GRICE E GIUCCIARDINI

 POLITICAL  PHILOSOPHY    EDITED BY  ANTHONY QUINTON    OXFORD READINGS IN PHILOSOPHY    OXFORD READINGS IN PHILOSOPHY    Series Editor G. J. Warnock    POLITICAL PHILOSOPHY    Oxford University Press, Walton Street, Oxford OX2 6DP  OXFORD 1967    CONTENTS    INTRODUCTION    I.    II.    Ill.    VII.    VIII.    X.    THE USE OF POLITICAL THEORY  by John Plamenawz    POLITICS, PHILOSOPHY, IDEOLOGY  by P. H. Partridge    ARE THERE ANY NATURAL RIGHTS?  by H. L. A. Hart    THE USES OF ‘SOVEREIGNTY’  by Stanley I. Benn    AUTHORITY  (1) by R. S. Peters  (2) by Peter Winch    THE PUBLIC INTEREST  by Brian Barry    LIBERTY AND EQUALITY  by E. F. Carritt    Two CONCEPTS OF LIBERTY  by Sir Isaiah Berlin    Two CONCEPTS OF DEMOCRACY  by Joseph Schumpeter    JUSTICE AND THE COMMON GOOD  by Brian Barry    NOTES ON THE CONTRIBUTORS    BIBLIOGRAPHY    INDEX OF NAMES    112    127    141    153    189  194  195  199    INTRODUCTION    1. The nature of the subject    The easiest and most uncontroversial way of defining political  philosophy is as the common topic of a series of famous books:  Plato’s Republic, Aristotle’s Politics, Machiavelli's Prince, Hobbes’s  Leviathan, Locke’s Treatises on Civil Government, Rousseau’s Contrat  Social, Hegel’s Philosophy of Right, The Communist Manifesto and Mill’s  Liberty. Such an enumeration, at any rate, defines a major continuing  strand in the history of Western thought, a great tradition of large-  scale reflection about politics. But a backward-looking list of this  kind is really no longer adequate to define political philosophy as a  going concern. What has changed the subject is the great increase in  methodological self-consciousness among recent philosophers  which has led them to accept a more limited conception of their  powers and, in consequence, of their responsibilities.   A comparatively definite place has now been marked out for philo-  sophy within the total range of man’s intellectual activities. It is  generally thought to stand in a very different relation to other modes  of thought from that in which they stand to each other. Where they are  substantive, concerned with some aspect or region of the world, it is  conceptual and critical, concerned with them rather than with the  reality they investigate. It should be conceived not as just another  mode of thought alongside them but rather as superimposed reflect-  ively on them. Very briefly, philosophy has the task of classifying and  analysing the terms, statements and arguments of the substantive,  first-order disciplines.   In the light of a conceptual interpretation of philosophy the works  that make up the great tradition of political thought are methodo-  logically very impure. They are only to a small, though commonly  crucial, extent works of philosophy in the strict sense. Besides con-  ceptual reasonings of the approved sort they contain two main kinds  of ingredient. First, there are factual or descriptive accounts of  political institutions and activities which may be collected under the  general heading of political science. Secondly, there are recommend-  ations about the ideal ends that political activity should pursue and  about the way political institutions should be designed in order to  serve these ends which may be called ideology.    2 POLITICAL PHILOSOPHY    A good deal of past political science has been somewhat formal or  legalistic; taking the rules which are the professed determinants of  the working of political institutions, in particular the formal con-  stitutions of sovereign states, at their face value and dealing only  as something of an afterthought with the deviations from these rules  that occur in actual practice. A further limiting tendency has been the  custom of political science of treating the political life of society as  comparatively autonomous. Both of these restrictions have been  largely removed by the development of political sociology which  investigates political behaviour as it actually occurs and seeks to  connect it with the general, non-political life of society, witlt class-  stratification, the economy, religious allegiances and so on.   Both the political-science and ideological components of works in  the great tradition of political theory were of high generality. The  details of local government or the hierarchical arrangement of courts  did not figure in them. Nor did the type of very concrete issue to be  found ina pamphlet or leading article about the reform ofa particular  law or the reconstruction of some part of the whole institutional  apparatus. But by current standards they are too all-inclusive to count  as works of political philosophy, strictly so-called, and their all-  inclusiveness has not been much imitated in recent years.   A sign of this change in the way the subject is conceived has been the  apparent petering-out of the great tradition. Surveys of the history of  political thought either come to an end with Marx and Mill in the mid-  nineteenth century or they wind up with apologetic chapters on the  major ideological movements of the most recent period and on the  highly engaged, rhetorical and practical thinking of the more arti-  culate political leaders. But an occasional magnificent dinosaur stalks  on to the scene, such as Hayek’s Constitution of Liberty, seemingly  impervious to the effects of natural selection.   Analytic philosophers have paid little attention to those problems  of political theory that do fall within their recognized field of interest.  Russell has been an active political ideologist, and, in his book Power,  something ofa political sociologist, but he has been very explicit about  the distinction between his work in these capacities and his work asa  philosopher proper. It has been widely held, indeed, that there really  is no such subject as political philosophy apart from the negative  business of revealing the conceptual errors and methodological mis-  understandings of those who have addressed themselves in a very  general way to political issues. (For an example of this see Weldon’s  Vocabulary of Politics.) A solid testimony to the width of this conviction    INTRODUCTION 3    has been the near-unanimity with which analytic philosophers have,  until very recently, avoided the subject altogether. Of course the great  tradition of political thought remains an important object of study in  its own right. But to study its members is only marginally to continue  the work they were doing. Many teachers of political philosophy are in  fact students of the history of very general, theoretical, political ideas.  But this no more makes them political philosophers than close attend-  ance at the bull-ring makes an aficionado into a bull-fighter.   The application of philosophical analysis to the fundamental  concepts and styles of reasoning that occur in political discourse  remains an open possibility. But until very recently the only extended  example of it has been the excellent Consent, Freedom and Political  Obligation of J. P. Plamenatz, first published in 1938 (Second edi-  tion 1968). In the first sentence of the introduction Plamenatz wrote:  ‘the purpose of this book is to provide definitions of a number of  words generally used by political thinkers’. The publication in 1959  of a substantial general survey of political theory from an analytic  point of view, Social Principles and the Democratic State by S. I. Benn  and R. S. Peters, was a sign of renewed interest and the suggestion  has been confirmed by the appearance in 1965 of B. M. Barry’s  Political Argument, perhaps the most uncompromisingly analytic  treatment of politics yct published.   The first task of an analytic philosophy of politics is to distinguish  the two main varieties of substantive political discourse: the factual  statements of political science and the evaluative affirmations of  ideology. Few would deny that such a distinction can be drawn, how-  ever critical they may be of the reasoning associated with it and the  irrationalistic conclusions about valuation that have often been  derived from it. Within each of these domains there are characteristic  concepts of a very general sort whose application is a matter of  frequent dispute. Philosophical analysis, it might be hoped, could  help the disputants to a better understanding of each other’s positions  and even, in some cases, of their own.   The central concept of political science is that of the state. Cor-  relative with the state is law. Positive law must have a state or sovereign  as its source and it is the first duty of a state to produce and maintain  law. A satisfactory account of the nature of law must trace its com-  plicated relations to morality which serves both to supply law witha  content and to be a standard for criticism of it. The traditional natural  law theory is an attempt to expound this relationship.   The most general concepts of ideology are those of the major    4 POLITICAL PHILOSOPHY    political values which are the more or less commonly recognized ends  of government: liberty, justice, security, prosperity and, perhaps,  democracy. It is by reference to such ends that particular schemes of  political institutions are recommended in preference to others. The  central problem of traditional political theory has been a kind of  generalized limiting case of the problem of justifying a particular  institutional scheme. This is the problem of political obligation which  is that of why, or under what circumstances, an individual should obey  any state at all, acknowledge any obligation to obey those who seek to  determine what he shall do, attribute any authority to those who claim  to be his rulers.   A theory of political obligation, by giving a rational answer to the  question ‘why should I obey the state’, must inevitably, itmight seem,  be ideology rather than analysis. But analysis can at least be used to  examine the form of arguments purporting to justify the state. Equally  ideological must be the endorsement of any major political value or  ideal as an end which governments ought to pursue. As it turns out  analysis and justification are harder to keep apart than formal  methodology might suggest, here even more than in other parts of  philosophy. Concepts of political ends are what have been called  essentially contested concepts (cf. W. B. Gallie in Proceedings of the  Aristotelian Society, 1955-61, pp. 167-98). The adherents of competing  ideologies try to interpret terms such as liberty and justice in such a way  that they apply to the realization of their own ideals. The methodo-  logical aim of a strictly neutral analysis of political terms, even of  the most general terms of political science, is hard to realize in practice  if the results are not to be trivial.   Some have argued, in the spirit of Popper’s remark that if the Soviet  Union is a democracy then he is against democracy, that the words in  which political discussion. is carried on do not matter and that a  political philosophy which regards the clarification of political terms  as its main task must thus be a waste of time and energy. This view is,  I think, doubtly mistaken. In the first place disputes that have some  ultimately verbal element are extremely influential. Furthermore the  verbal element in disputes is not generally so easy to identify and  dismiss as the Popperian example might suggest. And even in that  example the dispute is not merely verbal. Adherents of both liberal  and communist conceptions of democracy would agree that any  adequate conception of it must start from the notion of government by  the people. They disagree on the extent to which their competing views  are better adjusted to this basic and agreed requirement.    INTRODUCTION 5    The first two selections in this anthology are of the same broadly  methodological character as this section of the introduction. Mr  Plamenatz defends a moderately traditional view of the nature of  political theory which he defines as systematic thinking about the  purposes of government. Professor Partridge argues that analysis  cannot be kept wholly free from all ideological taint.    2. The state, law and morality   In the second of his Treatises on Civil Government Locke introduces  something that he calls the executive power of the law of nature. It has  three parts: the legislative power of deciding what the correct rules of  conduct are, the judicial power of applying these rules to particular  pieces of conduct and the penal power of administering sanctions to  those who have broken the rules. The function of this moderately com-  plicated idea in his theory is, in effect, to define the concept of the  state and, by implication, of the political generally. For it is by the  transfer of this executive power from free, natural individuals to a  common sovereign that a natural society is turned into a civil, or  politically organized, society.   It is not necessary to suppose that the transformation of natural into  civil society was brought about by some historically identifiable act of  transfer to find Locke’s notion a useful one. The executive power of the  law of nature can be used to distinguish political societies, as those in  which it is formally centralized, from non-political ones, in which itis  informally distributed amongst all individuals. There are two signifi-  cant implications of this use of the idea. The first is that it identifies the  essential functions of the state as those of maintaining law and order,  the second that it sees them as the responsibility of a state to the extent  that they are not left to individuals to exercise for themselves but are  remitted to a special person or set of people within the society asa  whole. To see the preservation of law and order as the essential  function of the state is not to regard it as the state’s only function. Nor  is it to say that a state could survive in practice if that was all it did.  The defence of societies from their external enemies is as ancient and  important a function of states as their defence from internal ones. Nor  does the requirement of centralization strictly imply that government  is, of necessity, oligarchical, though, in fact, no doubt, all governments  have been. But even in the most direct and Rousseau-like of demo-  cracies there would be a distinction between the weekly meeting during  which for an hour or two the citizens were acting in their sovereign  capacity as legislators and judges and the remainder of the week    6 POLITICAL PHILOSOPHY    during which they would obey, or disobey, the rules they had them-  selves laid down.   Two more recent definitions presuppose the Lockean concept of the  state but significantly extend it. The first of these is Austin’s well-  known definition of sovereignty. A sovereign, according to Austin, is  a determinate human superior, not in a habit of obedience to a like  superior, who receives habitual obedience from the bulk of a given  society. Austin defines law as the command of a sovereign so conceived.  Secondly there is Max Weber’s definition of the state. ‘A compulsory   olitical association with continuing organization’ he says ‘will be  called a state if and in so far as its administrative staff successfully  upholds a claim to the monopoly of the legitimate use of physical force  in the enforcement of its order’. Both Austin and Weber hold that a  large measure of effectiveness in imposing its rules is necessary toa  state. Weber adds that the means by which this effectiveness is secured  is wielding the only physical force that is generally recognized as  legitimate.   Putting these definitions together we may say that a society is  political, or has a state, if it contains a centralized agency for the  promulgation, application and enforcement of rules of conduct, if  these rules are generally obeyed and if only these rules are generally  recognized as legitimately sanctioned by physical force.   One point of these definitions is to distinguish the state conceived as  a politically organized society, as when we speak of nation-states, from  the state conceived as the rule-enforcing element within such a society.  If a society has a state within it in the second sense, then it is a state in  the first sense. In colloquial terms a country is one thing, a government  another, but what makes a collection of people into a country is the  fact that they all have the same government. We can speak of nations  that are not politically organized societies. Nationalism isan endeavour  to make states and nations coincide, initiated in circumstances where  they do not. We tend to think of the government as more particularly  the executive or administrative arm of the state as a whole. It remains  important to distinguish state from society just because some thinkers  have striven energetically to obliterate the difference between them. A  society is a collection of people who interact, persistently and in  characteristically human ways, cooperating and communicating with  each’ other. A society will persist only if there are generally accepted  rules of conduct but these need not be defined and enforced by any  centralized agency. At any rate social rules do not logically entail a  state even if they practically require one in most circumstances.    INTRODUCTION 7    Anarchism is not a self-contradiction, but at worst impracticable or  intolerable.   Law, in the most ordinary sense of the word, is a product of the  sovereign state. Theories of sovereignty can be understood as present-  ing criteria with which to decide about the rules prevailing in a society  which are laws proper, as contrasted with the private regulations of a  club, a family or a firm on the one hand and the prescriptions of  morality on the other. A familiar tradition in political theory dis-  tinguishes law in this sense as positive law from natural law. Many  theorists who derive political obligation from a contract, most notably  Locke, maintain that what in the end justifies obedience to the state is  its protection of the rights possessed by individuals under natural law.  Some adherents of natural law have gone further, saying that if arule  is in conflict with natural law it cannot be a positive law at all; but  perhaps their point could be less extravagantly made by regarding an  enactment contrary to natural law as giving a reason for withholding  obedience from the state responsible which may be overridden by  reasons for the opposite course.   One origin for the doctrine of natural law is the idea that God stands  to mankind at large in the relation of a monarch to his subjects. The  same analogy can, however, be turned upside down, as in the doctrine  of the divine right of kings, to show that there are no limits to the rights  of monarchs. The comparable relation of a father to his children is  ambiguous in the same way in its implications. It can be used as a  natural model for the right ordering of states but it can also be  exploited to argue that fathers, like sovereigns, owe their right to  obedience to the services they perform for those who obey them.   In an age like the present, with its apparently irreducible plurality of  conflicting moral beliefs, the doctrine of natural law has lost much of  its appeal. But there is a minimum interpretation in which only those  who take the service of the state to be the highest conceivable duty for  man could possibly reject it. Itcan be taken to say simply that there are  moral considerations by which the state’s claim to authority must be  judged. Unless one holds with Hegel that private morality is a crude,  primitive anticipation of the higher morality of positive law or,  attaching no meaning to moral discourse, abstains from it altogether,  one cannot consistently oppose this position.   A feature of natural law doctrine that has been much objected to,  notably by Bentham, is its claim that there are imprescriptible natural  rights, rights possessed by all men whose infringement by the state  strictly entails the forfeiture of the state’s authority. Utilitarians    8 POLITICAL PHILOSOPHY    deny that any specific moral principle is absolutely and unqualifiedly  valid in the way that this kind of natural right is held to be. They  would say that to ascribe a right to a man is to say that there is some-  thing that he ought to be allowed to do. A rightis natural if itis moral  and not positively conferred by a state. Now what people ought to be  allowed to do varies with the circumstances. It is always possible to  conceive circumstances in which it would be morally reasonable, in the  best interests of all concerned, to abridge freedom of speech or  movement or occupation.   Some analytic philosophers have made curiously heavy weather of  the word ‘natural’ in the phrase ‘natural rights’, taking it to imply  that the possession of these rights is deducible from the nature of man.  The nature of man is the set of defining characteristics in virtue of  having which things are identified as being men. These characteristics  are in fact empirical and so no conclusions about what men ought to  be allowed to do can be extricated from the concept itself. This is  rather laborious. ‘Natural’ here means simply ‘non-legal’. Natural  rights are those which men have by reason of being men and not in  virtue of their membership ina particular politically organized society  with its prevailing system of legal rights.   In practice there is a good deal of correspondence between the  content of positive law and natural law, if this is understood as the  broad moral consensus of the citizens. Unless the state in question  is very efficiently tyrannous, indeed, there must be, since if a state’s  positive law is morally repugnant to most ofits citizens they will havea  reason for disobeying it and in doing so they will remove the effective-  ness which is one of the state’s essential characteristics.   But there is always some divergence between positive law and the  generally accepted hard core of morality. In the first place much of  the moral consensus cannot be or need not be or should not be legally  enforced. It cannot be enforced if offences are very hard to detect (for  example, indulgence in sadistic fantasy). It need not be if the informal  sanctions of morality are sufficient to maintain good behaviour (for  example, ordinary politeness). It should not be if the type of conduct  involved derives most or all of its value from the fact that it is freely  undertaken.   Secondly, positive law needs, for a number of reasons, to be very  precise, something seldom true of the moral consensus. Thirdly, since  positive law needs to be stable it cannot be adjusted to correspond  with every apparent shift of the moral consensus. It follows that it  will be slow to adjust itself to real shifts. A final point is that the most    INTRODUCTION 9    generally interesting decisions and regulations of modern states  concern issues of a broadly economically distributive kind which do  not fall clearly within the moral concensus. The main political divisions  of a modern industrial society concern the proportion of the national  income that shall be taken and spent by the state and the objects on  which it should be spent. Complex conflicts of justice and efficiency,  of benevolence and personal freedom arise here which are the main  topic of everyday political controversy. One virtue of democratic  systems is that they provide machinery for the resolution of these  conflicts of interest and principle by pacificatory compromise and  without resort to violence.   Two items in this anthology deal with subjects discussed in this  section. Professor Hart argues that the existence of at least one  natural right must be admitted if it is allowed that there are any  moral rights at all. Mr Benn conducts a wide survey of the uses to  which the concept of sovereignty has been put.    3. Political obligation   The problem of political obligation—why should I, or anyone, obey  the state—has always been the fundamental problem of political  philosophy. The question it raises must be distinguished from. two  others with which it can easily be confused and it is also somewhat  ambiguous itself. First, to ask why I should obey the state is not to ask  why I do, though one answer to the latter question (viz. because I think  I ought to) raises the former. People obey governments and abide by  the law to a very great extent, no doubt, from force of habit and  because it does not often occur to them to do anything else. When the  possibility of disobedience does-occur to them, in cases where there is  an obvious clash between the demands of the state and private interests  or moral conviction, they are restrained by fear of the probable con-  sequences of disobedience. But people may also be prompted to obey  the state by the conviction that they are morally obliged to do so. To  inquire into the justification of this belief is to confront the problem  of political obligation.   The other question with which that of the justification of obedience  is often confused is that of how the state and its laws came into exist-  ence in the first place. The two issues look so very different at first  glance that it may be hard to understand how they can ever have been  run together. One explanation is that many obligations arise from  something that has happened in the past: from a positive undertaking,  for example, as in a marriage ceremony, or from the coming into    10 POLITICAL PHILOSOPHY    existence of a particular state of affairs, as when a man recognizes his  responsibility for injuries caused by his carelessness or for children  brought into the world by his sexual activities.   A related point, emphasized by Hume, is that people generally  regard prescription or customary acceptanceas the solidest foundation  ofa right, though this may be less true now than it was in the eighteenth  century. Another consideration is that when the state was created, if  there ever was such a moment, the question of whether to obey was a  live issue for everyone involved. More generally the confusion between  the two questions is assisted by the habit of describing the problem  of political obligation as concerned with the origins or foundations of the  state, a mode of expression which can be interpreted historically or  justificatorily. At any rate the two questions are entirely distinct. As  Hume argued, even if the first states did originate in a contractual  agreement between their members this has no bearing on our situation  now: we do not inherit our ancestors’ promissory obligations, and the  states we live under originated for the most part in violent seizure of  power.   With the question why I should obey the state extricated from others  with which it may be confused, we can go on to consider what precisely  it means. It has usually been taken to ask how it is that I am undera  moral obligation to obey the state. But this is not the only meaning it can  have, nor is it the most fruitful one. Morality, strictly so called, has  no proprietary hold over the word ‘ought’. We can also ask what  makes it reasonable, sensible or prudent for me to obey the state. This  goes to the opposite extreme from the narrowly moral interpretation  of the question and might seem to invite only such obvious and un-  illuminating answers as that I am likely to be sent to prison if I do not  obey. But between the two extremes there is a third possibility. We may  ask: what makes it a generally good or desirable thing for me, or  anyone, to obey the state? Here the rationality of political obedience  is identified neither with its moral obligatoriness nor with its con-  duciveness to strictly personal interest and advantage. A great deal of  what we ought to do is reasonable in this sense without being either  morally obligatory or immediately advantageous.   There are three main kinds of solution to the problem of political  obligation. First, there are what I shall call intrinsic theories, which  derive the rationality or obligatoriness of obedience from the intrinsic  character of the state. Secondly, at the other extreme, are extrinsic  theories, which justify the state by reference, direct or indirect, to the  purposes it serves, to the valuable consequences which flow from its    INTRODUCTION 11    possession of effective power. Finally, there are organic theories,  which transform the problem by arguing that it implies a mistaken,  ‘abstract’ conception of the relations between the state and the  individual citizen.   The simplest of intrinsic theories is traditionalism, the view that the  state ought to be obeyed because it always has been, Hume’s pre-  scription in its most elementary form. A historically important variant  is the divine right theory which holds that we should obey the state  because God has commanded us to do so. The theory of divine right  can take a legitimist form, in which the criterion of divine author-  ization is something other than the possession of effective power, or  it can be conformist, enjoining obedience to the powers that be  whoever they are and however they acquired their position.   A more intellectually appetizing kind of intrinsic theory is the  doctrine of aristocracy, which attributes intrinsic authority to the  best people, picked out by their wisdom, ancient lineage, heroic  qualities (as in fascism) or even wealth. In practice intrinsic theories  soon lose their formal purity since any rational argument to justify  obedience to traditional rulers, or to the best people, must rest on the  pre-eminent capacity of the recommended rulers for realizing ends  desired or valued by those called upon to obey them. This fate of  intrinsic theories shows their affinity to deontological accounts of  morally right action, which are liable to the same loss of identity.  According to the deontologist such moral principles of right action  as that one should keep promises or tell the truth are self-evident to  the moral intelligence. They do not need justification in terms of the  valuable results of general adherence to them and are only harmed and  enfeebled if such justification is attempted. They do not need it since  they retain their validity in cases where good results do not accrue:  one should keep a promise even though no-one will be better off for  one’s doing so. If such principles are made dependent on the  production of good consequences, it is argued, morality is degraded  into calculating expediency. But few deontologists are brazen enough  to insist that a trivial promise should be kept whatever happens, that  one should leave someone drowning in a lonely spot to his fate in  order to make a promised appearance at a tea-party. A rigidly de-  ontological theory of political obligation, one that holds the  principle that one should obey the state to be simply a self-evident  truth, is conceivable. But this will not be very plausible unless ‘state’  is redefined in terms which guarantee that only rulers who rule well  qualify for the descriptuon.    12 POLITICAL PHILOSOPHY    Extrinsic theories are the political correlates of teleological accounts  of morally right action which define a right action as one from which it  is reasonable to expect good consequences. In the doctrine of the  social contract, the most famous of extrinsic theories, the connexion  between obligation and good consequences is indirect. According to  the contractarian I ought to obey the state because I have somehow  promised or undertaken to do so. But the commitment from which my  obligation arises is not conceived as arbitrary, purposeless and un-  conditional. It is entered into for the sake of some ultimate end (for  example, security in Hobbes’s version, the protection of natural  rights in Locke’s). Its binding force is conditional on the effectiveness  of the state in realizing the end in question. For this reason a contract  theory can never be absolutist. It cannot, in the manner of some  intrinsic theories, assign unlimited authority to the state. Political  obligation may always lapse and the state’s authority be forfeited if  the conditions of the contract are not satisfied.   Two main objections to the contract theory should be mentioned.  First, since most people give no explicit undertaking to obey the state,  there is a difficulty about identifying the thing they do which is to be  interpreted as their making an implicit promise to obey. There is a  dilemma here. If the supposedly contractual act is not voluntary, such  as passively benefiting from the protection of the armed forces, it  cannot be regarded as a promise. If it is voluntary, such as voting in  an election, failure to perform it is not generally recognized as  relieving a man from his obligations as a citizen. Secondly, there is  Hume’s favourite objection that the good ends for which the promise  was made are sufficient to justify obedience to the state by themselves  and without the intermediary of a highly speculative act of moral com-  mitment. This leads to the conceptually more economical view of  utilitarianism, that obedience to the state is justified on directly  teleological grounds as a necessary condition of the general welfare,  the advantage of society at large.   The organic theory of political obligation is implied by the doctrine  of a general or real will advanced by Rousseau and Hegel. The theories  considered so far, intrinsic and extrinsic, conceive the fundamental  political situation as one in which some men, the citizens, are seen as  quite distinct from and wholly subordinate to others, the state. This,  the organic theorist maintains, is at any rate unnecessary and un-  desirable and perhaps, metaphysically considered, is an illusion. In  any properly constituted political system, perhaps in any effectively  functioning one, the state is or represents the better selves of the    INTRODUCTION 13    citizens, their real, general, impersonal, moral will as contrasted with  their private, particular, irrationally self-regarding will. In a political  system so conceived the citizens in obeying the state are following the  promptings of their real or better natures, subjecting their irrational  and self-interested passions to the control of their social and moral  reason. Rousseau thought that an organic and genuinely obligatory  political system was hard to attain, possible only in communities with  small populations and directly democratic institutions. Hegel believed  that it was approximated to in every effective state, to the extent at  least that it was historically possible that it should be. Rousseau’s  hyperdemocratic ideal seems as impossible of achievement as Hegel’s  bland redescription of the facts of political life seems unrealistically  complacent.   The analytic philosopher of politics does not give the general  problem of political obligation so central a place in the subject as his  traditional predecessors. It has the merit of raising conceptual  questions about arguments designed to establish the rightness of  action and of drawing attention to the difference between power, the  ability to secure obedience, and authority, the right to expect it. But  of more interest than the problem they have in common are the  different values which theorists of political obligation, to the extent  that their reasoning is teleological, see itas the state’s justifying function  to serve. The problem of political obligation represents the citizeri as  confronted by a single absolute choice between obedience and  resistance, between conformity and treason. Even in the least demo-  cratic societies the scope of an individual’s political action is seldom  so brutally circumscribed. Whether or not he has the formal right to  vote, to organize political associations and to convert others to his  way of thinking, he will have many means at his disposal for bringing  pressure to bear on the government, its acts, its composition, its  institutional form. The values that are relevant to the ultimate choice  between submission and rebellion are also relevant to a much more  extensive range of political choices. It is more profitable to consider  the ends of government on their own, detached from their traditional  involvement with a single extreme issue of political action.   The selections in this anthology from Professor Peters and Mr  Winch make up a debate about the correct interpretation of the  concept of authority which it is the aim of theories of political  obligation to explain. The paper by Mr Barry, in defending the  concept of the public interest, considers issues raised by Rousseau’s  notion of the general will.    14 POLITICAL PHILOSOPHY    4.The ends of government   An ideology prescribes ends for government. It lays down certain  ends as those to be pursued through political activity and through  political institutions. The simplest kind of ideology describes an  ideal society or utopia in which the ideologist’s values are fully  realized. Here the ideological aim is quite explicit. At the other  extreme a theory of political obligation can serve an ideological  purpose more indirectly. In it the preferred ends will appear as  necessary conditions for justifying the state’s authority.   There are objections to both procedures. Utopias, concentrating  on the long-range goal of political endeavour, neglect the problems  that arise about getting to the destination. Not all of these problems  are practical. The realization of one part of the ideal may bring un-  expected results in its train which obstruct the realization of the  remainder. One ideological defect of theories of political obligation  was pointed out at the end of the last section. The scope of an  individual’s political action is not confined to deciding whether or  not to obey the state. He can usually bring some influence to bear on  its selection of policies, its composition and its institutional form,  even in societies that are not formally democratic. Another defect is  that the essential conditions of political obligation, though they will be  included in one’s ideals, are not usually wholly coincident with them.  Only if I take the wildly extreme position of refusing to admit an  obligation to obey any government but a wholly ideal one will the  conditions of political obligation and the principles of political action  in general be identical.   The first task of the political philosopher in this field, and on one  view his whole responsibility there, is to clarify the concepts of  political ends. In the light of such a clarification he can critically  examine the arguments that are used to support the choice of political  ends. The conflicting conceptions that prevail of the political ends he  is concerned with express ideological disagreements and this makes  it hard to operate with strict neutrality and detachment. Opposing  ideologists try to pre-empt words like liberty, justice and democracy for  the type of political arrangement they favour. The political philo-  sopher can keep himself from being embroiled only if he confines  himself to articulating the way in which different ideological groups  use the terms in which they proclaim their ideals.   A plain example of this kind of ideological competition over a  concept is provided by liberty. The negative conception of liberty  favoured by liberal individualists is repudiated by collectivists in the    INTRODUCTION 15    interests of positive liberty. Negative liberty is absence of interference  by states, groups or individuals with the activities of individual men.  For interference to be an infringement of liberty it must be directed  against activities those interfered with actually want to carry out, it  must be intended to have this effect and it must work through dis-  incentives serious enough to be proper objects of fear. Positive liberty,  being commonly defined as the ability to do what I really want to do,  turns out to be very much like my ability to do what I ideally ought to  want to do. The conflict is not resolved by simply giving different  names to the two kinds of liberty and recognizing that one party  favours the one and its opponents the other. For both parties agree  that liberty ultimately consists in being able to do what you want to  do. But they disagree as to what this is and about how weare to find out  what it is.   There is a similar distinction between competing concepts of  democracy. Here, however, it is the positive conception that is the  more traditional: the view of Rousseau thata state is democratic to the  extent that its acts express the common will of its citizens. The  opposite view conceives democracy as a peaceful way of getting rid of  governments with which the majority of the citizens are dissatisfied  rather than as a means for the direct realization of their political aims.  Both parties agree that democracy is in some sense government by the  people. As for the rest of Lincoln’s formula: all government claims  to be for the people and all government is of the people—of whom else  could it be?) They disagree about how this agreed purpose is best  brought about.   There is some slight analogy between these opposed views of liberty  and democracy and two views about the nature of justice. The negative  view would be that the state ought not to treat its citizens differently  unless there is some relevant difference between them. Its positive  opposite number is that the state should seek to eliminate or com-  pensate for the natural inequalities of advantage that there are  amongst them. It could be argued that there is no real difference here,  since what one side sees as a natural inequality which the state ought  to do something about the other side could recognize as a relevant  difference justifying difference of treatment.   The most elementary form of justice is the impartial administration  of the law. This can be represented as a kind of equality since it  involves no account being taken in the judicial treatment of citizens of  those differences between them that are not mentioned in the law itself.  But laws that are justly administered can still be unjust in themselves    16 POLITICAL PHILOSOPHY    if the differences of judicial treatment they prescribe are in some way  unreasonable. While few would deny that equality is one principle,  perhaps the fundamental principle, of justice, few would maintain  that it is the whole of justice. The principle of equal treatment must  be qualified by the recognition that people have different needs and,  because of the services they have done, different deserts. Justice  might seem to be the most comprehensive of political ends, with the  possible exception of the common good, but on any definition it can  come into collision with other widely shared values. Unequal dis-  tribution of income or property may be defended on the ground that  it promotes general economic welfare. By according privileges to a  naturally well-endowed minority it calls forth specially productive  effort. Those who favour the maintenance of some productive in-  equalities—and there are, as the practice of professedly egalitarian  societies suggests, few who would wish to exclude them altogether—  are reluctant to say they approve a measure of injustice, But in this  they are perhaps as unreasonable as those who find it hard to admit  that the penal institutions of society are designed to reduce the liberty  of evildoers.   