we shall not discuss. All human action is ofcourse oriented to the conditions of its success. Thelesson we have to learn from our example is ratherthat a change in the mental climate may by itself,

L.M.W.—4

88 ON INSTITUTIONS

without the creation of any new institutions or thedisappearance of old landmarks in the institutionallandscape, turn out to be an important institutionalchange because it affects the way in which men usetheir existing institutions, An undesigned institutionwhich originally was able to operate in one of severalpossible ways (wage-rates could either fall or rise orremain constant) may, when one of these ways becomessocially obsolete and other institutional changes occurconcurrently, acquire an entirely new kind of impetusnever dreamt of by its pioneers. It is thus possible foran institution, without any change in its outer formof appearance and without anybody, even amongthose who make daily use of it, noticing it for a longtime, gradually to change its character, its modusoperandi and its place in the whole institutionalstructure.

VI

The time has come for us to cast a backward glanceat the road along which we have travelled and toattempt to draw some conclusions from what we havelearnt.

We came to see that a theory of action which aimsat intelligibility must rest on the parallelism betweenplan and action. Institutions serve to co-ordinate plansin large societies. To serve this purpose they mustform a structure to which coherence and permanencecan be attributed, as no institution stands by itself andall action extends into the future. But a changingworld also requires flexibility of plans and institutions.We saw that undesigned institutions in particular canbe regarded as successful plans which have crystallizedinto institutions through widespread imitation. It

ON INSTITUTIONS 89

seems therefore that the need for coherence andpermanence on the one hand and for flexibility on theother cannot be easily reconciled. But we need notdespair of our theory of institutional structure.

On the one hand, it would hardly be surprising ifthe range of possible disturbances of the institutionalorder generated by the need for flexibility which isdisclosed by our analysis were to considerably exceedthe actual range we are likely to find in any givensociety. An analytical scheme such as ours mustcomprehend the whole range of possible, and not justof probable occurrences. How many of these willbecome actual is another matter. The seriousness ofthe potential threat to institutional stability emanatingfrom the need for flexibility is very much a matter oftime. Slow change is less harmful than fast change.Almost any change takes time and so do its repercus-sions. The amount of change possible per unit of timeis also limited. As regards designed institutions, thereis a limit to the annual activity of skilled designers.Devices such as delegated legislation may widen theselimits but cannot erase them. In the case of undesignedinstitutions it takes time for successful modes of actionto crystallize into institutions. It takes time even forthe participants to find out which action was successfuland which was not. It takes further time for suchknowledge to become diffused among potentialimitators. Apart from the time aspect of the matter,the very looseness of the institutional structure wediscussed above tends to act as a protective device insuch cases. The lack of complementarity betweeninstitutions of different classes means here that thearea over which any given change will have re-percussions is limited.

On the other hand we clearly have to ask how

go ON INSTITUTIONS

in reality societies continue to cope with such prob-lems. How is the need for coherence and permanencereconciled with that for flexibility in the real world?While no doubt different devices have been employedfor this purpose in different societies and at differenttimes, four such devices appear to call for attentionin the context of our investigation and to be entitledto a place in an analytical scheme such as ours.

We are already familiar with the first two of thesefrom the model which we presented in the previoussection. The first device consists in granting toindividuals a fairly wide sphere of 'contractualfreedom', a sphere in which change must be expectedto be frequent and which may be regarded as the mainsource of undesigned institutions. The first deviceconsists then, briefly, in having institutions which arefrequently mutable in a definite sphere of action.

Its complement, our second device, consists inhaving a few 'fundamental' institutions which, bycontrast, are not mutable at all. In our formerterminology, these external institutions must providea firm outer structure in the interstices of which thesediments of individual efforts in the 'free and mutable'sphere can accumulate. We must stress again thatthese two devices are complementary to each other.Frequently mutable and (almost) immutable institu-tions require and support one another. As the classicaleconomists knew well, a market economy may adjustitself to changes of many kinds, but it rests un-conditionally on the institutions of property andcontract.

The third device, which is new to us, takes theform of meeting a situation requiring change not bythe creation of a new institution, nor by replacing anold by a new, but by 'widening' an existing institution

ON INSTITUTIONS gi

in such a way that it can serve new interests withoutupsetting the plans which have thus far made use of it.The widening of the concept of property in themodern company, in such a way that the relationshipsbetween directors and shareholders can be broughtwithin its province, appears a good example.

The fourth, and for more than one reason we mightsay the ultimate, device of which every societydisposes in order to defend itself against the desperatecases of dilemmas of this kind, is to prohibit changewhich threatens to upset the social order, and to actagainst the interests engendering it. Where institutionalchange prompted by the pursuit of interests threatensthe unity of the institutional order, it is the latterwhich has to be defended.

No society can stand more than a certain amount ofchange within any period. No doubt the limit oftolerance varies between one society and another, orbetween one period and another within the samesociety. A good deal will clearly depend on theextent to which the devices enumerated (and possiblyothers) are available as alternatives.

But there can be little doubt that some limit oftolerance of institutional change exists everywhere,and that every healthy society is able to call uponsocial forces of considerable strength when this limitis being approached. In this simple fact we have tosee a manifestation of cthe Rationality of Tradition'.Every social system is always jeopardized by thepluralism of contending interests and has to dependon the strength of its institutional order to defend itagainst such deformation of social space as wouldthreaten its continued existence.

On Political Institutions

IN THIS essay we shall endeavour to apply the insightwe have gained thus far into the nature and significanceof institutions and institutional structure in general tothe institutions of the political sphere. But at thesame time we shall have to relate whatever con-clusions we may reach in this field to Max Weber'sviews on these matters.

Prima facie it might appear that our first task canhardly be an altogether forbidding one. We havealready stressed the necessary existence of an institu-tional order, a structural relationship within whicheach individual institution must find its place. Ourtask would thus appear to be merely to indicate theprecise place which political institutions occupy withinthis general structure. Nor does it seem altogetherdifficult, at a first glance at least, to apply the conceptswe have devised to the institutions of the politicalsphere. One might even hope that the designednature of most political institutions will here relieveus of all those problems which, as we saw, are apt toarise from the possibly incoherent nature of un-designed institutions prompted by the pursuit ofdivergent interests. This should make our task easier.Moreover, are political institutions not externalinstitutions in the simple sense that most of the dailylife of society is taking place within the precinctsshaped by them without anybody, in normal cir-cumstances, taking much notice of their existence?Again, it seems that the complications which, as we

92

ON POLITICAL INSTITUTIONS 93

learnt, might arise from the difficulty of having todelimit the spheres of external and internal institutionsare here absent.

We shall soon see, however, that such optimism isquite unwarranted. In the first place, not all politicalinstitutions are designed institutions. Modern politicalparties as well as many of those organizations whichare capable of exercising political pressure in thepursuit of group interests were never 'designed' in thesense in which legal institutions are. Secondly, apolitical institution originally designed for one set ofpurposes may in the course of time and under thepressure of events and conflicting interests graduallychange its character and come to assume functionswhich were never assigned to it, as well as discardother functions which originally were.

Finally, the problem of 'coherence and permanence'versus 'flexibility' emerges in the political sphere in aform which is particularly acute. It is certainly truethat of all the institutions of society the political arethe most fundamental in the sense that it is they whichdesignate the holders of political power, which is tosay those within whose legitimate power it is to designnew institutions and modify existing ones. It is one oftheir functions to maintain rules for the politicalgame, from which the winners emerge as the power-holders. Such fundamental institutions, as we saw inthe second essay, must possess the attributes ofcoherence and permanence if the whole institutionaledifice is to rest securely on its foundations. But it isof the nature of democracy that such fundamentalinstitutions cannot permanently be entrusted to onegroup of persons. Free access to political power anccontinuous change of power-holders are of the essence*of democracy. The defence of these fundamental

94 ON POLITICAL INSTITUTIONS

institutions against the forces of change thus becomesa paramount problem since it is difficult, albeitperhaps not impossible, to devise rules to distinguishthose institutions which power-holders may changefrom those they may not. The ancient query 'quiscustodiet ipsos custodes?' thus assumes a new and morepoignant significance in this context.

Our second task, to relate our conclusions on thesematters to Max Weber's views, will prove to be evenharder, certainly harder than it has been thus farin this book. In our second essay, for example, wewere able to construct an analytical scheme of institu-tions and institutional structure from building blocksprovided by Weber's work, even though his workitself contained no such analytical scheme. Here, inthe political field, we face an altogether different situa-tion.

Weber held strong views on the political problemsof the Germany of his time and did not hesitate toexpress them with vigour, eloquence, and greatcourage. These views of his were supported by acoherent view of the political institutions of his con-temporary Germany, of what they were and of whatthey ought to have been. Behind this view, as weshould expect from so powerful a thinker, it is possibleto discern the nucleus of a general theory of politicswhich of course comprises a scheme of politicalinstitutions. But this theory is virtually embedded inpolitical writings of a strongly polemical nature.Naturally our task here would appear to be to distilthe theoretical content from the polemical materialand to try to give it a form consonant with our generalanalytical scheme. Unfortunately it is not as simpleas that. Two obstacles in particular present them-selves.


The first arises from Weber's famous insistence onthe separation between the sphere of science, in whichvalue judgements are inadmissible, and the sphere ofpolitics in which they are legitimate. His politicalwritings naturally abound in value judgements,usually of an adverse kind, on the political andeconomic elites of Wilhelminian Germany. Is it,then, a legitimate enterprise to draw what one hopeswill be tenable scientific generalizations from materialwhich its author had so clearly consigned to the non-scientific category? Even though sheer necessity maycompel us to embark upon this path we cannot do sowithout some misgivings.

Secondly, any attempt to discard the polemicalwrapping of his material and distil only the theoreticalcontent from it encounters the obstacle that a gooddeal of the material on which Weber drew for purposesof illustration and demonstration of his general thesislargely consists of facts seen in a certain perspective,which was Weber's perspective when he wrote butis not the only perspective in which these facts may beviewed. Moreover, in warning his readers of whatmight happen to a nation which lacked the politicalinstitutions appropriate to modern industrial society,he quite naturally assumed a knowledge of the socio-economic structure of his contemporary Germanywhich his original readers shared with him, butwhich non-German students of Weber half a centuryafter the original publication of these writings canhardly be expected to have. He was even entitled toexpect (and doubtless did) from the educated and well-to-do readers of the Frankfurter Zeitung to whom as arule he addressed himself in the first place, a degreeof sophistication which would permit them to discountsome of the rhetorical excrescences of his polemical

L.M.W.—4*


style, and nevertheless realize the very seriousnature of the issues at stake. All this no longerapplies to students of Weber's political work today,for whom Hohenzollern Germany belongs to pasthistory.

Since none the less some knowledge of the socialand political structure of Weber's Germany is in-dispensable for a proper appreciation of what he hadto say in his political writings, we propose to proceedin the following manner: In the first section of whatfollows we shall, very briefly, indicate the nature ofthe problems to which the need for the preservationof fundamental institutions typically gives rise in an'open' society of the modern kind. Next, we shall tryour hand at presenting a brief outline of the social andpolitical structure of Hohenzollern Germany, theindispensable frame of reference for Weber's politicalviews. It is to be hoped that to draw a reasonablyobjective picture of the Hohenzollern Empire fiftyyears after its downfall, with particular attention tothe constellation of social forces which sustained it,will not prove an altogether insuperable task. Thenin the third section we shall give an account ofWeber's own views on the political institutions of hiscontemporary Germany and the reforms they re-quired. This account will be followed, in the fourthsection, by a critical assessment of some of his viewsin terms of our own conclusions as well as by adiscussion of the effect which the revolution of 1918had on his views, as reflected in some of the work he didwhen he helped to draft the Weimar Constitution.In the final section we shall return to the main themeof this essay and endeavour to restate the problem ofthe political form of the institutional structure, con-sonant with and appropriate to an open society.


I

We encounter what we have come to regard as thecrucial problem of the institutional order, namely thepreservation of certain fundamental institutions in aworld of continuous change, in most spheres of sociallife and in many guises. In the political sphere anawareness of such a need is fairly widespread, to befound in most parts of the world and at almost alltimes. But so is awareness of the need for flexibility.How are the two to be reconciled? The distinctionbetween fundamental and secondary, immutable andmutable, institutions suggests itself as a way out, butit is a way out only on the high level of abstraction onwhich we have thus far dwelled. How is the distinctionto be drawn in concreto? And since it has to be drawnin practice by some individuals to whom this task hasbeen delegated, how are we to ensure that theiraction meets the needs of society as a whole ?

In most modern societies the preservation of funda-mental institutions usually takes the form of theirconstitutional 'entrenchment'. A constitution is abody of legal norms which cannot, as a rule, bechanged as easily as can ordinary norms. Certaininstitutions can thus no longer be changed by ordinarylegislation. But at the same time most constitutionsprovide for legal procedures which make constitutionalchange possible. A parliamentary majority of two-thirds, for example, or a plebiscite, may be required.Thus the protection of fundamental institutions is onlyrelative, not absolute.

Is there, then, no absolute protection againstconstitutional change? There certainly is againstunconstitutional change. No society of course will


permit its fundamental institutions, those whichdesignate the holders of legitimate power, to bethreatened by the use of force. Every society expectsits power-holders to marshal all the forces at theirdisposal for defending the existing social order againstviolent overthrow. Typically, in such cases of 'clearand present danger' to the existing constitutionalorder, all constitutions permit the temporary sus-pension of certain of their norms in order to defendthe existing order as a whole, be it by the declarationof a 'state of emergency' or in other ways. This, by theway, provides us with an interesting object-lesson inhow to distinguish between fundamental and lessfundamental institutions: the latter are those whichmay be suspended, in certain circumstances, if theformer are in danger. Fundamental institutions in oursense are those which must be defended at all costs,even at the temporary (for the period of danger)sacrifice of certain others also embodied in the con-stitution. Every constitution, it might be said, has tobe construed as containing an unwritten clause whichenjoins those who hold power under it to defend 'lawand order' at all costs when these are in jeopardy.

But all this amounts to is that certain channels ofchange are outlawed, and that certain temporarychanges may be necessary to safeguard the socialorder as a whole against change of a certain kind, e.g.change by force. It follows from the nature of theinstitutional order of an open society that, while theremay be relative protection for certain fundamentalinstitutions by making it difficult to change them,and even absolute protection against change of acertain kind, there can be no absolute protectionagainst all kinds of change. Some channels of changemust always remain open, however narrow they may be.


This may mean in practice that what cannot beachieved by illegal means, by force, may becomeattainable by legal means, by making use of thelegitimate channels of change. It is the great lessonwhich Hitler appears to have learnt from the failureof his Munich coup on 9 November 1923. This ofcourse is an extreme example. In reality any attemptto modify fundamental institutions by ostensibly legalmeans will at once encounter many obstacles, 'builtinto' the existing order or emanating from the pre-vailing climate of opinion. No mature society islikely to tolerate the destruction of its democracticinstitutions by the misuse of the very procedures whichthese institutions authorize. All the same, our exampledoes show the existence of an open problem.

But even apart from the case just mentioned which,on a level higher than that with which we are hereconcerned, raises the very important issue of legalityversus legitimacy, there remains the open question ofhow to safeguard fundamental institutions against theforces of slow erosion rather than rapid destruction,forces bound to be unleashed by the struggle forposition and power in society. The relative safeguardsmentioned above may render each such changedifficult by requiring a strong coalition of interests inorder to bring it about, a coalition such as may behard to bring together and even harder to maintainin a world where change is rapid and groups havemany interests to pursue, but they cannot render itimpossible. It has not been unknown for politicalparties of impeccable democratic principles to change,when in power, the age of franchise or the delimita-tion of constituencies in such a fashion as to entrenchtheir power. To choose an example from anotherfield, which forms nevertheless an equally fundamental


part of the institutional order, no parliament is likelyto pass an act invalidating all money debts and per-mitting debtors to enrich themselves at the expenseof their creditors. Nevertheless a continuous process ofinflation which lasts half a century will produce moreor less the same result.

II

The German Empire that came into existence in 1871did so, like most other political creations, as a result ofa series of compromises: between North and South,between federalism and unitarism, between Prussiaand the non-Prussians, between cultural pluralism andthe need for a strong political centre. But from thepoint of view which must be of primary interest tous here the outstanding feature of this complex ofcompromises was the fact that the Empire rested on acompromise between the Prussian state and bourgeoissociety. We shall call this the Grand Compromise.The Prussian state, the bureaucratic organizationwhich the Hohenzollern princes had created in orderto give administrative, military, and judicial unity totheir originally rather diffuse dominions, was ahierarchic structure with the King and his ministersat the top. The social substructure on the basis ofwhich it grew up to become a power in Europe hadbeen an agrarian society in which the East Elbiannobility held unquestioned elite status and in whicha profusion of fairly small market-towns were thecentres of trade and commerce. Naturally the nobilityprovided the state with its administrators and with itsofficer corps. But in the decades after the Congress ofVienna economic progress accelerated and in the

ON POLITICAL INSTITUTIONS IOI

newly acquired Rhineland, then Germany's mostprosperous and economically developed part, thePrussian administrators had to contend with the firstproblems of an industrial society, problems with whichthey were in no way equipped to deal.

By 1870 Germany had become an industrialcountry, at least by the standards of the time. Thecompromise mentioned above in its most significantaspect constituted a recognition of this fact. Thesubstructure of the new Empire was no longer anagrarian society, and the new elite which had broughtabout the change had to be given a place in the newpolitical structure.1 For the Prussian state, thedominant force in the new Empire, this meant thetransition from absolutism to constitutional monarchy.The Prussian monarchy had already (after 1848) givenup its right to unlimited legislation by Royal decreeand transferred the power of legislation to a two-chamber parliament. In the new imperial constitutionthis process was taken a step further, in fact this hadalready been done with the setting up of the NorthGerman Confederation in 1867.

For the liberal bourgeoisie, on the other hand, anda fortiori for its elite which rode on the crest ofeconomic success, acceptance of the compromiseentailed an intellectual reorientation which, as sub-sequent events were to show, was by no meanspainless. In 1848 and for two decades subsequent toit these people had been bitterly hostile to the Prussianmonarchy and to Bismarck, its 'strong man5. Theirsource of inspiration had been British, if not French,ideals of parliamentary government. But the revolu-

1 On this whole problem see John R. Gillis, 'Aristocracy andBureaucracy in Nineteenthrcentury Prussia5, Past and Present(December 1968).


tion of 1848 had failed, and after three victoriouswars the Prussian monarchy was stronger than ever.Not unnaturally, a compromise with the power-holders whose recent acts had tarnished liberalideals was at first widely resented, and accepting ashare of political responsibility in such conditions wasdistasteful to many. The trauma of the reorientationwhich acceptance of the compromise entailed hadmany serious consequences. One of them was thedisappearance of serious political thought from theGerman intellectual scene after the 1870s. Thepeculiar resentments which form the emotionalbackground of Max Weber's political thinking wereanother.

We must now cast a brief glance at the constitutionalstructure which rested on the foundation of this GrandCompromise. The legislative body consisted of twohouses, the Federal Council (Bundesrat) and theReichstag, the latter elected by universal equal adultmale franchise. But while there was a parliament,there was no parliamentary government. This fact,as we shall see, offered the main target for Weber'scritical attack. The Reich Chancellor and his Secre-taries of State were appointed by the Emperor and,while they were responsible to the Reichstag, did notrequire its confidence. No vote of censure couldremove them from power. In fact, an article of theConstitution which stipulated that nobody couldbelong to both Houses at the same time virtuallyprevented parliamentary government, since by tradi-tion the Reich Chancellor was also the Prime Ministerof Prussia and hence had to be in the Bundesrat.

In reality of course the need for a majority in theReichstag to pass legislation, in particular the annualbudget, entailed a situation not altogether dissimilar

ON POLITICAL INSTITUTIONS IO3

from one of parliamentary government with a coalitionof changing composition. Without such a majority nogovernment could govern. But it was possible to feel,as many did before Weber, that the permanent co-ordination between government and parliamentarymajority which is of the essence of parliamentarydemocracy would have offered a firmer and morestable basis for German politics than did the shiftingand usually short-lived arrangements to which mostChancellors of Wilhelminian Germany had to resort.

We now must take a closer look at what the com-promise meant in those terms that are of primaryimportance to us, that is in terms of fundamentalinstitutions and the institutional order. First of all, wehave to remember here that in the context of theGrand Compromise the market economy and theinstitutional framework appropriate to it were verymuch taken for granted. The Prussian civil servants,to whose mentality the compromise owed so much,never doubted that economic progress required abroad sphere of contractual freedom and the abilityof the economically active to shape institutions (not,to be sure, fundamental institutions but those of asecondary order) in accordance with their naturallyflexible needs. In fact, ever since the Stein-Hardenbergreforms of 1810 Prussian policy had pursued the pathof economic freedom. Before the return to protectionin 1878 the new Empire was even a free-trade country.A general presumption in favour of laissez faire ineconomic matters did thus not even have to form anexplicit part of the Grand Compromise we arestudying. It was simply taken for granted by allparties to the agreement.