Everyone agrees that it is an essential function of the state to  preserve the security of its citizens. Hobbes held its preservation by  a sovereign to be the sufficient condition of justified obedience to him,  thus placing it above all other political values. Later political theorists  have taken a less gloomy view of the costs of achieving it and have been  prepared to accept some risk to security for the sake of other political  ends. The general agreement there is about itaccounts for the fact that  it poses no serious conceptual problems.   One political value that has not yet been mentioned has had a ve  large influence on the course of political history but is seldom  emphasized in works of political theory. This is prosperity. In so far  as it does occur in theory it is as a slightly embarrassing aspect of the  common good. No doubt its somewhat unspiritual character is  responsible for this neglect. Until fairly recent times governments  have taken no very direct part in its pursuit. They have confined them-  selves to legal regulation of the conditions of economic activity by  controlling the currency, levying customs duties, limiting hours of  work, granting monopolies and so forth. Only in the last century have  they undertaken the direct management of productive enterprises and,  as a result of more extensive economic knowledge, taken up the  positive planning of the economy. The explicit ideological motive for  much of this extension of the state’s control of economic life has been    INTRODUCTION 17    socialistic, and has been based on considerations of justice rather  than of prosperity. A major problem here is to determine how largea  part of the common good material prosperity is. The ideology of  laisser-faire maintains that the common good will be most fully  realized in a society with a freely competitive economic system. But  the economic theory on which this ideology is based includes the  concept of social cost which applies to deprivations inflicted on the  community by the competitive pursuit of wealth which the market  mechanism does not correct.   The simplest way of recommending a political value is to assert  that men have a self-evident natural right that it should be secured  to them. The doctrine of axiomatic natural laws drew much of its  appeal from its connexion with the idea that the principles of morality  are divine commands. With the recession of that idea arguments of  a teleological kind have come to be generally relied on. In some cases  these arguments are utilitarian in the narrow, traditional sense. Such  is the inference that liberty is good because the kind of restraint in  whose absence it consists is unpleasant. On the other hand, in his  famous defence of liberty John Stuart Mill, a professed utilitarian,  recommended it as the indispensable condition for the discovery of  new truths and the preservation of old ones, without stopping to con-  sider the bearing of truth on utility in the sense of happiness.   Political theorists have very often fastened on one political end or  other as supremely valuable and have argued that everything a reason-  able man would consider good will be achieved by its pursuit. In doing  so they have been led to extend the concept of their prime value so  that it covers things far outside the original field of its application.  Socialists have represented poverty as a kind of unfreedom while  conservatives have objected to limitations of the privileges of wealth  as cases of injustice. But there is no need to assume that all political  ends are ultimately identical, that in pursuing any of them to the  limit we must in the end realize all the others. It certainly seems that  there are direct conflicts between them. Liberty and equality are often  at odds with one another, as are liberty and security, or prosperity  and justice. If the concepts of political ends are clearly articulated  and understood, an effective kind of rational discussion about them  is possible which has no real point if they are all so stretched that  they run into one another.   Four of the selections of this anthology concern the ends of govern-  ment. Professor Berlin discusses positive and negative conceptions of  liberty, Professor Schumpeter positive and negative conceptions of    18 POLITICAL PHILOSOPHY    democracy. Mr Carritt examines the relations, and particularly the  tensions, between liberty and equality, as Mr Barry does in the cases of  justice and the common good.    I    THE USE OF POLITICAL THEORY’  JOHN PLAMENATZ    J    EVEN IN OXFORD, which more perhaps than any other place in the  English-speaking world is the home of political theory or philo-  sophy, it is often said that the subject is dead or sadly diminished in  importance. I happen to have a professional interest in assuming  that it is still alive, and as likely to remain so as any other subject as  long as man continues to be a speculative and enterprising animal. J  do not think I am biased; I do not think I need to be. The importance  of the subject seems to me so obvious, and the reasons for questioning  that importance so muddled, that I do not look upon myself as  defending a lost or difficult cause. Political philosophy is dead, I have  heard men say, killed by the logical positivists and their successors  who have shown that many of the problems which exercised the great  political thinkers of the past were spurious, resting on confusions of  thought and the misuse of language. Apply the solvent of linguistic  analysis to these pretentious systems, they say, and when the dross  has melted away, little that is valuable remains. I think that this isa  mistake, and I want to explain why I think so.   I admit that the great political thinkers have raised many spurious   roblems, that they have been confused and have misused language. I  believe that those who study their theories ought to subject them to  close and rigorous criticism. J believe that they made many mistakes;  but I do not believe that they were mistaken in trying to do what they  did. I do not believe that the progress of science and philosophy has  left no room for their kind of activity.   By political theory 1 do not mean explanations of how governments  function; 1 mean systematic thinking about the purposes of govern-  ment. Perhaps it would be better to speak of political philosophy  rather than of political theory, keeping the second expression for what   From Political Studies, Vol. 8 (Clarendon Press, 1960) pp. 37-47. Reprinted by    permission of the author and the Clarendon Press.  ‘This article is based ona lecture given at the University of Exeter on 13 March 1959.    20 JOHN PLAMENATZ    purport to be explanations of the facts. If I have not done this, it is  because the word philosophy is nowadays used in a narrower sense  than it used to be, especially in English-speaking countries. The  political theory that I wish to speak about is emphatically not linguistic  analysis. It is a form of practical philosophy; it is practical philo-  sophy as it relates to government. I want to argue that it is a serious  and difficult intellectual activity, and that the need for it, in modern  times, is as great as ever it was—indeed much greater. It is not aneed  which disappears with the progress of science (and especially of the  social sciences), and is in no way weakened by the achievements of  contemporary philosophy.   It is not a less urgent need than it was; it is only a need less easily  satisfied.    ‘    I]    The belief that political theory or philosophy is dead rests on several  misconceptions.   1. In the past, political theory has often been a mixture of two  activities: it has sought to explain how government functions or how  it arose or why it is obeyed, and it has also put forward opinions  about what government should aim at and how it should be organized  to achieve those aims. These two quite different activities have not  always been kept distinct. Indeed, the Utilitarians were among the  first to insist that they ought to be so kept, though they did not always  take their own advice. Both these activities are useful. But, for reasons  which are not far to seek, the fact that they have so often been confused  has brought discredit on one of them much more than on the other.  In this scientific age, the explanation of what actually happens is  always respectable. We must have theories about how this or that form  of government functions; we must even have theories about govern-  ment in general, we must take notice of what is common to all forms  of government. These are all theories that can be verified; they are  attempts, more or less successful, to extend our knowledge. But, it  is said, theories about what government should aim at and how it  should be organized do not extend our knowledge; they merely  express preferences, while pretending to do much more. They vary  from age to age, from country to country, from party to party, from  person to person. It is conceded that they affect action and that  therefore we need to know what they are and how they arise. But it  is also taken for granted that what they are matters less than how they  affect men’s behaviour; that it is more important to inquire into    THE USE OF POLITICAL THEORY 21    their origins and consequences than to study them for their own sake.  They have to be studied because people have in fact taken them  seriously and been influenced by them, but reasonable men can do  without them.   There has been in recent times some resentment of, and contempt  for, political philosophy. It is said of it that it not only pretends  to give us knowledge but also stands in the way of our getting it.  Durkheim, in his Rules of Sociological Method, argued that political  theorists, in order to reach the conclusions they want to reach, define  the terms they use in such a way as to make it seem to follow from  their definitions of the state or of law or even of human nature that  government should aimat this rather than thatand should be organized  in one way rather than another. Political theorists, it is said, have   roduced concepts which stand in the way of a scientific explanation  of the facts because their real (though unacknowledged) function is to  justify what the theorists happen to think desirable.   It is certainly true that political theory or philosophy does not  produce the same kind of knowledge as political science, and itis also  true that it has stood in the way of political science.’ But even if  political philosophy has stood in the way of political science, that is  no reason for dismissing it as fantasy or the mere airing of preferences.  It is only a reason for distinguishing it from intellectual activities  of other kinds.   2. What has gone by the name of political philosophy in the past  has been shown to be remarkably confused. This has caused some  people impressed by the confusion to speak as if what the political  thinkers of the past attempted were not worth doing, and as if the only  useful function of political philosophy were to dissipate confusion.  Political philosophy, they say, is properly the analysis of political  concepts.   I do not deny the need for this analysis, and though I should wish to  use the expression political philosophy in a wider sense, I do not quarrel  with its being used thus narrowly. But if this is to be called political  philosophy, there still remains another intellectual activity, which is  neither political science nor political philosophy, which is more  important than the second and not less important than the first,  and which is likely to endure when political philosophy, in this narrow  sense, has lost what importance it now has. I should not wish to   11 think this second charge exaggerated: I suspect that it was much more ignorance    than the failure to distinguish between explanation and advocacy which impeded the  progress of political science. But I am not concerned to argue this point.    22 JOHN PLAMENATZ    quarrel about names. If, for instance, Mr. Weldon had wanted to do  no more than make a narrow use of the expression political philosophy,  I should neither have followed his example nor condemned him for  trying to set it. But I suspect that he wanted to do more than this; I  suspect that he wanted to suggest that, apart from political philo-  sophy, as he understood and practised it, and political science, there  was nothing important, difficult, and useful to be done by rigorous  and systematic theorists in the field of politics.   Political philosophy, understood in Mr. Weldon’s sense, is not  likely to remain important for long. At the moment, because political  thinkers still use ambiguous concepts, the careful analysis of these  concepts is still needed to show that many traditional problems are  spurious, arising only because the men who put them have fallen  victims to the confusions and intricacies of language. Since these  concepts are often borrowed from, or shared with, other studies  traditionally known as philosophy, the philosopher is better placed  than other people to show how they generate spurious problems. This  is an important service which the philosopher, in the narrow sense,  can still do for the student of politics. But those who practise this  kind of political philosophy should notice their own limitations.  When they show us what confusions of thought there are in Rousseau’s  doctrine of the general will or in Hegel’s doctrine of the state,  we have cause to be grateful to them. They see the nonsense in these  doctrines, and they explain what makes it nonsense. So far their work  is useful. If, however, they go further, they risk doing harm. They  are too ready to assume that where they have seen nonsense there is  no sense which they have not seen. Just as it takes some skill in  linguistic analysis to see the nonsense in Rousseau, so perhaps it takes  some knowledge of sociology and psychology to see the sense in him.   The philosopher in this narrow sense already does no service to  the natural scientist. He studies scientific method as the scientist  does not study it; and there is therefore a sense in which he under-  stands what the scientist does better than the scientist himself under-  stands it. He knows better than the natural scientist how science  differs from other kinds of intellectual activity. Yet he has nothing  to teach the scientist, for what he knows about science that the  scientist does not know is not knowledge needed to make a good  scientist. The philosopher does not help the natural scientist to either  his ideas or his methods. And so it will eventually be with the political  scientist; the time will come when he will need no more help from the  analytical philosopher. He needs him, even now, only to rescue him    THE USE OF POLITICAL THEORY 23    from confusions of thought; he does not need him, any more than the  natural scientist does, to produce the ideas he uses to explain the  events he studies.   3. The great variety of theories about what government should  aim at and how it should be organized has discredited these theories.  It is said of them that they do no more than expound the preferences  of their makers, and that in any case they are socially determined.   I do not see the force of these objections. What does the variety  of these theories prove about them? That they are not true? But if they  expound preferences, the objection is out of place. Itis unreasonable  to argue that they are not scientific, and then to object to them that  they are not true. They are neither true nor false.   Does the variety of these theories prove them unimportant? In just  what sense? Will anyone deny that they have had a large influence on  the course of history? The fact that they have not served as blue-prints  for the reconstruction of society is no evidence that they have not been  important. They have powerfully affected men’s images of themselves  and of society, and have profoundly influenced their behaviour.   Does the variety of these theories prove that we no longer need  them? I do not see that it does. What are we to have in their place?  Political science? But its function is not the same. It does notattempt  what these theories attempt. Why then should it supersede them? And  we can say the same of political philosophy as Mr. Weldon understood  it. Its function is different. It does not satisfy the same need. And  just as political science and the analysis of political concepts do  not satisfy this need, so they do not remove it. It is still there, no  matter how active and successful they may be.   Are these theories unimportant because they are socially deter-  mined? The production of such theories is an activity of man in  society, and is therefore affected by his other social activities. All  social activities limit one another. What men can do or even imagine  in one direction is limited and affected by what they can do and  imagine in other directions. We may agree that a theory like Marxism  could not have been produced in the Dark Ages. But then neither  could the steam engine have been produced then. The feasible and  the imaginable are limited by the actual. This is as true of industry  and science as of political theory.   We soon get into difficulties if, like Marx, we treat political theory  as of secondary importance. Marx called it a form of ideology or  false consciousness, contrasting it with science, which gives us real  knowledge; and he looked forward to the day when we should have    24 JOHN PLAMENATZ    true social science and be able to dispense with ideology. Yet he could  not help attributing great importance to ideology. A class, to be  politically effective, must have an ideology; and unless it is politically  effective it has no active role to play in history. Ideology is illusion,  and yet, unless men had these illusions, the course of social evolution  would not be what it is.   We have here an example of a type of simple and false reasoning to  which many people—and sometimes even philosophers—are still  prone. They show that one kind of theory is mistaken by its producers  for another kind, and then conclude that the second kind supersedes  the first. Marx’s version of it is this: the makers of ideology mistake  it for science, and therefore when science shall have come into its  own, there will be no room for ideology.   Marx made one kind of mistake, Burke made another. He thought  that political theory, except when it justifies the established order  is harmful. That, at least, is to admit its importance. Yet Burke,  because he did not see clearly the function of political theory, mis-  understood the French revolution. He saw the revolution as a disaster  caused by people’s being misled by the philosophers. Its immediate  cause was that the unprivileged classes were making new claims on  society, claims which could not be met unless society was greatly  changed. The philosophers did not create the conditions that disposed  the unprivileged to make these claims; their task was rather to  formulate the claims, toexpound them systematically, and to condemn  the old society which could not meet them. It was useless to rail at  the philosophers for disturbing society. It is true that there was no  overt demand for the theories they produced. There never is a  demand for such theories in advance of their appearance. But there  was a readiness to accept them when they appeared. There was, in that  sense, a need for them. Burke’s mistake was in not understanding this  need, and Marx’s in speaking as if the need would disappear when  the social studies had become scientific.   It may be true that the need is more difficult to satisfy the more  the social studies become scientific. It may also be true that, because  of the discredit into which traditional political theory has fallen,  the need is less widely recognized, especially among intellectuals,  than it used to be. The old political theorists did so many things  which they ought not to have done that we are tempted to conclude  that there is no longer a need to do anything that they did. We may  admire their fantasies, and yet say that the ime for fantasy is over.  We may say: By all means, let us state our preferences if we feel so    THE USE OF POLITICAL THEORY 25    inclined ; let us make explicit the rules of conduct and the ideals which  we accept. But this is something altogether more modest than what the  old political theorists attempted.   There is some truth in this way of thinking, but it falls so far short of  the whole truth as to be profoundly misleading. I want to explain  why this is so.    Ill    In primitive societies, custom and prejudice are perhaps sufficient  guides to conduct. And by prejudice I mean here what Burke meant by  it; I mean a belief about right conduct which the believer takes on  trust. In primitive societies, men can perhaps do without a systematic  practical philosophy, just as they can do without a dogmatic religion.  In the eyes of a sophisticated student of a primitive society, the customs  and beliefs belonging to it may forma coherent whole; he may see how  they fit together to make it the peaceful and contented society which it  is. But in the eyes of the primitive man, they are not a coherent but  only a familiar whole; he does not see how they fit together, he merely  lives comfortably with his neighbours and with himself because in fact  they do fit together.   The sophisticated man needs more than a set of customs and  prejudices which are in fact coherent, though he does not see that they  are; he needs a practical philosophy. He lives in a changing society,  and he is socially mobile in that society; he is not exposed to change  which is so slow that he cannot perceive it. He lives in a society where  men strive deliberately to change their institutions. If he is not to  feel lost in society, he needs to be able to take his bearings in it; which  involves more than understanding what society is like and how it is  changing. It also involves having a coherent set of values and knowing  how to use them to estimate what is happening; it involves having a  practical philosophy, which cannot, in the modern world, be adequate  unless it is also a social and political philosophy.   In the past practical philosophy was rooted in religion and meta-  physics; men derived, or purported to derive, their beliefs about  how they should behave and how government should be organized  from God’s intentions for man or from the nature of the world or  from man’s being a rational creature. But many of the teachings of  religion and metaphysics have been undermined by science or by  logic; they have been shown to be incompatible with the facts or to  rest on confusions of thought and bad argument. Not all religious    26 JOHN PLAMENATZ    and metaphysical doctrines have been directly controverted; for  many have referred to an order of realities supposed to be beyond  the realm of ordinary experience, with which alone science is con-  cerned. They are beyond the reach of science, and logic cannot touch  them if they are self-consistent. Yet the spread of science disposes  many people to reject even these doctrines. They reject not only what  science can show to be false, they also reject what science does not  show to be true. Though there is, perhaps, nothing irrational about  having both unverifiable and verifiable beliefs about the world,   rovided the first beliefs do not conflict with the second, many  people find it difficult to do this, and feel the need to reject all  beliefs for which there is no evidence. They may, of course, reject  them consciously, and yet also behave as if they believed them; which  is irrational. But that possibility does not concern us.   With the decay of religion and metaphysics there has gone a de-  preciation of the practical philosophies so long connected with them.  There has even been a change of attitude to the moral principles  contained in these philosophies. Let me give an example to illustrate  my meaning. ‘All men are equal in the sight of God’ is a statement  about God’s feelings and intentions for man; it purports to be a  statement of fact. It is not, on the face of it, a value-judgement; it  is descriptive and not prescriptive, and yet it is unverifiable. Connected  with this statement are beliefs about how men should behave. These  beliefs do not follow logically from the bare statement about God’s  feelings and intentions; they follow only if it is assumed (as of course  it always is) that men ought to behave in ways that further God’s  intentions for them. Though, when a man ceases to believe in God,  he is not committed to rejecting these beliefs, since they do not follow  logically from the statement that God has certain intentions, he is  inclined to feel less strongly about them. It is only when these beliefs  are put before him in some other connexion, as parts of some other  intellectual structure, that he is again disposed to accept them as  fervently as he did before.   The attempt to derive moral principles from theology or meta-  physics is a time-honoured way of putting them forward as principles  which all men everywhere do or ought to accept. Therefore, when  this manoeuvre is rejected, so too is the idea that there are universal  principles. It is admitted that there-always are moral principles and  that there always will be, it is admitted that the study of what they  are and how they arise is valuable. But the task of elaborating a  systematic practical philosophy is depreciated, it is what the theo-    THE USE OF POLITICAL THEORY 27    logians and metaphysicians used to do. It is what they still do, though  with less conviction now than when their labours were not greeted  with scepticism. The task made sense to them; but how can it make  sense to men thoroughly imbued with the scientific spirit?   Now, this attitude to practical philosophy is quite irrational. The  need for it is there, whether or not it is possible to derive universal  principles from beliefs about God or the world or man; it is there,  whether or not it can be shown that there are principles which men do  or ought to accept everywhere. Man today, much more than in the  past, must get his own bearings in the world; he must make himself  at home in the world, for he can no longer be at home in it merely by  conforming to the conventions and acquiring the prejudices of his  station in society. Indeed, he no longer has a station, as his ancestors  did; he is much more socially mobile in a much more quickly  changing society.   Self-conscious, sophisticated man’s conception of himself does not  consist only of what he knows about himself or thinks he knows; it  consists also of what he aspires to be. Admittedly, he is not what he  aspires to be; he is what he is. But the kind of image he has of himself  depends largely on what he aspires to be. He does not get his aspira-  tions from the sciences, not even the social sciences; he gets them,  directly or indirectly, from practical philosophy, whether or not  that philosophy is tied to religion or to metaphysics. He cannot live  from hand to mouth, following custom and accepting all current  prejudices as they come. He lives in a kind of society which makes him  critical and self-critical. To be happy, he must have aspirations,  and must also feel that he can live up to them; he must be true to  some image of himself. If he wants what he cannot get, or wants in-  compatible things, or has ambitions that bring him into conflict  with other men, he cannot be happy.   Not everyone is capable of acquiring for himself a coherent prac-  tical philosophy. Not everyone feels the need for it. There are doubt-  less some people—and who knows how many they are?—who are  quite content to drift through life. There are others who need guidance  but are incapable of philosophy. They seek guidance from churches,  from political parties, and from other organizations, and also from  friends. There are still others who make for themselves a practical  philosophy without engaging in controversy or adding anything to  the stock of ideas and arguments. But some there must be who do  the systematic thinking which goes to the making of practical philo-  sophies. They are not scientists; their business is not to explain what    28 JOHN PLAMENATZ    happens in the world. And they are not philosophers in the rather  narrow contemporary sense; their business is not to explain how  we use language or how we get knowledge or what exactly it is that  we are doing when we pass moral or aesthetic judgements or when we  make decisions. They are philosophers in a quite different sense: they  try to produce a coherent system of principles and to establish what  needs to be done to enable men to live in conformity with them. They  do not merely examine and compare the principles, showing where  they are incompatible and explaining their consequences; they do not,  like honest shopkeepers, display a large variety of goods, describing  them all accurately and leaving it to the customer to choose what  pleases him best. They produce a hierarchy of principles, and try to  explain how men should use them to make their choices. This is how  they help to provide them with a practical philosophy.   If the producers of these theories were like honest shopkeepers,  if they were mere purveyors of ideas, they could not meet the need  which it is their function to meet. If their business were merely to  explain what this or that principle amounts to, how it fits in with  other principles, and what is likely to happen when it is acted upon;  if their business were to offer a large variety of principles, or even  philosophies, for consideration, inviting every man to make his own  choice among them, they would only bewilder and annoy. But they are  not mere purveyors of ideas; they are preachers and propagandists.  They are people who have, or who believe they have, discovered how  men should live; and they will not be listened to unless they speak  with conviction. They need not all speak with one voice, but each of  them must take his stand. This is a condition of their effectiveness.  If every missionary were to explain several different religions to his  listeners, leaving it to them to make a choice, religion would take  no hold. A man must already be committed before he can do much to  help other people to commit themselves. As it is with missionaries,  so it is with philosophers of this kind. Their business is to help people  commit themselves. Freedom of thought is preserved, not because  each thinker offers several theories for inspection and choice, but  because different thinkers offer different theories with equal con-  viction. It is not the variety of strongly held convictions among the  intellectual ¢dlite which is bewildering and depressing; it is the  lack of conviction among them. Strong convictions attract and repel;  they do not leave people indifferent. They encourage those who have  the ability to do so to make up their own minds, to know where they  stand. They do what science and linguistic analysis cannot do.    THE USE OF POLITICAL THEORY 29    It is not enough that practical philosophies should be strongly  held; they should also be well thought out and realistic. They should  aim at self-consistency and at taking account of the facts.! The more  thoughtful they are, the more they encourage thought in the persons  who take stock of them. It does not matter that very few people should  swallow them whole. Whoever considers them seriously will usually  want to do more than establish their merits and defects, he will also  want to construct a practical philosophy for himself, and the more  they challenge thought in him, the more thoughtful that philosophy  will be.   The more men live in societies which change quickly, the more  mobile they are in those societies, and the more accustomed to the  idea that they can, by taking thought, change their social environment  to come closer to their ideals, the greater the part of social and poli-  tical thought in practical philosophy. Its business is to relate a  coherent body of principles to government; its business is to tell us  what government should do to realize those principles and how it  should be organized to do it. Political theory, as distinct from  political science, is not fantasy or the parading of prejudices; nor  is it an intellectual game. Still less is it linguistic analysis. It is an  elaborate, rigorous, difficult, and useful undertaking. It is as much  needed as any of the sciences. Its purpose is not to tell us how things  happen in the world, inside our minds or outside them; its purpose  is to help us decide what to do and how to go about doing it. To  achieve that purpose, it must be systematic, self-consistent, and real-  istic. We learn to cope with the world, not by collecting principles at  random, but by acquiring a coherent practical philosophy, which we  acquire largely in the process of considering other philosophies of  the same kind.    Practical bisenee gS is deeply affected by psychology and the social sciences.  Though we do not logically derive our values bent what we know (or think we know)  about ourselves and our social environment, we do change them as we change our  minds about the facts. No one has done more than Freud to change our standards of  sexual morality. Though these changed standards do not follow logically from his  psychological theories, people who accept the theories are more disposed than they  would otherwise be to accept the standards. But this detracts nothing from the im-   ortance or the distinctive Beir of practical philosophy. Art, too, is deeply affected   y science and by practical philosophy, and yet it is an activity of a quite different  kind which seems unimportant only to people who do not understand what it is. The  more our standards are liable to change, the greater our need for practical philosophy.  The greater our need, not just to understand how they have changed, but to introduce  order among them. The need for practical philosophy is part of man’s need to be his  own master, to make up his own mind how he shall live and what he shall be.    30 JOHN PLAMENATZ    No doubt only a small minority acquire, or are capable of acquir-  ing, a coherent practical philosophy. But then only a small minority  are capable of becoming scientists. We do not show that an activity is  unnecessary or useless by showing that onlya few persons engage in it.    IV    It may well be that no practical philosophy, and therefore no  political theory, is universally acceptable. There may be no set of  principles of which we can say: if men understood these principles,  and also understood what human nature is and might be, they would  accept them. I suspect that Marx and Engels believed the contrary.  They denied that a practical philosophy can be derived logically  from theology or from the nature of man, but they believed, none  the less, that the fundamental rules and values of the classless society  are universally acceptable, in the sense that men who understand  what man and society are and might become do accept them. They  expected the morality of classless societies to be everywhere the  same and unchanging.   To defend my thesis I need not go as far as Marx and Engels went.  I say only that the need for practical philosophy exists in all sophis-  ticated societies. Just as sophisticated man is a scientist and an artist  and an analytical philosopher, so too is he a practical philosopher  and a political theorist. Most men, of course, are not so, but some  are. Modern society creates a need for what they do which can neither  be destroyed nor met by science and analytical philosophy.   There is nothing illiberal about practical philosophy and political  theory, thus conceived. Admittedly, they are indoctrination; they are  not the mere sorting out of ideas and their implications. But there need  be no monopoly of indoctrination. In a liberal society there are some  principles common to all or most of the political theories currentin it.  There is both community and variety of beliefs. But the beliefs held  in common are as much open to question as the others. For society to  remain liberal, it is not necessary that these beliefs should not be  questioned; all that is needed is that they should be widely accepted.  The more men differ, and the longer they have been accustomed to  differing, the more likely they are to accept principles which make it  possible for those who differ to live peacefully together. The principles  commonly accepted are not more strongly held than the others; they  are merely held along with the others. The Catholic or the Protestant  who believes in toleration is not a liberal first and a Catholic or a    THE USE OF POLITICAL THEORY 31    Protestant afterwards, nor is he a less fervent believer than he would  be if he were intolerant. So it is also with political creeds; they are not  the less strongly held merely because those who hold them are  tolerant.    U  POLITICS, PHILOSOPHY, IDEOLOGY!    P. H. PARTRIDGE    My object is to comment upon what seem to me to be some typical  trends in current English and American political theory, having in  mind the view that has recently been constantly asserted, that political  theory has been in decline or may even have expired during the last  few decades.? I will argue that the report of death, even of decline,  is grossly exaggerated, that in fact the present period is unusually  fertile in thinking about politics that is not only original and  important, but is also, at any rate in many significant respects, entirely  in the traditions of ‘classical’ political theory. That is one half of  my thesis; the other half is that during the past few years some very  important shifts in interest, approach, and emphasis have certainly  occurred; and I shall make some suggestions about the character of  these shifts, and the reasons for them.   Unfortunately, those who have beer; announcing die decay or death  of classical political theory have as a rule taken less trouble to  establish the fact of death or decay than to assert its causes. I do not  know what kind or amount of evidence is necessary to prove that  political theory has declined in volume or quality, but the assertion  does not seem to be very plausible on its face. For one thing, changes  of this somewhat radical kind do not occur quite so dramatically as   From Political Studies, Vol. 9 (Clarendon Press, 1961), pp. 217-35. Reprinted by  permission of the author and the Clarendon Press.    'The writer is very deeply indebted to Professor Wilfrid Harrison for many sug-  gestions for the improvement of the form of this article.    'This is a selection from some of the more recent books and articles which have  discussed the present condition of political theory: A. Cobban. ‘The Decline of  Political Theory’ (Pol. Sa. Q., vol. Ixviii, no. 3, cae J. C. Rees, ‘The Limitations of  Political Theory’ (Pol. Studies, vol. ii, no. 3, 1954); P. Laslett (ed.), Philosophy, Politics and  Society (1956); G. C. Field, ‘What is Political Theory?’ (Proc. Arist. Society, vol. liv);  G. E. G. Catlin, ‘Political Theory: What is It?’ (Pol. Sc. Q., vol Ixxi, no. 1. 1957):  D. Braybrooke, ‘The Expanding Universe of Political Philosophy’ (Review of Metaphysics,  vol. xi, no. 4, 1958); J. P. Plamenatz, ‘The Use of Political Theory’ (Pol. Studtes, vol.  vii, no. 1, 1960); H. R. Greaves, ‘Political Theory Today’ (Pol. Sc. Quarterly, vol.  Ixxv, no. 1, 1960); A. Brecht, Political Theory (1950); H. V. Jaffa, ‘The Case Against  Political Theory’ (Journal of Politics, vol. 22. no. 2. 1960); L. Strauss, What is Political  Philosophy, and Other Studies (1959).    POLITICS, PHILOSOPHY, IDEOLOGY 33    this one is sometimes supposed to have occurred. Or, again, if we call  to mind the very large number of books that have dealt quite recently  with such central problems as the nature and conditions of democracy,  the group theory of political action and structure, and the theory  of bureaucracy, one might have thought that there has never beena  time when so much theoretical speculation about politics has been  going on. Clearly, when writers nevertheless maintain that political  theory is in decline, they must have something else in mind. Hence,  we should perhaps begin by considering briefly what are some of the  different things that political theory has meant, what different sorts  of thinking about politics the name has been applied to. Some kinds  may have become unfashionable, while others have continued to  flourish.    I   Classical political theory has usually been a mixture of different  kinds of inquiry or speculation. One could distinguish three different  impulses—philosophical, sociological, ideological. This paper will  be mainly about the third, but I want to say something first about  all three. The political speculation of Plato, Hobbes, Locke, or Hegel  is philosophical chiefly because each of these writers has tried to  connect his conclusions about political organization, or about the  ‘ends’ of political life, with a wider philosophical system. He has  tried to derive political and social conclusions from more general  beliefs about the nature of reality, to show that every sphere of  reality, including the political, possesses certain common features  or ‘categories’, that all these spheres can be spoken about in the same  logical language, that, in short, political conclusions follow from or  are supported by more general logical and metaphysical principles.  And one obvious reason for the current impression that classical  political theory is in decay is that there is not so much of this sort  of argument now: many philosophers now insist that one cannot  deduce the ‘rightness’ or ‘rationality’ of a form of political organiz-  ation, or of a political policy, from more ultimate principles. For  example, that section of Weldon’s Vocabulary of Politics which deals  with what he calls ‘foundations’ has this main philosophical purpose.   As I have said, one special form of the traditional connexion  between philosophy and politics starts from the conception that it is  the task of philosophy to exhibit what is common between the social  and other ‘spheres’ of reality. Historically, of course, this has been  a most important link between philosophy and politics; for instance,    34 P. H. PARTRIDGE    philosophical atomism has lent support to social individualism, the  dialectic and the concrete universal are as important in Hegel’s social  theory as in the more philosophical parts of his work; and more  recently some of the earlier political pluralists gained some support  from the criticisms of philosophical monism developed by William  James and others.   