As regards fundamental institutions the matter wasdifferent. From the point of view which is of interest

104 T H E LEGACY OF MAX WEBER

to us the transformation of the absolute state into aconstitutional monarchy meant that the fundamentalinstitutions were now more secure than they werebefore: the state had abdicated its right to interferewith the legal and institutional order at will. It istrue of course that the transfer of legislative powerfrom the Prussian king and his council of ministersto the Reichstag, a body elected by universal franchise,would by itself do little to help entrench fundamentalinstitutions. But the Reichstag could not legislatewithout the Bundesrat, and in this latter bodyPrussia had a right of veto on all matters concerningdefence, indirect taxes, and customs tariffs. Funda-mental institutions were thus fairly well protected.The separation of powers inherent in the GrandCompromise made it more difficult to change them.

The German word which was generally used todenote the compromise we have tried to describe, thereconciliation between the Prussian state and bour-geois society, was Rechtsstaat, which is also the Germanword for 'rule of law'. Under the new dispensation thestate in effect, if not in words, promised to devote allits power to upholding the legal order and thefundamental institutions on which it rested, and toabstain from arbitrary interference with it.1 To besure, the legal order could be changed by legislation,but the constitutional arrangements appeared to besuch that no fundamental changes could be broughtabout without the consent of all groups affected bythem.

Rechtsstaat, briefly then, meant that the main purpose1 For an authoritative, if somewhat stylized, exposition of this

notion see Rudolf Gneist, Der Rechtsstaat (Berlin, 1872). Weberrarely used the word, and never in the sense it has in our text.Bismarck appears to have disliked it.


of the state was now to uphold the legal order, andthat all actions of those who held power under it hadto be ultimately justified in terms of this purpose.This was a far cry from the mercantilism and pater-nalism of Frederician Prussia. But not until the 1880swas the ideal of the Rechtsstaat seriously challenged. Itwas only then that, under the leadership of theeconomists of the Verein fur Sozialpolitik, educatedGermany gradually began to turn to the oppositeideal of the Welfare State.

The main defect of this Grand Compromise provedto be the fact that the industrial working class had nopart in it. As Germany became an industrial countryand the working class rapidly increased in numbers,this weakness made itself more and more strongly felt.It was this fact, more than any other single fact, whichcaused the decline of the Rechtsstaat ideal in Germanyand the rise of various ideologies extolling the WelfareState. It seemed to many educated Germans thatwhat was then known as the 'labour question' couldnot be solved within the existing framework butrequired state intervention in favour of the workers.

To bring trade unions within the existing legalframework was still tolerably easy. Collective bargain-ing, however, did not exist in Germany until theFirst World War, when it was started with the officialblessing of the Ministry of Supply—the Kriegsamt. Butthe Social Democratic Party, the chief protagonist inthe political field of working-class aspirations, pre-sented altogether new problems. In the official view,with its pretension to 'overthrow capitalistic society',its intransigence, and its internationalism, the partywas simply 'unassimilable'. It just did not fit into theexisting political structure. Since the situation of theGerman socialist party under the Hohenzollern


Empire has been described by Schumpeter in whatmust be one of the most perceptive historical sketcheswritten by this brilliant author, we need here do nomore than refer the reader to it.1

In concluding this brief bird's eye picture of thepolitical landscape of Hohenzollern Germany we haveto draw attention to one of its outstanding features,which is today often not well understood, but whichis of crucial significance if we are to properly assessthe purport of Weber's polemics. It is undeniable thatthe power and influence of the East Elbian nobility(the Junkers) was out of all proportion to theireconomic and social significance. The Germaneconomy would have been better off without theRittergut. In this sense Max Weber's main criticism ofthe German political system, namely that its leadershipfailed to reflect the true balance of forces within thenation, was perfectly valid. Nevertheless, this politicalelite, small in numbers, weak in its economic base,enjoyed a vast amount of tacit support within thenation and, what is more, could have commanded andmade use of this support in a crisis. This support wasforthcoming from people of all social classes, fromregions far beyond the borders of the Kingdom ofPrussia and from all religious communities. Therewere to be found, between the Moselle and theVistula, many men, big and small, in the mostunexpected quarters, who would have stood by thePrussian elite if it had been threatened. It is of coursethe hallmark of a true political elite that in a crisis,on the ultimate testing-ground, it is able to drawsupport from all layers of society and is not confined

1 Joseph A. Schumpeter, Capitalism, Socialism, and Democracy(Allen and Unwin, 1942), ch. XXVI, V: 'The German Partyand Revisionism'.


to a narrow social base. In a serious conflict such anelite is much stronger than its rivals enjoying a merehomogeneous following, as many parties of the lefthave discovered to their dismay.

The fact just stated was in Hohenzollern Germanya fact of the highest political significance. The SocialDemocratic leaders knew it well. The striking dis-crepancy between their ability to command millionsof votes at elections and their inability to affect anypolitical decisions is largely explained by it. So did theshrewd trade-union leaders who firmly refused to lettheir organizations be used for political ends. Weberknew it, too, and so of course did his original readers.The fact kindled his wrath, and the bitterness of hispolemical tone owes not a little to his knowledge of it.Only defeat in the First World War destroyed thewidespread support the political elite had thus farenjoyed.

Some of the consequences of the prestige thePrussian elite had enjoyed even outlasted WilhelminianGermany and clouded the political life of the WeimarRepublic. Many of the people whose attitude wehave attempted to describe felt a sense of shock anddeprivation when the Hohenzollern Empire fell. Theirlasting resentment provided a most valuable treasureto the enemies of the Republic.

We shall now make an attempt to test the efficacy ofthe few conceptual tools we have found useful in theanalysis of the political structure of imperial Germanyby applying them to the Weimar Republic, itssuccessor.

The Weimar Republic, too, rested on a complex ofcompromises. Politically it rested on a compromisebetween the trade unions, the Roman CatholicChurch, and the liberal intelligentsia, that is, between


precisely those forces which under the Empire hadbeen remote from the seats of power. The politicalparties which came to be known as the 'WeimarCoalition5 (Social Democrats, Centre Party, Demo-crats) reflected precisely these social forces.

Our survey will be brief, even briefer than it was inthe case of the Empire.1 A compromise of the kindindicated naturally gave rise to many difficultproblems which we cannot deal with here. Thestrength of the Republic lay in the simple fact that thesocial forces which now shared political power un-doubtedly were the strongest forces in German society.The new leaders went about their tasks in a workman-like fashion. They successfully overcame the difficultproblems of post-war adjustment and, if very late,inflation. They might have coped with the greatdepression and mass unemployment had they under-stood the economic implications of what we can nowsee was one of the great social changes of the 1920s:the emergence of downward rigidity of the wage-level.

Outside the economic sphere there were of coursemany problems. Civil servants and soldiers, havinglost the natural apex of their hierarchies, had toundergo a process of intellectual reorientation whichto many of them was no less painful than had been thecorresponding process to the liberal bourgeoisie of the1870s. But since, in the absence of any serious pre-tender to the throne, the possibility of a return to themonarchy became more and more remote as timewent by, the problem would have solved itself.Lastly, it is permissible to point out that the presentBonn Republic is in all important respects virtually aunilineal continuation of the Weimar Republic. Its

1 Max Weber died in June 1920, in the second year of theRepublic.

ON POLITICAL INSTITUTIONS

intellectual atmosphere and the composition of thesocial forces supporting its political structure areidentical with those of Weimar. The mere fact thatafter the holocaust of Nazi rule and the Second WorldWar the new German state was again erected on thesame foundations as the Weimar Republic had been,because there were no others, seems to us to attest theinherent strength of the social forces underlying both.

Where, then, lay the weakness of the WeimarRepublic? It so happens that the critical source of itsweakness lay in precisely that sphere which is ofparticular interest to us: its fundamental institutionsrested on no firm basis. The compromise was re-garded by too many of the participants as a temporaryrather than a permanent one, not as a Grand Com-promise but rather as a petit compromis.

To understand this we have to understand thepeculiar dilemma the socialist leaders confronted.When the Empire collapsed they suddenly had toshoulder a responsibility for which they were in noway prepared. They had grown up in an atmosphereof Marxism, after 1900 increasingly diluted, it is true,by generous doses of Revisionism. They had assumed,in accordance with Marxist teaching, that thetriumph of Socialism would follow the collapse ofcapitalism, and nobody doubted until 1914 that thisday was a long way off. In November 1918 theysuddenly found themselves in the position that, as aresult of a culmination of events in which they hadlittle part, namely Germany's defeat in the war, theyhad political power, but that they had to govern acapitalistic society. Even worse, they had to govern itin coalition with bourgeois parties, with people forwhose 'petty bourgeois5 outlook they had never had,or so at least they had to pretend outside the


Reichstag, anything but contempt. There was nothingin the book of Marxian teaching to give guidance forsuch a situation.

It speaks for the character of the socialist leadersthat they set about the practical task of reconstructionwithout flinching and with a fair measure of success.But, as often happens in history, it proved actuallyeasier to cope with the new reality than to adjust oldideas to it.

In coming to terms with the reality of the 1920sGerman socialists adopted an ideology which restedon a clear distinction between the dubious capitalisticpresent and the glorious socialist future, and whichcame to regard the Weimar Republic merely as astage of transition to the latter. But how exactly thetransition would come about was a question left open.Nobody doubted that in the short run the SocialDemocratic Party's main task was to promote thewelfare of the working class and that, in the cir-cumstances given, this aim required the promotion ofvigorous growth of capitalistic industry. But how soc-ialism was to be promoted in the long run was less clear.

In the prevailing atmosphere of uncertainty aboutthe path to the future, socialist thinkers contrived tocling to one article of Marxist faith: Capitalism isever-changing and all such change is to the goodsince it brings the dawn of socialism nearer. Economicand social change must therefore not be resisted. Wemay not at once understand exactly how it will worktowards socialism, but we are entitled to presume it.Only he who defends outmoded ways of thought andobsolete institutions, the 'reactionary', is the enemy.An 'open horizon' thus came to occupy the centre ofthe stage in the mentality which received officialsanction.

ON POLITICAL INSTITUTIONS III

This ideology of the 'open horizon' had certainimmediate effects. Naturally it filled the otherpartners of the Weimar Coalition with suspicion andweakened them. What matters from our point of viewis that it prevented any serious discussion of con-stitutional questions.1 To these 'advanced' thinkersquestions about the strength of the foundations of theRepublic were questions of little interest. Why botherto strengthen these foundations if every change was,somehow, a step towards socialism? In the WeimarRepublic, to show an interest in questions aboutfundamental institutions stamped one unmistakablyas a 'reactionary', a thinker in outmoded static ways.

The most disastrous consequences of this type of'advanced' thought only became apparent withHitler's rise to power. The Nazi ideologists found theprevailing climate of advanced opinion ideally suitedto their purposes. Whoever opposed them was ofcourse a 'reactionary'. Whoever accused them ofsubverting the institutions of the Weimar Constitutionunder whose protection they were able to rise topower 'resisted social change'. It can hardly surpriseus to learn that, when the new Caesar appeared onthe 'open horizon', many people who for a decade

1 Proportional Representation offers a good example. Whenit was introduced in 1919, Weber opposed it. Its abolition andreturn to single-member constituencies would actually havebeen to the benefit of the Social Democratic Party and of coursewould have made the initial success of the Nazi Party moredifficult. But the type of mind mentioned above spurned suchmundane considerations. The efforts of some of the moreperspicacious among the younger members of the party hier-archy, such as Mierendorff, were in vain. See J. Schauff (ed.),Neues Wahlrecht (Berlin, 1929). The reform of the federal structurewas another urgent problem which for similar reasons wasneglected.


and a half had been exhorted by advanced thinkersnot to regard institutions as 'ends in themselves' andnot, from bourgeois prejudice, to resist necessarysocial change, behaved at the moment of crisis inexact conformity with these precepts. Who was toknow, in 1933, that the Fiihrer did not point the wayto the future ?

I l l

We now turn to the task of trying to extract Weber'spolitical sociology from his political writings. For thispurpose the writings of his last years are the mostrewarding, partly because they show his thought atits most mature, but also because in commenting onthe dramatic changes which occurred in Germanybetween 1917 and 1920 he felt it necessary to show thathis political views were based on a solid analysis of thesocial and economic situation. We shall draw mostlyupon a series of articles published originally in theFrankfurter £eitung in the summer of 1917 and, slightlyrevised, as a pamphlet in May 19181 because itcontains a long discussion of the political institutions,and their relationships to one another, that areappropriate to modern industrial society. But we shallof course also have to take account of his proposalsfor the Weimar Constitution and the arguments bywhich he supported them.

Weber grew up in an atmosphere of that moderateliberalism which reflected the Grand Compromisedescribed above. His father, a high municipal officialin Berlin, was also a National Liberal member of thePrussian Landtag. He was thus 'born into polities',

1 Tarlament und Regierung im neugeordneten Deutschland',Gesammelte Politische Schriften (Second Edition, 1958), pp. 294-431.(Translation of all passages quoted is ours.)

ON POLITICAL INSTITUTIONS I I3

and politics remained one of his major intereststhroughout his life.1 In his political views he soonmoved away from his paternal moorings, at first tothe right, but on recovering from his illness after 1902increasingly towards the left. He regarded socialismas Utopian, at least in his time, but strongly felt theneed for giving the working class a share in politicalpower. His sympathies were with the (left liberal)Progressive Party.

Two Reichstag members of that party, Haussmannand Naumann, were his close friends. The oftenirresponsible pronouncements of the Kaiser on mattersof foreign and internal policy filled him with gloomand dismay. He felt that only a constitutional reformwhich made an end to the 'personal regime' of themonarch could save Germany from disaster.

For the first two and a half years of the First WorldWar an uneasy political truce prevailed in Germany.But in 1917 the parties in the Reichstag becamerestive and began to urge constitutional reforms in thedirection of parliamentary government. Weber ap-parently timed his series of articles in the summer of1917 to support these moves.2

1 Our account of Weber's political activity follows in generalWolfgang J. Mommsen, Max Weber und die deutsche Politik,i8go-ig2o (J. G. B. Mohr, 1959).

This does not mean that we accept Professor Mommsen'sinterpretation of Weber's political thought in all respects. Inour view he is a little too strongly inclined to see in Weber atypical educated German of the Wilhelminian era. Max Weber,to be sure, was a Wilhelminian German. He also was a Europeanpolitical thinker in the great tradition of Machiavelli andHobbes.

2 For an excellent account of the internal political situationin Germany at that time see Gerald D. Feldman, Army, Industryand Labor in Germany, igi4~igi8 (Princeton University Press1966).

114 O N POLITICAL INSTITUTIONS

In advocating parliamentary government for Ger-many Weber was not moved by traditional democraticideals. 'Sovereignty of the people', for example,meant nothing to him. As a positivist he rejected andscorned all arguments derived from a body of ideasultimately founded on a belief in natural law. For himthe creation or reform of political institutions wasentirely a matter of expediency. The only criterionacceptable to him was: Did a certain institution helpor hinder a given nation in the struggle for power andsurvival? For Weber was what we might call an'agonistic5 thinker for whom the continuous strugglebetween human groups (not necessarily only nation-states) constituted the ultimate reality of all politicalexistence.

He based the case for parliamentary government inGermany on his belief that this system provided anideal mechanism of selection of political leaders and anecessary mechanism for the control of bureaucracy.The first claim was, in his view, adequately supportedby British experience. To appreciate the second wehave to understand his attitude towards bureaucracy.

For Weber the rise of bureaucratic organization,not merely in state administration but also in businessand in fact in all sections of society, was an outstandingcharacteristic of modern society. Its ineluctable naturestems from its efficiency. It creates uniformity andpredictability in large-scale societies because the actsof thousands of officials are all oriented to identicalnorms. It increases efficiency owing to the division offunctions it makes possible. But as an institution itrequires other complementary institutions to supportit. It must be part of an institutional order if it is tofunction well.

On the one hand, in an ever-changing and uncertain

ON POLITICAL INSTITUTIONS I I5

world certain decisions have to be made for which therules of the bureaucratic hierarchy can provide noguidance. They have to be made by an altogetherdifferent type of man, a 'leader5. The training whichthe members of the upper strata of the bureaucratichierarchy receive is likely to impede rather than tohelp them in the exercise of the decision-makingfunction. Yet, the political leaders of HohenzollernGermany were in fact at best successful administrators.Weber held that the successful functioning of thebureaucratic hierarchy requires that the men at thetop are recruited from outside the hierarchy. Theideal training-ground for such men was the 'politicalstruggle' in parliament and in election battles.

On the other hand, society has to be protectedagainst the abuse of bureaucratic power. Weber, whowas doubtful about the power of a bureaucracy tocheck such abuses from within the system, believedthat only control by independent politicians en-trenched in a parliament with strong prerogatives, inparticular the right to set up committees, wouldachieve this end.

It may be tempting to see in these views of Weber'sthe influence of the old idea of 'checks and balances'within the political system, the reflection of a desire tocheck the overwhelming power of the 'state' by the'countervailing power' of a parliamentary forceemanating from the body of 'society'. But such aninterpretation would be quite wrong. Weber did notthink in these static terms. He wanted a strong, not aweak state, but a state which would be more effectiveas an instrument of the German nation than theexisting state was. Liberal preoccupation with theseparation of powers and the equilibrium of politicalforces was far from his mind.

I l 6 ON POLITICAL INSTITUTIONS

Like Veblen, Weber regarded the political structureof Hohenzollern Germany as a 'feudal relic5. Themain task facing German political thinkers was tosubstitute for it something better suited to the needsof an industrial country in the twentieth century.Weber had no doubt that parliamentary governmentwas the answer.

He made no attempt to hide the low esteem in whichhe held most of the Reichstag leaders, in particular theSocial Democrats and those of the Centre Party (thetwo strongest parties without which parliamentarygovernment would hardly have been possible).1 Hetherefore faced the difficulty of having to explain whyafter forty-five years of its existence the parliamentarymechanism of selection of leaders had produced nobetter results. His explanation was historical: Bismarckhad reduced the Reichstag to political impotence bykeeping all decision-making power in his own hands.The Reichstag of the 1870s had still contained arespectable number of real leaders, the Reichstag of1912 after forty years of political impotence hadvirtually none. By fashioning German political in-stitutions to serve as tools in his own autocratic handsBismarck had deprived the system of its capacity to pro-duce leaders who might have been his worthy successors.

He left a nation without any and every political education,far below the level which, in this regard, it had reachedtwenty years earlier. And, what is more, a nation withoutany political will, accustomed to seeing the great states-man at its top looking after its politics . . . But a politicaltradition the great statesman did not leave at all. In-

1 On the German parliamentary scene before 1914 see KlausEpstein, Matthias Erzberger and the dilemma of German Democracy(Princeton University Press, 1959).


dependent minds, let alone men of character, he neitherattracted nor even tolerated.1

Weber had to contend with another problem here.In arguing that bureaucratic organizations extendedinto all corners of modern society he had to admitthat the same applied to modern political parties.They, too, are being dominated by their respective'machines'.

As regards their internal structure, all parties in the courseof the last few decades, with increasing rationalization ofthe technique of electioneering, went over to bureau-cratic organization. The stages of this development whichthe various parties have reached may differ; the generaldirection of the road is, at least in mass states, un-ambiguous, (p. 316)

In fact, the description of the process by whichnineteenth-century parties, led by notables, haveevolved into the modern mass parties with theirstaff of permanent party officials (Parteibeamten) is oneof Weber's most original contributions to politicalsociology, far transcending in its importance theephemeral political purpose it originally served. Weberalso fully realized the changes in the significance of thevarious parliamentary activities which have ac-companied this metamorphosis of the party structure:parliamentary debate means very little, and the reallyimportant work is done in the committees. 'Speecheswhich a deputy makes are today no longer personalstatements, even less attempts to persuade the op-ponents. They are official party declarations, addressedto the country 'through the window5 (pp. 332-3).

What, then, is the position of the political leadervis-d-vis the formidable power of the party hierarchy?

1 Gesammelte Politische Schriften, p . 307.

I l8 ON POLITICAL INSTITUTIONS

Will he not have to succumb to it? In reply to suchquestions Weber pointed to one of the limitations ofbureaucratic organization with which we are al-ready familiar: its inability to make decisions, whichhave to be made 'at the top', i.e. outside it. He alsostressed that, if the party wants to win elections andgain the spoils of power, the party bureaucracy needsthe leader who can win votes. It is this more thananything else, the democratic leader's appeal to theelectorate, which makes him indispensable to his party,safeguards his power against the party bureaucracy,and distinguishes him from the successful head of anadministrative bureaucracy, such as were the states-men of Wilhelminian Germany.