But clearly, a great deal of political theory has not been philoso-  phical in this sense. Although de Tocqueville or Graham Wallas  discussed some of the problems that political philosophers also  discussed, their political views were not systematically connected  with a philosophical position. A lot of de Tocqueville’s writing we  should now describe as sociological; he asserts generalizations,  including causal generalizations, about the behaviour of social  phenomena. Naturally, this is not only true of the de Tocquevilles,  Bagehots or Maines; even in the writings of the political philosophers  in the more technical sense, there is much sociological general-  ization—Hobbes is a notable example. Political philosophy was, of  course, one of the parents of contemporary sociology. And no one  doubts that the development of sociology as a specialized, altogether  more rigorous, subject has also affected general political and social  speculation.   By the ideological impulse I mean merely the form of political  thinking in which the emphasis falls neither on philosophical analysis  and deduction, nor on sociological generalization, but on moral  reflection—on elaborating and advocating conceptions of the good  life, and of describing the forms of social action and organization  necessary for their achievement. Of course one cannot draw a sharp  line between ideological and philosophical political writing: almost  all political philosophy has been ‘practical philosophy’ in that it has  had the practical object of persuading readers of the ‘rationality’ or  moral superiority of some specific form of social organization. Even  the Leviathan has such a practical aim, although Hobbes’s practical  conclusions are.grounded upon philosophical and other argument  which has usually been taken to be of greater interest than the  practical purpose. But Rousseau, unlike Hobbes and Hegel, was not  primarily a philosopher. The Social Contract introduces some general  distinctions and conceptions of a philosophical kind, the significance  and limitations of which were more developed by later writers, includ-  ing Green and Bosanquet. But one would not expect much to be said  about Rousseau in a history of philosophy—not, at least, in one  written by an Englishman. In his case it would be very easy to detach    POLITICS, PHILOSOPHY, IDEOLOGY 35    and stress the ideological element. In fact, this is what seems to have  happened: the image or model of democratic society drawn from the  Social Contract has been very much more influential historically than  any philosophical conceptualization or argument that the book  contains.   But no matter how we choose to classify any theorist, I think it will  be agreed that we can distinguish these three different impulses and  interests, and these three ‘orders’ of political thinking. Nor will any-  one dispute that one very powerful interest throughout the course  of European social thinking has been the ‘ideological’—moral  argument about ends and ways of life, and about the institutional  conditions of the good life.    IJ   Possibly each one of these impulses has grown weaker in recent  years. Undoubtedly the progress of detailed empirical political and  social inquiry has shaken the habit of speculative sociological general -  ization of the ‘philosophical’ kind, the concoction of what Dahl has  called ‘macrotheory’ in politics. Part of what is to be found in  classical political philosophy has now been absorbed into political  science and sociology. Again, many philosophers now reject the  conception of philosophy, and of the resulting connexion between  philosophy and politics, on which most of classical political philo-  sophy (in the strict sense) rests. I cannot here discuss the technical  reasons for the rejection: I shall make just one remark—addressed  to many political writers who have been lamenting the decay of  traditional political philosophy: this lament is idle if you have no  answer to the technical philosophical arguments which can be dep-  loyed against the practice of supporting conclusions about the  ‘functions’ of the State or about the rational ordering of social life  by resting upon ‘higher’ or ‘more ultimate’ logical and philosophical  principles.   But does it follow that these philosophical arguments have  produced a decay of philosophical political theorizing? I should  argue that there has not been such a decay. While philosophers have  been disclaiming competence in political discussion, political scient-  ists have been delighted to accept their disclaimer; and it can be  argued that what is now happening is only an example of the familiar  separation of a distinct subject from the parent philosophical stem.    ‘See his review article on de Jouvenel’s Sovereignty—‘Political Theory: Truth and  Consequences’ (World Politics, vol xi, no 1, 1958).    36 P. H. PARTRIDGE    Political science, like economics, and other sciences, has reached  the point where the specialists must themselves deal with the very  general matters, including conceptual analysis, popularly or tradi-  tionally called ‘philosophical’; and they do so at a level of sophistic-  ation and complexity which the philosopher qua philosopher could  hardly approach. The boundaries between philosophy and politics  are being redrawn. A number of twentieth-century philosophers have  drawn a very sharp line indeed between ‘philosophical’ questions  (or conceptual analysis) and ‘empirical’ questions; between ‘second  order’ and ‘first order’ statements about politics. By and large, the  political scientists have been sensible enough to see that the drawing  of this sharp boundary is hopelessly disabling for the study of social  theory, and have ignored it. On the other hand, the philosophers who  have imposed it have been left in occupation of only a wafer-thin  slice of the territory of politics—as The Vocabulary of Politics clearly  demonstrates.   The philosophical impulse, in fact, is not the most important part  of our inquiry; technical philosophy, after all, has only a limited  influence; many of the most important political theorists of the last  three centuries were not much influenced, or not influenced at all, by  the technical philosophy of their own time. The philosophical impulse  or influence is only one of those which have sustained political  speculation; others have been equally or even more important. This  brings me, then, to the ‘ideological impulse’ with which, in one way  or another, the rest of this essay will be concerned.    iil   The writers who have drawn attention to what they interpret as a  languishing of philosophical thinking about politics have mainly in  mind the political philosophy which has been an extension of moral  theory—which inquires into the ends of the State and its morally  right organization (What is the State that I should obey it?)—into  the morally necessary or morally justifiable ordering of political  society. The ‘decline of political theory’ is taken to be the decline of  the moral interest in politics. ‘On the one hand, there is a great deal  of eagerness to deal with politics in mora] terms; on the other, the  insights of psychology, anthropology and of political observation  have silenced the urge.’! It is this fact—if it is a fact—that is most in  need of explanation.    ‘Judith Shklar, Beyond Utopia (Princeton, 1957), p. 272. J..P. Plamenatz’s article  ‘The Use of Political Theory’ (Pol. Studies, vol. viii, no. 1, 1960) and H. R. Greaves,    POLITICS, PHILOSOPHY, IDEOLOGY 37    Is it a fact that there has been a ‘slackening’ of the ‘moral urge”?  Once again it is not hard to construct a case against. There is, after  all, a very large amount of contemporary discussion of the moral  foundations and dimensions of politics; for instance, the many  fashionable criticisms of secular or positivistic liberalism such as  are examined by Frankel in his Case for Modern Man. Nor should we  overlook the moral discussion to be found in the pages of sociologists  who are interested in the non-political areas of social life. One striking  (and perhaps strange) feature of recent American sociology has been  the popularization of the notion of ‘alienation’. This, together with  the closely connected themes of anomie, depersonalization, the  atomization allegedly inherent in large-scale industrial society, the  supposed dissolution of ‘community’ and aggrandizement of the  State and other depersonalized, bureaucratized forms of organiza-  tion, is surely one of the most widely-followed fashions of social  thought at present. Perhaps there is a good reason for this; perhaps  the social conditions and changes which appear to be morally most  significant are not changes, or possible changes, in political arrange-  ments or institutions, but those at other levels of society. The kind  of lives people live in modern cities, the demands that industrial and  other great organizations make upon them, the effects of commer-  cialism, industrialization, and ‘mass media’ on popular culture—one  could argue thatitis in such contexts thatthe more importantoccasions  for moral disquiet and reflection are to be found. In other words, if  the moral interest in politics has declined, one reason could be that  questions of political organization, of allocation of political rights  and powers, &c., are not at present generally felt to be morally  critical.!    IV   I want to develop this. There now prevails in England and the  United States and in several other Western-type democracies a quite  unusual degree of political relaxation and consensus. J shall not try to  state carefully what this consensus embraces; but it obviously  embraces the fundamental constitution of the liberal-democratic  order. There is no significant social or intellectual movement which  calls into question the broad structure of rights and powers under-    ‘Political Theory Today’ (Pol. Sc. Quarterly, vol. Ixxv, no. 1, 1960) are examples of  this line of argument.    'I am speaking of course of the Western democracies; the reader will immediately  protest against this remark if it is applied to contemporary Africa.    38 P. H. PARTRIDGE    stood to constitute or define a democratic polity. There are no new  classes struggling to win a share in political power, none struggling  for an enlargement of power in ways that would entail substantial  modification of political foundations. In Western Europe since, let  us say, the beginning of the seventeenth century, it is not often that  this could have been said. One could point, of course, to important  contemporary controversies about rights or liberties (e.g. contro-  versies about limitations upon the freedom of action of trade unions),  but such controversies tend less and less to raise issues of great  generality, and generality has normally been taken to be the work of  philosophical issue in politics.   Now, if this is roughly true, it is plausible to suppose that this  consensus (and not technical changes in philosophy or the growth of  empirical social science, or other developments of extremely  circumscribed influence) is the main factor affecting the character of  contemporary political theory. If classical political theory has died,  perhaps it has been killed by the triumph of democracy. At any rate,  this seems to me to bea very relevant consideration which has not been  sufficiently noticed by those who have written about the decay of  political theory.’   In fact, the consensus appears to include more than the general  system of powers and rights and the legally established institutions  which give effect to them. It seems to embrace also objectives and  justifications of policy. Is there not an all but universal acceptance    ‘May we not suspect the influence of ideological consensus even in some of the  arguments employed by recent English philosophers? The late Miss Margaret  Macdonald, for example, in her well-known essay, The Language of Political Theory,  argues that the orthodox question Why should I obey the State, or Why should I  obey the law? is in principle unanswerable. I can give reasons why I should obey  (or disobey) this particular law; but there is no answer to the general question  Why should I obey the law (or the State)? Now, apart from the question whether  this argument does justice to what such a philosopher as T. H. Green was really  saying, we may ask whether the argument would in fact always hold. If we give  the State a content (let us take it to be, for example, a secular political authority  claiming ultimate or sovereign political power over all other institutions), have  there not been theocrats and anarchists to whom it was by no means obvious that  the State, as distinct from other institutions, was entitled to claim final obedience?  If there is, then, a real general issue, is it pointless to try to make a case in general  terms for the supremacy of the State ae its law, though we may admit that it is  a general case only, and that there may quite well be conditions—as Green, along  with other political philosophers admitted—under which the case for the State’s  final authority does not hold? May not Miss Macdonald’s argument derive some  plausibility from the fact that the State and its legal supremacy are now so firml  established as a matter of historical fact that to ask in general, Why should I obey  the State? seems as sensible as to ask why I should obey the laws of gravity?    POLITICS, PHILOSOPHY, IDEOLOGY 39    of the belief that continuous technological and economic innovation,  uninterrupted expansion of economic resources, a continuously rising  standard of ‘material welfare’, are the main purposes of social life  and political action, and also the main criteria for judging the success  and validity of a social order? No doubt, there is in any of the  societies I have referred to some public questioning and criticism of  these objectives and criteria; nevertheless, these are the objectives  and criteria which define the course of contemporary democratic  politics. They are the ‘built-in’ criteria which render irrelevant and  impotent any alternative social philosophy.’   To speak of unusual ideological consensus when it is popularly  supposed that the world is divided by warring ideologies may seem  paradoxical; but the career of communism in England, the U.S.A.,  and some other democracies reinforces the argument. Communism  has had no important effect in these countries as an alternative  political philosophy; in England, as Miss Iris Murdoch has put it,  communism has been left to the communists: one of the chief forces  in creating and consolidating democratic consensus has been the  repudiation of the consequences of communist revolutions. And one  might document the growth of consensus in England by examining  the history of socialist thought over the past three or four decades.  There are now few socialist writers who advocate any systematic alter-  native to the basic political assumptions and arrangements of a  liberal-democratic system. Most of the specifically socialist notions  of extended democratic rights and institutions which had some  currency earlier in the century—industrial democracy, workers’ parti-  cipation in management, guild socialism, and so on—are not much  heard of now in serious politics. One of the most interesting points  about such recent English works as Crosland’s The Future of Socialism  and Strachey’s Contemporary Capitalism is that they disclose very plainly  indeed how the standard institutions and procedures of liberal  parliamentary democracy are now accepted as common ground.    Nt is possible, of course, that the political and moral consensus may be more  superficial than appears; and that there may be conflicts or frustrations growing  in the deeper social soils that most of us are not sensitive enough to perceive. If  the sociologists to whom I earlier referred are right in much of what they say about  the psychological dissatisfactions and social dislocations of industrial society, then  this may be so. At any rate, they have not yet become significant in practical  politics, and, except for a few writers of the ‘new Left’ in England and America,  they have not provided material for new political formulations. In his Beyond the  Welfare State, Myrdal tries to describe in more detail the consensus that exists in  the stable democracies; he seems to have no doubt about the stability of the  prevailing agreement upon the arrangements and objectives of the Welfare State.    40 P. H. PARTRIDGE    Vv   But, further, there is the even more interesting and important fact  that some of the most influential political theorists of the day have  become consciously anti-ideological. A closer look at some of the  arguments which have been brought to bear against ideological  politics will help us to see more clearly what is happening in con-  temporary English and American political theory.   By and large, the ruling trend of contemporary theory has been  reacting against the more optimistic philosophies or ideologies of the  past two centuries; consciously or implicitly, it has set about deflating  the larger ideas of human possibilities that recommended themselves  to many thinkers in the past, and has engaged in the job of cutting  down our notions of man’s nature to size.   The argument against ‘ideological politics’ has taken a number of  different lines. It may simply assert that ideological ways of formulat-  ing political attitudes and objectives have declined in the course of  this century as a matter of fact. But most writers who have touched on  this theme have intended to do more; they have produced an account  of what they believe to be rational political action. One argument is  that ideologically-dominated thinking has no relevance to the  controlling facts of contemporary social structure and change. ‘Grand  alternatives’ like capitalism and socialis™ are irrelevant, because our  choice is not between all-inclusive and mutually exclusive alternatives ;  in any society, there may be an indefinite number of ways in which  different institutions and social mechanisms may be arranged and  administered. Sometimes, this point is connected with a wider point—  that ideological thinking has usually been totalistic; that is, that it  has assumed that every important characteristic of a society is con-  nected with a single governing mechanism, and that the whole of  human life can be transformed from a central point. Thus, it is  suggested, ideological thinking tends to adopt global views of social  structure and political action.!   Again, it has been argued that this totalistic illusion has been  responsible, in those countries which have suffered ideologically-  inspired revolution, for much of the barbarism of the twentieth-  century upheavals. The attempt to transform society globally can  never be successful, and demands the employment of force on a  monstrous scale and ina never-ending process. The logic of the idea   'This is suggested by the way in which some followers of Marx have talked about    ‘the’ revolution; as if there were one revolution that would transform society, and  the eradication of all social evils waited upon ‘the’ revolution.    POLITICS, PHILOSOPHY, IDEOLOGY 41    of total transformation leads to perpetual force, apart from the fact  that ideological conviction is often associated with moral and political  fanaticism.   Finally, I must mention the criticism of ideological politics that  Edward Shils has developed because it is theoretically the most  penetrating and interesting. Shils’s criticism connects with his theory  of social groups and kinds of social cohesion. His argument is meant  to suggest, I think, that those who have defined political action (or  the change to be accomplished by political action) ideologically have  erred by imposing upon civil society a character repugnant to it, one  which rather characterizes other types of social grouping and adhesion.  Civil society, according to Shils, is characterized by a plurality of  groups, interests, and values, and the attachment of the members of  civil society to the common set of values is normally moderate, luke-  warm, sporadic, and intermittent. Thus, those who envisage a civil  society as embodying a shared and intensely experienced set of  common values are really imputing to civil society the emotional,  psychological, or moral qualities that are characteristic of quite  different types of social group, for instance, the sect or the community  bound by ‘primordial’ ties of blood or propinquity. The attempt to  create a civil society possessing a heightened emotionalism, and a  more intense, inclusive, and continuous integration around a  common ‘centre’, is thus bound to be destructive of liberty and other  political values of civil society.’   For such reasons, much recent British and American political  theory has been concerned with the devaluation of ideology and  ideologies, with showing the importance of ‘technique’ as opposed  to ideology, or with showing that ‘incrementalism’ (Dahl and  Lindblom), or ‘piecemeal engineering’ (Popper), are the most rational  methods of political change. Now, this quite considerable body of  recent writing obviously raises some very important questions (the  question of the nature and function of ideologies in politics has  certainly not been exhausted); and it seems odd, therefore, that  there should be so much talk about the decline or decay of political    'The reference is primarily to the article, ‘Primordial, Personal, Sacred and Civil  Ties’, in British Journal of Sociology, vol. 8, no. 2. Some of the bearings of this view  upon questions of political philosophy are suggested in The Torment of Secrecy. Mary  other lines of contemporary political thinking are allied to those I have summarized.  For example, Talmon’s studies of the history of political ideas, Totalitarian Democracy  and Political Messianism, have a similar tendency; as have the various articles that  have recently appeared discussing the positive functions of political apathy in  democracies, for example, W. H. Morris Jones, ‘In Defence of Apathy’ (Pol. Studies,  vol. iv, no. 6, 1954).    42 P. H. PARTRIDGE    theory. Nor does this particular line of thought justify the suggestion  that political theory has been entirely supplanted by factual or purely  descriptive and explanatory political ‘science’; for these writers are  concerned with the justification of forms of political action and  organization, just as political philosophers have always been. The  method of justification is no doubt different; writers like Shils ground  their conclusions on sociological premises rather than on more  general philosophical ones. But this is but another illustration of the  point already made that in twentieth-century political theory the  discussion of the general issues is being detached from ‘philosophy’  and more and more closely linked with empirical social inquiry.   It could be argued, I think, that the thorough-going pluralism  of present-day Anglo-American political theory has tended strongly  to inhibit the formulation of general principles, values, or objectives  of political life. It is the pervasive belief of current English and  American political science that the ‘essence’ of democratic politics  is a process of bargaining and of finding adjustments between compet-  ing demands, interests, values, ways of life-adjustments that will be  more or less temporary and shifting as conditions change within and  without society. In the more stable and affluent democracies this is  the character that present political life has assumed. And this, surely,  is enough to account for the relaxed atmosphere of modern demo-  cratic politics, the absence of political ideas of general range and  importance, and of comprehensive political doctrine. In general,  the politics of adjustment is one that directs itself to separate and  limited issues, most of which affect only limited sections of a com-  munity, and any one of which engages rarely, and usually only  marginally and casually, the interests or passions of a large span of  the public, and which are unlikely, therefore, to generate political  movements or coherent bodies of theory which aim to articulate a  whole cluster of issues. It is evident that this corresponds very closely  to Shils’s account of the nature of ‘civil society’.    4It is not only in political thinking that particularism or piecemealism is the  prevailing habic of thought. Wright Mill’s recent book, The Sociological Imagination,  may be read as a protest against this habit of thought in a wider context. If I follow  him correctly, when he advocates the ‘sociological imagination’, he is protesting  against the practice of many sociologists—the ‘abstracted empiricists’—of concentrating  upon the separate and circumscribed social phenomenon or problem. He still believes  that one can explore the structure of society as a whole; that there are controlling  areas (or at least especially strategic areas) through which a wide range of social life can  be affected; and that contemporary social theorists fail to attend to those strategic areas  within the social system which have the widest influence on freedom and other values  of democratic society.    POLITICS, PHILOSOPHY, IDEOLOGY 43    It would take me too far afield to try to examine in detail the  now standard arguments against ‘ideological politics’; I shall make  only one or two observations. One might argue, to begin with,  that the writers to whom I am referring are incautious in their  acclaim of the passing of ideologies; that they are generalizing too  boldly from the special conditions that now happen to obtain ina  few societies. Again, it may even be misleading to say that in  these societies the function of ideology has vanished. It can be  argued that the politics of ‘incrementalism’, of bargaining and  adjustment, of the pursuit of limited objectives, can itself operate  as it does only because of the strong and wide ideological con-  sensus that happens to rule in these societies. In his Preface to  Democratic Theory, Dah! makes the point that the processes of  bargaining and adjustment of claims involve agreed values and  principles which keep political conflicts within bounds, limit the  demands which minorities will seek to have granted, define the  range within which acceptable solutions are known to lie, and so  on. But this is a point of general importance. Recent political  theorists have been apt to under-emphasize the extent to which  all the elements which enter into the consensus operate as a  necessary condition of effective political bargaining and com-  promise. Ideology may be no less an important element in a  political and social system because it lies below the level of general  political controversy.   Again, just because of the special circumstances of our time, we  may be too ready to conclude that ideology is a false, irrational, and  even disastrous guide to political action. We look back over recent  history, and we see that the aspirations and expectations of ideological  and utopian thinkers or agitators differ ludicrously from the states  of affairs that actually came to be: such highly-charged ideas as  equality, fraternity, ‘positive’ freedom in the sense of general partici-  pation in the control of social affairs, and notions about a classless  society, common ownership and the like, seem to have come to very  little. And these large moral intimations have apparently not only  been held to be irrelevant to the actual course of events: it is often  argued that they have been pernicious in their effects: they have  encouraged colossal blundering, they have blinded men to the under-  standing of their own limitations, to the reality of original sin  (Reinhold Niebuhr’s account of the human situation accords with  many of the other fashionable currents of present-day social thought) ;  they have provided facile justifications for ruthlessness and terror.    44 P. H. PARTRIDGE    Michael Polanyi has somewhere argued that the revolutionary excesses  of this century have resulted partly from ‘the excess of theoretical  aspiration over practical wisdom’. Consequently, it is not hard to find  grounds for arguing, as Popper does in effect in The Open Society and  its Enemies, that gradual, piecemeal, ‘experimental’ attack on limited  and particular problems, the pragmatic alleviation of particular  evils, is the only rational method of political change.   In a sense, this sort of argument is incontrovertible. Political  ideology has often been mainly faith, myth, superstition, political  and moral dogmatism and fanaticism; when at full flood it has  sometimes produced the most appalling destruction of existing  institutions, traditions, and values. It would indeed be difficult to  hold that there was anything rational about such ideology-  impregnated social upheavals: to say that they are a ‘rational’ way  of bringing about desired social change would be as strange as to  say that an earthquake is a good way of producing a lake.   Yet it seems to me that none of these arguments suffices to dismiss  ‘ideological politics’ as an obsolete and irrational method of social  change. They cannot be ‘justified’; but they may be inevitable in some  circumstances; and we may also be able to argue—with the wisdom of  hindsight and from the point of view of historical determinism—  that they were the necessary condition for the production of certain  social changes now accepted as desirable. ‘Justification’ and ‘ration-  ality’ are categories applicable only within spheres of social action  and change where calculated choice can be made; they apply only  within the means-end model of social change and explanation. But  could we not argue that there are certain types of social change which  are (or were) desirable changes, but which could not have come about  as a result of rational calculation and piecemeal adjustment, but only  as the short- or long-term consequences of widespread ideological  and even utopian upsurge and agitation?   I am aware that in these very swift remarks I have run together  with. nothing like sufficiently careful definition and discrimination  some exceedingly hard questions of political analysis and theory. And  I confess that I do not know how to prove that phases or periods of  intense ideological activity (such as the activity of the Levellers and  the Diggers in the English civil war, or the ideological ferment that  preceded and accompanied the French Revolution, or the swell of  socialist myth-making, moralizing, and social criticism that grew  throughout the nineteenth century) are a necessary lever of political  change, even though they never succeed in bringing about the results    POLITICS, PHILOSOPHY, IDEOLOGY 45    that ideologists believe they are bringing about. But one can argue  negatively. One can argue that the ‘unideological’ politics of adjust-  ment and piecemeal change must necessarily accept limited goals and  types of social change, or again, that the many built-in pressures  that support order and stability, the natural need and desire to  go on living the daily life, operate to sift and narrow the objectives  that are likely to be put on the agenda of responsible politics. (This, in  fact, was one of the stock Marxist arguments against the politics of  gradualism and reform—that the policy of incrementalism or rational  calculation will involve us in accepting as constants certain structural  features of the existing system, and certain moral or ideological  elements that are embedded in that structure.)   To put this in another way—in any given state of society there will  be well-established institutions and habits of moral thinking, which  are central in the sense that they protect important elements, and  which operate to limit the objectives, methods and types of change  which are accepted as matters for political policy and governmental  action, so that at any given time that part of the social structure that  is at all generally recognized as subject to political action and change  is always comparatively small. But it is in relation to what must be  called the institutional and ideological infrastructure that ideological  ferment and ideological politics have a very important function. They  have their important effects below the level of ‘rational’ or program-  matic political action, in eroding or loosening established moral and  ideological habits and certainties, in producing the climate of opinion  in which it is ultimately possible for new sorts of political or social  objectives, new forms of social control and organization, new tech-  niques of social action, to be accepted as parts of the ordinary  programmes of political parties.’   However, it is not my intention in this article to embark upona  thorough examination of this question. My comments are simply  meant to illustrate two main points: (a) That recent political scientists    'The only support that one could give to this assertion would be in the form of  extended historical analysis. One would have to show how prolonged periods of  ideological ferment, no matter how ‘utopian’ the ideologies might have been,  ultimately produced important political results in the form of quite common-place  programmes and policies. Examples are the revolutionary current in France from  the late eighteenth century to the end of the nineteenth, or the nineteenth-century  British movement of utopian socialism. A more extreme, and large-scale, example  would be the rise, spread and ultimate political effects of the great world religions.  There are some interesting remarks about some of the topics I have touched on in  the last few paragraphs in Howard Horsburgh’s article, ‘The Relevance of the Utopian’  (Ethics, vol. Ixvii, no. 2).    46 P. H. PARTRIDGE    have in fact been raising interesting and important problems of  political theory and their discussion is a continuation of discussions  that have been going on in Western political theory for centuries.  (It is evident that some of the issues raised are very close to those  raised by Burke about the French Revolution or by Marxists in their  controversies with ‘gradualists’ and reformers.) (b) Yet, at the same  time, this example does illustrate an important shift of interest in  recent political speculation. In the democratic countries practical  politics are mainly concerned for the time being with limited object-  ives and adjustments: many political scientists have come to accept  this style of politics as the rational norm for all political activity: the  discounting of ideology has been accompanied by a scepticism  concerning general speculation about the moral issues of politics, a  disposition to assume that thinking about ‘techniques’ alone qualifies  as really rational or practical political thinking.    VI    Let me take two other examples. The anti-ideological trend of so  much recent political theory (‘anti-ideological’ both in the sense of  discounting the importance of ideologies in politics, and in the sense  of attacking and debunking political ideologies that have been very  influential) is very plain in much of what is now being said about the  nature of democracy and about the working of democratic institutions,  especially political parties.   I will start with Schumpeter’s brilliant discussion in Capitalism,  Socialism and Democracy which expresses a line of thought and an  emphasis which reappears substantially in later books.’ Schumpeter  begins, it will be remembered, by rejecting ‘The Classical Theory of  Democracy’ which he formulates in this way: “The democratic method  is that institutional arrangement for arriving at political decisions  which realizes the common good by making the people itself decide  issues through the election of individuals who are to assemble in  order to carry out its will.’ He rejects this view because it involves  assumptions and concepts about non-existents—a common good, a  popular will, &c. He goes on to expound and defend ‘Another Theory  of Democracy’ of which the formula is: ‘The democratic method is  that institutional way for arriving at political decisions in which  individuals acquire the power to decide by means of a competitive  struggle for the people’s vote.’    ‘For example, Dahl’s Preface to Democratic Theory and Down’s An Economic Theory of  Government Decision-Making in a Democracy.    POLITICS, PHILOSOPHY, IDEOLOGY 47    Obviously, this involves a very different notion of the functions  of the ‘elements’ in a democratic system and the relations between  them. The leaders of the political parties decide, not ‘the people’.  It is the more or less organized groups of men competing for power  who have the initiative and supply the political drive: ‘so far as there  are genuine group-wise volitions at all... [they] do not assert them-  selves directly... | but} remain latent, often for decades, until they  are called to life by some political leader who turns them into political  factors’ (p. 270, 2nd edition). Policies and programmes are to be  viewed as weapons employed in the competition for office, taken up  or discarded according as they help or prejudice the party’s compet-  itive position; some question will become an ‘jssue’ in politics when  a leader or party judges it to be ‘a good horse to back’. And a political  party ‘is a group whose members propose to act in concert in the  competitive struggle for political power’ (p. 283).'   A model like Schumpeter’s is not to be criticized for being un-  realistic. Any model involves selection and simplification: it is to be  judged by its capacity to explain (and perhaps predict) the facts itis  intended to explain. No doubt we could explain within the limits of  Schumpeter’s assumptions a great many of the political events ofa  democratic state. Nevertheless, I want to comment on a couple of the  interesting features of his simplification.   In the first place, his model places heavy emphasis on manipulation  and leadership—on the making and propagating of policy—and its  tendency is to draw attention away from the ‘infrastructure’. I do  not. intend to imply that he would deny the importance of the  infrastructure; nevertheless, it is important to be quite explicit that  this special emphasis is there.   Now, if we like, we may apply the word ‘politics’ to one level of  activity only—to the level where policy operates, where individuals  and groups more or less explicitly and deliberately seek adjustments  and arrangements of their particular interests and activities. But, if  we define the political in this way, it is very important to remember  that the range of the political is always oscillating, and may at times  oscillate somewhat violently.? And for Schumpeter to say that ‘so far    ‘$0 also Downs, on Economic Theory and Democracy, p. 28: ‘parties formulate policies  in order to win elections, rather than win elections in order to formulate policies’. This  is described as ‘the fundamental hypothesis of our model’.   2It seems to me that recent political science has suffered from concentrating so  heavily on the study of short-term political events (e.g. the study of the single election),  on the act of decision-making’, on ‘bargaining models’ and so on. This may have had  advantages from the point of view of concreteness, empirical exactness and rigour.    48 P. H. PARTRIDGE    as there are genuine group-wise volitions at all . . . they do not assert  themselves directly... but remain latent often for decades until they  are called to life by some political leader who turns them into  political factors’ is to be guilty of a very considerable over-simplifica-  tion. It is true, of course, that this is how things often happen;  political leaders formulate policies which are not formulated by  other social groups; they crystallize or focus attitudes or demands that  would otherwise remain uncrystallized; they may propose solutions  that no one else has proposed. But to leave the matter there would  be to postulate a gap between the politician (and the political party)  and other social organizations which may be wider in some societies  than in others, but which is not an invariable feature of all democracies.  The party politician who appears in the political theory of Schumpeter  and many other present-day political theorists is an abstraction. In  ‘real life’ not all politicians are members only of a political party, or  committed only to the success of a party (as ‘success’ is defined in the  sort of model I am discussing). They are often also members of other  social organizations; and very often they may be said to belong to  a movement—and a movement is a social as well as a political pheno-  menon, and something it would be difficult to define except (at least  partly) by reference to doctrine.   Now it is a matter of common observation that politicians are  very often in the grip of a conflict between the beliefs or interest  of their organizations or movements and of their party. And how  they act in the end will not always be explicable solely by reference  to the conditions of party success in the competition for power.  Moreover, democratic societies differ from one another as regards  the latitude for manoeuvre that parties enjoy. Within the one society  there is an oscillation over time: the relations between parties and  ‘society’ change, and there are occasions when the impact of demands,  or of more general social and political ideas generated within social  movements or organizations, upon the life and actions of parties is  much greater than at others.! This being so, the model of democratic    But it has brought about a very drastic abstraction from a great deal of political  reality. Rostow, it will be remembered, in his British Economy in the Nineteenth Century  suggests an outline of the relations between economic change and political action. His  schema involves two inter-related tripartite divisions: one between long term, trend  or medium term, and short-term economic processes, the other a division between  economy, society and politics; his suggestions concerning the different ways in which  the three types of economic change operate at the political level could be profitably  explored further by political theorists.   ‘And sometimes a difference as regards the current ‘issues’ of politics: as regards  some issues, a party may have no alternative but to obey pressure ‘from below’;    POLITICS, PHILOSOPHY, IDEOLOGY 49    politics which Schumpeter proposes will be helpful for explanation  only in certan cases. And, even with this limitation, it is more likely  to be helpful in the understanding of short-term runs of political  decisions than in the understanding of a long-term direction of  political change.   