To Weber the successful democratic leader wasalways something of a Caesar.1 He therefore re-pudiated the traditional case for democracy.

It is not the many-headed assembly of parliament assuch that can 'govern' and 'make' a policy. Nowhere in theworld is there any question of that, not even in England.The whole broad mass of deputies function only as afollowing for the 'leader', or the few leaders who formthe Cabinet, and obey them blindly as long as they aresuccessful. This is as it ought to be. It is always the 'principleof the small number5, i.e. the superior capacity for politicalmanoeuvre of the smaller leading groups, which dominates

1 It is a question of some interest which actual historicalfigures Weber had in mind when he drew his picture of thedemocratic leader. We have here of course a composite portraitof the modern British statesman as seen through Weber's eyes,in which it is possible to discern certain traits of PalmerstonGladstone and Disraeli.

There can also be little doubt that Weber, writing in 1917,was much impressed by the vigorous leadership of Lloyd George,but, in addressing the German public in wartime, thought itwiser not to mention the enemy leader by name.

ON POLITICAL INSTITUTIONS II9

political action. This 'Caesaristic' ingredient is (in massstates) indestructible, (p. 336)x

The November revolution and the need for a newGerman constitution afforded Max Weber the op-portunity for once to take an active part in the shapingof political institutions for the new German republic.Hugo Preuss, an eminent constitutional lawyer whomthe provisional government had made Secretary ofthe Interior, set up a committee to produce a draft of anew constitution. Weber was a member of it, actuallyits only non-official member. The committee met inBerlin from 9 to 12 December 1918. For Weber as apolitical reformer this was, we may say, his greatesthour: for once he was able to exercise direct influenceon constitutional development. As Professor Mommsenhas rightly stressed, this committee, though it lackedofficial status and its deliberations were confidential,was in reality the 'birth chamber' of the Weimar Con-stitution. Almost all its fundamental features origin-ated here, and Weber took a most vigorous andsuccessful part in the work of the committee.

But for him the hour of triumph was tinged bytragedy, and it was to be brief. When Weber went toBerlin for the committee meetings in early December,he was firmly convinced that he would be elected tothe National Assembly the following month, and couldreasonably hope in this case to be a member of itsConstitution Committee. No doubt he regarded the

1 The similarity of Weber's ideas to those of Mosca and Paretois here striking indeed. Yet, to the best of our knowledge, Weberdid not know their writings. Pareto's main work in this field, theTrattato di Sociologia Generate, was published in Florence duringthe war. Weber can hardly have seen it. But Pareto had espousedsimilar ideas since 1900. It is possible that Weber had heard ofthem through Michels who until 1914 was a close friend of his.

L.M.W.—5


committee work in Berlin as only the first step in hiswork for the constitution. These hopes were shattered.The delegates of the Democratic Party who drew upthe list of candidates for the Hesse-Nassau con-stituency put Weber's name on the list, but his placein the order of candidates was a hopeless one. Sothese few days in December 1918 remained his onlyperiod of active political work.

In the deliberations of the committee Weber soughtto strengthen the position of the President as againstthat of the Reichstag. He persuaded his colleagues tohave the President elected by popular vote. He evenwanted to give him a part in legislation by enablinghim to put bills the Reichstag had rejected to areferendum of the electorate. He pointed out thatparliaments had lost a good deal of credit in themodern world (of 1918!). Today, he said, one couldscarcely hand over all power to a parliament withoutmisgivings. A countervailing power had becomenecessary.

Two months later, in an article published inBerliner Bbrsenzeitung on 25 February 1919, he waseven more outspoken.1

Only the election of a Reich President by the peopleaffords opportunity and cause for a selection of leadersand thus for a new organization of the parties. . . . Theelection has shown that everywhere the old professionalpoliticians succeed, contrary to the wish of the mass ofelectors, in eliminating the men who are enjoying theconfidence of the latter in favour of political stock figures(Politischer Ladenhiiter). The result has been that just thebest minds have turned away from all politics.

Weber then pointed out that proportional representa-1 Gesammelte Politische Schriften, pp. 486-9.


tion for the Reichstag was bound to strengthen thistrend.

Parliament will thus become a body in which those setthe tone to whom national interests mean nothing, butwho essentially are carrying a mandate for economicinterests: a parliament of narrow minds (ein Banausen-parlament) incapable of providing a field of selection ofpolitical leaders in any sense. This much must be said foronce plainly and openly (p. 487).

We need not doubt that such a statement was tosome extent prompted by anger at the discomfiturehe had so recently suffered at the hands of theDemocratic Party delegates of Hesse-Nassau. Even so,can we really believe that no such dark thought hadever crossed his mind before then?

IV

At the beginning of this essay we pointed out that oneof the obstacles which an attempt to distil theoreticalcontent from Weber's political writings encounters,lies in the nature of the historical material he used. Todisregard polemics and concentrate on salient pointsof the argument is rather difficult where the facts onwhich the argument rests admit of more than oneinterpretation and the polemical purpose is alreadyinherent in the perspective in which these facts areviewed. In making a number of critical comments onWeber's thought it thus seems to us legitimate tostart with a few matters of fact which, in the per-spective of the half century which has elapsed since,look different to us from what they did to Weber.

We now know, for example, that after the DailyTelegraph interview of 1908 Wilhelm II on the whole


refrained from interfering with political decisions ofhis government and kept strictly within his con-stitutional role. There seems little reason to doubtthat Weber's strong personal hatred of Wilhelmplayed him false and that his conception of theKaiser's part in events was largely a misconception.

Weber owed his immense success as a politicalcritic in 1918 to the events of the time. The regimethe defects of which he had so mercilessly laid barelost the war and collapsed. Naturally he acquiredsomething of the reputation of a prophet. But lookingat it now, after half a century, it seems permissible tobe less impressed by this concomitance of thought andevents than were his contemporaries. If parliamentarygovernment had existed in Germany in 1914, is theremuch reason to believe that events might have takena different turn? From what we now know aboutwhat happened in the various capitals of Europe inJuly 1914, this seems rather doubtful. Nor do we seemuch reason to believe that, once the war had brokenout, a democratic German government, given thestate of public opinion and Ludendorff 's prestige aslong as he was successful, could have acted verydifferently from that of Bethmann-Hollweg. Soldiersin countries with parliamentary government have notfound it impossible to influence decisions.

On the other hand, we would not make too muchof the discrepancy between Weber's strong plea forparliamentary democracy in 1917 and his proposalsfor the Weimar Constitution, which actually aimed atstrengthening the power of the President at theexpense of the Reichstag. After all, with the fall of themonarchy the whole existing power-structure hadvanished. While in 1917 Weber's main aim was towrest some power from the existing power-centre and


have it transferred to the party leaders, one year laterhis main preoccupation had become the idea thatthere might be no power-centre left outside theReichstag.1

The changed circumstances here fully account forhis change of view. It is of course noteworthy that sosoon after his original pronouncements about theunique properties of parliaments as training groundsfor political leaders he found it necessary to look for asecond string. It certainly indicates that even in 1917his trust in the Reichstag may have been weaker thanhe permitted his readers to infer.

There can of course be no question whatever ofblaming Weber, who died in 1920, for the collapse ofthe Weimar Republic in 1933. We should not evenmention this point were it not for the fact that Germancritics of Weber have of late presumed to see in the'great demagogue', the 'caesaristic' leader of thequotation we gave above, a prototype of Hitler.

This is of course a complete misconception. Weber's'demagogue' is playing the game within the demo-cratic set of rules the continued existence of whichour author took for granted. The most that might besaid in this context is that certain weaknesses in thestructure of the institutions which supported theWeimar Republic bear some similarity to certainweaknesses in Weber's thought.

1 In the Borsenzeitung article quoted above Weber actuallysaid as much. 'Formerly, under the old regime {im Obrigkeitsstaat)one had to work for the enhanced power of the parliamentarymajority so that the importance, and thus the level, of parliamentmight be raised. Today the position is that all constitutionaldrafts reflect a blind faith in the infallibility and omnipotence ofthe majority, not of the people, but of the parliamentarians: theopposite and equally undemocratic extreme.' Gesammelte PolitischeSchriften, p. 488.


We must now turn to matters which concern usmore closely. We noted in our second essay thatWeber has 'no theory of institutions', though he dis-cusses institutions of many kinds. We, on the otherhand, attempted to show that Weber's view of humanaction as being free within the bounds of certainconstraints, oriented towards purposes, means, andobstacles, actually presupposes such a conception ofthe institutional order as will fit into a comprehensiveanalytical scheme. We now have to examine what theabsence of such a general scheme in Weber's workentails for the problems of the political sphere. Weberwas of course well aware of the fact that within agiven political order each institution must have aclearly defined function and that a certain degree ofcoherence must exist between them if they are tofunction properly. It goes without saying that suchawareness on the part of a political thinker does notwarrant the label of 'functionalism'. The functionalcoherence of the institutions which form the politicalorder is one thing, a structural-functional view of thesocial world as a whole (seeing it as a 'social system')is quite another.

Nevertheless, we find Weber here in the curiousposition that he, an opponent of functionalism, whenhe has to apply functional analysis to the relationshipsbetween institutions of the political and economicspheres, is making use of a particularly crude functionalmodel. He demands a high degree of uniformity of itscomponent elements. His notion of such an order iscouched in terms of homogeneity, of certain commonproperties shared by all institutions of a society. It ispossible to see in this notion a legacy of the HistoricalSchool, of the intellectual environment in which hehad grown up. What he lacked, by contrast, was an


adequate notion of the necessary degree of hetero-geneity which must exist at the same time if eachdifferent institution is to play its own part within thesocial order. Such heterogeneity is evidently thenecessary basis of that division of functions whichmust exist if the system is to function as a whole.Complementarity between the various institutionswhich together constitute an institutional order thusrequires a measure of heterogeneity in certain respects^which does not detract from the need for somehomogeneity in other respects. Weber, however,lacked an analytical organon which would have per-mitted him to examine such questions properly.Apparently he failed to see any need for this dualset of relationships between the elements of aninstitutional order. He had a conception of such anorder, to be sure, but an incomplete one. The absenceof an analytical scheme for the study of institutions inhis work manifests itself in defects such as this. He whospurns complex analytical models often has to pay theprice of having to make unconscious use of the crudestof models.

When he criticized the Prussian state and thepolitical structure of Hohenzollern Germany as a'feudal relic' his argument rested on the need forhomogeneity among all the institutions of modernindustrial society. Germany, in most other respects anindustrial country, lacked in his view the politicalinstitutions she required in order to cope with theintricate problems of the twentieth century. It is clearthat in this argument 'society' is conceived in terms ofhomogeneity: all the institutions of an industrialsociety must partake of certain common character-istics. They must, for instance, all bear the imprintof the spirit of rationality and of formal legality. The


economists of the Historical School, in their attackson the classical economists, had made use of exactlythe same type of argument. Economic institutions hadto conform to the general pattern (the 'spirit') of theinstitutions of the society they served. It was thus afallacy to believe that their character could bedetermined in terms of their economic functions only.It is interesting to observe how in his political polemicsWeber (probably unwittingly) resorts to a method ofanalysing institutional relationships which in hismethodological writings he had emphatically re-pudiated !

In the brief picture we drew above of the politicalstructure of Hohenzollern Germany we attempted toindicate the judicious blend of stability and flexibilitywhich characterized its fundamental institutions, andthe nature of the social forces on which these rested.Weber saw a different picture. To some extent nodoubt the discrepancy is the result of differences invalue judgements that it would be futile to discuss. Tosome extent it may be due to differences in the degreeof significance attributed to various facts, a matter ofhistorical perspective in the sense mentioned above.But to a certain extent, which we are here inclined toregard as significant, the discrepancy may also be dueto a difference of theoretical perspective. The theoryof institutional order we have attempted to outlinecompels us to judge any given structure in terms of itsfunctional interrelatedness, the complementarity ofthe institutions of which it is composed, as well as interms of the forces of human action which are de-ployed within this structure. It thus may well be thatwhat looks at the outset like a lack of institutionalcohesion, manifesting itself as a compromise ofinterests, actually provides the institutional order as a


whole with a greater degree of flexibility than it couldotherwise possess. Weber lacked this particular theo-retical perspective and thus had to adopt another one.No doubt he would have claimed that he chose thatperspective 'which the German situation at the timerequired'. In our view this must be a matter ofdispute.

The conceptual structure of Weber's politicalthought, the implied analytical scheme he adopted,seems to us a matter of sufficient importance towarrant further examination. Weber's mind wasalways trained, on the one hand, to the range of thepossible and, on the other hand, to what actuallyhappened in history. His conceptual system was'typological', that is, it was essentially designed topermit the classification of a large number of actualinstances within the wider framework of the range ofpossibilities. Within this conceptual framework therewas little scope for an examination of the 'range ofnecessity', of how A must work in conjunction with Bif the system of which both form part is to functionsmoothly. But problems of the institutional order, andin particular of political structure, are essentiallyproblems involving such complementarity. We mustnot be surprised to learn that, with due allowancefor human nature, Weber was inclined to disregardthem.

To understand this attitude of his we have toremember his strong aversion to everything that, tohim, smacked of Natural Law doctrines. To point outwhat is 'necessary' had, in the legal and institutionalsphere, for so long been the characteristic contributionof thinkers in the Natural Law tradition that it is atleast understandable, if not pardonable, that theiropponents had come to view with some suspicion the


very category of 'necessity', even where no meta-physical grounds for such necessity were involved.Such suspicion was in Weber simply another legacyof the Historical School, and so was the tendency togo to the other extreme and to deny the existence ofany necessary relationships at all. Nor did the tendencyof thinkers in the Natural Law tradition, as time wentby, to rely less and less on metaphysical sources of thelaw, and more and more on arguments derived from'the nature of the law as such5 help to still the sus-picions of their opponents who claimed to see in sucharguments little more than attempts to 'secularize' aninherently untenable doctrine. It was on such groundsas this that Weber refused to discuss questions con-cerning the natural limits as well as the functions ofinstitutions.

In this strong aversion to anything faintly suggestiveof Natural Law teaching we see at least one reason forWeber's failure to appreciate the nature of the GrandCompromise on which the Bismarckian Reich rested,as well as of that on the basis of which the edifice of theWeimar Republic was being erected in his own day.Another reason may well be his temperamentalattitude, his inclination to see all social life in terms ofconflict and contest for power, and consequently toregard all compromises as nothing but temporarymakeshifts in the continuous struggle for power. Itseems to us that the two types of aversion were at leastcompatible, if not perhaps closely related to oneanother. Together they explain why such notions asseparation of powers, balance of social forces, equili-brium and stability formed no part of his conceptualinventory.

Weber saw quite clearly the need for the dailycompromises of political life. He actually stressed it

ON POLITICAL INSTITUTIONS I2g

in his account of the work of parliamentary com-mittees, or in explaining the consequences of theabsence of a two-party system (which he thought wasin any case impossible in Germany). But the largercomplex of compromises upon which all these dailycompromises must come to rest, and without whicha social order capable of satisfying the need forstability and flexibility is impossible, never seems tohave moved into his field of vision. Perhaps, however,it is less than fair to blame Weber for failing to havesystematically pursued the implications of his assump-tions in writings which, as we pointed out, were meantto be polemical and persuasive rather than scientificand systematic.

It is, in our view, impossible to appreciate properlythe nature of the role Weber assigns to his 'politicalleader5 unless we bear in mind this gap in his imageof the political structure. But we must make an effortto locate this gap correctly. It might be tempting tosay that for Weber the leader is, in a sense, a deus exmachina, a man who brings about what social forcesand the institutions in which they are crystallized areunable to accomplish, namely to co-ordinate thevarious conflicting interests and unify the groups whichrepresent them. The figure of the leader, we mightsay, is designed to fill an institutional gap.

But this would not be an accurate account ofWeber's view. He emphasized that the democraticleader is the product of a process engendered byparliamentary democracy. Thus he cannot be ac-cused of neglecting the role of institutions in themaking of political decisions. It is only when we askhow the existence of parliamentary institutionsvouchsafes a continuous supply of leaders of requisitestature, and how the system would function if no such


men were forthcoming, that the gap becomes visible.Weber actually took pains to describe the mechan-

ism of institutional pressures to which a politician hasto respond if he wants to become a national leader.We are given to understand that the leader as nationalleader will practise later on what he has first learntas party leader, namely the co-ordination of differentgroup interests and their subjection to the interestsof the greater whole, be it party or nation. Thesegroups have to accept such subjection since withoutthe victorious leader they can hope for no share inpower at all.

Two aspects of this process call for particular notice.In the first place the function of political parties is hereessentially to serve as training grounds for leaders. Inthis respect they all appear to be homogeneous, atleast in the sense that they all provide the same kindof training for their aspirant leader. There is nomention here of the fact that each party is differentfrom any other party and enjoys an individualitycharacterized by the peculiar constellation of groupinterests, denominational, regional, economic, etc. towhich it lends expression. We notice again Weber'spreference for structural concepts couched in terms ofhomogeneity rather than specialization, of whatDurkheim called mechanical solidarity rather thanorganic solidarity.

But we should hardly be able to do justice toWeber's conception of political matters if we failed todraw attention to the dynamic background in frontof which the leader has to perform his duties, thecontinuous stream of unexpected events accompaniedby problems to which they give rise. These problemsare always new, each is a problem sui generis whichcannot be solved by reference to routine—diplomatic,


bureaucratic, or otherwise. It is precisely his abilityto improvise solutions for such new problems whichdistinguishes the Weberian political leader from theable bureaucrat. It is an activity which requiresversatility, originality, and an eye for those aspectsof a situation which permit of being manipulated—allqualities of mind not often found among those whoare masters in performing routine duties. In otherwords, political decision-making is an activity whichdefies all static schemes and calls for adaptability to arapidly changing world. Hence Weber thought itunprofitable to attempt to describe the activity whichpertains to the solution of practical problems in suchcircumstances in terms of a universally applicablescheme.

This tendency to stress the plastic nature of allsocial relationships, and hence of institutions, is notconfined to his political writings. It can also be foundin his systematic work, in particular in those parts of itin which he has to discuss the relationships which linkthe various sectors of the institutional order. In thechapter on 'The Economic System and the NormativeOrders'1 the emphasis is on the tenuous nature of thelink between legal and economic order.

In theory, a socialist system of production could bebrought about without the change of even a singleparagraph of our laws, simply by the gradual, freecontractual acquisition of all the means of production bythe political authority. This example is extreme, but forthe purpose of theoretical speculation extreme examplesare most useful, (p. 36)

On the other hand, there exists a dual relationshipbetween legal and economic order.

1 Sociology of Law, ch. II.


Obviously to a very large extent any legal guaranty isdirectly at the service of economic interests. Even wherethis does not seem to be, or actually is not, the case,economic interests are among the strongest factors in-fluencing the creation of law. For, any authority guaran-teeing a legal order depends, in some way, upon theconsensual action of the constituent social groups, and theformation of social groups depends, to a large extent,upon constellations of material interests, (p. 37)

Beyond this loose and cautiously worded generaliza-tion Weber will not go.

But as regards modern capitalistic society inparticular, something more specific can be said aboutthe relationship between legal and economic order,between state and society. 'Class interests have cometo diverge more sharply than ever before. The tempoof modern business communication requires a promptlyand predictably functioning legal system, i.e. onewhich is guaranteed by the strongest coercive power'(p. 39). The implication is here clearly that the needfor coercive power is linked to the increasing frictionbetween class interests. The emphasis is on thespecific historical character of the circumstances whichhave here forged a link that in other historicalconditions may not exist.

This has been the result of the development of the market.The universal predominance of the market consociationrequires on the one hand a legal system the functioning ofwhich is calculable in accordance with rational rules. Onthe other hand, the constant expansion of the marketconsociation has favoured the monopolization and re-gulation of all 'legitimate' coercive power by one universalcoercive institution through the disintegration of allparticular status-determined and other coercive structures,which have been resting mainly on economic monopolies.(P. 40)

ON POLITICAL INSTITUTIONS I33

It is possible to welcome the wholesome emphasislaid here on the limited nature of all social generaliza-tions, and yet, at the same time, to regret that thewider problems of institutional structure which liebeyond the range within which Weber is here keepinghis discussion, have remained unexplored.

V

Three problems emerge from our discussion ofWeber's political thought, which are of sufficientimportance to us to warrant further discussion. Allthree reflect different aspects of the phenomenon ofinstitutional change in time and are thus germane toour major theme.

With the first of these problems we are alreadyfamiliar from our earlier discussion. This is theproblem of how to safeguard fundamental institutionsagainst the forces of slow erosion that are bound to beunleashed by the continuous struggle for power andposition in society. Those political institutions whichgovern access to the seats of power, define themechanism of selection of power-holders, and cir-cumscribe, and hence limit, the functions of power-holders are clearly the most fundamental politicalinstitutions. They should be designed to last, since theplans of millions of people and the multiplicity of otherinstitutions towards which the actions of these plannersare oriented must all come to rest on this firm basis.