This, then, is the kind of emphasis and selection which the now  fashionable Schumpeterian model of democracy contains. It is  relatively uninterested in the important Marxian notion of ‘shifts in  the foundation’ or infrastructure and their long-run political effects;  or, if we may use a different image, it largely ignores the series of  concentric circles within which, in some societies and at some times,  the generation of social attitudes, ideas, and ideologies takes place.  Perhaps the most important point I want to make is that these circles  oscillate as conditions change: they grow wider or narrower, as  regards both the range of social institutions or conditions which  become subjects of social questioning and idea-making, and the  numbers of those who are to some degree caught up in social  criticism.!   One thing that may be said about Schumpeter’s account is that it  tends to stress the specialized and professional character of political  activity; that its intention and effect is one of ideological deflation.  And this brings me to the other comment I wish to make about his  theory. Most of the so-called theories of democracy in the history of  political thought have been primarily normative or prescriptive.    as regards others, it may be able to play the electoral market in the manner  Schumpeter describes.   ‘In his review of de Jouvenel’s Sovereignty in World Politics to which I have already  referred, Dahl employs this concept of oscillation in speculating about the nature of  democratic ‘consensus’. His point, of course, is very closely related to mine. It would  be a legitimate comment that Schumpeter does not pretend to be providing a model of  any political system but only of democratic systems. And he might argue—indeed, there  are plain indications that he would argue—that a democratic system will operate  successfully (in the sense of maintaining over a long period its character or structure)  only if conditions are as I have described them, in particular, only if the range of  matters that come up for political decision and action remains narrow. This is now,  of course, a widely-held view; one aspect of the reaction against socialist ideologies,  or against notions of ‘holistic’ planning, is the argument that democracy is a method  and system of government which requires as a condition that ‘the effective range of  political decision should not be extended too far’ (Schumpeter). It is dubious whether  this particular hypothesis can be given any hard meaning: what I have already said  about the way in which consensus oscillates suggests that how much a government can  do and get away with—or a party propose and get away with—will be affected by many  different conditions which are far from being stable. However, it will be clear that  this particular theory about democracy is closely allied with Shils’s account of the  nature of ‘civil society’ to which I referred earlier.    50 P. H. PARTRIDGE    One important strain of democratic thought has connected demo-  cracy with the enlargement and greater equalization of opportunities  for participating in the control or management of common affairs  for sharing effective social responsibility. On this view, part of the  task of a theory of democracy would be to investigate the means by  which self-government or participation in the direction of affairs  could be more widely extended. Mill puts very succinctly what has  been a pretty constant problem and theme in the growth of democratic  theory: ‘A democratic constitution not supported by democratic  institutions in detail, but confined to the central government, not only  is not political freedom, but often creates a spirit precisely the reverse,  carrying down to the lowest grade in society the desire and ambition  of political domination.’ This is not only succinct, but also percep-  tive—as the post-Millian growth of party discipline and party  machines, of political bosses, and of publicity agents and carefully  managed publicity campaigns remind us.   Perhaps there is no logical incompatibility between this and the  Schumpeterian model; one might, for instance, say that Schumpeter  is specifying the minimum conditions of democracy at the level of  central government, and that his specification is not affected by how  much or how little decentralization and diversification of control  there might be down below. But I doubt it. I suspect that the more  the two positions were developed, the more theoretical conflict would  become apparent. On Schumpeter’s view, it is the function of political  parties to compete for votes by raising issues and proposing policies;  it is the function of the public to express a preference between  competing leaders and would-be governments. He appears to be  saying that so far as democracy is concerned, this is not merely how  democratic systems happen to work now, but that it is the most that can  be expected. In effect, his model has a normative or prescriptive ring  about it; and this is the more so when it is examined in the contextin  which he places it, viz. the rejection of the ‘classical theory of  democracy’ with its very different conception of the meaning of self-  government and of individual political participation and respons-  ability.   My second example is Dahl’s model in Preface to Democratic Theory.  This differs a little from Schumpeter’s in that Dahl assigns a less  prominent place to parties and the electoral competition and a more  prominent place to organized minorities and pressure groups. He  warns us against attributing too rich a function to elections; in his  view, the main function of elections in the democratic system is that    POLITICS, PHILOSOPHY, IDEOLOGY 51    they extend the range of the minorities which exert an influence on   governmental decisions. According to Dahl, democracy is neither»  rule by the majority nor rule by a minority, but rule by minorities.   ‘Thus the making of governmental decisions is not a majestic march   of great majorities united on certain matters of basic policy. It   is the steady appeasement of relatively small groups’ (p. 146).   (Dahl at least has it in common with Schumpeter that he rejects the   majority and anything that smacks of the notion of the general   will.)   One might say that Dahl’s assertion is just not true as a universal  proposition; that there are not a few occasions and issues on which  a majority does form and exert its influence. Recent political theory  has no doubt gained in getting rid of loose and muddled concepts like  ‘the general will’ and the “common good’, and uncriticized, question-  begging ones like ‘the majority’; but it is another matter whether it  is fortunate in what it has substituted. In Dahl (and in the whole  tribe of ‘pressure group’ analysts) the emphasis falls heavily on the  notion of the determinate, relatively ‘given’, or impermeable,  minority or group, possessing its own clear and determinate interests  which have to be attended to by parties and governments. This, no  one will want to deny, is one sub-system within the very complex system  of democratic politics; but it exists in interplay with other sub-  systems, including the interaction of interests, institutions, movements,  ways of life within a process of general influence, and ‘discussion’  (if I may fall back on one of the governing conceptions of older  theorists of democracy), and the resulting slow spread of general  attitudes or bodies of belief which certainly affect the course of  politics in the long run. In rejecting the categories and assumptions  of the older philosophical idealists and monists, some of the  more recent school of pluralists have tended to play down the  social processes by means of which some sort of common deno-  minator of sentiment and idea is created within a society, pro-  cesses which are undoubtedly important for politics, and which  their idealist predecessors rightly took to be pretty central for  political theory, even if they misrepresented or unduly magnified  them.   In an earlier section I illustrated the general reaction against  ideology which has marked recent political and sociological thought.  Correspondingly, then, in some recent democratic theory there has  been an undermining of versions of democratic ideology which have  been very prominent until quite recently.    52 P. H. PARTRIDGE    VII    I have already explained that it is not my object to examine the  theoretical strength of the views I have been considering. My  main object has been to draw attention to some very common  characteristics of recent political thought with a view to explaining  the widely shared impression that political theory or political  philosophy has lost its inspiration. If we take the work of such writers  as Schumpeter and Dahl, it is absurd to say that the energy or the  rigour of political theorizing have declined; on the contrary, it has  acquired .an analytical thoroughness and sharpness, a closeness in  argument, that is pretty new. But what this more recent work does  show is a narrowing of moral interests and expectations, a dismissal  of wider notions of equality, freedom, participation, &c.,! and,  accompanying this, the tendency to be most interested in the existing  machinery of democratic systems. This is not simply a matter of  science replacing the more philosophical interest in principles, values,  or objectives. The ambition to lay the foundations of an empirical  science of politics is no doubt a very important intellectual influence;  but the present tren. 1s also critical: it expresses an ideological or  philosophical standpoint of its own, an inclination to accept as  inevitable, or at least as more rational than any alternative, the  broad types of organization, the distribution of rights and roles, the  methods of adjusting existing interests, which have by now come to  define democracy in the Anglo-Saxon democracies. In short, as I have  said before, the current feeling that there is no very persuasive  alternative to the prevailing methods and orthodoxies of Anglo-  American democracy is at least one of the reasons for the shortcomings  (if shortcomings they are) of contemporary political theory.    ‘Instead of illustrating my general thesis by dealing with recent theories of demo-  cracy, I might have taken recent writing about liberty, one striking feature of which has  been an emphasis on the value or importance of ‘negative freedom’ and the distrust  of notions of ‘positive freedom’. Berlin’s Two Concepts of Liberty is quite typical in this  respect. Berlin’s way of reviewing somewhat indiscriminately many different concepts  of ‘positive freedom’, ranging from the special views of Hegel to much more prosaic  attempts to connect liberty with the exercise of political initiative, is surprising  and questionable; all the same, in his sensitiveness to the possibly illusory and dangerous  character of ideas of ‘positive freedom’, he is very much in tune with his time. Ona  different point: the trend of thought I describe in the text is not of course entirely  novel, as readers of Michels’s Political Parties will be well aware; the argument between  the oe of a radical democratic ideology and those who insist upon the hard  logic of social organization has been going on for some time. Whatis perhaps character-  istic of more recent years is the unusual weakness of the strain of moral criticism and  speculation.    8  ARE THERE ANY NATURAL RIGHTS?!  H. L. A. HART    it follows that there is at least one natural right, the equal right of    all men to be free. By saying that there is this right, I mean that in the  absence of certain special conditions which are consistent with the  right being an equal right, any adult human being capable of choice  (1) has the right to forbearance on the part of all others from the use  of coercion or restraint against him save to hinder coercion or re-  straint and (2) is at liberty to do (i.e., is under no obligation to  abstain from) any action which is not one coercing or restraining or  designed to injure other persons.’   I have two reasons for describing the equal right of all men  to be free as a natural right; both of them were always emphasized  by the classical theorists of natural rights. (1) This right is one which  all men have if they are capable of choice: they have it qua men and not  only if they are members of some society or stand in some special  relation to each other. (2) This right is not created or conferred by    From Philosophical Review, Vol. 64 (1955), pp. 175-91- Reprinted by permission of  the author and the Philosophical Review.    1] was first stimulated to think along these lines by Mr. Stuart Hampshire, and I  have reached by different routes a conclusion similar to his.   ?Further explanation of the perplexing terminology of freedom is, I fear,  necessary. Coercion includes, besides preventing a person from doing what he chooses,  making his choice less eligible by threats; restraint includes any action designed to  make the exercise of choice impossible and so includes killing or enslaving a person.  But neither coercion nor restraint includes competition. In terms of the distinction  between ‘having a right to’ and ‘being at liberty to’, used above and further dis-  cussed in Section I, B, all men may have, consistently with the obligation to forbear  from coercion, the Liberty to satisfy if they can such at least of their desires as are  not designed to coerce or injure others, even though in fact, owing to scarcity, one  man’s satisfaction causes another’s frustration. In conditions of extreme scarcity  this distinction between competition and coercion will not be worth drawing;  natural rights are only of importance ‘where peace is possible’ (Locke). Further,  freedom (the absence of coercion) can be valueless to those victims of unrestricted  competition too poor to make use of it; so it will be pedantic to point out to them  that though starving they are free. This is the truth exaggerated by the Marxists  whose identification of poverty with lack of freedom confuses two different evils.    54 H. L. A. HART    men’s voluntary action; other moral rights are.' Of course, it is quite  obvious that my thesis is not as ambitious as the traditional theories  of natural rights; for although on my view all men are equally en-  titled to be free in the sense explained, no man has an absolute or  unconditional right to do or not to do any particular thing or to be  treated in any particular way; coercion or restraint of any action may  be justified in special conditions consistently with the general prin-  ciple. So my argument will not show that men have any right (save the  equal right of all to be free) which is ‘absolute’, ‘indefeasible’, or  ‘imprescriptble’. This may for many reduce the importance of my  contention, but I think that the principle that all men have an equal  right to be free, meagre as it may seem, is probably all that the poli-  tical philosophers of the liberal tradition need have claimed to  support any programme of action even if they have claimed more. But  my contention that there is this one natural right may appear unsatis-  fying in another respect; it is only the conditional assertion thati/there  are any moral rights then there must be this one natural right. Perhaps  few would now deny, as some have, that there are moral rights; for the  point of that denial was usually to object to some philosophical claim  as to the ‘ontological status’ of rights, and this objection is now  expressed not as a denial that there are any moral rights but as a  denial of some assumed logical similarity between sentences used to  assert the existence of rights and other kinds of sentences. But it is  still important to remember that there may be codes of conduct quite  properly termed moral codes (though we can of course say they are  ‘imperfect’) which do not employ the notion of a right, and there is  nothing contradictory or otherwise absurd in a code or morality con-  sisting wholly of prescriptions or ina code which prescribed only what  should be done for the realization of happiness or some ideal of per-  sonal perfection.?, Human actions in such systems would be evaluated  or criticized as compliances with prescriptions or as good or bad,  night or wrong, wise or foolish, fitting or unfitting, but no one in sucha  system would have, exercise, or claim rights, or violate or infringe  them. So those who lived by such systems could not of course be    ‘Save those general rights (cf. Section II, B) which are particular exempli-  fications of the right of all men to be free.    7Is the notion of a right found in either Plato or Aristotle? There seems to be  no Greek word for it as distinct from ‘right’ or ‘just’ (Stxaiov), thought expressions  like ta tua Sixava are I believe fourth-century legal idioms. The natural ex-  pressions in Plato are 16 éavrov (éxew) or ta tiwi ddetAdueva, but these seem  confined to property or debts. There is no place for a moral right unless the moral  value of individual freedom is recognized.    ARE THERE ANY NATURAL RIGHTS? 55    committed to the recognition of the equal right of all to be free; nor,  1 think (and this is one respect in which the notion of a right differs  from other moral notions), could any parallel argument be con-  structed to show that, from the bare fact that actions were recognized  as ones which ought or ought not to be done, as right, wrong, good  or bad, it followed that some specific kind of conduct fell under these  categories.    I    (A) Lawyers have for their own purposes carried the dissection of  the notion of a legal right some distance, and some of their results!  are of value in the elucidation of statements of the form ‘X has a  right to...’ outside legal contexts. There is of course no simple  identification to be made between moral and legal rights, but there  is an intimate connexion between the two, and this itself is one  feature which distinguishes a moral right from other fundamental  moral concepts. It is not merely that as a matter of fact men speak of  their moral rights mainly when advocating their incorporation in a  legal system, but that the concept of a right belongs to that branch  of morality which is specifically concerned to determine when one  person’s freedom may be limited by another’s? and so to determine  what actions may appropriately be made the subject of coercive  legal rules. The words ‘droit’, ‘diritto’, and ‘Recht’, used by con-  tinental jurists, have no simple English translation and seem to  English jurists to hover uncertainly between law and morals, but they  do in fact mark off an area of morality (the morality of law) which  has special characteristics. It is occupied by the concepts of justice,  fairness, rights, and obligation (if this last is not used as it is by  many moral philosophers as an obscuring general label to cover every  action that morally we ought to do or forbear from doing). The most  important common characteristic of this group of moral concepts is  that there is no incongruity, but a special congruity in the use of    1As W. D. Lamont has seen: cf. his Principles of Moral Judgement (Oxford, 1946);  for the jurists, cf. Hohfeld’s Fundamental Legal Conceptions (New Haven, 1923).    2Here and subsequently I use ‘interfere with another’s freedom’, ‘limit an-  other’s freedom’, ‘determine how another shall act’, to mean either the use of  ccercion or demanding that a person shall do or not do some action. The connexion  between these two types of ‘interference’ is too complex for discussion here; I  think it is enough for present purposes to point out that having a justification  for demanding that a person shall or shall not do some action is a necessary  though not a sufficient condition for justifying coercion.    56 H. L. A. HART    force or the threat of force to secure that what is just or fair or some-  one’s right to have done shall in fact be done; for it is in just these  circumstances that coercion of another human being is legitimate.  Kant, in the Rechtslehre, discusses the obligations which arise in this  branch of morality under the title of officia juris, ‘which do not require  that respect for duty shall be of itself the determining principle of  the will’, and contrasts them with officia virtutis, which have no moral  worth unless done for the sake of the moral principle. His point is,  I think, that we must distinguish from the rest of morality those  principles regulating the proper distribution of human freedom  which alone make it morally legitimate for one human being to  determine by his choice how another should act; and a certain specific  moral value is secured (to be distinguished from moral virtue in which  the good will is manifested) if human relationships are conducted  in accordance with these principles even though coercion has to be  used to secure this, for only if these principles are regarded will free-  dom be distributed among human beings as it should be. And it is I  think a very important feature of a moral right that the possessor of  it is conceived as having a moral justification for limiting the freedom  of another and that he has this justification not because the action he  is entitled to require of another has some moral quality but simply  because in the circumstances a certain distribution of human freedom  will be maintained if he by his choice is allowed to determine how that  other shall act.   (B) I can best exhibit this feature of a moral right by reconsidering  the question whether moral rights and ‘duties’! are correlative. The  contention that they are means, presumably, that every statement of  the form ‘X has a right to...’ entails and is entailed by ‘Y has a duty  (not) to ...’, and at this stage we must not assume that the values of  the name-variables “X’ and ‘Y’ must be different persons. Now there  is certainly one sense of ‘a right’ (which J have already mentioned) such  that it does not follow from X’s having a right that X or someone else  has any duty. Jurists have isolated rights in this sense and have re-   ‘1 write ‘duties’ here because one factor obscuring the nature of a right is the  philosophical use of ‘duty’ and ‘obligation’ for all cases where there are moral  reasons for saying an action ought to be done or not done. In fact ‘duty’, ‘obligation’,  ‘right’, and ‘good’ come from different segments of morality, concern different  types of conduct, and make different es of moral criticism or evaluation.  Most important are the points (1) that obligations may be voluntarily incurred  or created, (2) that they are owed to special persons (who have rights), (3) that  they do not arise out of the character of the actions which are obligatory but out    of the relationship of the parties. Language roughly though not consistently confines  the use of ‘having an obligation’ to such cases.    ARE THERE ANY NATURAL RIGHTS? 57    ferred to them as ‘liberties’ just to distinguish them from rights in the  centrally important sense of ‘right’ which has ‘duty’ as a correlative.  The former sense of ‘right’ is needed to describe those areas of social  life where competition is at least morally unobjectionable. Two  people walking along both see a ten-dollar bill in the road twenty  yards away, and there is no clue as to the owner. Neither of the two  are under a ‘duty’ to allow the other to pick it up; each has in this  sense a right to pick it up. Of course there may be many things which  each has a ‘duty’ not to do in the course of the race to the spot—  neither may kill or wound the other—and corresponding to these  ‘duties’ there are rights to forbearances. The moral propriety of all  economic competition implies this minimum sense of ‘a right’ in  which to say that ‘X has a right to’ means merely that X is under  no ‘duty’ not to. Hobbes saw that the expression ‘a right’ could have  this sense but he was wrong if he thought that there is no sense in which  it does folluw from X’s having a right that Y has a duty or at any rate  an obligation.   (C) More important for our purpose is the question whether for  all moral ‘duties’ there are correlative moral rights, because those  who have given an affirmative answer to this question have usually  assumed without adequate scrutiny that to have a right is simply  to be capable of benefiting by the performance of a ‘duty’; whereas  in fact this is not a sufficient condition (and probably not a necessary  condition) of having a right. Thus animals and babies who stand to  benefit by our performance of our ‘duty’ not to ill-treat them are  said therefore to have rights to proper treatment. The full consequence  of this reasoning is not usually followed out; most have shrunk from  saying that we have rights against ourselves because we stand to bene-  fit from our performance of our ‘duty’ to keep ourselves alive or  develop our talents. But the moral situation which arises from a  promise (where the legal-sounding terminology of rights and obli-  gations is most appropriate) illustrates most clearly that the notion  of having a right and that of benefiting by the performance of a ‘duty’  are not identical. X promises Y in return for some favour that he will  look after Y’s aged mother in his absence. Rights arise out of this  transaction, but it is surely Y to whom the promise has been made and  not his mother who has or possesses these rights. Certainly Y’s mother  is a person concerning whom X has an obligation and a person who  will benefit by its performance, but the person ¢o whom he has an obli-  gation to look after her is Y. This is something due to or owed to Y,  so it is Y, not his mother, whose right X will disregard and to whom    58 H. L. A. HART    X will have done wrong if he fails to keep his promise, though the  mother may be physically injured. And it is Y who has a moral claim  upon X, is entitled to have his mother looked after, and who can waive  the claim and release Y from the obligation. Y is, in other words,  morally in a position to determine by his choice how X shall act and  in this way to limit X’s freedom of choice; and it is this fact, not  the fact that he stands to benefit, that makes it appropriate to say  that he has a nght. Of course often the person to whom a promise has  been made will be the only person who stands to benefit by its per-  formance, but this does not justify the identification of ‘having a  right’ with ‘benefiting by the performance of a duty’. It is important  for the whole logic of rights that, while the person who stands to  benefit by the performance of a duty is discovered by considering what  will happen if the duty is not performed, the person who has a right  (to whom performance is owed or due) is discovered by examining  the transaction or antecedent situation or relations of the parties out  of which the ‘duty’ arises. These considerations should incline us not  to extend to animals and babies whom it is wrong to ill-treat the  notion of a right to proper treatment, for the moral situation can be  simply and adequately described here by saying that it is wrong or that  we ought not to ill-treat them or, in the philosopher’s generalized  sense of ‘duty’, that we have a duty not to ill-treat them.! If common  usage sanctions talk of the rights of animals or babies it makes an idle  use of the expression ‘a right’, which will confuse the situation with  other different moral situations where the expression ‘a right’ hasa  specific force and cannot be replaced by the other moral expressions  which I have mentioned. Perhaps some clarity on this matter is to be  gained by considering the force of the preposition ‘to’ in the ex-  pression ‘having a duty to Y’ or ‘being under an obligation to Y  (where ‘Y’ is the name of a person); for it is significantly different  from the meaning of ‘to’ in ‘doing something to Y’ or ‘doing harm  to Y’, where it indicates the person affected by some action. In the  first pair of expressions, ‘to’ obviously does not have this force,  but indicates the person to whom the person morally bound is  bound. This is an intelligible development of the figure of a bond  (vinculum juris : obligare); the precise figure is not that of two persons  bound by a chain, but of one person bound, the other end of the chain  lying in the hands of another to use if he chooses.? So it appears   'The use here of the generalized ‘duty’ is apt to prejudice the question whether  animals and babies have rights.   2Cf. A. H. Campbell, The Structure of Statr’s Institutes (Glasgow, 1954), p. 31-    ARE THERE ANY NATURAL RIGHTS? 59    absurd to speak of having duties or owing obligations to ourselves—  of course we may have ‘duties’ not to do harm to ourselves, but what  could be meant (once the distinction between these different meanings  of ‘to’ has been grasped) by insisting that we have duties or obligations  to ourselves not to do harm to ourselves?   (D) The essential connexion between the notion of a right and  the justified limitation of one person’s freedom by another may be  thrown into relief if we consider codes of behaviour which do not  purport to confer rights but only to prescribe what shall be done.  Most natural law thinkers down to Hooker conceived of natural law  in this way: there were natural duties compliance with which would  certainly benefit man—things to be done to achieve man’s natural  end—but not natural rights. And there are of course many types of  codes of behaviour which only prescribe what is to be done, e.g.,  those regulating certain ceremonies. It would be absurd to regard  these codes as conferring rights, but illuminating to contrast them  with rules of games, which often create rights, though not, of  course, moral rights. But even a code which is plainly a moral code  need not establish rights; the Decalogue is perhaps the most im-  portant example. Of course, quite apart from heavenly rewards  human beings stand to benefit by general obedience to the Ten  Commandments: disobedience is wrong and will certainly harm in-  dividuals. But it would be a surprising interpretation of them that  treated them as conferring rights. In such an interpretation obedience  to the Ten Commandments would have to be conceived as due to or  owed to individuals, not merely to God, and disobedience not merely  as wrong but as a wrong to (as well as harm to) individuals. The  Commandments would cease to read like penal statutes designed  only to rule out certain types of behaviour and would have to be  thought of as rules placed at the disposal of individuals and regulating  the extent to which they may demand certain behaviour from others.  Rights are typically conceived of as possessed or owned by or belonging to  individuals, and these expressions reflect the conception of moral  rules as not only prescribing conduct but as forming a kind of moral  property of individuals to which they are as individuals entitled;  only when rules are conceived in this way can we speak of rights and  wrongs as well as right and wrong actions.’    ‘Continental jurists distinguish between ‘subjektives’ and ‘objektives Recht’, which  parpesponde very well to the distinction between a right, which an individual] has,  and what it is right to do.    60 H. L. A. HART    I    So far I have sought to establish that to have a right entails having  a moral justification for limiting the freedom of another person and  for determining how he should act; it is now important to see that  the moral justification must be of a special kind if it is to constitute  a right, and this will emerge most clearly from an examination of the  circumstances in which rights are asserted with the typical ex-  pression ‘I have a right to... ’. It is I think the case that this form of  words is used in two main types of situations: (A) when the claimant  has some special justification for interference with another’s freedom  which other persons do not have (‘J have a right to be paid what you  promised for my services’); (B) when the claimant is concerned to resist  or object to some interference by another person as having no  justification (‘7 have a right to say what I think’).   (A) Special rights. When rights arise out of special transactions  between individuals or out of some special relationship in which they  stand to each other, both the persons who have the right and those  who have the corresponding obligation are limited to the parties to  the special transaction or relationship. I call such rights special  rights to distinguish them from those moral rights which are thought  of as rights against (i.e., as imposing obligations upon)! everyone,  such as those that are asserted when some unjustified interference is  made or threatened as in (B) above.   (i) The most obvious cases of special rights are those that arise  from promises. By promising to do or not to do something, we  voluntarily incur obligations and create or confer rights on those  to whom we promise; we alter the existing moral independence of the  parties’ freedom of choice in relation to some action and create  a new moral relationship between them, so that it becomes morally  legitimate for the person to whom the promise is given to determine  how the promisor shall act. The promisee has a temporary authority  or sovereignty in relation to some specific matter over the other’s  will which we express by saying that the promisor is under an  obligation to the promisee to do what he has promised. To some  philosophers the notion that moral phenomena—rights and duties  or obligations—can be brought into existence by the voluntary action  of individuals has appeared utterly mysterious; but this I think has  been so because they have not clearly seen how special the moral  notions of a right and an obligation are, nor how peculiarly they are    1 Cf Section (B} below.    ARE THERE ANY NATURAL RIGHTS? 61    connected with the distribution of freedom of choice; it would  indeed be mysterious if we could make actions morally good or bad  by voluntary choice. The simplest case of promising illustrates  two points characteristic of all special rights: (1) the right and  obligation arise not because the promised action has itself any  particular moral quality, but just because of the voluntary trans-  action between the parties; (2) the identity of the parties concerned  is vital—only this person (the promisee) has the moral justification  for determining how the promisor shall act. It is his right; only in  relation to him is the promisor’s freedom of choice diminished, so  that if he chooses to release the promisor no one else can complain.  (ii) But a promise is not the only kind of transaction whereby  rights are conferred. They may be accorded by a person consenting  or authorizing another to interfere in matters which but for this  consent or authorization he would be free to determine for himself.  If I consent to your taking precautions for my health or happiness  or authorize you to look after my interests, then you have a right  which others have not, and I cannot complain of your interference if  it is within the sphere of your authority. This is what is meant bya  person surrendering his rights to another; and again the typical  characteristics of a right are present in this situation: the person  authorized has the right to interfere not because of its intrinsic  character but because these persons have stood in this relationship.  No one else (not similarly authorized) has any mght' to interfere  in theory even if the person authorized does not exercise his right.  (iii) Special rights are not only those created by the deliberate  choice of the party on whom the obligation falls, as they are when  they are accorded or spring from promises, and not all obligations to  other persons are deliberately incurred, though I think it is true of  all special rights that they arise from previous voluntary actions.  A third very important source of special rights and obligations which  we recognize in many spheres of life is what may be termed mutuality  of restrictions, and I think political obligation is intelligible only if  we see what precisely this is and how it differs from the other right-  creating transactions (consent, promising) to which philosophers  have assimilated it. In its bare schematic outline-it is this: when  a number of persons conduct any joint enterprise according to rules  and thus restrict their liberty, those who have submitted to these  restrictions when required have a right to a similar submission from  those who have benefited by their submission. The rules may provide  ' Though it may be better (the lesser of two evils) that he should: cf. p. 62 below.    62 H. L. A. HART    that officials should have authority to enforce obedience and make  further rules, and this will create a structure of legal rights and duties,  but the moral obligation to obey the rules in such circumstances is  due to the co-operating members of the society, and they have the  correlative moral right to obedience. In social situations of this  sort (of which political society is the most complex example) the  obligation to obey the rules is something distinct from whatever  other moral reasons there may be for obedience in terms of good  consequences (e.g., the prevention of suffering); the obligation is  due to the co-operating members of the society as such and not  because they are human beings on whom it would be wrong to in-  flict. suffering. The utilitarian explanation of political obligation  fails to take account of this feature of the situation both in its  simple version that the obligation exists because and only if the  direct consequences of a particular act of disobedience are worse  than obedience, and also in its more sophisticated version that the  obligation exists even when this is not so, if disobedience increases  the probability that the law in question or other laws will be dis-  obeyed on other occasions when the direct consequences of obedience  are better than those of disobedience.   Of course to say that there is such a moral obligation upon those  who have benefited by the submission of other members of society  to restrictive rules to obey these rules in their turn does not entail  either that this is the only kind of moral reason for obedience or that  there can be no cases where disobedience will be morally justified.  There is no contradiction or other impropriety in saying ‘I have an  obligation to do X, someone has a right to ask me to, but I now see  I ought not to do it’. It will in painful situations sometimes be the  lesser of two moral evils to disregard what really are people’s rights  and not perform our obligations to them. This seems to me parti-  cularly obvious from the case of promises: I may promise to do some-  thing and thereby incur an obligation just because that is one way in  which obligations (to be distinguished from other forms of moral  reasons for acting) are created; reflection may show that it would in  the circumstances be wrong to keep this promise because of the suf-  fering it might cause, and we can express this by saying ‘J ought not to  do it though J have an obligation to him to do it’ just because the  italicized expressions are not synonyms but come from different  dimensions of morality. The attempt to explain this situation by  saying that our real obligation here is to avoid the suffering and  that there is only a prima facie obligation to keep the promise seems    ARE THERE ANY NATURAL RIGHTS? 63    to me to confuse two quite different kinds of moral reason, and in  practice such a terminology obscures the precise character of what  is at stake when ‘for some greater good’ we infringe people’s rights  or do not perform our obligations to them.   The social-contract theorists rightly fastened on the fact that  the obligation to obey the law is not merely a special case of bene-  volence (direct or indirect), but something which arises between  members of a particular political society out of their mutual relation-  ship. Their mistake was to identify this right-creating situation of  mutual restrictions with the paradigm case of promising; there are  of course important similarities, and these are just the points which  all special rights have in common, viz., that they arise out of special  relationships between human beings and not out of the character of  the action to be done or its effects.   (iv) There remains a type of situation which may be thought of as  creating rights and obligations: where the parties have a special  natural relationship, as in the case of parent and child. The parent’s  moral right to obedience from his child would I suppose now be  thought to terminate when the child reaches the age ‘of discretion’,  but the case is worth mentioning because some political philosophies  have had recourse to analogies with this case as an explanation of  political obligation, and also because even this case has some of the  features we have distinguished in special rights, viz., the right arises  out of the special relationship of the parties (though it is in this case  a natural relationship) and not out of the character of the actions to  the performance of which there is a right.   (v) To be distinguished from special rights, of course, are special  liberties, where, exceptionally, one person is exempted from obli-  gations to which most are subject but does not thereby acquire a nght  to which there is a correlative obligation. If you catch me reading  your brother’s diary, you say, ‘You have no right to read itv’. I say, ‘I  have a right to read it—your brother said I might unless he told me  not to, and he has not told me not to’. Here I have been specially  licensed by your brother who had a right to require me not to read  his diary, so lam exempted from the moral obligation not to read it,  but your brother is under no obligation to let me go on reading it.  Cases where rights, not liberties, are accorded to manage or inter-  fere with another person’s affairs are those where the licence is not  revocable at will by the person according the right.   (B) General rights. In contrast with special rights, which constitute  a justification peculiar to the holder of the right for interfering    64 H. L. A. HART    with another’s freedom, are general rights, which are asserted defen-  sively, when some unjustified interference is anticipated or threatened,  in order to point out that the interference is unjustified. ‘I have the  right to say what I think’.! ‘I have the right to worship as I please’.  Such rights share two important characteristics with special rights.  (1) To have them is to have a moral justification for determining how  another shall act, viz., that he shall not interfere.? (2) The moral  justification does not arise from the character of the particular action  to the performance of which the claimant has a right; what justifies  the claim is simply—there being no special relation between him and  those who are threatening to interfere to justify that interference—  that this is a particular exemplification of the equal right to be free.  But there are of course striking differences between such defensive  general rights and special rights. (1) General rights do not arise out  of any special relationship or transaction between men. (2) They are  not rights which are peculiar to those who have them but are rights  which all men capable of choice have in the absence of those special  conditions which give rise to special rights. (3) General rights have  as correlatives obligations not to interfere to which everyone else is  subject and not merely the parties to some special relationship or  transaction, though of course they will often be asserted when some  particular persons threaten to interfere as a moral objection to that  interference. To assert a general right is to claim in relation to some  particular action the equal right of all men to be free in the absence  of any of those special conditions which constitute a special right  to limit another’s freedom; to assert a special right is to assert in  relation to some particular action a right constituted by such special  conditions to limit another’s freedom. The assertion of general rights  directly invokes the principle that all men equally have the right to  be free; the assertion of a special right (as I attempt to show in  Section III) invokes it indirectly.   1In speech the difference between general and special rights if often marked  by stressing the pronoun where a special right is claimed or where the special! right  is denied. ‘You have no right to stop him reading that book’ refers to the reader’s    general right. ‘You have no right to stop him reading that book’ denies that the  person addressed has a special right to interfere though others may have.    ?Strictly, in the assertion of a general right both the right to forbearance from  coercion and the iberty to do the specified action are asserted, the first in the face  of actual or threatened coercion, the second as an objection to an actual or anti-  cipated demand that the action should not be done. The first has as its correlative  an obligation upon everyone to forbear from coercion; the second the absence in any  one of a justification for such a demand. Here, in Hohfeld’s words, the correlative  is not an obligation but a ‘no-right’.    ARE THERE ANY NATURAL RIGHTS? 65  Il    It is, I hope, clear that unless it is recognized that interference with  another’s freedom requires a moral justification the notion of a right  could have no place in morals; for to assert a right is to assert that  there is such a justification. The characteristic function in moral  discourse of those sentences in which the meaning of the expression  ‘a right’ is to be found—‘I have a right to...’, ‘You have no right  to...’, ‘What right have you to...?’—is to bring to bear on inter-  eeees with another’s freedom, or on claims to interfere, a type of  moral evaluation or criticism specially appropriate to interference  with freedom and characteristically different from the moral criticism  of actions made with the use of expressions like ‘right’, ‘wrong’,  ‘good’, and ‘bad’. And this is only one of many different types of  moral ground for saying “You ought...’ or ‘You ought not...’. The  use of the expression ‘What right have: you to...?’ shows this more  clearly, perhaps, than the others; for we use it , just atthe point where  interference is actual or threatened, to call for the moral title of the  person addressed to interfere; and we do this often without any  suggestion at all that what he proposes to do is otherwise wrong and  sometimes with the implication that the same interference on the  part of another person would be unobjectionable.   But though our use in moral discourse of ‘a right’ does pre-  suppose the recognition that interference with another’s freedom  requires a moral justification, this would not itself suffice to establish,  except in a sense easily trivialized, that in the recognition of moral  rights there is implied the recognition that all men have a right to  equal freedom; for unless there is some restriction inherent in the  meaning of ‘a right’ on the type of moral justification for interference  which can constitute a right, the principle could be made wholly  vacuous. It would, for example, be possible to adopt the principle  and then assert that some characteristic or behaviour of some human  beings (that they are improvident, or atheists, or Jews, or Negroes)  constitutes a moral justification for interfering with their freedom;  any differences between men could, so far as my argument has yet  gone, be treated as a moral justification for interference and so  constitute a right, so that the equal right of all men to be free would  be compatible with gross inequality. It may well be that the expression  ‘moral’ itself imports some restriction on what can constitute a moral  justification for interference which would avoid this consequence,  but I cannot myself yet show that this is so. It is, on the other hand,    66 H. L. A. HART    clear to me that the moral justification for interference which is to  constitute a right to interfere (as distinct from merely making it  morally good or desirable to interfere) is restricted to certain special  conditions and that this is inherent in the meaning of ‘a right’ (unless  this is used so loosely that it could be replaced by the other moral  expressions mentioned). Claims to interfere with another’s freedom  based on the general character of the activities interfered with (e.g.,  the folly or cruelty of ‘native’ practices) or the general character of  the parties (‘We are Germans; they are Jews’) even when well founded  are not matters of morat right or obligation. Submission in such cases  even where proper is not due to or owed to the individuals who inter-  fere; it would be equally proper whoever of the same class of persons  interfered. Hence other elements in our moral vocabulary suffice to  describe this case, and it is confusing here to talk of rights. We saw  in Section II that the types of justification for interference involved  in special rights was independent of the character of the action to the  performance of which there was a right but depended upon certain  previous transactions and relations between individuals (such as  promises, consent, authorization, submission to mutual restrictions).  Two questions here suggest themselves: (1) On what intelligible  principle could these bare forms of promising, consenting, sub-  mission to mutual restrictions, be either necessary or sufficient,  irrespective of their content, to justify interference with another’s  freedom? (2) What characteristics have these types of transaction or  relationship incommon? The answer to both these questions is I think  this: If we justify interference on such grounds as we give when we  claim a moral right, we are in fact indirectly invoking as our justi-  fication the principle that all men have an equal right to be free.  For we are in fact saying in the case of promises and consents or  authorizations that this claim to interfere with another’s freedom is  justified because he has, in exercise of his equal right to be free,  freely chosen to create this claim; and in the case of mutual restrict-  ions we are in fact saying that this claim to interfere with another's  freedom is justified because it is fair; and it is fair because only so  will there be an equal distribution of restrictions and so of freedom  among this group of men. So in the case of special rights as well as  of general rights recognition of them implies the recognition of the  equal right of all men to be free.    IV  THE USES OF ‘SOVEREIGNTY’    STANLEY I. BENN    JEAN BODIN defined ‘sovereignty’ as ‘supreme power over citizens and  subjects, unrestrained by law’. Since then criticisms of theories in  which the term has been employed have led to repeated attempts to  redefine it and to distinguish different kinds of ‘supreme power’ and  examine the relations between them. For Austin the sovereign is ‘a  determinate human superior, not_in a habit of obedience to a like.  superior, (receiving) habitual obedience from the bulk of a given  society .! Applying this notion to the British Constitution Dicey finds  it necessary to distinguish ‘legal sovereignty’ and ‘political sove-  reignty’.? Lord Bryce employs a different distinction. ‘Legal sove-  reignty’, he says, is primarily the concern of the lawyer: “The sovereign  authority is to him the person (or body) to whose directions the law  attributes legal force.’? This kind of sovereignty, Bryce says, “is created  by and concerned with law, and law only’.+ But it is also possible to  detect a ‘practical sovereign’: ‘The person (or body of persons) who  can make his (or their) will prevail whether with the law or against  the law. He (or they) is the de facto ruler.’> More recently Mr. W. J.  Rees has attempted an exhaustive analysis of the ways in which  ‘sovereignty’ has been used and has tried to establish three possible  senses. He begins with ‘power’: ‘To exercise power .. . is to deter-  mine the actions of persons in certain intended ways. There are,  however, different species of power, and these may be distinguished  according to the means used to determine persons’ actions.” He    From Political Studies, Vol. 3 (Clarendon Press, 1955), pp. 109-22. Reprinted by  permission of the author and the Clarendon Press.   J. Austin, Lectures on Jurisprudence (5th ed.), p. 221.   2A. V. Dicey, Law of the Constitution (gth ed.), p. 72.   3Lord Bryce, Studies in History and Jurisprudence, vol. ii (1901), p. 51-   *Tbid., p. 56.   *Ibid., p. 59-60.   Sw. J. Rees, ‘The Theory of Sovereignty Restated’, in Mind, vol. lix (1950).   "Ibid., p. 511.    68 STANLEY I. BENN    distinguishes three such species to each of which corresponds a species  of supreme power, or sovereignty.    ‘Legal sovereignty’ is a capacity ‘to determine the actions of persons in  certain intended ways by means of a law. . . where the actions of those who  exercise the authority, in those respects in which they do exercise it, are  not subject to any exercise by other persons of the kind of authority  which they are exercising.' ‘A person or a body of persons may be said  to exercise coercive Sovereignty, or supreme coercive power, if it determines  the actions of persons in certain intended ways by means of force or the  threat of force, and if the actions of the persons who exercise the power,  in those respects in which they do exercise it, are not themselves capable  of being similarly determined.? ‘To exercise political influence ...is to  determine in certain intended ways the actions, jointly or severally, of  the legal and coercive sovereigns, provided always that their actions are  determined by some means other than a rule of law or a threat of force... .  To exercise sovereignty in this sense is to exercise political influence, as  now defined, to a greater degree than anyone else... .’8    ‘Legal sovereignty’, it seems, might be attributed to Parliaments  or amending organs or constitutions ;* ‘coercive sovereignty’ to armies  or similar organized forces or a socially coercive power such as  existed under the frank-pledge system;> ‘influential sovereignty’ to  a ruling class, the majority of the electorate, a priesthood, or some  other such group.®   I propose in this paper to isolate and examine these and other  usages, to try to discover in what kinds of study, if any, each is likely  to be useful; and to determine whether they possess any common  element that would justify the use of the one word ‘sovereignty’ to  cover them all.    II. Legal Sovereignty    It has often been said that a ‘legal sovereign’ is necessary in every    'W. J. Rees, ‘The Theory of Sovereignty Restated’, in Mind, vol. lix (1950), p- 508.  (My italics $.1.B.)   *tbid., p. 511. (My italics S.1.B.)   3Ibid., p. 514. (My italics S.1.B.)   *Ybid., pp. 516-17. It is not clear from this passage that Mr. Rees would ascribe  legal sovereignty to a constitution. Such an ascription has been made by other  writers, however (e.g. by Sir Ernest Barker in Principles of Social and Political Theory,  p- 61, and Lord Lindsay in The Modern Democratic State, pp. 222-9), and I propose to  examine the implications of this usage.   SW. J. Rees, op. cit., p. 509.   *Ibid., p. 513.    THE USES OF ‘SOVEREIGNTY’ 69    state if legal issues are to be settled with certainty and finality.’ From  one point of view, this necessity derives from the nature of a judicial  decision understood as one determining a dispute within the frame-  work of established rules (as distinct from one made according to  subjective criteria). The judge called upon co settle a dispute sees  law as a system of rules to guide his decision; and sucha system needs  criteria of validity determining which rules belong to it; it needs  a supreme norm, providing directly or indirectly the criteria of  validity of all other norms, and not itself open to challenge.’   Where a written constitution exists, it is approximately true to  say that the constitution itself provides such a supreme norm; and  in this sense one may speak of the ‘legal sovereignty of the consti-  tution’. An amendable written constitution will provide criteria for   ‘E.g. Sir Ernest Barker, op. cit., p. 59: “There must exist in the State, as a legal  association, a power of final legal adjustment of all legal issues which arise in its ambit.’  W. J. Rees, op. cit., p. 501: ‘Laws can only be effectively administered if there exists    some final legal authority beyond which there is no further legal appeal. In the absence  of such a final legal authority no legal issue could ever be certainly decided, and    government would become impossible.’ J. W. Salmond: Jurisprudence (10th ed. by -  Glanville Williams), App. I, p. 490: ‘It seems clear that every political society involves ”    the presence of supreme power... . For otherwise all power would be subordinate,  and this supposition involves the absurdity of a series of superiors and inferiors ad  infinitum.’ But contrast John Chipman Gray, The Nature and Sources of the Law, p. 79  (quoted by W. Friedmann, Legal Theory, 2nd ed., p. 147): ‘The real rulers of a political  society are undiscoverable. They are the persons who dominate over the wills of their  fellows. In every political society we find the machinery of government . .. . We have  to postulate one ideal entity to which to attach this machinery, but why insist on  interposing another entity, that of a sovereign? Nothing seems gained by it, and to  introduce it is to place at the threshold of Jurisprudence a very difficult, a purely  academic, and an irrelevant question.’ Gray seems to argue (a) that the influential  sovereign is undiscoverable; (6) that the jurist is needlessly multiplying the entities  by postulating a legal sovereign. But (b) is not a necessary inference from (a).    2Cf. H. Kelsen, General Theory of Law and State (1945), p. 124: “The legal order...  is therefore not a system of norms coérdinated to each other, standing, so to speak,  side by side on the same level, but a hierarchy of different levels of norms. The unity  of these norms is constituted by the fact that the creation of one norm—the lower one—  is determined by another—the higher—the creation of which is determined by a still  higher norm, and that this regressus is terminated by a highest, the basic norm which,  being the supremé reason of validity of the whole legal order, constitutes its unity.’  And Salmond, op. cit., sec. 50: ‘It is requisite that the law should postulate one or  more first causes whose operation is ultimate, and whose authority is underived. In  other words, there must be found in every legal system certain ultimate principles from  which all others are derived, but which are themselves self-existent. ... Whence  comes the rule that Acts of Parliament have the force of law? This is legally ultimate;  its source is historical only, not legal....No Statute lays it down. It is certainly  recognized by many precedents, but no precedent can confer authority upon pre-  cedent. It must first possess authority before it can confer it. If we enquire as to the  number of these ultimate principles, the answer is that a legal system is free to  recognize any number of them, but it is not bound to recognize more than one.’    70 STANLEY I. BENN    identifying valid amendment. But even so the constitution may not be  altogether identified with the supreme norm; for there may be rules  for its interpretation which judges accept as binding but which are  not prescribed in the constitution. Effectively, therefore, it is the  traditional judicial interpretation of the constitution that is the  supreme norm.   The absence of a written document does not vastly alter the situ-  ation. The supreme norm in English law is provided in part by the  maxim ‘Parliament is sovereign’. But this leaves open the question  ‘What is an Act of Parliament?’ A judge must be able to refer to a  criterion superior in status to an Act, which will establish which rules  are Acts. (In a recent article! Mr. Geoffrey Marshall has drawn at-  tention to the way in which the interpretation of Parliamentary  sovereignty is changing. The critical question, in his view, is: “What is  Parliament?’ This seems to me to put the problem the wrong way. A  judge requires not a definition of the organ Parliament, but a  criterion by which to recognize a norm of the type ‘Act of Parliament’.  For judicial decisions are reached in the light of norms, not of organs.  Mr. Marshall seems to argue that there is a difference in principle  between the view typified by Lord Campbell’s dictum that the Parlia-  ment roll provides conclusive evidence of a statute’s validity? and the  rule in Harris v. Dénges which implies that a rule issuing from  Parliament by a procedure other than that legally prescribed is not  an Act.? But the difference is not that Parliament is held, in the one  view, to be above the law, and, in the other, to be subject to law; it  lies in the stringency of the criteria which, in the view of the court,  a rule must satisfy in order to be deemed an Act of Parliament.)   An Act of Parliament, therefore, is subordinate to the supreme  constitutional norm. It is, however, a rule of a special type in that its  binding force cannot be challenged on the grounds that it is in  substantial conflict with any superior norm. (In this respect it differs  from an Act of Congress or a statutory instrument.) In view of this    'G. Marshall, ‘What is Parliament? The Changing Concept of Parliamentary  Sovereignty’, in Political Studies, vol. ii, no. 3 (1954), pp. 193-209.    2In Edinburgh and Dalkeith Rly. Co. v. Wauchope (1842): ‘All that a Court of Justice  can do is to look to the Parliamentary roll: if from that it should appear that a bill  has passed both Houses and received the Royal assent, no Court of Justice can inquire  into the mode in which it was introduced into Parliament, nor into what was done  previous to its introduction, of what passed in Parliament during its progress in its  various stages through both Houses.’    3See G. Marshall, op. cit., for a full discussion of this and other relevant cases.    THE USES OF ‘SOVEREIGNTY’ 71    peculiarity, it might be useful to ascribe to Acts of Parliament  immediate supremacy as decisive rules in questions of substance, while  the norm from which their validity derives might be termed ultimately  supreme. This is a reinterpretation ofa distinction made by Sir Ernest  Barker, who ascribes ultimate sovereignty to the constitution and  immediate sovereignty to a supreme legislative organ; but it avoids  the awkward asymmetry of his ascription:' the word ‘sovereignty’  can scarcely be precisely and unambiguously defined and yet fit with  equal comfort both an organ and a norm.’   The interpretation of ‘legal sovereignty’ I have offered has, I  believe, the advantage that while meeting the judicial need for an  ultimate point of reference, it avoids the criticisms directed against  the command theory with which the notion of sovereignty has  traditionally been associated. Whether law is command is irrelevant.  For the judge is interested in the ‘source’ of a law only if by ‘source’  is meant the higher norm from which its validity derives; its legis-  lative origin is a fact to be assessed according to established legal  criteria. Further, ‘legal sovereignty’, as I conceive it, need not imply  that law is ‘effective’, i.e. generally observed in an actual conimunity.  A student might apply ancient legal principles to hypothetical cases;  in so doing he would be acting in a way closely parallel to a judge on  an English bench, and would find the same necessity for a supreme  norm. The same would apply to the student of Utopian or Erewhon-  ian law. Again, ‘sanction’ is non-essential to ‘sovereignty’ in this sense,  and the difficulties which arise in applying some legal theories of  sovereignty to constitutional and administrative law thus do not  arise here.’   This notion of the ‘supreme norm’ is essential to any study of the  rules governing decisions within a normative order. It is of primary  importance for the practising lawyer, and for the jurist. It is also    ‘Sir Ernest Barker, op. cit., bk. ii, sec. 5. Sir Ernest recognizes the asymmetry, but  considers it ‘inherent in the nature of the case’ (p. 63).  *Ibid., loc. cit.; and W. J. Rees, op. cit., pp. 516-17.    3Cf. L. Duguit, Law and the Modern State, p. 31: ‘In those great state services which  increase every day... the state... intervenes in a manner that has to be regulated  and ordered by a system of public law. But this system can no longer be based on the  theory of sovereignty. It is applied to acts where no trace of power to command is  to be found. Of necessity a new system is being built, attached indeed by close bonds  to the old, but founded on an entirely new theory. Modern institutions . . . take their  origin not from the theory of sovereignty, but from the notion of public service.’  (Quoted in H. E. Cohen, Recent Theories of Sovereignty, p. 40.) This notion is in no way  incompatible with the view of sovereignty I am suggesting.    72 STANLEY I. BENN    of significance to the administrator, and to the student of adminis-  tration interested in the legal sources and limitations of admin-  istrative discretion rather than in the motives which determine the  exercise of discretion.   In historical or sociological studies and those concerned with  moral questions the notion of a supreme norm is at most only  indirectly relevant. If we ask such questions as ‘How do laws  develop?’, “What governs the content of law in this (or any) com-  munity?’, “What is the role of law in this (or any) society ?’,’ we shall  need a way of distinguishing law from other modes of social control,  but the judicial criterion of validity will not necessarily be an element  in such a principle of differentiation. Of course, any description of  the life of a community must, to be complete, include an account of  its judicial system, and so of the assumptions made by the men whose  business it is to reach decisions within this normative order; but the  supreme norm will figure in a sociologist’s account as a feature of the  conceptual apparatus employed by lawyers, not as part of his own.   Similarly, in asking the moral question ‘What ought laws to be like?’  we need to distinguish laws from, say, conventional moral rules. But  the principle of differentiation must now be related to those aspects  of law which constitute it a distinct problem (e.g. the coercive  sanction, or the presumption that most people will obey it), and the  judicial criterion of validity will not necessarily figure as part of it.   The questions of political science are both normative and des-  criptive. If the political scientist is concerned with the state as a  normative order, the idea of the supreme norm will have the same  relevance for him as it has for the lawyer; but if his questions concern  men’s actual political behaviour, his view of law will be much more  that of the sociologist.    III. Legislative Sovereignty    The approach to ‘legal sovereignty’ that I have suggested derives  from reflection upon the activities of a judge, for whom the law  appears, at any particular moment, as a body of given rules to guide  his judgement. For the political scientist, however, law appears in the  process of creation; he is concerned with law-making and law-    'Cf. R. Wollheim’s distinction between questions about law which are in Juris-  rudence, and those which are not, in ‘The Nature of Law’, in Political Studies, vol. ii    1954), Pp. 139-40.    THE USES OF ‘SOVEREIGNTY’ 73    makers;! he is interested in ‘legislative organs’, and not merely in  ‘legal norms’. I propose accordingly to inquire now whether there is  a place in the political scientist’s vocabulary for ‘supreme legis-  lative organ’, and what it might mean to attribute ‘supremacy’ in  this way. To distinguish the supremacy of a norm from that of a  legislative organ, I propose to use ‘legal sovereignty’ for the former  and ‘legislative sovereignty’ for the latter.   A political scientist might significantly classify legislative organs  in a legal order into superior and inferior (or subordinate), and he  might arrange them hierarchically as a sort of reflection of the judge's  hierarchy of norms. The judge will deem an Act of Parliament super-  ior in status to a statutory instrument; the political scientist will  deem Parliament superior in competence to a minister acting as  legislator. But does it follow that the necessity which leads the judge  to postulate a supreme norm is paralleled by a similar necessity  leading the political scientist to postulate a supreme legislative organ?  Such an organ would be omnicompetent, that is, competent to legis-  late on all matters without the possibility that any of its rules might be  invalidated by reason of conflict with some other rule not of its own  making.’ It might reasonably, therefore, be called ‘legislative sove-  reign’. But such a sovereign is not logically necessary to a legal order.  A constitution might allocate fields of legislative competence between  co-ordinate organs, or place certain matters beyond the competence  of any organ (e.g. by a Bill of Rights); and in respect of such limit-  ations the constitution might be unamendable. (This qualification  is important, since the competence to amend the constitution in these  respects would be, on an ultimate analysis, omnicompetence.) In such  a case, there would be no omnicompetent organ. On the other hand,  one might speak of one organ with supreme competence in a particular  field or of several such organs; and that would mean that though the  rules of such an organ might be invalidated by reason of conflict with  the constitution they could not be invalidated through conflict with  the rules of any other organ. But one cannot say a priori that every  legal order must possess one or more ‘supreme legislative organs’ even    'Cf. Kelsen, op. cit., p. 39: ‘If we adopt a static point of view, that is, if we consider  the legal order oniy in its completed form or im a state of rest, then we notice  only the norms by which the legal acts are determined. If, on the other hand, we adopt  a dynamic outlook, if we consider the process through which the legal order is created  and executed, then we see only the law-creating and law-executing acts.’    Except for a rule of another organ to which this one had expressly delegated a  limited competence to make rules, in a given field, of equal status to its own (e.g. by  a ‘Henry VIII’ clause).    74 STANLEY I. BENN    in this sense. A constitution that is unamendable (at least in respect  of its allocation of fields of competence) might constitute two (or  more) organs co-ordinate in the same field, so that a rule enacted by  either might set aside a rule of the other. A judge operating such an  order would require only some general prescription to show which  of conflicting rules enacted by different organs he should deem  binding; and this could be met by the principle that in case of  conflict a later rule should repeal an earlier. This might be highly  inconvenient if the co-ordinate organs were operated by men  of different opinions, and competition developed for the latest  place. But this could be avoided without making one organ supreme  in each field, if, for example, co-ordinate organs were operated by  members of one highly disciplined political party, or by men who  reached decisions by mutual agreement before legislating. The  judge need not then be faced with conflicts any oftener than he is  in England.   There is thus neither logical nor practical necessity for a legislative  sovereign in every state, though there may be states in which such  organs are discoverable. But it should be stressed that to ascribe  ‘sovereignty’ to a legislative organ in either of the senses just con-  sidered is to attribute to it not ‘power’, in the sense of ability ‘to  determine the actions of persons in certain intended ways’, but legal  capacity or ‘competence’; it is to say no more than that a judge  will set an organ’s rules in a particular kind of relation to the rules  of other organs. It is, indeed, a statement about the formal structure  of a legal order. It does not presuppose any actual ability possessed by  the men acting through an organ to determine the actions of other  persons in intended ways. It does not even require that the action of  the judge himself should be so determined; for the person occupying  judicial office may disregard the law. Law is normative: it prescribes  how a person must act to function as a judge within the legal order;  it does not predict that he will so act.   Yet it is true that law-making is one way of ‘determining the  actions of persons in certain intended ways’. A sociologist seeking  to explain behaviour in a community would need to take its statutes    ‘*Power’ suffers from a systematic ambiguity. When we refer to ‘the powers of Local  Authorities’, ‘Parliament’s power to legislate on any subject whatsoever’, or ‘legisla-  tive powers of Ministers’, we mean ‘competence’ or ‘entitlement’—i.e. that they are  ‘empowered’ to act in this or that way. This is a quite different sense from that implied  by Mr. Rees’s definition: ‘to determine the actions of persons in certain intended ways’.  The ‘power’ possessed by a Local Authority to orginize concerts is clearly not power  in this second sense. Neither is it a species of a ‘power’ genus. Mr. Rees’s argument  suffers from his failure to make this distinction. Vide op. cit., p. 511.    THE USES OF ‘SOVEREIGNTY’ 75    into account, since the knowledge that a particular rule is a statute  may condition the behaviour of those subject to it. Consequently,  legislators can often be regarded as determining the actions of persons  in intended ways. But there is no warrant for automatically trans-  ferring ‘supremacy’ as applied to competence to any power deriving  from such competence. It is not, for instance, necessarily true that  the men who operate the organ termed ‘supreme’ receive more  obedience than those operating a ‘subordinate’ organ. The amending  organ of the U.S.A.—Congress together with three-fourths of the  States—is omnicompetent (or very nearly so), yet the Eighteenth  Amendment was much less effective than most Acts of Congress.'  ‘Supremacy’, then, is relevant, when applied to legislative organs,  only when a legislative act is considered as a directive to a judge: in  all other contexts it is out of place.    IV. The Sovereignty of the State in its international aspect    There remains to be considered, before leaving the juristic field,  the sense of ‘sovereignty’ as applied in international relations.   It has often been argued that state ‘sovereignty’ is incompatible  with international law. The term implies that the state is a self-  sufficient legal order; and this must mean that a judge operating that  order need seek no further than its own supreme norm. The tra-  ditional problem then arising is put by Kelsen in the following terms:    ‘That the State is sovereign means that the national legal order is an  order above which there is no higher order. The only order that could be  assumed to be superior to that of the national legal order is the inter-.  national legal order. The question whether the State is sovereign or not  thus coincides with the question whether or not international law is an  order superior to national law.”?    1This is not to suggest that ‘supreme legislative power’ is necessarily meaningless.  It could conceivably be used in historical and sociological studies. To attribute it to  A might mean (1) that all the laws he made were invariably effective and could not be  overturned (which would be the ‘power’ equivalent to ‘supreme competence’); (2) that  they were more generally effective than anyone else’s (though the use of ‘superior’  rather than ‘supreme’ might accord better with common usage); or (3) that his laws  were usually effective, and his conduct was not determined by laws made - others.  Examples of (1) probably cannot be found; (2) would be useful only if the effec-  tiveness of laws depended on their sources, which seems improbable; (g) might be  true of a few autocrats, but must be unusual. A fourth apparent possibility, viz. that  in consequence of A’s possessing supreme legislative competence his laws are more  likely to be effective, ceteris paribus, than rules liable to invalidation, is really only  another way of saying that the legal order is effective. None of these senses seems  important and I shall not consider them further.    ?Kelsen, op. cit., p. 384.    76 STANLEY I. BENN    A pluralistic position, he argues, is inadmissible: two legal orders  with conflicting norms cannot be simultaneously valid for the same  territory.’ The choice lies, therefore, between the primacy of inter-  national law, with non-‘sovereign’ national legal orders deriving  validity from it, and the primacy of national law endowing inter-  national law with validity to the extent that it recognizes it. But the  consequence of the second view is ‘state solipsism’,? for now only one  State can be held to be sovereign; other legal orders exist for it only  as derivatives of itself, either directly or inuirectly through its re-    cognition of international law. Kelsen adds:    ‘It is, however, logically possible that different theorists interpret the  world of law by proceeding from the sovereignty of different States. Each  theorist may presuppose the sovereignty of his own State, that is to say, he  may accept the hypothesis of the primacy of his own national legal order.  Then he has to consider the international law which establishes the relations  to the legal orders of the other States and these national legal orders as  parts of the legal order of his own State, conceived of as a universal legal  order. This means that the picture of the world of law would vary according  to what State is made the basis of the interpretation. Within each of these  systems, erected on the hypothesis of the primacy of national law, one State  only i is sovereign, but in no two of them would this be the same State.”    Kelsen appears to regard this solution as irrefutable but unsatis-  factory. I believe, however, that otherwise stated it can throw light on  the place of ‘sovereignty’ in international law, and of international law  within the structure of ‘sovereign’ national orders.   Within the English legal framework an English judge will take  cognizance of international law as a part of English law to the extent  that its rules do not conflict with other rules of English law; the  national laws of other states will equally be subject to the criteria  of validity of the English legal order, and in so far as they are re-  cognized by a judge will become parts of that order. In this sense,  then, it is true that for the English judge, the only sovereign order is  his own. But mutatis mutandis the same is true of a French or any other  national judge. Each can operate only within his own order, and for  him it is self-sufficient. This is again true of the international lawver.  His order is a self-sufficient order embracing national orders as sub-  ordinate parts. A given rule may well be valid in one of these orders    ‘Ibid., p. 363.  *Ibid., p. 387.  3Ibid., p. 386.    THE USES OF ‘SOVEREIGNTY’ 77    (national or international) and invalid in another. But there is not  here, as Kelsen supposes, any contradiction, and if it involves ‘State  solipsism’, this need cause no embarrassment. Kelsen’s argument,  that ‘two norms which by their significance contradict and hence  logically exclude one another, cannot be simultaneously valid’, misses  the point. He requires that there shall be only one objectively valid  legal order. But to ascribe ‘self-sufficiency’ to an order rules out the  ascription to it of ‘validity’, which for Kelsen is meaningful only  within an order. Accordingly, many such systems can logically exist,  side by side, and none can claim greater legal validity than another.   It follows, as a corollary to this analysis, that if the international  lawyer refers to ‘sovereign orders’, or the national lawyer to ‘other  sovereign states’, then the sense of the word ‘sovereign’ as here used  must be different from that in which either applies the term to his  own order, as self-sufficient. He is now using it of a particular type  of partial order, analogous to other partial orders, like ‘corporations’,  recognized by various legal orders. The precise definition of ‘a  sovereign state’ in any given legal order is a question of particular  not of general jurisprudence and cannot be settled by reflection upon  the nature of legal systems in general.    V. Sovereignty as ‘supreme coercive power’    ‘Sovereignty’ as ‘supreme coercive power’ is not, I believe, relevant  to or meaningful in a normative study of political institutions.   If we begin by defining the state as a coercive order, that is, as an  order maintained by the exercise or threat of physical force, then  coercive power is, by definition, necessary to it. If we define it in  some other way, then,.in Mr. Rees’s phrase, coercive power is  ‘causally necessary’ to it, if it is to be capable of surviving violent  opposition. In either case, the coercive power attributed is a mode of  operation, or an institutional framework, within which action is  undertaken by. whatever men happen to occupy ‘the appropriate  offices or to fit the constitutional categories which the order provides.   For instance, to say that ina particular state the coercive power is  exercised by the Army is to say that this mode of state action is the  proper function of any group which satisfies a set of legal or con-  ventional conditions constituting it the coercive organ of the state,  and which acts according to the procedures proper to such an organ  (e.g. under orders from the Minister for War or the Commander-in-  Chief). In this context only one coercive power is possible in a state:    78 STANLEY I. BENN    for the term must refer either to a mode of action within the single   _ normative order, or to the organs whose mode of action it is. If  several organs employ this mode, they all operate within the same   _order, and so jointly constitute its coercive power. The state’s coercive  power may therefore in a sense be divided, but so long as we think of  the state as an order there is no point in saying of one or other organ,  or of a group of organs, that itis supreme. For ‘supremacy’ implies the  possibility of conflict, and a conflict of coercive organs is incompatible  with the conception of the state as an order. Thus if conflict does  arise between groups qualified to act as coercive organs, then at least  one group must be acting otherwise than as a state organ. For  example, an army in rebellion against the established Government  is not acting as a state organ. (Of course, in any territory at any  moment there may well be more than one actual coercive organ-  ization: in 1932, besides the coercive forces of the German Republic  there existed the Brown Shirts. But the Brown Shirts were not part  of the state order.)   One further point—the distinction drawn by Mr. Rees between  political orders in which coercive sovereignty is exercised by an  institutionally coercive power and those in which itis exercised bya  socially coercive power! is misleading in two ways. To reserve the  term ‘institutional’ for coercive power exercised by professional  armies, &c., obscures the fact that where all, or nearly all, the  members of a community collectively constitute the coercive organ,  their function is ‘institutional’. Secondly, inasmuch as Mr. Rees has  in mind the classification of political orders, the attribution of  ‘supremacy’ to coercive organs is redundant, and nothing is lost by  abandoning the term.   In historical and sociological studies, ‘supreme coercive power’  may well be meaningfully used. A statement like: “By 1649, the New  Model Army had emerged as the supreme coercive power in England’  is not concerned with institutional relations in the English constitu-  tional order, but with power relations between groups of men ina  particular territory. In such studies we may well compare the coercive   ower of one group with that of another. For the historian, Brown  Shirts and Communists are as much factors of the 1932 German  situation as were the armed forces of the Republic, and he might  declare one of them ‘supreme’, in the sense that, had armed conflict  developed, it could have defeated its rivals. It is doubtful, however,  whether the term is helpful in describing any but the simplest situa-  1W. J. Rees, op, cit., pp. 509-10, and Section I above.    THE USES OF ‘SOVEREIGNTY’ 79    tions. In peaceful conditions we could say that coercive sovereignty  is exercised by the coercive organs of the state; and after a civil war  we could attribute it to a victorious army that remained united. But  it would rather mislead to try to apply it, for example, to Sicily in  the days of the Mafia or France in the days of the Maquis.’ In any  case, even if the seat of coercive sovereignty can be located, the  possibilities of inquiry opened up are limited. It will give us no way  of understanding the importance in the determination of policy of  those controlling the coercive organs of the state. Taken collectively,  the Germany army, navy, air force, Gestapo, and S.S. may have con-  stituted the coercive sovereign in Hitler’s Germany; but to understand  the part played by those controlling these organs in shaping political  events we have to consider them separately, not collectively, and  to examine their mutual relations and rivalries and the power each  exercised at any given moment. To lump in problems of this sort is  to obscure rather than to illumine.    VI. Sovereignty as ‘the strongest political influence’    The first question here concerns the type of discussion to which  ‘influence’ may be relevant. I drewa distinction earlier? between words  of two logical types—‘competence’ and ‘power’ (as ability to deter-  mine the actions of persons in intended ways). ‘Influence’ is a word  of the same type as ‘power’. To establish A’s competence we examine  his status in a normative order; to establish his influence we must  observe how men behave in relation to him, whether for instance  they act on his suggestions or consult his wishes. ‘Influence’, and  consequently ‘the strongest political influence’, have thus no place  in a normative study. It is only in historical or sociological studies  that they can be meaningfully employed. It is of course true that a  man’s status in a normative order may be a source of influence; but  the extent of that influence, and, indeed, its very existence, cannot  be established by normative study.   Is the search for ‘an influential sovereign’ likely to be fruitful in  historical and sociological studies? We must distinguish, first, two  senses of ‘influence’: as in (1) ‘Climate influences vegetation’ and (2)  ‘Rasputin’s influence over Nicholas IJ’. In (1) no more than ‘effect’  is implied: there is no suggestion of intention; in(2) the effect produced  is one intended. When we speak of ‘the strongest political influence’   1Cf. Lord Bryce, op. cit., p. 63.   2Section III above.    80 STANLEY I. BENN    we are presumably thinking of some group which can shape govern-  mental policy to its own purposes. We are using ‘influence’ therefore  in sense (g).   Now we should not say that a group was the ‘influential sovereign’  merely because it had occasionally shaped government policy as it  intended. That would multiply sovereigns endlessly and deprive the  term of all point. As Mr. Rees has pointed out, ‘sovereignty’ r resembles  dispositional words in that it implies recurrent capacity to determine  policy in tended ways under understood conditions.’ In seeking an  influential sovereign, therefore, we should be seeking a stable domi-  nant influence over a fairly wide range of political issues.   In states of one type a single group (e.g. a ruling class), able  decisively to influence policy whenever it operated, could be regarded  as such an influence. The value of this approach, however, would  depend on the range of common interests from which the group’s  identity derived and which therefore constituted its field of operation.  In states of another type Governments are sensitive to pressures  from diverse interests, and political decisions are thus the outcome of  an interplay of influences rather than expressions ofa single dominant  influence. If we seek an influential sovereign here, then, we are likely  to be seriously misled; terms like ‘lobby’ and ‘pressure group’ will  be much more appropriate analytical concepts.   Sometimes influence is attributed to ‘the electorate’ or ‘the  majority’ as one might attribute it to ‘the bankers’ or a ‘ruling class’.  This is a mistake. Groups such as ‘the bankers’ or ‘the ruling class’  derive identity from common interest and homogeneity of intention;  ‘the electorate’ denotes a state organ. All that electors have in common  is the right to vote. Severally, or in groups, they may exercise influence  deriving from electoral competence; but there is not therefore one  super-influence of the electorate as such. An election is a procedure  in which influences are pitted against one other; what emerges is a  result, or an ‘effect’, not a new influence. We cannot say of the elec-  torate that it influences policy as it intends; it has no single intention,  only a multitude of intentions given different weights by the electoral  process. It is no more accurate to assign influential sovereignty to  ‘the majority’ (as Mr. Rees seems to do).? In any election a certain  aggregate of interests is more numerous than another, and this  arithmetical relation, corresponding to a recognized legal procedure,  is a source of influence for the groups concerned. But the aggregates   1W. J. Rees, op. cit., pp. 514-15.    ?Tbid., pp. 512-13.    THE USES OF ‘SOVEREIGNTY’ 81    at the next election will be differently constituted; in five elections  there will be five majorities; and we should not treat them as though  they were one group, the majority, exercising a stable dominant  influence. Consequently, the inference to be drawn from ‘The majority  (or the electorate) is sovereign’ is not that Government is sensitive to  a specified influence but that it is sensitive to all influences.   Finally, ‘influential sovereignty’ might be applied to an organiza-  tion, like a Church or a Communist party, which has a policy on all,  or most, matters, and is able to make it effective. Bur here again the  policy is not the intention of a group identifiable by common interest,  but the result of an interplay of influences within the organization.  The internal politics of influential organizations need to be inter-  preted in terms of pressure groups just as much as do the politics  of states. To attribute sovereignty to the Communist party is not to  provide an explanation of the changes in Soviet policy since 1917:  it is the struggle for power within the party that is the point of interest  for the student of Soviet history.   The concept ‘influential sovereignty’ has the disadvantage, then,  that it may direct attention to the wrong questions, or conceal the  need for inquiry beyond the point where the influential sovereign has  been identified.    Vil    In this paper I have identified six senses in which ‘sovereignty’   might be meaningfully employed:   (a) to express the supremacy of a norm in a legal hierarchy, as  viewed by a lawyer, or by a student concerned with the legal  limits of discretion;   (b) in a study of constitutions as normative orders, to refer either  to the omnicompetence, or to the supreme competence within its  field, of a legislative organ;   (c) to express the self-sufficency of a legal order from the point of  view of a lawyer operating within it;   (d) to refer to a particular kind of partial order, the definition of  which may vary from one legal order to another (its utility in this  sense being limited to particular jurisprudence);   (e) to express the ability of bodies such as armed forces to defeat  all probable rivals;   (f) to express the ability of a sectional interest decisively to influence    policy.    82 STANLEY I. BENN    The first four senses are relevant to normative studies and cannot  be directly utilized in historical or sociological studies without  confusion. Each of them is a useful concept in its own field, but they  seem to have little in common. The first two share the idea of ‘su-  premacy’ but in slightly different senses of that word; the third is  an expression of totality, rather than supremacy; the fourth implies  neither notion. The fifth and sixth senses, unlike the first four, do  imply ability to determine other people’s conduct; and it is in these  senses alone that sovereignty implies supreme power. These two  senses may be relevant to historical or sociological studies, and are  not relevant to normative studies; their usefulness where they are  relevant is limited, for they can be seriously misleading.   In the light of this analysis it would appear to be a mistake to treat  ‘sovereignty’ as denoting a genus of which the species can be distin-  guished by suitable adjectives, and there would seem to be a strong  case for giving up so Protean a word.    Vv    AUTHORITY  (1) R. S. PETERS    1. Authority and Artifice.   THERE are good reasons as well as personal excuses for ushering in  Hobbes at the outset of a discussion on ‘authority’ ; for Hobbes him-  self introduced the concept to deal with difficult problems connected  with the analysis of human institutions. And there is little point  in making a list of the different ways in which the term ‘authority’  can be used unless the distinctions are made with an eye on the  problem or cluster of problems that can be clarified by means of  them.   Hobbes was impressed by the fact that a civil society is not a  natural whole like a rook or a beehive; yet it is not a mere multitude  of men. A multitude of men becomes an artificial person when each  man authorizes the actions of a representative. ‘Of persons artificial,  some have their words and actions owned by those whom they  represent. And then the person is the actor; and he that owneth his  words and actions is the AUTHOR: in which case the actor acteth by  authority ...and as the right of possession, 1s called dominion; so  the right of doing any action, is called AUTHORITY. So that by  authority, is always understood a right of doing any act; and done by  authority, done by commission, or licence, from him whose right itis.’  (Leviathan, Ed. Oakeshott pp. 105-6.) De Jouvenel, also, uses the  concept of ‘authority’ in the context of the same type of problem.  Having rejected the view that civil societies come into being through  voluntary association or through domination from without, he  claims that authority is ‘the efficient cause of voluntary asso-  ciations’ .. . ‘Everywhere and at all levels social life offers us the  daily spectacle of authority fulfilling its primary function—of man  leading man on, of the ascendancy of a settled will which summons  and orients uncertain wills ..’. Society in fact exists only because    Symposium by R. S. Peters and Peter Winch. From Proceedings of the Aristotelian Society,  Supp. Vol. 32 (1958), pp. 207-40. Reprinted by courtesy of the authors and the Editor of  the Aristotelian Society, with a postscript to his paper by Peter Winch.    84 R. S. PETERS    man is capable of proposing and affecting by his proposals an-  other’s dispositions; it is by the acceptance of proposals that  contracts are clinched, disputes settled and alliances formed  between individuals... What I mean by “authority” is the  ability of a man to get his proposals accepted’. (Sovereignty pp.    29-31.)    2. The de jure and de facto senses of ‘authority’.   I have chosen to start off with these quotations from Hobbes and  de Jouvenel partly because they both introduce the concept of ‘autho-  rity’ in the context of the attempt to elucidate what is meant by a  society as distinct from a multitude of men, and partly because the  two quotations illustrate an important difference in the ways in which  the term ‘authority’ is used in the context of the same sort of problem.  For Hobbes ‘authority’ is what might be called a de jure concept; for  de Jouvenel it seems to be a de facto one. In other words, for Hobbes  the term indicates or proclaims that someone has a right to do some-  thing. ‘Done by authority’ means ‘done by commission or licence from  him whose right it is’. Now I am not concerned to defend Hobbes’  odd conception of the handing over of rights or his account of  ‘authorization’. But, whatever the correct analysis of the connexion  between ‘authority’ and ‘right’, it is quite clear that there is a very  important use of the term ‘authority’, which is favoured by Hobbes,  which connects the two concepts. A man who is ‘in authority’ for  instance, clearly has a right to do certain sorts of things. This use of  ‘authority’ is to be contrasted with the de facto use favoured by de  Jouvenel. For he says “What I mean by “authority” is the ability of  a man to get his proposals accepted’. The Oxford English Dictionary  seems to permit both usages; for it gives ‘power or right to enforce  obedience’. It also speaks of ‘power to influence the conduct and  actions of others; ... personal or practical influence; power over the  opinion of others; intellectual influence’; as well as ‘moral or legal  supremacy; the right to command or give an ultimate decision. . .  title to be believed’. And in ordinary conversation the two senses  can be used without danger of misunderstanding in one sentence when  we say things like ‘The headmaster and others in authority had,  unfortunately, no authority with the boys’. The question quite natur-  ally arises how these two senses of ‘authority’ are related and whether  both senses are important, as Hobbes and de Jouvenel maintain, for  saying certain sorts of things about specifically human relationships  and organizations.    AUTHORITY 85    3. Hobbes’ rendering of the de jure sense.   The de jure concept of authority presupposes a system of rules which  determine who may legitimately take certain types of decision, make  certain sorts of pronouncements, issue commands of a certain sort,  and perform certain types of symbolic acts. Hobbes brings this out  by saying that the actions of a representative are authorized. He  relies on the sense of ‘authorize’ which assimilates it to commissioning  or giving a warrant to a man to do certain types of things. The  subjects are conceived of as having words and actions which they own,  of which they are the ‘authors’, and to which they have a right. They  then appoint a representative to whom they transfer their right. He  is now commissioned or ‘authorized’ to act on their behalf. “So that  by authority is always understood a right of doing any act; and done by  authority, done by commission or licence trom him whose rightitis.’  Now Hobbes, as is well known, and as Mr. Warrender has recently  shown in such stimulating detail, had a very strange view of natural  rights which permeates this picture of authority. He was led by it to  conceive of authority in general in terms of the particular case where  a man is the author of a word or act, to which he also has aright, and  where he commissions someone else to act in this matter on his behalf.  This is indeed a case of an authorized act; but there is a more general  meaning of ‘authorize’ which is to set up or acknowledge as authorit-  ative; to give legal force or formal approval to. Similarly ‘authorized’  in its most general meaning is equivalent to ‘possessed of authority’.  ‘Authorization’ is better understood in terms of the general concept  of ‘authority’ rather than vice-versa. Hobbes pictured ‘authority’ in  terms of ‘authorization’ which is one of its derivatives. But he did bring  out the obvious connexion between ‘authority’ and the existence of an  ‘author’ in the realm of acts and words, which is the key to seeing how  the concept works.    4. ‘Auctoritas’ as the key to ‘authority’.   The concept of ‘authority’ is obviously derived from the old  concepts of ‘auctor’ and ‘auctoritas’. An ‘auctor’ was, to quote Lewis  and Short, ‘he that brings about the existence of any object, or  promotes the increase or prosperity of it, whether he first originates  it, or by his efforts gives greater permanence or continuance to it’.  ‘Auctoritas’ which is a producing, invention, or cause, can be  exercised in the spheres of opinion, counsel or command. The point  of this little excursion into philology is to stress not only the sphere of  opinion, command and so on, in which ‘auctoritas’ 1s regarded as    86 R. S. PETERS    being exercised, but also the connexion of the concept with ‘produc-  ing’, ‘originating’, ‘inventing’—in short, with there being an author.   Now in some spheres of social life it is imperative to have such  ‘auctores’ who are producers or originators of orders, pronounce-  ments, decisions and so on. It is also the case that in social life,  whether we like it or not, there are such ‘auctores’ to whom commands,  decisions and pronouncements are to be traced back in any factual  survey of how social regulation is brought about. This is the sense of  ‘authority’ stressed by de Jouvenel. The notion of ‘authority’ involves  therefore either a set of rules which determine who shall be the auctor  and about what, or, in its de facto sense, a reference to a man whose  word in fact goes in these spheres. The de jure sense of ‘authority’  proclaims that a man has a right to be an ‘auctor’; the de facto sense  states that he is a matter of fact one. Hobbes’ account of ‘authoriza-  tion’ relates to the particular case where a man has a right to bean  ‘auctor’, as laid down by a set of rules, and where he commissions  someone else to do what he himself has a right to do. Indeed, often, as  in a bureaucratic system, there are subordinate sets of rules which lay  down procedures for the granting of such warrants and commissions.  But all authority cannot adequately be conceived in this fashion.    5.Weber’s legal-rational and traditional rules for determining who is IN  authority.   Indeed, one of the great services done by the sociologist, Max  Weber, has been to stress the different types of normative systems  which are connected with different types of authority. For legitimacy  may be bestowed in different ways on the commands or decisions or  pronouncements issuing from an ‘auctor’. In what he calls a legal-  rational system the claim to legitimacy rests on ‘a belief in the  “legality” of patterns of normative rules and the right of those  elevated to authority under such rules to issue commands’. (Theory of  Economic and Social Organization, Ed. Talcot Parsons, pp. 300/1.) There  is also, however, traditional authority ‘resting on an established  belief in the sanctity of immemorial traditions and the legitimacy of  the status of those exercising authority under them’.   There are most important and interesting differences between these  types of authority but this is not the place to investigate the difference  between traditional and legal rules, or to comment on the adequacy  of Weber’s analysis—but in both cases to speak of ‘the authorities’ or  ‘those in authority’ or those who ‘hold authority’ is to proclaim that  on certain matters certain people are entitled, licensed, commissioned    AUTHORITY 87    or have a right to be auctores. And the right is bestowed by a set  pattern of rules.    6.Weber’s charismatic authority.   This type of authority is to be distinguished clearly from other  types of authority where the right derives from personal history,  personal credentials, and personal achievements, which, as will be  argued later, are intimately bound up with the exercise of authority  in its de facto sense. There is a gradation from the pure de jure sense  of ‘authority’ as when we say that ‘Wittgenstein held a position of  authority in Cambridge’, through the notion of ‘an authority’ as  when we say ‘Wittgenstein was an authority on William James’ to the  de facto sense as when we say ‘Wittgenstein exerted considerable  authority over the Moral Science Club’. Both the last two senses of  ‘authority’, unlike the first, imply something about the attributes  or qualifications of the individual in question. But the details of  this transition are very difficult to make explicit.   Weber, as a matter of fact, made much of authority deriving from  personal characteristics when he spoke of ‘charismatic authority’—  ‘resting on devotion to the specific and exceptional sanctity, heroism  or exemplary character of an individual person, and of the normative  patterns or order revealed or ordained by him’ (op. cit. p. 301). He  was thinking primarily of the outstanding religious and military  leaders like Jesus and Napoleon. He therefore pitched his account  rather high and personal ‘authority’ is decked with the trappings of  vocation, miracles and revelation. Nevertheless, there is something  distinctive about the charismatic leader which he shares in an exag-  gerated form with other ‘natural’ leaders who exercise authority in  virtue of personal claims and personal characteristics. For he is  unlike the moral reformer who gives reasons of a general kind for his  innovations, reasons which he expects everyone to appreciate. He  appeals to revelation or claims that he has a call. These are not  really justifications of his innovations; they are ways of stressing  that he need give no justification because he is a spectal sort of man.    7. Gradations in the concept of ‘AN authority’.   This notion of presenting credentials of a personal sort is an  intermediary between the purely de jure and the de facto senses of  ‘authority’. For the reference to personal characteristics is a way  of establishing that a man has a right to make pronouncements and  issue commands because he is a special sort of person. And, although    88 R. S. PETERS    in some societies a man who sees visions and goes into trance states  is in danger of electric shock treatment, in other societies pointing  to such peculiarities of personal biography are ways of establishing  a man as an authority in certain spheres. In societies where the  claim to vocation or revelation is acceptable there are also, usually,  collateral tests for eliminating charlatans and the mentally deranged.  But his claims rest, as it were, on some kind of personal initiation  into mysteries that are a closed book to most men. In a similar way  years of study of inaccessible manuscripts would establish a man  as ‘an authority’ on a special period of history, or years spent in  Peru might establish a man as ‘an authority’ on the Incas. Collateral  tests would, of course, be necessary to vouch for his trustworthiness.  But in many fields people become ‘authorities’ by some process of  personal absorption in matters that are generally held to be either  inaccessible or inscrutable. Dodds suggested that the Forms were  objects of this sort for Plato—objects which the initiated had to  scrutinize by a kind of bi-location of personality as practised by  shamans. And the scrutiny of such objects gave the philosopher kings  a right to make decisions and issue commands—in short, made them  authorities. (The Greeks and the Irrational, pp. 210/11.)   Weber stresses the importance of success as a necessary condition  for the maintenance of charismatic authority. If success deserts  the leader he tends to think of his god as having deserted him or his  exceptional powers as failing him. And his authority will be corres-  pondingly reduced. The disciples, it is said, were in despair when  Jesus had been crucified. It was only when he accomplished the  supreme feat of rising from the dead that they recovered their faith  in him and in his claims. To a certain extent the charismatic leader  is in the position of a man who keeps spotting Derby winners without  a system. His authority depends on always being right by virtue of a  ‘flair’ or a ‘hunch’—words which point to his inability to give grounds  for his pronouncements. It is because his authority derives from such  personal pecularities that failure tends to be fatal. This is a very  important empirical generalization about a necessary condition for  the exercise of authority which applies at much more mundane levels.’   The point, however, is that in the case of these extreme types of  charismatic authority revelation and success are not simply necessary  conditions for the exercise of authority de facto. They are also grounds    ‘Ernest Gellner has pointed out to me that in many societies there are institutional  devices for covering up failure so that the authority can’t be wrong.    AUTHORITY 89    for establishing the right to be an auctor. This can be shown, too, in  more mundane spheres where we speak of a person being an authoritv.  He has not been putin authority; he does not hold authority according  to any system of rules. But because of his training, competence and  past success in this sphere he comes to be regarded as an authority.  He has a right to speak. It may be the case that people do not exercise  authority in various spheres unless they are competent and successful  as a matter of fact; but it is also the case that they come to be regarded  as authorities because these necessary conditions come to be regarded  as grounds for a right. The notion of an authority, therefore, implies,  as it were, a self-generating system of entitlement which is confined  to specific spheres of pronouncement and decision. We speak ofan  authority on art, music or nuclear physics. The grounds which entitle  a manare directly connected with his personal history and achievements  in a specific sphere. These grounds vary from the extremes of revel-  ation, initiation and vocation, through less esoteric grounds like study  of inaccessible material in history, to the more public and accessible  training of a scientist. But in all these spheres success seems to be  a usual ground of entitlement.    8. De facto authority: its necessary conditions and meaning.   It was suggested by reference to the Wittgenstein example, that  there was a gradation from the purely de jure sense of ‘authority’,  through the concept of ‘an authority’ to the de facto sense of ‘authority’.  The analysis of de facto authority must now be tackled and the question  faced whether the term ‘authority’ can ever be used properly if there  is no suggestion of a right to make decisions and issue commands or  pronouncements. Does the exercise of authority de facto presuppose  that the person who exercises it must be in authority or an authority?  In the Admirable Crichton situation the butler, in fact, exercised  authority, though the lord was in authority. Are we to assume that, in  some sense the butler had a right to make decisions? Or does saying  that the butler had authority over the lord mean simply that the lord  accepted the butler’s decisions just because they issued from a parti-  cular man in whose presence his ‘genius was rebuked’?   Of course most people who exert authority de facto do so because  of the deference paid to their office or status rather than because of  any outstanding personal characteristics. But there is often a mixture  of both as in the case of Julius Caesar or Queen Elizabeth the First.  Indeed there is subtle interweaving of these institutional and personal  conditions for the exercise of authority de facto. For, as we say, the office    go R. S. PETERS    makes the man; and often the man gives dignity to the office. The same  tendency is to be observed in cases where it is more appropriate to  speak of there being an authority. The entitlement accorded has a  snowball effect. Often the outcome is disastrous—portentous pro-  nouncements which are unquestionably accepted but which turn out  to be erroneous. The generalization to other spheres is also a well-  known phenomenon—one which Socrates spent so much time  attacking.   There is, therefore, a widespread connexion between being in  authority or an authority and the de facto exercise of authority. But  this is a contingent connexion, not a necessary one. And as Admirable  Crichton situations are not unusual, it looks as if being in authority  or an authority are only frequently conditions for exercising authority;  it does not look as if they are even necessary conditions.   What then of the cases where a man exercises authority de facto  purely because of certain personal characteristics—when either there  is no deference paid to his office if he is an official, or when he is not  in a position of authority at all? There are two questions here which  need to be distinguished. The first is about the conditions other than  being in authority or an authority which are necessary to the exercise of  authority de facto. The second is the logical question of what it means to  exercise authority de facto in this tenuous sense. Is it the case that always  the exercise of authority implies that in some sense, a man must be  regarded as entitled to command, make decisions and so on? Are there  necessary conditions which, as in the case of ‘anauthority’ come to be  regarded as grounds for aright? To answer this it will be as well to deal  briefly with the sorts of things which might be suggested as necessary  conditions.   A variety of generalizations can be made about necessary conditions  for bringing about unquestioning conformity—for instance, that a  man’s decisions tend to be accepted in proportion to the extent to  which he has been proved right before. Success, too, strengthens  another necessary condition for the exercise of authority—the expect-  ation of being believed, followed or obeyed. People will tend to accept  decisions and obey orders in proportion as the man who makes them  or gives them expects that they will. The more successful he is, the less  questioning there will be and the greater will be the confidence sa  which he utters them. We have phrases like ‘an air of authority’, ‘  authoritative voice’, and Jesus, it is said, produced ae oh  because as a boy he spoke ‘with authority’ in the temple. Such descrip-  tions draw attention to the outward signs of the inner certitude which    AUTHORITY gi    is usually necessary for the exercise of authority. For itis not sufficient  for a man to be in fact wise or shrewd or a felicitous prophet, if he is  to exercise authority. He must also be known to be so. It is said that  Attlee’s authority in the country suffered in his early days as Prime  Minister because he did not have a good public relations officer. A  man cannot exercise authority if he hides his light under a bushel.  Such empirical generalizations are the province of the social  psychologist. The question of philosophical interest is whether any  such empirical conditions must come to be regarded as grounds for  a right if a man is to be said properly to exercise authority without  being in authority or an authority. A concrete case will help here.  Suppose there is an explosion ina street ora fire ina cinema. Someone  comes forward who is not a policeman or a fireman or manager of the  cinema and who is quite unknown toall present—.e. he is not regard-  ed as ‘an authority’ in virtue of his personal history or known  competence in an emergency. Suppose he starts issuing orders and  making announcements. And suppose that he is unquestioningly  obeyed and believed. Would we say that such a man exerted authority  in a crisis? I think we would only say so if we thought that his orders  were obeyed simply -because they were his. There would have to be some-  thing about him in virtue of which his orders or pronouncements were  _regarded as being in some way legitimately issued. Maybe it would be  his features; maybe it would be the tone of his voice.! Maybe he would  have a habit of command. But those who heard him would have to  think in an embryonic way that he was the sort of man who could be  trusted. It would put the matter altogether too strongly to say that they  thought he had a right to take control. For obviously, in any useful  sense of ‘right’, he has not got a right. He has not been appointed;  he is nota status-holder; he possesses no credentials of a more person-  al sort. All that can be said is that there is something about him which  people recognize in virtue of which they do what he says simply  because he says it. Perhaps the word ‘faith’ is required here; for, as  Hobbes put it, the word ‘faith’ is required when our reasons for assent  derive ‘not from the proposition itself but from the person pro-  pounding’.  It may be, however, that the search for some vague ground for the  acceptance of orders in this unquestioning way is to approach the    'Cf. King Lear, Act. 1 Sc. IV. Lear: Who wouldst thou serve? Kent: You.  Lear: Dost thou know me fellow? Kent: No sir: but you have that in your countenance  which I would fain call master. Lear: What’s that? Kent: Authority.    92 R. S. PETERS    analysis of ‘authority’ in its de facto sense in too positive a manner.  Perhaps the use of the term ‘authority’ is to deny certain characteristic  suggestions rather than to assert a positive ground for unquestioning  obedience. People often do what theyare told because they are threat-  ened or bribed or physically forced. After all, obedience in a crisis can  be produced by a fire-hose or machine gun, irrespective of who is  manning it. Maybe the term ‘authority’ is necessary for describing  those situations where conformity is brought about without recourse  to force, bribes, incentives or propaganda and without a lot of argu-  ment and discussion, as in moral situations. We describe such situa-  tions by saying that an order is obeyed or a decision is accepted simply  because X gave it or made it. This is a way of excluding both that action  was taken on moral grounds and that the person acted under con-  straint or pressure or influence. The use of authority, in other words,  is a manner of regulating human behaviour which is an intermediary  between moral argument and the use of force, incentives and pro-  paganda.    g. Common features of all uses of ‘authority’.   There are, therefore, features which all uses of the term ‘authority’  have in common. In so far as the de facto sense implies that, in an  indeterminate and embryonic sense, the person who exercises author-  ity is regarded as ‘having a right’ to be obeyed, and so on, the de facto  sense is parasitic on the de jure sense. But the common features of both  senses are, perhaps, best brought out by summarizing and making  explicit the peculiar nature and réle of authority in the regulation of  human behaviour—the point at which I embarked on this analysis in  the company of Hobbes and de Jouvenel.   (a) In contrast to ‘power’. The first feature to stress is the connexion  between ‘authority’ and the use of certain types of regulatory utter-  ances, gestures and symbolic acts. A person in authority has a right to  make decisions, isstie pronouncements, give commands and perhaps  perform certain sorts of symbolic significant acts. To have authority  with another man is to get him to do things by giving orders to him,  by making pronouncements and decisions.   The main function of the term ‘authority’ in the analysis of a social  situation is to stress these ways of regulating behaviour by certain  types of utterance in contrast to other ways of regulating behaviour.  This is to reject the more usual attempts to analyse ‘authority’ in terms  of ‘power’ as exemplified by Weldon, for instance, who claims that  ‘authority’ means power exercised with the general approval of those    AUTHORITY 93    concerned. (Vocabulary of Politics, p. 56.) This, of course, is not to deny  that it may be important, as Warrender stresses, to distinguish physical  power from political power, the latter being confined to cases where an  element of ‘consent’ is involved, as when a man does something  because he is threatened, cajoled or duped, in contrast to when he is  physically coerced—e.g., bound and put into prison. (See The Political  Philosophy of Hobbes pp. 312/3.) It might, therefore, be tempting to  regard the exercise of authority as a species of the exercise of political  power distinguished by approval as opposed to mere acceptance on  the part of the victim. But this, surely, is an over-simplification. For  often what we want to bring out when we say that men are in authority  or exert authority over other men is that they get their way or ought  to get their way by means other than those of force, threats, incentives and  propaganda, which are the usual ways of exercising power. It is only  when a system of authority breaks down ora given individual loses his  authority that there must be recourse to power if conformity is to be  ensured. The concept of ‘authority’ is necessary to bring out the  ways in which behaviour is regulated without recourse to power—to  force, incentives and propaganda. These ways are intimately bound  up with issuing pronouncements, making decisions and giving com-  mands. I suppose the concept of ‘power’ can be extended to cover  these ways of influencing people. But my claim is that ‘power’ usually  has meaning by contrast with ‘authority’ rather than as a generic  term of which ‘authority’ is just one species.   In so far as there is a positive connexion between ‘power’ and  ‘authority’ it is better conceived along other lines. For instance,  it might well be true that a common condition for the exercise of  authority de facto is the ability to dispose of overwhelming power,  if necessary. Or, alternatively, power might be regarded as a ground  of entitlement. The old saying that there can be no legitimacy without  power might be interpreted in this second way—as claiming that one  of the grounds which give a man a right to command must always be,  directly or indirectly, the ability to dispose of power, if necessary. Or  it could be interpreted in the first way as an assertion that the posses-  sion of power is a necessary condition for the de facto exercise of  authority, the legitimacy of which might be established in other ways.  And, of course, this necessary condition, like others which I have  mentioned before, can come to be regarded as a ground of entitle-  ment. There is, however, no need to explore this positive connexion  in detail. For my claim is that these are answers to other questions—  questions about the grounds of entitlement or about the necessary    94 R. S. PETERS    conditions for the exercise of authority, not questions about the  meaning of ‘authority’.   There is little mystery about why authority should be so intimately  connected with the problem of the analysis of human institutions. For  men, pace Aristotle, are rule-following animals; they talk and regulate  their own behaviour and that of others by means of speech. Men  perform predictably in relation to each other and form what is called  a social system to a large extent because they accept systems of rules  which are infinitely variable and alterable by human decision. Such  systems can only be maintained if there is general acceptance of  procedural rules which lay down who is to originate rules, who is to  decide about their concrete application to concrete cases, and who is  entitled to introduce changes. In other words, if this peculiarly human  type of order is to be maintained there are spheres where it is essential  that decisions should be accepted simply because somebody, specified  by rules of procedure, has made them. It is very difficult to play cricket  without an umpire, just as it is difficult to conceive of an army working  without a hierarchical system of command. The term ‘authority’ is  essential in those contexts where a pronouncement, decision or com-  mand must be accepted simply because some person, conforming to  specifications laid down by the normative system, has made or given  it—where there must be a recognized ‘auctor’. More liberal societies,  of course, guard against injustice and stupidity by instituting further  procedures for appealing against decisions of those in authority. But  this ismerely a device whereby a higher authority is instituted tocorrect  the mistakes of a lower one. It is still a regulatory device which relies  on the institution of authority and in no way abrogates the duty of  obedience to the lower authority, provided that the lower authority is  acting intra vires.   (b) Incontrast tomoral and sctentific regulation of conduct and opinions. This  analysis of ‘authority’ accounts also for a long tradition which stresses  the incompatibility between authority and certain specific human  enterprises like science and morality. For it would be held that in  science the importance of the ‘auctor’ or originator is at a minimum,  it never being justifiable in scientific institutions to set up individuals  or bodies who will either be the originators of pronouncements or who  will decide finally on the truth of pronouncements made. The pro-  cedural rules of science lay it down, roughly speaking, that hypotheses  must be decided by looking at the evidence, not by appealing to a man.  There are also, and can be, no rules to decide who will be the origin-  ators of scientific theories. In a similar way, it would be held thata rule    AUTHORITY 95    cannot be a moral one if it is to be accepted just because someone has  laid it down or made a decision between competing alternatives.  Reasons must be given for it, not originators or umpires produced. Of  course, in both enterprises provisional authorities can be consulted.  But there are usually good reasons for their choice and their pro-  nouncements are never to be regarded as final just because they have  made them. In science and morality there are no appointed law- givers  or judges or policemen. This is one of the ways in which life in the  laboratory differs from life in the army and law-courts.   This analysis of ‘authority’ readily explains, too, the connexions  so often made between ‘authority’ and ‘command’. For commands,  roughly speaking, are the sorts of regulatory utterances for which no  reasons need to be given. A man can only give a command ifhe is ina  position of authority or if he exerts authority i in a de facto sense. For  as an occupant of an office or as a status holder he has a right to make  decisions which are binding and to issue orders. Similarly, if the de facto  sense of authority is being used, to say that a man has authority over  other men is to say, amongst other things, that they will do what they  are told without questioning the prudence, wisdom and good sense of  the decision. They may, of course, question its legality; for questions  can be raised about a man’s right to issue commands in general or in  a particular sphere. These are questions about his right to an office  or status, or about the sphere of its competence or his prerogative. But  once it is granted that he occupies an office or holds a status legitim-  ately, and once it is made clear that he is not straying from its sphere  of competence or exceeding his prerogative, there can be no further  question of justifying his commands. For commands just are the type  of regulatory utterance where questions of justification are ruled out.   Authority, however, is not exercised only in the giving of commands.  There are also the spheres of making pronouncements and decisions  and the performance of symbolic acts. Behaviour or opinion in these  spheres is regulated by the utterance of a man which carries with it the  obligation for others to accept, follow or obey. The claim put forward  by Hobbes and Austin, that law is command, is right in stressing the  connexion between law and authority but wrong in conceiving of  commands as the only form of authoritative utterance. Similarly those  who speak of ‘the authority of the individual conscience’ cannot be  supposed merely to be saying that in moral matters a man must give  himself orders, which sounds, in any case, a little quaint; rather they  are saying that in moral matters a man must decide himself between  conflicting claims and principles and not accept the pronouncements    96 R. S$. PETERS    and decisions of others simply because they issue from determinate  sources. In morals a man must be his own ‘auctor’.    