But in an open society every institution is of coursein principle revisable. The unrestricted circulation ofthe component groups of the political elite is of thevery essence of democracy. We typically find that inthe continuous struggle for power parties will makepromises to the electorate which cannot be redeemed


unless the power of the power-holders is increased andthe institutions limiting such power are weakened.The history of economic policy in all Western countriesduring the last half-century provides a continuousseries of examples of this kind. In this way thestruggle for power leads to the gradual erosion of thoseinstitutions which circumscribe and limit the exerciseof power.

Whether Weber saw this as a serious problem ishard to say. We have to remember that for most ofhis active life his main concern was to bring aboutprecisely such a change in the political structure ofGermany as would facilitate the unrestricted circula-tion of political leaders. He could thus hardly beexpected to give a sympathetic hearing to the case forthe defence. But it is perhaps permissible to seeevidence of an incipient preoccupation with thisproblem in his continuing adherence to the monarchy,until its actual fall, and later on in his attempt, in hisproposals for the Weimar Constitution, to give widerpower to the President as the guardian of the Con-stitution and to limit the power of the Reichstag. Inhis Sociology of Law, on the other hand, we findpassages suggesting that in his view all fundamentalinstitutions are, at least primarily, manifestations ofclass interests.

Freedom of contract and all the propositions regardingas legitimate the property derived therefrom obviouslybelong to the natural law of the groups interested inmarket transactions, i.e. those interested in the ultimateappropriation of the means of production.1

Perhaps, in the light of what has happened in theyears since Weber died, it is today unnecessary to

1 Ibid., p. 294.

ON POLITICAL INSTITUTIONS I35

argue at length that this is a fallacy, and that every-body has an interest in the existence of some funda-mental institutions which limit the exercise of politicalpower since nobody can know whether the nextCaesar will not turn against him or his children. Butthe question of how to make the long-run needs ofsociety as a whole prevail over the short-run needs ofpoliticians who have to win elections by gaining votesfrom electors who are unable to understand suchsubtle issues (and probably would not care aboutthem if they did), is indeed very much an openquestion.

Our first problem thus concerned what we mightterm the vertical complementarity of the institutionalorder, the link between fundamental and secondaryinstitutions. It arose from the growing tendency towiden the discretionary authority of power-holdersand add to the number of functions assigned topolitical institutions. Our second problem, by contrast,concerns the link between purely political institutionsand is thus a problem of what we might call horizontalcomplementarity. It arises in cases in which an existinginstitution 'sheds' one of its functions without thisfunction being assumed by any other institution,either existing or new. We thus have a cgap' emergingin the institutional order. The fact that such cases canoccur without any obvious remedy being available,and without the disappearance of the social need whichthe function now discarded had so far satisfied,strengthens our reluctance, explained above in oursecond essay, to ascribe to the institutional order acloser texture than it actually possesses and to see a'social system' in it. An institution may cease toexercise a function, not because the demand for it hasdisappeared, but because nobody is capable of

I36 ON POLITICAL INSTITUTIONS

supplying it any longer. We find here another reasonfor the inadequacy of structural functionalism as abasis for a theory of the institutional order.

The great contemporary example is the decline ofparliamentary debate. In our world parliamentaryspeeches are no longer addressed to other members inorder to sway their voting. They contain no 'argument'in the sense in which legal argument, for instance, isdesigned to sway judicial decision, or an 'argument' indaily life is meant to be 'won' by us. To be sure,modern parliaments have assumed many otherfunctions the need for which in modern society is notin doubt. The question we have to ask here is: Canthere be a substitute for public discussion on mattersof legislation and policy which takes place, not merelybetween men whose education permits them to speakwith authority on them, not between experts with noresponsibility for the making of decisions, but betweenmen who share the same practical experience ofgovernment? Press conferences? Party congresses? Ifin fact no such substitute has come into existence wehave to conclude that the need is not met. The reasonfor this phenomenon has to be sought of course in thedisappearance from the political life of 'advanced'Western societies of men who, only loosely tied to theapparatus of a political party, knew how to hold theirown in a debate.

Another interesting example is the permanentinflation of our age, the disappearance of a money ofconstant purchasing power which former generationstook for granted and which certainly constituted, inthe world before 1914, a fundamental economicinstitution. Our contemporary money has lost thisquality. Some of the reasons for this process wedescribed in our second essay.


From these examples we may learn that it ispossible for 'gaps' to open in the institutional orderwhich it is impossible to fill, and yet this order as awhole survives. The relationship between needs andinstitutions is thus once more seen to be by no means asimple matter. It is a fallacy to regard the institutionalorder as a seamless web, a faultless structure governedby a simple one-to-one relationship between needs andthe institutions designed to meet them. By the sametoken, we should make a very careful study of thesituation at hand every time before we pronounce anygiven institution 'obsolete5. It may be that tomorrowanother one of our institutions, overloaded with urgenttasks, will 'shed' a function which our allegedly'obsolete' contemporary one is excellently qualified tofulfil. The reception of Roman Law in medievalEurope is here an obvious example.

Our third problem emerges within the samecontext of the dynamics of institutional change. Wepointed out in an earlier essay that in a sense whichis significant for us some institutions are relics of thepioneering efforts of former generations, from whichwe are still drawing benefit. In this they are likeancient buildings in a modern town. Each carries theimprint of its origin and history. Its origin may havebeen a Grand Compromise of social groups long sincedefunct. In the course of history various interests willhave made use of it, clustered around it, and may haveformed a crust later generations find hard to erase.Habits of action, aspirations regarded as sociallylegitimate, traditions and values may have come intoexistence in which it plays a significant part. Whathas been said thus far applies to institutions of everytype.

But when we begin to look more closely at the forms

I38 ON POLITICAL INSTITUTIONS

which such changes may take, the difference betweendesigned and undesigned institutions at once assumesconsiderable significance. Undesigned institutions willrise, move and fall, prompted by the genius of theirinitiators, the perspicuity of their imitators, theadaptive skill of their users and the destructive abilityof those who fashion new institutions which supersedethe old.

With designed institutions this whole process ofcreative destruction1 does not exist, or at least it neednot come into existence. A designed institution bearsthe imprint of the social situation that gave birth toit and gradually acquires the kind of 'crust5 wedescribed. But it does not have to compete with others.It will shape others, namely the secondary institutionswhich will settle in its 'interstices', but will not beshaped by them. A good deal will therefore depend onthe quality of its original design. A designed institutioncannot be better than the quality of its design.

We noted above that institutions can only bedesigned to meet certain known situations, or alimited number of possible, that is conceivablesituations, but not an unlimited number of unknownsituations. Nevertheless the only test available to usto determine the quality of the design of an institutionis the test of history.

Two conclusions appear to emerge from thisdiscussion:(1) With designed institutions it matters a good deal

who designs them. A Grand Compromise may bea great success in its day and lead to the solutionof a large number of urgent social problems, butthe institutions to which it gives birth may be of

1 We borrow this apt expression from J. A. Schumpeter,Capitalism, Socialism and Democracy, ch. V I I I .


shoddy design and their mechanism of poorworkmanship. Those who are good at conciliatingconflicting group interests and devising compro-mises which minimize social friction are notnecessarily qualified as designers of institutions.Successful manipulators of interests are unlikelyto excel in the quality of craftsmanship. Thosewhose minds are too absorbed by the problems ofthe day are unlikely to be good prophets.

(2) With designed as with undesigned institutionsadaptive skill is a matter of importance. Asregards the reform of old institutions, good timingis often of the essence of success. Since the re-former will as a rule have a number of ways andmeans open to him, his task is in many respectssimilar to that of the original designer. Again,it is by no means obvious that those most sensitiveto present needs of reform will be the best judgesof the needs of the future.

In what, then, does the legacy of Max Weberconsist, in the field of political thought? The actualpolitical issues to which he made his contributions, thecontroversies on Germany's political structure in whichhe participated with such vigour between 1917 and1920, now belong to history. They do not interest us.What must interest us, however, are the presupposi-tions of the arguments Weber advanced in thesecontroversies of the past. For these presuppositionscontain, albeit in rudimentary form, a general theoryof political institutions and their modus operandi inmodern industrial society. Weber demanded forGermany a political form which was adapted to therequirements of such a society.

Political institutions do not exist in isolation from


the rest of the social structure, and in particular fromthe rest of the institutional order. They must fit theirsocial environment. This is no new discovery. ThatBritish parliamentary institutions are 'not for export'is by now fairly well known. That 'democracy' means,and has meant, different things in different countriesand at different times is equally well understood. Anearlier generation had already come to learn thatGerman princes set up as kings of Balkan countriesdid not, by virtue of their role as monarchs, necessarilymake these countries into stable elements of theEuropean balance of power.

In this essay we have endeavoured to show that thecomplex of relationships existing between thoseinstitutions which embody the political form of societyand the rest of its institutional order is a matter ofsome intricacy. It is not simply a matter of the formerfitting a pre-existent latter, as Weber sometimesappears to imply, or of rinding the requisite mode ofadaptation.

On the one hand, political institutions are funda-mental institutions as they help to determine thedistribution of power, and this includes institution-making power, in society. Not merely the permanenceof the legal order, but the coherence and flexibility ofthe institutional order as a whole rest upon thisfoundation and depend on the stability of the politicalorder.

On the other hand, such coherence as links thelatter to the rest of society must not be conceived inthe all too simple terms of homogeneity. In complexsocieties some Grand Compromise of the kind wehave discussed is often necessary to keep politicalinstitutions functioning. This means that the variousparties to the compromise will view not merely the

ON POLITICAL INSTITUTIONS I4I

terms of the compromise, mostly unwritten and notexactly proclaimed to the electorate, but each of thevery institutions in which these terms are embodied,in its own perspective—and these perspectives maydiffer widely. A change in the terms of the GrandCompromise will often produce changes in themodus operandi of political institutions, but need haveno repercussions in the rest of the institutional order.Some stability of the latter in the face of changes inthe balance of political power is simply a socialnecessity.

In the last resort all institutions are vehicles ofhuman action, which is the ultimate reality of sociallife. This insight of Weber we made our starting-point.Institutions are at the same time instruments of, andconstraints upon, human action. What we haveendeavoured to show in this book is that it is possibleto spin a thread which ties the complex edifice of theinstitutional order in its political form to the simpleunit act of the individual actor and the plan whichguides it. This insight is part of the legacy which weowe to Max Weber.