10. Conclusion.   To conclude: my thesis is that the concept of ‘authority’ can  be used in a de jure and a de facto sense. Amongst the former uses  it is very important to distinguish the kind of entitlement implied  in being i authority from that implied in being an authority.  Authority in a de facto sense is parasitic on the de jure sense in  that it implies that decisions are in fact accepted or commands  obeyed simply because they issue from a certain person whose  attributes are in some way regarded as bestowing legitimacy on  them. The grounds for this legitimacy are often much more indeter-  minate than those more impersonal grounds characteristic of de jure  authority. There are, however, more general negative features which  all senses of ‘authority’ share. The term is always used to speak of  ways in which conduct is regulated as distinct from the mere use  of power—e.g. the giving of commands, the making of decisions and  pronouncements, as distinct from the use of force, incentives and  propaganda. Secondly, within the sphere of decisions, pronounce-  ments and other such regulatory utterances, authority is confined to  those which are or must be obeyed simply because someone has made  them. This second feature of ‘authority’ brings out the contrast  between laws, commands and religious utterances, on the one hand,  and those of science and morality on the other. Both these features  of ‘authority’ are rooted in the Latin word ‘auctoritas’ which implies  an originator in the sphere of opinion, counsel and command.    (2) PETER WINCH    The concept of authority does not merely give rise to isolated  philosophical difficulties of its own. It is intimately connected with  some of the most central issues in philosophy. Hence Dr. Peters is  right to start with Hobbes: for Hobbes’ account of authority is closely  bound up with his general philosophical account of the nature of  human life, thought and society. Indeed, the connexions between the  philosophy of society and politics on the one hand and metaphysics  and epistemology on the other have probably never been so clearly  brought out as they were by Hobbes. (Thus I think itis a great mistake  to try, as some have lately tried, to treat Hobbes’ account of politics  as if it had nothing logically to do with his epistemological presup-  positions. It is not merely a mistaken interpretation of Hobbes but  is also a symptom of wrong ideas about the relevance of philosophy  to politics. Of course, in saying this I should not be taken as endorsing  the specific account Hobbes gives of epistemology and of politics.)   But my agreement with Dr. Peters, fortunately for the future of  this symposium, ends there. Although I think that he starts off with  a genuine philosophical problem, and one which an analysis of the  concept of authority should have much to contribute to, I do not  accept the analysis which he offers; I should like to suggest, moreover,  that the defects in it which I hope to be able to point out arise out of  a failure on Peters’ part to keep his initial problem clearly in mind  and go deeply enough into it. My method, therefore, will be to develop  my argument independently from the same starting point as Peters’,  trying to show en passant what seems to me wrong with the account of  authority which he offers.   What light then does the notion of authority throw on the nature  of the cohesion or unity which is characteristic of societies of human  beings as opposed to what Peters calls ‘natural wholes’? As Peters  notes, Hobbes uses the notion in a legalistic way: for him, the unity  of a society is a sort of legal fiction, involving the quasi-legal notion  of representation, which he regards as closely analogous to author-  ization and hence as involving the notion of authority.    A multitude of men, are made one person, when they are by one man, or one  person, represented; so that it be done with the consent of every one of that  multitude in particular. For it is the unity of the representer, not the unity of    98 PETER WINCH    the represented, that maketh the person one. And it is the representer that  beareth the person, and but one person; and unity, cannot otherwise be under-  stood in multitude.   (Leviathan, Ed. Oakeshott, p. 107.)    What is important here for my purposes is that the real unit, in  Hobbes’ conception, is the individual will. His problem is to say how  a large number of wills can be conceived as co-ordinated in the way  with which we are familiar in human societies. Now M. de Jouvenel  equally, in those passages which Peters cites, seems to take the indivi-  dual will as his starting point: he thinks of a society in terms of the  mutual influence of such wills. Social movements, for him, start as  the projects of individual wills; authority is the faculty of interesting  the wills of others in one’s own projects.   Peters himself does not, I think, explicitly declare himself on  this issue. Nevertheless, I think it is fair to ascribe to him too the  view that the starting point in the analysis of authority should be  the success of the individual in getting his decisions accepted by other  individuals. This is implicit, for instance, in his account of the notion  of ‘natural’, ‘de facto’ authority, with his stress on the importance of  purely personal qualities.   My paper starts from a point of view which is opposed to this.  Although a man who exercises authority does indeed influence the  wills of other men, authority cannot be understood as a peculiar sort  of influence of one will upon another. If that sounds paradoxical, let  us recall that although a man who has knowledge does indeed believe  something, knowledge cannot be understood as a peculiar kind of  belief. This analogy points to a parallel, which I think may be quite  illuminating, between the problem in political philosophy of giving  an account of the distinction between authority and power and the  problem in epistemology of giving an account of the distinction  between knowledge and belief. The connexions between these ques-  tions are brought out in the argument of Plato’s Gorgias; I shall return  to the parallel subsequently.   Authority is not a sort of influence. It is not a kind of causal relation  between individual wills but an internal relation. The very notion of a  human will, capable of deliberating and making decisions, presup-  poses the notion of authority. I shall try to show this by considering  the whole question of the nature of the unity of a human society from  a different point of view.   I want to say first that the chief way in which this unity differs from    AUTHORITY 99    that of what Peters calls ‘natural’ wholes is that it is a unity essentially  involving concepts. It would obviously be going much too far to say  that a human society is a conceptual unity in the sense in which one  can say this of a system of ideas; but there are analogies. For the  interaction of human beings in society, unlike that of animals, involves  communication, speech and mutual understanding (or, of course,  misunderstanding). It is a type of interaction which can be accounted  for adequately neither in terms of instinct nor of conditioned reflex.’   It follows from this that one cannot give a full account of the nature  of a human society without giving an account of the way in which con-  cepts enter into the relations which men have to each other in sucha  society.   Wittgenstein has shown how notions like communication and  understanding presuppose the notion of following a rule. He has also  offered an account of this latter notion which brings out the peculiar  kind of social interaction which it involves: what he calls “agreement  to go on in the same way’. Now Peters mentions, in Section g of his  paper, that activities which are governed by rules can be carried on  only if there is agreement that somebody should be in authority to  make crucial decisions. But he does not seem to me to see the full  bearing of this fact on the analysis of the concept of authority. Itis not  clear whether he regards the connexion between rule-governed activ-  ities and authority as merely contingent (arising out of the tendency of  men to come into mutual conflict), or whether he is making a gram-  matical statement about what is involved in the.very notion of a rule.  1 think it is important to see that the connexion is conceptual rather  than contingent.   The acceptance of authority is not just something which, as a matter  of fact, you cannot get along without if you want to participate in rule-  governed activities; rather, to participate in rule-governed activities  is, in a certain way, to accept authority. For to participate in such an  activity is to accept that there is a right and a wrong way of doing things,  and the decision as to what is right and wrong ina given case can never  depend completely on one’s own caprice. (Cf. Wittgenstein: Philoso-  phical Investigations, I, 258.) For instance, pace Humpty Dumpty, J  cannot (at least in general) make words mean what J want them to mean:  I can use them meaningfully only if other people can come to under-  stand how I am using them. Of course, I can decide, in a certain  context, to make the sound ‘red’ mean what is commonly meant by the    ‘For reasons of space much must be taken for granted here. I have argued the point  at greater length in The Idea of a Social Science (Routledge & Kegan Paul, 1958).    100 PETER WINCH    sound ‘blue’; but I can do this only in so far as I also understand the  meanings of a great many words which J have not decided upon. In  other words, when it comes to following rules I must (as a matter of  logic) accept what certain other people say or do as authoritative.   This approach suggests that there is an intimate conceptual con-  nexion between the notion of authority on the one hand and, on the  other hand, the notion of there being a right and a wrong way of  doing things. That is the position that I propose to maintain and  develop in what follows.   It may sound far-fetched to start a discussion of the concept of  authority at this point; for the activity of speaking a language is not  one in which the exercise of authority is at all obtrusive.’ When we  use words in the right way we do not think of ourselves as bowing to the  dictates of an alien will. No; but then I want to say that to submit to  authority (as opposed to being subjected to power) is not to be subject  to an alien will. What one does is directed rather by the idea of the  right way of doing things in connexion with the activity one is perform-  ing; and the authoritative character of an individual’s will derives  from its connexion with that idea of aright way of doing things. (This,  I think, is partof the truth beind Collingwood’s odd, but in some ways  illuminating, definition of authority in The New Leviathan, 20. 45:  “Something capable of ruling itself sometimes appears to be (but is not  in fact) ruled by something else. I refer to the case in which one thing  is said to have authority over another.’)   The réle of authority in activities like speaking a language is ob-  scured by the fact that the authority in question is not esoteric. All  educated Englishmen are authorities on the correct speaking of  English. This makes it particularly easy to, and important not to,  overlook the interwovenness of the idea of there being a correct way  of speaking, on the one hand, and the established practice of a certain  group of people (the ‘authorities’: in this case a very wide group), on  the other.   All characteristically human activities involve a reference to an  established way of doing things. The idea of such an established way of  doing things in its turn presupposes that the practices and pronounce-  ments of a certain group of people shall be authoritative in connexion  with the activity in question. Further, we can give no account of the    ‘Though, as Max Weber several times emphasizes, the exercise of authority (in a more  obvious sense) is certainly essential to the maintenance of a language. Cf. Wirtschaft und  Gesellschaft. Kapitel III: ‘It is the authority exercised in school which puts the seal on  what counts as the orthodox, correct way of writing and speaking’ etc.    AUTHORITY 101    nature of the wholes which we call human societies, as opposed to that  of ‘natural’ wholes, except by giving an account of what is involved in  characteristically human activity. It is in this way, I suggest, that the  notion of authority is important to the conception of a human society.  It should be noted that I have made no explicit reference here to the  idea of one individual human will’s influencing another. A relation of  authority, as opposed to one of power, is an indirect relation between X  and ¥ involving as an intermediary the established way of performing  the activity on which X and Y are engaged.   I can now amplify my earlier remark that authority is an internal  rather than a causal relation. It is so because of its connexion with  the ideas embodied in the form of activity within which it is exercised.  (I use the notion of a ‘form of activity’ here in an extended sense to  include not merely activities like tree-felling, chess-playing, etc., but  also moral and political behaviour, which constitute forms of activity  in a somewhat different sense. De Jouvenel’s distinction, to which |  shall return shortly, between the ‘team of action’ and the ‘milieu of  existence’, may be helpful in explicating this distinction.) If N is trying  toteach mechess and Iam trying to learn, N and I are internally related  by way of my acceptance of his authority on the right way to play chess.  Again, if Nis a judge trying a case in which I am litigating, his authority  over me is an internal relation which can only be understood in terms  of the system of (legal, moral and political) ideas which give such legal  processes their sense in our society. In neither of these two examples  can the relation of authority between N and me be understood in  purely causal (sociological or psychological) terms.   Much of Peters’ argument turns on his belief that ‘it is very impor-  tant to distinguish the kind of entitlement implied in being in  authority from that implied in being an authority’ (Section 10). His  Weberian idea of ‘natural’ authority, depending on purely personal  qualities, commits him to a denial of my assertion that the notion of  an ‘established’ way of doing things 1s essential to the notion of  authority as such. I am saying, in a sense, however, that someone  who is in authority is always an authority on something. I am aware  of the difficulties this way of speaking raises, especially in connexion  with those situations where Weber speaks of the exercise of ‘charis-  matic’ authority; I shall reserve discussion of these until later. I  should like first to show that the kind of analysis I propose is capable  of easing certain long-standing philosophical difficulties which the  notion of authority gives rise to.   Earlier I cited a remark of Collingwood’s to the effect that to    102 PETER WINCH    be subject to authority is not to be in somebody else’s power. This  brings us face to face with the whole question of how the necessity  for authority in human affairs fits in with men’s freedom of choice.  To be subject to somebody else’s will is for one’s own freedom  of choice to be reduced; but there is a powerful philosophical  tradition to the effect that the exercise of legitimate authority is not  a curtailment of this freedom. De Jouvenel, for instance, writes:  ‘Authority is the faculty of inducing assent. To follow an authority  is a voluntary act. Authority ends where voluntary assent ends. There  is in every state a margin of obedience which is won only by the use  of force or the threat of force: it is this margin which breaches  liberty and demonstrates the failure of authority.’ (Sovereignty, Pp. 33.)  This line of thought seems to me one which it is important to empha-  size. I shall now try to show that authority, according to the acceunt  I have been giving of it, is not by any means a curtailment of liberty  but is, on the contrary, a precondition of it.   The liberty in question is the liberty to choose. Now choice, as  Hobbes (though in a misleading way) emphasized, goes together with  deliberation (Leviathan, Ch. 6). To be able to choose is to be able to  consider reasons for and against. But to consider reasons is not, as  Hobbes supposed, to be subject to the influence of forces. Considering  reasons is a function of acting according to rules; reasons are  intelligible only in the context of the rules governing the kind of  activity in which one is participating. Only human beings are capable  of participating in rule-governed activities, hence other animals  cannot be said to deliberate and choose, though Hobbes, consistently  with his premises, maintained otherwise. Thus it is only in the context  of rule-governed activities that it makes sense to speak of freedom of  choice; to eschew all rules—supposing for a moment that we under-  stood what that meant—would not be to gain perfect freedom, but to  create a situation in which the notion of freedom could no longer find  a foothold. But I have already tried to show that the acceptance of  authority is conceptually inseparable from participation in rule-  governed activities. It follows that this acceptance is a precondition  of the possibility of freedom of choice. Somebody who said that he was  going to renounce all authority in order to ensure that he had perfect  freedom of choice would thus be contradicting himself. (A conceptual  version of the man who thought that he could fly more easily if only  he could escape the inhibiting pressure of the atmosphere.)   Consider an example. If 1 am being taught chess, then the pro-  nouncements of my teacher are authoritative for me because of my    AUTHORITY 103    recognition of the fact that he is telling me the correct way to move  the pieces. If I make a wrong move and he corrects me, this is not in  any intelligible sense an encroachment on my freedom of action. Until  I know how to play chess the question of my being free or not to play  the sort of game I choose cannot arise. And I can only learn how to  play by accepting the pronouncements or example of some mentor or  mentors as authoritative.   I realize that this example loads all the dice in my favour, and it is  time now to consider some cases which give me more difliculty. In this  connexion I must draw attention to two aspects of the chess example.  (1) Playing or not playing chess is itself a matter of choice; but it  is certainly not true of all cases of authority that it is accepted volun-  tarily in this sense. Very often authority cannot be accepted or  rejected at will because it is not a matter of choice for us to participate  or not to participate in the form of activity within which it is exercised.  Indeed, one of the most telling criticisms frequently made of social  contract theories of authority is precisely that they overlook this point.  (2) In the chess example the meaning of the expression ‘the right way to  proceed’ is clear and unambiguous. People who know how to play  chess do not dispute about what moves of the various pieces are  legitimate (and this fact belongs to our idea of the game of chess).  But this feature is lacking from many of the cases in which authority  is exercised, and particularly from those which give the most  philosophical trouble. There will, for example, certainly be no general  agreement about whether or not a given exercise of political authority  was ‘right’ or not (and this belongs to our idea of politics). Peters  is so impressed by this that he is led to think that the concept of  authority becomes applicable precisely where the concept of ‘the  right way to proceed’ ceases to apply. That, at any rate, is how I  interpret his insistence that an appeal to authority is a way of avoiding  having to give reasons for what one does or says. Now I agree that  this lack of agreement about what is right creates philosophical diffi-  culties; but I do not think that Peters’ way of dealing with them is  satisfactory.   I shall now deal with these two objections in turn.   (1) Consider the following two cases in which the authority  exercised over a person clearly does not depend on his choice to  participate in any particular form of activity: (a) the authority of  adults over children; (b) political authority.   (a) The point about children is that they are not yet ina position  to exercise freedom of choice in the full sense, because they have not    104 PETER WINCH    yet been sufficiently educated in modes of social life to be able to  deliberate. The exercise of authority over them, therefore, cannot be  an encroachment on their freedom: it is via the exercise of authority  that they will be inducted into modes of social life and thus be made  capable of deliberating and exercising choice. A child is obviously  not in a position to choose to do this or that until he has learned how  to do this and that.   (6) The difficulty raised by political authority is quite different.  It is not, characteristically, exercised over children (and any attempt to  derive it from the notion of paternal authority is, I think, completely  misconceived). But still, like the liability of children to adult authority,  one’s liability to political authority does not depend on a decision to  subject oneself to it; in this way it is unlike the case of someone who  subjects himself to the authority of those who know howto play chess  in deciding to learn chess himself, or the case of someone who  subjects himself to the authority of the priest by deciding to become  a Roman Catholic.   To deal with this I shall use a modified version of de Jouvenel’s  distinction, already alluded to, between the ‘milieu of existence’ and  the ‘team of action’ (op. cit., Chapter 4). Activities like playing  chess, building bridges, performing religious duties, going to war,  etc., do not take place in isolation. Thev presuppose an established  social framework. No society can be understood as just one big action  group. But neither can it be properly understood as just the sum of  the various action groups which compose it. For new (political)  problems (that is, problems not specific to any particular action  group) arise out of the fact that action groups influence each other:  either by mutual assistance or by conflict. Moreover, no indivi-  dual will belong to just one action group, with the result that  (moral) problems of divided loyalty occur within the life of the  individual. Along with those new problems go specific ways of  treating and thinking about them: conventions dealing with right  and wrong ways of settling conflicts, for example. And the carrying  out of those conventions will, in the public sphere, involve the  exercise of authority. This is the sphere of Jouvenel’s ‘milieu of  existence’ and the authority exercised within it is what he calls the  authority of the rex as opposed to that of the dux who leads the team  of action.   For my purposes it is important to emphasize that our very idea  of the kind of activity carried on by the action group carries with  it the idea of a milieu of existence in which some kind of political    AUTHORITY 105    authority is exercised in the settling of conflicts. We do not know  what it would be like for such a mode of activity to be carried on in  complete isolation: apart, that is, from other modes of activity with  which it is in contact, with which it may conflict, and for which  conflicts there must be conventions governing their equitable settle-  ment. Although, therefore, one does not choose to accept political  authority; although its applicability to one does not depend on any  decision one may or may not have made to ‘engage in politics’;  nevertheless, the fact that one is a human social being, engaged in  rule-governed activities and on that account able to deliberate and to  choose, is in itself sufficient to commit one to the acceptance of  legitimate political authority. For the exercise of such authority isa  precondition of rule-governed activities. There would, therefore, bea  sort of inconsistency in “choosing to reject’ all such authority.  And since the acceptance of such authority is implied in the kind of  behaviour to which alone the category of freedom of choice is  applicable, it would be absurd to regard it as a derogation from a  man’s freedom of choice.    (2) I turn now to the other objection: that whereas there is  general agreement on what counts as ‘right’ when we are dealing with  the moves of chessmen, this is not true of other situations in which  authority is exercised—of politics, for instance. Now I agree that  here—and in some other contexts too—general agreement on the right  course of action is lacking; I agree, too, that it is precisely here  that it is necessary to have someone in authority. That is, where we  have no agreement about what is to be done, we must, unless we are  to lapse into chaos, have some agreement about who is to decide what  is to be done. But I still wish to maintain, in opposition to what I  take to be implied by Peters’ position, that we have to deal with  genuine authority, as opposed to bare power or ability to influence,  only where he who decides does so under the idea of what he conceives  to be the right decision. This fundamental fact is not altered by the  controversial character of the distinction between right and wrong  here.    Consider for instance the authority of the Pope over Roman  Catholics which is, in a sense, absolute in religious matters. All the  same, if a Pope were to issue an Encyclical denying the existence of  God and advocating the practice of free love, 1 doubt whether this  would be recognized as carrying the papal authority along with it.  Papal authority, that is, is not completely beyond the possibility    106 PETER WINCH    .of all criticism;! and this, I want to say, is true of all authority,  because authority is essentially bound up with systems of ideas, and  systems of ideas essentially involve the possibility of discussion and  criticism.   Again, certainly not everyone would agree that the Labour  Government acted rightly in nationalizing the steel industry. Never-  theless part of that act’s authoritative character derived from the  fact that it was claimed to be the right thing to do in the circumstances  (and some sort of case had to be made out for it). A great deal of  authority would have been lost if the action had been generally and  seriously regarded as an arbitrary act of dispossession for the sake of  personal enrichment or for the sake of a social grudge. An authority  can be allowed to make mistakes (up to a certain point) about what  is the right course to follow, and still retain its authoritative character:  but for it to be thought that it no longer cares about what is right  and what is wrong (in the sense appropriate to the context in which  it operates) is for it to degenerate from authority into force. For  reasons connected with this, M. de Jouvenel seems to me to be saying  something of considerable philosophical importance, in his book on  Sovereignty, when he recalls the attention of political philosophers  from the problem of who is to decide to the problem of what is to be  decided. This second problem is not merely pragmatically important;  it is conceptually interwoven with the first problem via the concept  of authority.   I shall now consider the implications of what I have been say-  ing for the distinctions which Dr. Peters attempts to draw in his  paper. His position appears to be the following. On the one hand  we have authority of what Weber called the ‘traditional’ and ‘legal-  rational’ types. It is characteristic of these that the authority in  question attaches to a status or an office defined and held accord-  ing to some more or less explicit system of rules. On the other  hand we have ‘de facto authority’ a watered-down version of Weber’s  ‘charismatic’ authority, which attaches to a specific person in  view of certain personal qualities which he exhibits—as in Peters’  quotation from King Lear. Intermediately, we have authority which   ' Peters said, in an earlier version of his paper, that a man who supports what he says  by claiming to speak with the voice of God (cf. Hobbes, op. cit., p. 243) is trying to  rule out the need to produce arguments. Perhaps he is ruling out arguments of acertain    sort; but what he is producing is itself an argument of a different sort, which a religious  man may give (religious) reasons for accepting or rejecting.    2Cf. G. C. Homans: The Human Group, p. 171 for an illuminating illustration of the  close connexion between the idea of authority and that of the right way to behave.    AUTHORITY 107    is accepted by virtue of what Peters calls ‘credentials of a personal  sort’—a history of outstanding success, for example, in a given field  of activity.   Peters has difficulty with the notion of ‘de facto authority’; for  while, on the one hand, he is unwilling to say that this may depend on  a set of rules of some sort—as this would seem to endanger its distinct-  ness from ‘de jure authority’; yet, on the other hand, in order to  distinguish it from mere power, he has somehow to bring in the notion  of there being some right to exercise it on the part of the person who  does so. He concludes: ‘All that can be said is that there is something  about him (sc.the wielder of de facto authority) which people recognize  in virtue of which they do what he says simply because he says it.’  This sounds to me suspiciously evasive. Either people do what he says  simply because he says it, or else they do what he says in virtue of  something else about him, which they recognize. If the latter, then it  will be part of the philosopher’s job to say what that ‘something else’  is. I suggest that the way in which Peters has drawn his distinctions  precludes him from doing this.   I can show this better by considering the distinctions of Max  Weber’s on which Peters leans. According to Weber ‘traditional’,  ‘legal-rational’, and ‘charismatic’ authority represent three ‘ideal  types’. That is, they are conceptually distinct though seldom, if ever,  found in their pure forms in actuality. I want to suggest, on the  contrary, that these three types are not even conceptually distinct.  Both the idea of the ‘legal-rational’ and that of the ‘charismatic’  presuppose the idea of a tradition.   I will concentrate here on the notion of charisma. Weber says  quite explicitly that charismatic authority is not at all tied to a  tradition. (Cf. Wirtschaft und Gesellschaft, p. 555.) In the same strain he  remarks that the characteristic attitude of the charismatic leader  is: ‘It is written that....But I say unto you....(Ibid., p. 141.)  Charismatic authority is conceived as a revolutionary force, as one  of the main agencies by which new ways of living and thinking are  introduced into a society.   Granted that this is so, it is still very misleading to oppose  charisma to tradition. The point about it is not that it stands apart  from established ways of doing things but that it stands to them in  a very special relation. Apart from the tradition to which it stands  in such a relation it is quite unintelligible and inconceivable.   Jesus Christ certainly revolutionized the religion of the Jews.  The authority that he exercised was clearly very different from that    108 PETER WINCH    of an orthodox rabbi. But what Jesus was, what he did and said, and  the kind of authority he exercised, are completely unintelligible  apart from the Jewish religious tradition. He came to fulfil the Law.  When he opposed what he said to what was written he cannot sensibly  be taken to have meant that he was replacing what was written by  something completely different. Rather, he threw new light on what  was written; and what he said could not be understood as it was  intended except by someone who had some knowledge of what was  written. (This question is discussed with great illumination by  Schweitzer in his The Psychiatric Study of Jesus.)   Or, to take a very different example, William Webb Ellis when in  1823 he picked up the ball and ran and thereby created the game of  Rugby Football, exercised a sort of charismatic authority over his  fellow-participants in the game: an authority very different in kind  from the legal-rational sort wielded by the duly appointed captain of  a football team. But still, Ellis could not have done what he did apart  from the rules of the game as they then existed; and we can only  understand the nature of the authority which he exercised by  considering what he did, and the effects of what he did, in relation to  those rules.   Peters says that the charismatic or natural de facto leader is unable  to give reasons for what he does. Perhaps so. Nevertheless, what he  does or proposes has a sense; and it derives this sense from the tra-  dition of activity in the context of which it occurs, whether this bea  context of religious thought and practice, rules of football, or  whatever.   To say that X is exercising de facto authority when his decisions are  accepted simply because they are his does not go far enough. As Peters  implicitly recognizes when he says that his decisions are accepted  because people recognize ‘something about him’, we must look  further than the mere fact that his decisions are accepted, if we are to  account for his authority. We must ask what lies behind that accep-  tance. And what will be found to lie behind it is the tradition of  activity which gives his proposals and decisions, and other people’s  acceptance of them, their sense. His authority consists in the fact that  his followers trust him to show them the right course to pursue in the  context of that activity. And his exercise of that authority. may, in  genuinely charismatic cases, result in the giving of a new sense to the  notion of ‘the right course to pursue’.   Let us consider a little more closely the external marks of natural  authority which Peters emphasizes in Section 8 of his paper. Let us    AUTHORITY 109    ask what makes a certain air, a certain tone of voice, a certain demean-  our, a sign of authority. Certainly not anything intrinsic to the  demeanour itself. In the film about the exploits of H.M.S. Amethyst,  ‘Yangtse Incident’, the ‘authoritative’ demeanour of the rating who  dressed up as an officer in order to bluff the Chinese commandant  was merely laughed at by his fellow ratings. A given demeanour can  only be a sign of authority in a special sort of context.   If we try to explain what (even de facto) authority 1s in terms of  these external marks—as when Peters wonders ‘whether any such  empirical conditions must come to be regarded as grounds fora right’ —  we shall get into the same kind of difficulty as Descartes got into in  trying to account tor knowledge in terms of the clearness and distinct-  ness of ideas. The clearness and distinctness of one’s ideas may bea  sign that one knows what one is talking about, but will not by itself  serve as a criterion of the truth of what one is saying. Similarly,  the confidence of one’s demeanour may be a sign that one knows the  right thing to do and that what one proposes may therefore be  accepted as authoritative. But even the success of one’s confident  demeanour in inducing others to do what one proposes is not the  ultimate test of one’s authority. The test of whether or not other  people were right to accept one’s authority will be the subsequent  assessment of the rectitude of what was done at one’s instigation.   The parallels are close between the misguided attempts in political  philosophy to account for authority in terms of the properties and  relations between individual wills and the equally misguided attempts  in epistemology to account for knowledge as a property of the  individual mind. Authority is no more a sociological concept than  knowledge is a psychological one.    Postscript (1967)    The subject of this symposium was authority sans phrase, rather  than political authority, and I want to emphasize that my paper was  concerned with the more general question. This is important for the  following reason: what I wrote about specifically political authority  is badly mistaken {as I realized as soon as the original paper had  gone to the press) and the mistake tends to some extent to give a  misleading appearance to my whole argument.   It is a central point of Hobbes’s thesis that the answer to the  question, “What constitutes the unity of the state?’ also provides  the answer to the question, ‘What are the conditions under which we    110 PETER WINCH    are entitled to call a collection of individuals a human society?’ Now  I believe it is a consequence of my line of argument that Hobbes is  quite mistaken in what he says about this. But this is obscured by the  impression my paper gives that it is concerned with just the same  question as Hobbes was concerned with in the passages quoted by  Peters and myself. The question to which my remarks are relevant is,  ‘What is it about human life that makes the concept of authority  applicable at all?’; and I tried to show that the answer to this question  is also part of the answer to the question ‘What is a human society ?’  But though these latter questions have to be answered by anyone  who wishes to clarify the peculiar nature of political authority,  their answers will not in themselves provide such a clarification,  which requires an analysis not just of the way concepts in general  enter into life in human societies, but of the way a particular  set of political concepts enters into the life of a body politic  and into the binding together of its members under a common  regime.   My failure, in my paper, to make these distinctions clear enough  led me to make the following quite false remark: ‘the fact that one is  a human social being, engaged in rule-governed activities and on that  account able to deliberate and to choose, is in itself sufficient to  commit one to the acceptance of legitimate political authority’. This  position is close to Locke’s analogy between a political ruler and an  umpire; and there are two considerations which show it to be quite  mistaken. (1) There is, it now seems to me, no good reason to  suppose that social life is impossible in the absence of anything like  the authority of the state. (2) The authority of the state, where it  exists, is suz generis and somehow imposed from without on other  social institutions. (This is one of the main points of Hume’s criticism  of Locke in his essay ‘Of the Original Contract’, the argument of which  I think is by and large correct.)   But J still think that Hume’s argument needs to be supplemented  by something like the main argument of my paper. Though the state  faces other social institutions as something like an external force  with its own, in a way independent, sources of authority, still this  force and this authority are what they are by virtue of the fact that  there exists a concept of the state in the society within which they  are exercised—a concept which enters into what subjects will and will  not submit to from the state and into the activities of the officers of  the state. This concept is not itself imposed by the state; it manifests  itself in the spontaneous life of the society, even though its existence    AUTHORITY 111    makes possible the imposition of certain things in a way which would  not otherwise be possible.   To say this much is to do no more than state a problem: what is the  peculiar character of this concept and what difference does its  existence make to the life of a society? I do not attempt to answer this  question here; the purpose of this postscript is simply to correct the  misleading appearance of my original paper to claim to have provided  an answer.    VI  THE PUBLIC INTEREST    BRIAN BARRY  I    A TRIBUNAL of Enquiry claims that the public interest requires  journalists to disclose their sources of information; the Restrictive  Practices Court invalidates an agreement among certain manufac-  turers as contrary to the Restrictive Practices Act and therefore  contrary to the public interest; the National Incomes Commission  says that a proposed rise for the workers in an industry would be  against the public interest. These examples could be multiplied end-  lessly. Each day’s newspaper brings fresh ones. In arguments about  concrete issues (as opposed to general rhetoric in favour of political  parties or entire societies) ‘the public interest’ is more popular than  ‘justice’, ‘fairness’, ‘equality’, or ‘freedom’.   Why is this? Roughly, there are two possible answers. One is that  ‘the public interest’ points to a fairly clearly definable range of  considerations in support ofa policy and ifitisa very popular concept  at the moment all this shows is that (for better or worse) these  considerations are highly valued by many people at the moment. This  is my own view. The other answer is that politicians and civil servants  find it a handy smoke-screen to cover their decisions, which are  actually designed to conciliate the most effectively deployed interest.   These sceptics often buttress their arguments by pointing out that  most theoretical writing about ‘the public interest’ is vague and  confused. This theme is copiously illustrated by Frank J. Soraufin his  article ‘The Public Interest Reconsidered’, Journal of Politics, XIX  (Nov. 1957) and by Glendon Schubert in his book The Public Interest :  Critique of a Concept. But it is a familiar idea that people who are  perfectly well able to use a concept may nevertheless talk rubbish  about it, so even if many of the writings about the concept are confused  it does not follow that the concept itself is. A more cogent line of  argument is to construct a definition of ‘the public interest’ and then    From Proceedings of the Aristotelian Society, Supp. Vol. 38 (1964), pp. 1-18. Reprinted  by courtesy of the author and the Editor of the Aristotelian Society.    THE PUBLIC INTEREST 113    show that, so defined, nothing (or not much) satisfies it. From this, it  can be deduced that most uses of the phrase in political discussion  must be either fraudulent or vacuous. Like Sorauf and Schubert, the  best known expositors of the view are Americans—one may mention  A. F. Bentley’s The Process of Government and D. B. Truman’s The  Governmental Process. But the most succinct and recent treatment is to  be found in Chapters Three and Four of The Nature of Politics byJ.D.B.  Miller, and it is to a criticism of these chapters that I now turn.    