Biographical Note

MAX WEBER (1864-1920), German economist, sociologistand philosopher, was born into a well-to-do family, whosehouse was a rendezvous for liberal politicians and dons fromBerlin University. His father was a Berlin city official andparliamentarian. On leaving school in 1882 he studied lawfor a year at Heidelberg University and (after militaryservice) for two further years at Berlin and Gottingen, takinghis examination in 1886. His doctoral thesis of 1889 wasentitled A Contribution to the History of Medieval BusinessOrganizations. Further training for the bar followed, duringwhich he made a study of legal institutions and wrote hisRoman Agrarian History and its Significance for Public andPrivate Law (1891). He became a Privatdozent in the Facultyof Law of the University of Berlin, combining lecturing withresearch and consultancy to government agencies. Hemarried Marianne Schnitger in 1893, and in 1894 becameProfessor of Economics at Freiburg University. In 1896 hewent to Heidelberg University in a similar capacity.

His load of work however proved too much: in 1897,when he was 33, he became ill, suffering from acute anxietyand exhaustion. He was forced to give up teaching. Afterfour years he made a partial recovery, and became in-terested in the relation between religious beliefs andeconomic activity. In 1903 he was appointed associateeditor of the Archivfiir Sozialwissenschaft und Sozialpolitik, andlater resumed part-time academic work. The Protestant Ethicand the Spirit of Capitalism was published in 1904.

In 1907 a legacy made him financially independent. Thushe was able to pursue his work as a private scholar for anumber of years. During the First World War he establisheda reputation as a political writer. After the NovemberRevolution he helped to produce the first draft of the

H3

144 BIOGRAPHICAL NOTE

Weimar Constitution. Although a prominent member of theGerman Democratic Party, he failed to be elected to theNational Assembly. He was a guest professor in Vienna for afew months in 1918, and afterwards became Professor ofEconomics in the University of Munich. In June 1920, atthe age of 56, he died of pneumonia.

Max Weber was a strange and in many ways a somewhatcontradictory personality. On the one hand he had a strongdesire to participate actively in political and academic life,a desire which remained largely unfulfilled. On the otherhand he was withdrawn, uncompromising, puritanical. Inthe last thirty years his reputation as a scholar and thinkerof the first order has continuously increased. His scholarlyoutput was extremely impressive in both range and quality;he is now recognized as one of the great minds of thetwentieth century, and both his influence and the volumeof literature about him are still growing.

Bibliographical Note

A BIBLIOGRAPHY of Weber's writings is given in MarianneWeber's biography of her husband: Max Weber, EinLebensbild (J. C. B. Mohr, Tubingen 1926; LambertSchneider, Heidelberg 1950). According to ReinhardBendix (see p. 147), this is complete except for the followingtwo items:

1. 'Die drei reinen Typen der legitimen Herrschaft',Preussische Jahrbiicher, vol. 187, pp. 1-12 (1922).

2. Jugendbrife, ed. Marianne Weber (J. C. B. Mohr,Tubingen, 1936).

The most important editions and collections not mentionedin Marianne Weber's 1926 bibliography, or subsequent to it,are the following (first editions unless otherwise stated):

3. Wirtschaft und Gesellschaft, 2nd edn., 2 vols. (J. C. B.Mohr, Tubingen, 1925): see no. 26.

4. Gesammelte Aufsatze zur Religionssoziologie, 3 vols. (J. G. B.Mohr, Tubingen, 1920-21).

5. Gesammelte Aufsatze zur Wissenschqftslehre (J. C. B. Mohr,Tubingen, 1922).

6. Gesammelte Aufsatze zur Social- und Wirtschqftsgeschichte(J. C. B. Mohr, Tubingen, 1924).

7. Gesammelte Aufsatze zur Soziologie und Socialpolitik (J. C. B.Mohr, Tubingen, 1924).

8. Gesammelte Politische Schriften (Drei Masken Verlag,Munich, 1921).

9. Staatssoziologie (Duncker & Humblot, Berlin, 1956).

The English translations of Max Weber's works that areavailable to date are as follows:10. The Protestant Ethic and the Spirit of Capitalism, tr. Talcott

Parsons; foreword by R. H. Tawney (George Allen145

146 BIBLIOGRAPHICAL NOTE

and Unwin, London, 1930; Charles Scribner's Sons,New York, 1958; Unwin University Books, London,

1965).11. From Max Weber: Essays in Sociology, tr., ed. and with

an introduction by H. H. Gerth and C. Wright Mills(Kegan Paul, Trench, Trubner & Co., London, 1947;Oxford University Press, New York, 1947).

12. Max Weber on the Methodology of the Social Sciences, tr. anded. by Edward A. Shils and Henry A. Finch (FreePress of Glencoe, New York, 1949; Collier-Macmillan,London, 1950).

13. The Theory of Social and Economic Organization, tr. A. M.Henderson and Talcott Parsons (William Hodge & Co.,London, 1947; Oxford University Press, New York,1947; Free Press of Glencoe, New York, 1964; Collier-Macmillan, London, 1964).

14. The Religion of China: Confucianism and Taoism, tr. and ed.by Hans H. Gerth, with an introduction by C. K. Yang(Free Press, Glencoe, 111., 1951; Macmillan, New York,1964; Collier-Macmillan, London, 1964).

15. The Religion of India: the Sociology of Hinduism andBuddhism, tr. [from articles published in Archiv fiirSozialwissenschaft, Bd. 41] and ed. by Hans H. Gerth andDon Martindale (Free Press, Glencoe, 111., 1958, 1967;Collier-Macmillan, London, 1967).

16. Ancient Judaism, tr. Hans H. Gerth and Don Martindale(Free Press, Glencoe, 1952; Collier-Macmillan, London,

17. Max Weber on Law in Economy and Society, tr. EdwardShils and Max Rheinstein, ed. Max Rheinstein (HarvardUniversity Press, Cambridge, Mass., 1954). Referredto in the present volume as Sociology of Law.

18. General Economic History, tr. Frank H. Knight, ed. S.Hellman and M. Palyi (George Allen and Unwin,London, 1927; Free Press, Glencoe, 1950).

19. The Social Causes of the Decay of Ancient Civilization',Journal of General Education, vol. v, pp. 75-88 (1950).

20. The City, tr. and ed. by Don Martindale and Gertrud

BIBLIOGRAPHICAL NOTE I47

Neuwirth (Free Press, Glencoe, 1958, 1966; Heinemann,London, i960; Collier-Macmillan, London, 1966).

21. The Rational and Social Foundations of Music, tr. and ed. byDon Martindale, Johannes Riedel and Gertrud Neu-wirth (Southern Illinois Press, Carbondale, 1958).

22. 'The Three Types of Legitimate Rule', Berkeley Publica-tions in Society and Institutions, vol. vi, pp. 1-11 (1958).

23. The Sociology of Religion, tr. Ephraim Fischoff [from 4thedn. of Wirtschaft und Gesellschaft], introduction byTalcott Parsons (Beacon Press, Boston, 1964; Methuen,London, 1965).

24. On charisma and institution building: selected papers, ed.with an introduction by S. N. Eisenstadt (Universityof Chicago Press, Chicago and London, 1968).

25. Basic Concepts in Sociology, tr. and introduced by H. P.Secher (Peter Owen, London, 1962; Citadel, NewYork, 1962).

26. Economy and Society, tr. E. Fischoff, ed. Guenther Rothand Claus Wittich: 4th edn, 3 vols. (Bedminster,Totowa, N.J., 1968).

27. Politics as a Vocation, tr. and ed. by H. H. Gerthand C. Wright Mills (Fortress, Philadelphia, 1965):included in no. 11.

28. Max Weber: Readings, ed. S. M. Miller (Crowell, NewYork, 1963).

See R E I N H A R D BENDIX, Max Weber, an IntellectualPortrait (Doubleday, New York, 1959), p. vii, for a helpfulbibliographical note on the writings of Max Weber (includ-ing a guide to the relation between the German originalsand English translations), and also H. H. and H. I. Gerth,'Bibliography on Max Weber', Social Research, vol. XVI,pp. 70-89 (New York, March 1949).

Index

Austrian school of economics,see Menger, G.

Berliner Borsenzeitung, i2ossBismarck, O., 101, 10411, 116

Collective bargaining, 83

Durkheim, E., 130

Epstein, K., n6nEucken, W., 28n

Feldman, G.,Frankfurter geitung, 95, 112Friedman, M.,

fundamental and secondary,13, i5,9<>,93,98

external and internal, 81, 92

Leader, political, 14, 117Lloyd George, D., n8n

Mach, E., 334-5Machlup, F., 28nMenger, G., 13, 24-5, 53, 56-

60, 64Method, scientific, 10-11, 34,

praxeological, 7, 9, 23, 37-8,

43, 49, 57, 60, 64historical, 19-22

„,.„. T Methodenstreit, 24, 55GJhs.J :om Michek, R., i .JnGneist, R., io4n Mierendorff, C , i n nGrand Compromise, i6, .00, M o m m s e n w I I 3 n I i g

125, 128, 137-8, 140-. M 6 8 ^ g 9

Growth, economic, 5 Mosca/G., 6 2 / n a n

Hicks, J., 6n, 85 , T , .Historical School, 10, 17, 24, Natural Law doctrines, 127-8

26, 54-5, 58, 124-6, 128 . . . . o

Hohenzollern Germany, 16, Order, institutional, 8, 51, 75-96, 100, 125-6 77

Orientation, 37, 62Ideal Type, 3, 11, 26Inflation, 84, 100, 126 Pareto, V., 42, 62,Institutions designed and un- Parsons, T., 1, 9,

designed, 13, 56, 67-8, 72, Plan, 9, 12, 20, 29, 34, 38-9,139 43-8, 68

149

15°Poincare, H., 34-5Popper, K., 10Preuss, H., 119

INDEX

Sombart, W., 28Stammler, R., 61Structural Functionalism, 7-8,

73-4, 124System, legal, 64, 70, 72, 75, 77Rechtsstaat, 104-5

Reichstag, 102, 113, 116, 120-2,134 Verstehen, see Method, praxeo-

Rex, J., 7411 logicalRunciman, W., 7411 Volksgeist, 58, 60, 63Roscher, W., 58, 60

Weimar Constitution, 96, 111,Schmoller, G., 24 119, 134Schumpeter, J., 24, 106, i38n Weimar Republic, 16, 107SS,Schutz, A., 3on 119, 123, 128Social Systems Theories, 73, Wilhelminian Germany, see

124 Hohenzollern Germany

the legacy of max weber - mises.at · the case of weber, a good deal of his work has been left to...

Documents