I]    Miller defines ‘interest’ as follows: ‘we can say that an interest  exists when we see some body of persons showing a common concern  about particular matters’ (p. 39). On the basis of this he later puts  forward two propositions. First, one is not ‘justified in going beyond  people’s own inclinations in order to tell them that their true interest  lies somewhere else’ (p. 41). It ‘seems absurd’ to suppose that an  interest can exist if those whose interest it is are not aware of it  (p. 40). And secondly, a ‘common concern. . . must be present if we  are to say that a general interest exists’. “A common concern will  sometimes be found in the society at large, and sometimes not. More  often it will not be there’ (p. 54).   Apart from the last point, which is a statement of fact and one I  shall not query here, these propositions follow analytically from the  original definition of ‘interest’, though Miller does not see this clearly.  Everything hinges on that slippery word ‘concern’ which plays sucha  crucial part in the definition. One can be concerned at (a state of  affairs) or concerned about (an issue) or concerned with (an organiza-  tion or activity) or, finally, concerned dy (an action, policy, rule, eéc.).  The noun, as in ‘so-and-so’s concerns’ can correspond to any of the  first three constructions, and it seems plain enough that in these three  kinds of use nobody can be concerned without knowing it. In the  fourth use, where ‘concerned by’ is roughly equivalent to ‘affected  by’, this is not so: someone might well be affected by an economic  policy of which he had never heard. But the noun ‘concern’ does not  have a sense corresponding to this, nor does Miller stretch it to cover  it. Naturally, if ‘interest’ is understood in terms of actual striving,  no sense can be given to the idea of someone’s having an interest  but not pursuing it. Similarly, if ‘interest’ is defined as ‘concern’ it  hardly needs several pages of huffing and puffing against rival  conceptions (pp. 52-54) to establish that “common or general    114 BRIAN BARRY    interest’ must be equivalent to ‘common or general concern’.   Since, then, Miller’s conclusions follow analytically from his  definition of ‘interest’, with the addition of a factual premise which  1 am not here disputing. I must, if I am to reject his conclusions,  reject his definition. Miller can, of course, define ‘interest’ any way  he likes; but if he chooses a completely idiosyncratic definition he  can hardly claim to have proved much if it turns out that most of the  things that people have traditionally said about interests then become  false or meaningless. He clearly believes himself to be taking part  in a debate with previous writers and it is because of this that he is  open to criticism.   Let us start from the other end. Let us begin by considering the  things we normally want to say about interests, the distinctions which  we normally want to draw by using the concept, and then see whether  it is not possible to construct a definition of ‘interest’ which  will make sense of these ordinary speech habits.   The first part of Miller’s definition, which makes interests shared  concerns, conflicts with our normal wish to drawa distinction between  someone’s private or personal interests on the one hand and the  interests which he shares with various groups of people on the other  hand. Simply to rule out the former by fiat as Miller does seems to  have nothing to recommend it. It might perhaps be argued in defence  of the limitation that only interests shared among a number of people  are politically important, but it can surely be validly replied that this  is neither a necessary nor a sufficient condition.   The second part of the definition equates a man’s interests with  his concerns. This conflicts with a great many things we ordinarily  want to say about interests. We want to say that people can mistake  their interests, and that while some conflicts are conflicts of interests,  others (e.g., “conflicts of principle’) are not. We distinguish between  ‘disinterested’ concern and ‘interested’ concern ina particular matter;  we find it convenient to distinguish ‘interest groups’ (e.g., The  National Farmers’ Union) from ‘cause’ or ‘promotional’ groups(e.g.,  The Abortion Law Reform Association). ‘They co-operate because  they have a common interest’ is ordinarily taken as a genuine explana-  tion, rather than a pseudo-explanation of the ‘vis dormitiva’ type, as  it would be if co-operation were identified with (or regarded as a direct  manifestation of) acommon interest. We allow that one can recognize  something as being in one’s interest without pursuing it. Finally, we do  not regard it as a contradiction in terms to say, ‘I realize that so-and-  so would be in my interests but nevertheless I am against it’. These    THE PUBLIC INTEREST 115    points are all inconsistent with Miller’s definition, and in addition  the last of them is inconsistent with any attempt such as that of S. I.  Benn to define a man’s interests as ‘something he thought he could  reasonably ask for’ (‘“Interest’’ in Politics’, Proceedings of the Aristotelian  Society, 1960, p. 127).   Can a definition be found which will make sense of all these uses  of ‘interest’ ? I suggest this: a policy, law or institution is in someone’s  interest if it increases his opportunities to get what he wants—what-  ever that may be. Notice that this is a definition of ‘in so-and-so’s  interests’. Other uses of ‘interest’ all seem to me either irrelevant or  reducible to sentences with this construction. Thus, the only unforced  sense that one can give to “What are your interests ?’, which Benn  imagines being put seriously to a farmer, is that it is an enquiry into  his favourite intellectual preoccupations or perhaps into his leisure  activities—applications of ‘interest’ whose irrelevance Benn himself  affirms. Otherwise, it has no normal application, though a ‘plain man’  with an analytical turn of mind (such as John Locke) might reply:    ‘Civil interest I call life, liberty, health and indolency of body; and the  possession of outward things, such as money, lands, houses, furniture and  the like’ (Letter Concerning Toleration).    This might be regarded as a specification of the kinds of ways in which  a policy, law or institution must impinge on someone before it can be  said to be ‘in his interests’. Unpacked into more logically trans-  parent (if more long-winded) terms it might read: ‘A policy, law or  institution may be said to be in someone’s interests if it satisfies the  following conditions... .’   The main point about my proposed definition, however, is that it  is always a policy that is said to be ‘in so-and-so’s interest’—not the  actual manner in which he is impinged upon. (From now on I shall  use ‘policy’ to cover ‘policy, law or institution’.) There are straight-  forward criteria specifying the way in which someone has to be  affected by a policy before that policy can be truly described as being  ‘in his interests’; but whether or not a given policy will bring about  such results may quite often be an open question.   It is this feature of ‘interest’ which explains how people can  ‘mistake their interests’—item number one on the list of ‘things we  want to say about interests’. The stock argument against this possibi-  lity is that if you assert it you must commit yourself to the view that  ‘some people know what’s good for other people better than they do  themselves’. But this can now be seen to rest on a gross equivocation.    116 BRIAN BARRY    The presumably illiberal, and therefore damaging, view to be saddled  with would be the view that policies which impinge on people in  ways which they dislike may nevertheless be said to be ‘in their  interests’. But this is not entailed by the statement that people  may ‘mistake their interests’. All that one has to believe is that  they may think a policy will impinge upon them in a way which will  increase their opportunities to get what they want when in fact it  will do the opposite. Whether his opportunities are increased or  narrowed by being unemployed is something each man may judge for  himself; but it is surely only sensible to recognize that most people’s  opinions about the most effective economic policies for securing  given ends are likely to be worthless. In his Fireside Chat on June 28,  1934, President Roosevelt said:    ‘The simplest way for each of you to judge recovery lies in the plain facts  of your own individual situation. Are you better off than you were last year?  Are your debts less burdensome? Is your bank account more secure? Are  your working conditions better? Is your faith in your own individual future  more firmly grounded?’    It is quite consistusc to say that people can ‘judge recovery for  themselves’ without respecting their opinions about the efficacy of  deficit financing.   The other ‘things we normally want to say’ also fit the proposed  definition. People may want policies other than those calculated to  increase their opportunities—hence the possibility of ‘disinterested  action’ and ‘promotional groups’. Similarly, a man may definitely not  want a policy which will increase his opportunities (perhaps because  he thinks that the policy is unfair and that others should get the  increase instead). Hence the possibility of someone’s not wanting  something that he acknowledges would be in his interests. Finally,  nothing is more common than for someone to agree that a policy  would increase his opportunities if adopted, and to want it to be  adopted, but at the same time to say that the addition of his own  efforts to the campaign to secure its adoption would have such a small  probability of making the decisive difference between success and  failure for the campaign that it is simply not worth making the effort;  and of course if everyone is in the habit of reasoning like this a policy  which is in the interests of a great many people, but not greatly in  the interests of any of them, may well fail to receive any organized  support at all.   No doubt there is room for amplification of my definition of what    THE PUBLIC INTEREST 117    it is for a policy to be in someone’s interest. In particular the phrase  ‘opportunities to get what he wants’ needs closer analysis, and  account should be taken of the expression ‘so-and-so’s best interests’  which tends to be used where it is thought that the person in question  would make such an unwise use of increased opportunities that he  would be better off without them (e.g., a heavy drinker winning  a first dividend on the football pools). However, 1 doubt whether  refinements in the definition of ‘interest’ would alter the correctness  or incorrectness of what I have to say about ‘the public interest’, so  I turn now to that expression.    iil    If ‘interest’ is defined in such a way that ‘this policy is in A’s interest’  is equivalent to ‘A is trying to get this policy adopted’ it is decisive  evidence against there being in any but a few cases a ‘public interest’  that there is conflict over the adoption of nearly all policies in a state.  But on the definition of ‘interest’ I have proposed this would no longer  be so. A policy might be truly describable as ‘in the public interest’  even though some people opposed it. This could come about ina way  already mentioned: those who oppose the policy might have ‘mistaken  their interests’. In other words, they may think the policy in question  is not in their interests when it really is. Most opposition in the  U.S.A. to unbalanced budgets can be explained in this way, for  example. Disagreements about defence and disarmament policy are  also largely disagreements about the most effective means to fairly  obvious common goals such as national survival and (if possible)  independence.   There are two other possibilities. One is that the group opposing  the measure is doing so in order to further a different measure which  is outside the range of relevant comparisons. The other possibility  is that the opposing group have a special interest in the matter which  counteracts their interest as members of the public. I do not expect  these two descriptions to be clear; I shall devote the remainder of the  paper to trying to make them so, taking up the former in this section  and IV, and the latter in V.   Comparison enters into any evaluation in terms of interests. To  say that a policy would be in someone’s interests is implicitly to,  compare it with some other policy—often simply the continuance of  the status quo. So if you say that a number of people have a common  interest in something you must have in mind some alternative to it    118 BRIAN BARRY    which you believe would be worse for all of them. The selection of  alternatives for comparison thus assumes a position of crucial impor-  tance. Any policy can be made ‘preferable’ by arbitrarily contrasting it  with one sufficiently unpleasant. Unemployment and stagnation look  rosy compared with nuclear war; common interests in the most  unlikely proposals can be manufactured by putting forward as the  alternative a simultaneous attack by our so-called ‘independent  deterrent’ on Russia and the U.S.A. All this need do is remind one that  one thing may be ‘in somebody’s interest’ compared with something else  but still undesirable compared with other possibilities. The problem  remains: is there (in most matters) any one course of action which is  better for everyone than any other? Fairly obviously, the answer is:  No. Any ordinary proposal would be less in my interest than a poll  tax of a pound a head, the proceeds to be given to me. And this can  be repeated for everybody else, taking each person one at a time.  This, however, seems as thin a reason for denying the possibility of  common interests as the parallel manoeuvre in reverse was for assert-  ing their ubiquity. In both cases the comparison is really irrelevant.  But what are the criteria for relevance? The simplest answer (which  will later have to have complications added) is that the only proposals  to be taken into account when estimating ‘common interests’ should  be proposals which treat everyone affected in exactly the same way.  Take the traditional example of a law prohibiting assault (including  murder). If no limitation is imposed upon the range of alternatives it  is easy to show that there is no ‘common interest’ among all the  members of a society in having such a law directed equally at everyone.  For one could always propose that instead the society should be  divided into two classes, the members of the first class being allowed  to assault the members of the second class with impunity but not vice  versa, as with Spartans and Helots; or each member of the first group  might be put in this position only vis-d-vis particular members of  the second group. (Examples of this can be drawn from slave-holding,  patriarchal, or racially discriminatory systems such as the ante-  bellum South, ancient Rome and Nazi Germany respectively.) It could  perhaps be argued that the ‘beneficiaries’ under such an unequal  system become brutalized and are therefore in some sense ‘worse off’  than they would be under a regime of equality. But the whole point of  ‘interest’—and its great claim in the eyes of liberals—is that the  concept is indifferent to moral character and looks only at oppor-  tunities.   Yet even the most sceptical writers often admit that a law prohibiting    THE PUBLIC INTEREST 119    assault by anyone against anyone is a genuine example of something  which is ‘in the public interest’ or ‘in everyone’s interest’. This  becomes perfectly true when the alternatives are restricted to those  which affect all equally, for then the most obvious possibilities are  (a) that nobody should assault anybody else and (b) that anybody  should be allowed to assault anybody else. And of these two itis hardly  necessary to call on the authority of Hobbes to establish that, given  the natural equality of strength and vulnerability which prevents any-  one from having reasonable hopes of gaining from the latter set-up,  the former is ‘everyone’s interest’.    IV    A convenient way of examining some of the ramifications of this  theory is to work over some of the things Rousseau says in the Social  Contract about the ‘General Will’. Judging from critiques in which  Rousseau figures as a charlatan whose philosophical emptiness is  disguised by his superficial rhetoric, it is hard to see why we should  waste time reading him, except perhaps on account of his supposedly  malign influence on Robespierre. I doubt the fairness of this estimate,  and J am also inclined to deprecate the tendency (often though not  always combined with the other) to look on Rousseau through  Hegelian spectacles. We need to dismantle the implausible psycholo-  gical and metaphysical theories (e.g., ‘compulsory rational freedom’  and ‘group mind’) which have been foisted on Rousseau by taking  certain phrases and sentences (e.g., ‘forced to be free’ and ‘moral  person’) out of context. As a small contribution to this process of  demythologiziny Rousseau I want to suggest here that what he says  about ‘the general will’ forms a coherent and ingenious unity if it  is understood as a treatment of the theme of common interests.   Rousseau’s starting point, which he frequently makes use of, is  that any group will have a will that is general in relation to its  constituent members, but particular with respect to groups in which  it in turn is included. Translating this into talk about interests it means  that any policy which is equally favourable to all the members of a  given group will be less favourable to member A than the policy most  favourable to A, less favourable to member B than the policy most  favourable to B, and so on; but it will be more favourable to each of  the members of the group than any policy which has to be equally  beneficial to an even larger number of people. Suppose, for example,  that a fixed sum—say a million pounds—is available for wage increases    120 BRIAN BARRY    in a certain industry. If each kind of employee had a separate trade  union one might expect as many incompatible claims as there were  unions, each seeking to appropriate most of the increase for its own  members. If for example there were a hundred unions with a  thousand members apiece each employee might have a thousand  pounds (a thousandth of the total) claimed on his behalf, and the  total claims would add up to a hundred million pounds. At the  other extreme if there were only one union, there would be no point  in its putting in a claim totalling more than a million pounds (we  assume for convenience that the union accepts the unalterability of  this amount) and if it made an equal claim on behaif of each of its  members this would come to only ten pounds a head. Intermediate  numbers of unions would produce intermediate results.   Rousseau’s distinction between the ‘will of all’ and the ‘general  will’ now fits in neatly. The ‘will of all’ is simply shorthand for ‘the  policy most in A’s interests, taking A in isolation; the policy most  in B’s interests, taking B in isolation; and so on’. (These will of  course normally be different policies for A, B and the rest.) The  ‘general will’ is a single policy which is equally in the interests  of all the members of the group. It will usually be different from  any of the policies mentioned before, and less beneficial to anyone  than the policy most beneficial to himself alone.   We can throw light on some of the other things Rousseau says in  the one-page chapter II.iii. of the Social Contract by returning to the  trade union example. Suppose now that the leaders of the hundred  trade unions are told that the money will be forthcoming only ifa  majority of them can reach agreement on a way of dividing it up. A  possible method would be for each leader to write down his preferred  solution on a slip of paper, and for these to be compared, the process  continuing until a requisite number of papers have the same proposal  written on them. If each started by writing down his maximum  demand there would be as many proposals as leaders— the total result  would be the ‘will of all’. This is obviously a dead end, and if no  discussion is allowed among the leaders, there is a good chance that  they would all propose, as a second best, an equal division of the  money. (There is some experimental evidence for this, presented in  Chapter 3 of Thomas Schelling’s The Strategy of Conflict.) Such a  solution would be in accordance with the ‘general will’ and represents  a sort of highest common factor of agreement. As Rousseau puts it, it  arises when the pluses and minuses of the conflicting first choices are  cancelled out.    THE PUBLIC INTEREST 121    If instead of these arrangements communication is allowed, and  even more if the groups are fewer and some leaders contro} large  block votes, it becomes less likely that an equal solution will be  everyone’s second choice. It will be possible for some leaders to  agree together to support a proposal which is less favourable to any  of their members than each leader’s first choice was to his own  members, but still more favourable than any solution equally bene-  ficial to all the participants. Thus, as Rousseau says, a ‘less general  will’ prevails.   In IL.iii. Rousseau suggests that this should be prevented by not  allowing groups to form or, if they do form, by seeing that they are  many and small. In the less optimistic mood of IV.i., when he returns  to the question, he places less faith in mechanical methods and more in  widespread civic virtue. He now says that the real answer is for  everyone to ask himself ‘the right question’, i.e., ‘What measure will  benefit me in common with everyone else, rather than me at the  expense of everyone else?’ (I have never seen attention drawn to the  fact that this famous doctrine is something of an afterthought whose  first and only occurrence in the Social Contract is towards the end.)  However, this is a difference only about the most effective means of  getting a majority to vote for what is in the common interest of all.  The essential point remains the same: that only where all are equally  affected by the policy adopted can an equitable solution be expected.    ‘The undertakings which bind us to the social body are obligatory only  because they are mutual; and their nature is such that in fulfilling them we  we cannot work for others without working for ourselves... ... What makes  the will general is less the number of voters than the common interest  uniting them; for, under this system, each necessarily submits to the conditions  he imposes on others: and this admirable agreement between interest and  justice gives to the common deliberations an equitable character which at  once vanishes when any particular question is discussed, in the absence of a  common interest to unite and identify the ruling of the judge with that of the    party. (I1.iv. Cole’s translation.)    Provided this condition is met, nobody will deliberately vote for a  burdensome law because it will be burdensome to him too: this is why  no specific limitations on ‘the general will’ are needed. Disagreements  can then be due only to conflicts of opinion—not to conflicts of  interest. Among the various policies which would affect everyone in  the same way, each person has to decide which would benefit himself  most—and, since everyone else is similarly circumstanced, he is    122 BRIAN BARRY    automatically deciding at the same time which would benefit everyone  else most. Thus, to go back to our example of a law prohibiting  assault: disagreement will arise, if at all, because some think  they (in common with everyone else) would make a net gain of  opportunities from the absence of any law against assault, while  others think the opposite. This is, in principle, a dispute with a  right and a wrong answer; and evervone benefits from the right  answer’s being reached rather than the wrong one. Rousseau claims  that a majority is more likely to be right than any given voter, so that  someone in the minority will in fact gain from the majority’s decision  carrying the day. This has often been regarded as sophistical or  paradoxical, but it is quite reasonable once one allows Rousseau his  definition of the situation as one in which everyone is co-operating to  find a mutually beneficial answer, for so long ; as everyone is taken as  having an equal, better than even chance of giving the right answer,  the majority view will (in the long run) be right more often than that  of any given voter. (Of course, the same thing applies in reverse:  if each one has on average a less than even chance of being right,  the majority will be wrong more often than any given voter.) The  formula for this was discovered by Condorcet and has been presented  by Duncan Black on page 164 of his Theory of Committees and Elections.  To illustrate its power, here is an example: if we have a voting body  of a thousand, each member of which is right on average fifty-one per  cent of the time, what is the probability in any particular instance  that a fifty-one per cent majority has the right answer? The answer,  rather surprisingly perhaps, is: better than two to one (69%). More-  over, if the required majority is kept at fifty-one per cent and  the number of voters raised to ten thousand, or if the number of  voters stays at one thousand and the required majority is raised to  sixty per cent, the probability that the majority (5,100 to 4,900  in the first case or 600 to 400 in. the second) has the right  answer rises virtually to unity (99.97%). None of this, of course,  shows that ‘Rousseau was right’ but it does suggest that he was no  simpleton.   To sum up, Rousseau calls for the citizen’s deliberations to com-  prise two elements: (a) the decision to forgo (either as unattainable  or as immoral) policies which would be in one’s own personal  interest alone, or in the common interest of a group smaller than the  whole, and (b) the attempt to calculate which, of the various lines of  policy that would affect oneself equally with all others, is best for  him (and, since others are like him, for others). This kind of two-step  deliberation is obviously reminiscent of the method recommended in    THE PUBLIC INTEREST 123    Mr Hare’s Freedom and Reason, with the crucial difference that whereas  Mr Hare will settle for a willingness to be affected by the policy in  certain hypothetical circumstances, Rousseau insists that my being  affected by the policy must actually be in prospect. There is no need  to construct a special planet to test my good faith—my bluff is called  every time. By the same token, the theory I have attributed to Rousseau  requires far more stringent conditions to be met before something can  be said to be in the common interest of all than the vague requirement  of ‘equal consideration’ put forward by Benn and Peters in their  Social Principles and the Democratic State.    Vv    Even if Rousseau can be shown to be consistent it does not follow  that the doctrine of the Social Contract has wide application. Rousseau  himself sets out a number of requirements that have to be met before  it applies at all: political virtue (reinforced by a civil religion),  smallness of state, and rough economic equality among the citizens.  And even then, as he points out plainly, it is only a few questions  which allow solutions that touch all in the same way. If only some are  affected by a matter the ‘general will’ cannot operate. It is no longera  case of each man legislating for himself along with others, but merely  one of some men legislating for others. It is fairly obvious that  Rousseau’s requirements are not met in a great modern nation  state—a conclusion that would not have worried him. But since lam  trying to show that ‘the public interest’ is applicable in just such a  state it does have to worry me. It is here that I must introduce my  remaining explanation of the way in which something can be ‘in the  public interest’ while still arousing opposition from some.   Think again of the examples with which I began this paper. The  thing that is claimed to be ‘in the public interest’ is not prima face in  the interests of the journalist whose sources may dry up, the workers  whose rise is condemned or the businessmen whose restrictive prac-  tices are outlawed. But do first appearances mislead? After all, the  journalist along with the rest gains from national security, and  workers or industrialists gain along with the rest from lower prices.  To avoid a flat contradiction we need more refined tools; and they  exist in ordinary speech. Instead of simply saying that some measure  is ‘in his interests’ a man will often specify some réle or capacity in  which it is favourable to him: ‘as a parent’, ‘as a businessman’, ‘asa  house owner’ and so on. One of the capacities in which everyone finds  himself is that of ‘a member of the public’. Some issues allow a    124 BRIAN BARRY    policy to be produced which will affect everyone in his capacity as  a ‘member of the public’ and nobody in any other capacity. This is  the pure ‘Rousseau’ situation. Then there are other issues which  lack this simplicity but still do not raise any problems because those  who are affected in a capacity other than that of ‘member of the public’  are either affected in that capacity in the same direction as they are  in their other capacity of ‘member of the public’ or at least are not  affected so strongly in the contrary direction as to tip the overall  balance of their interest (what I shall call their ‘net interest’) that way.  Although this is not quite what I have called the ‘Rousseau’ situation,  the ‘Rousseau’ formula still works. Indeed, Rousseau sometimes  seems explicitly to accept this kind of situation as satisfactory,  as when he says (III.xv.) that ina well-ordered state ‘the aggregate of  the common happiness furnishes a greater proportion of that of each  individual’.   Finally, we have the familiar case where for some people a special  interest outweighs their share in the public interest. The journalist  may think, for example, that compulsory disclosure of sources would  indeed be in the public interest but at the same time conclude that  his own special interest as a journalist in getting information from  people who want to stay anonymous is more important to him than  the marginal gain in security that is at ‘stake. In such cases as this  Rousseau’s formula will not work, for although everyone still has a  common interest in the adoption of a certain policy gua ‘member of  the public’, some have a net interest in opposing it.   To adopt the policy which is ‘in the public interest’ in such a case  is still different from deliberately passing over an available policy  which would treat everyone equally, for in the present case there 1s  no such policy available. Even so, it involves favouring some at the  expense of others, which makes it reasonable to ask whether it is  justifiable to recommend it. Various lines of justification are possible.  Bentham seerris to have assumed that in most matters there was a  public interest on one side (e.g., in cheap and speedy legal procedures)  and on the other side the ‘sinister’ interest of those who stood to  gain on balance from abuses (e.g., ‘Judge & Co.’) and to have believed  (what is surely not unreasonable) that a utilitarian calculation would  generally give the verdict to the policy favouring ‘the public’. Ona  different tack, it might be argued that it is inequitable for anyone  to benefit from ‘special privileges’ at the expense of the rest of the  community. But unfortunately neither of these is as clear as it looks  because a hidden process of evaluation has already gone on to decide    THE PUBLIC INTEREST 125    at what point an interest becomes ‘sinister’ and how well placed  someone must be to be ‘privileged’. The cheapest and speediest dis-  pensation of law could be obtained by conscripting the legal profes-  sion and making them work twelve hours a day for subsistence rations;  but this would no doubt be ruled out by a utilitarian as imposing  ‘hardship’ and by the believer in distributive justice as not giving a  ‘just reward’ for the work done. Thus, by the time one has fixed the  level of rewards beyond which one is going to say that ‘privilege’ and  ‘sinister interest’ lie it is virtually analytic that one has defined a  ‘good’ solution (whether the criteria be utilitarian or those of dis-  tributive justice).   It is clearer to say that in these ‘non-Rousseauan’ situations the  public interest has to be balanced against the special interests  involved and cannot therefore be followed exclusively. But ‘the public  interest’ remains of prime importance in politics, even when it runs  against the net interest of some, because interests which are shared  by few can be promoted by them whereas interests shared by many  have to be furthered by the state if they are to be furthered at all.  Only the state has the universality and the coercive power necessary  to prevent people from doing what they want to do when it harms the  public and to raise money to provide benefits for the public which  cannot, or cannot conveniently, be sold on the market: and these  are the two main ways in which ‘the public interest’ is promoted.  This line of thought brings us into touch with the long tradition that  finds in the advancement of the interests common to all ane of the  main tasks of the state. The peculiarity of the last two centuries or  so has lain in the widespread view that the other traditional candi-  dates—the promotion of True Religion or the enforcement of the Laws  of Nature and God—should be eliminated. This naturally increases  the relative importance of ‘the public interest’.   A contributory factor to this tendency is the still continuing process  of social and economic change which one writer has dubbed the  ‘organizational revolution’. These developments have in many-ways  made for a more humane society than the smaller-scale, more loosely  articulated, nineteenth-century pattern of organization could   rovide. But they have had the incidental result of making obsolete  a good deal of our inherited conceptual equipment. Among the  victims of this technological unemployment are ‘public opinion’  and ‘the will of the people’. On most of the bills, statutory instruments  and questions of administrative policy which come before Parliament  there is little corresponding to the nineteenth-century construct of    126 BRIAN BARRY    ‘public opinion’: the bulk of the electorate holding well-informed,  principled, serious views. Even when an issue is sufficiently defined  and publicized for there to be a widespread body of ‘opinion’ about  it these opinions are likely to be based on such a small proportion  of the relevant data that any government which conceived its job as  one of automatically implementing the majority opinion would be  inviting disaster.   This does not entail that voting with universal suffrage is not a  better way of choosing political leaders than any alternative; but if  ‘public opinion’ is a horse that won’t run this means that ‘public  interest’ has to run all the harder to make up, since as we have seen  it has the advantage of operating where those affected by the policy  in question have not even heard of it and would not understand it if  they did. Consider for example the arrangements which enable the  staffs of organizations whose members are affected by impending or  existing legislation to consult with their opposite numbers in govern-  ment departments about its drafting and administration. This system  of ‘functional representation’, which now has almost constitutional  status, would not get far if each side tried to argue from the opinions  of its clients (the organization members and ‘the public’ respectively)  on the matter; but their interests do provide a basis for discussion,  a basis which leaves room for the uncomfortable fact that in a large  organization (whether it be a trade union, a limited company or a  state) information and expertise are just as likely to be concentrated  in a few hands as is the formal power to make decisions.    VI    At the beginning of this paper I suggested that the popularity of  ‘the public interest’ as a political justification could be attributed  either to its vacuity or to its being used to adduce in support of  policies definite considerations of a kind which are as a matter of  fact valued highly by many people. If my analysis of ‘the public  interest’ is correct, it may be expected to flourish in a society where  the state is expected to produce affluence and maintain civil liberties  but not virtue or religious conformity, a society which has no distinc-  tion between different grades of citizen, and a society with large  complex organizations exhibiting a high degree of rank and file apathy.  I do not think it is necessary to look any further for an explanation  of the concept’s present popularity.    VII  LIBERTY AND EQUALITY.  E. F. CARRITT    IN a recently published collection of essays called ‘Why am I a Demo-  crat?’!' Mr. Ronald Cartland says: ‘What we must settle at once is  whether we rate freedom? above equality.’ ‘Equality involves subjug-  ation and repression.’ I select this statement only as a candid and  contemporary expression of a doctrine that has always seemed to me  both paradoxical and muddled. Left to myself I should have thought  that liberty and equality involved one another; indeed I should have  found it hard to separate them. Mr. Cartland himself seems not quite  free from confusion here for, between the two remarks which I have  just quoted, he says: “Toleration and equal justice are possible only ina  democracy’, where toleration, I suppose, means freedom of speech,—  equal freedom,—and democracy means political equality; equal  justice, I suppose, is simply justice, for unequal justice would be  injustice. So it is implied that freedom should be equal, and that itand  justice are only possible with equal political power. What then is the  equality with which freedom is supposed to be incompatible? To  answer this question I think we must go back to the history of the  doctrine. For it is no new one.   Burke paid tribute to liberty, which he thought was conferred and  safeguarded by the British Constitution of his day, but to that  constitution he thought democracy or political equality was abhorrent.  With the enlargement of the franchise during the nineteenth century  it began to be assumed, at first by revolutionaries, later by Whigs,  young Tories and Tory Democrats, finally by almost all public  speakers, not only that we desired liberty but that what conferred and  safeguarded it was democracy, that is political equality, which they  identified with the British Constitution as revised. Consequently  Burke on their view had been wrong. But in the spirit of Burke it was    From Law Quarterly Review, Vol. 56 (1940), pp. 61-74. Reprinted by permission  of the author’s executors and the Law Quarterly Review.    ‘Edited by R. Acland (Lawrencé and Wishart).  71 use the words liberty and freedom in the same sense.    128 E. F. CARRITT    still declared, for instance by both Gladstone and Disraeli, that this  freedom was incompatible with some other equality.   These are platform politics, but if we look at the considered  statements of political theorists and especially of historians, we  find the same thing.   Acton in his Lectures on Liberty says that in the course of the French  Revolution ‘the passion for equality made vain the hope of freedom’.  Lecky in Democracy and Liberty (1, 212-215) says that ‘Democracy (ie.  equality of political power) may often prove, the direct opposite of  liberty... it destroys the balance of classes’ (i.e. introduces class  equality). Bagehot in The English Constitution and Erskine May in  Democracy in Europe (11, 333) work the same theme, but perhaps the  most striking exposition is in Sir James Fitzjames Stephen’s Liberty,  Equality and Fraternity (p. 250):   ‘I doubt much whether the power of particular persons over their  neighbours has ever in any age of the world been so well defined and  safely exercised as it is at present. If, in old times, a slave was  inattentive, his master might no doubt have him maimed or put to  death or flogged; but he had to consider that in doing so he was  damaging his own property, that when the slave had been flogged he  would continue to be his slave; and that the flogging might make him  mischievous or revengeful and so forth. If a modern servant mis-  conducts himself he can be turned out of the house on the spot, and  another hired as easily as you would calla cab. To refuse the dismissed  person a character may very likely be equivalent to sentencing him to  months of suffering and to a permanent fall in the social scale.’   Now what can Sir James have been driving at? I think it is clear  that he deplored the power which employers have over their servants, as  being so great that the latter have less freedom than slaves. And he  seems to attribute this inequality of freedom, which of course is the  correlative of an inequality of power, to their equality in some other  respect, I suppose to their political equality. He gives no grounds for  this attribution, except the suggestion that it is seldom to the interest  of a slave-owner or cattle-breeder to injure his chattels. I can hardly  think he would have proposed to remedy the insecurity of wage-    'M. Arnold (Mixed Essays, Equality) cites a speech of Lord Beaconsfield to  Glasgow students about 1856 and quotes Mr. Gladstone as ‘in his copious and  eloquent way’ saying: ‘Call this love of inequality by what name you please,—the  complement of the love of freedom, or its negative pole, or the shadow which the  love of freedom casts, or the reverberation of its voice in the halls of the constitution,—  it is an active, living, and life-giving power.’    LIBERTY AND EQUALITY 129    earners by substituting an extreme servitude with a legal right in the  owner to kill, flog or maim his slaves. Nor does it seem likely that he  is consciously arguing for the more obvious remedy, that of increasing  freedom by adding economic to political equality. I cannot tell what  he wants. All he has shown is, what nobody can have doubted, that  majority government does not necessarily and immediately secure the  maximum of equal freedom. Nor does any form of government. He  might have gone further and said that it is possible for a majority  government not only to allow economic inequality and consequent  interferences with liberty, but actually itself to be as intolerant as  any other government of free speech and action. James Mill (Govern-  ment, Encyclopedia Britannica) is brutally frank : “Whenever the powers  of the Government are placed in any hands other than those of the  community, whether those of one man, of a few, or of several, those  principles of human nature which imply that Government is at all  necessary, imply that those persons will make use of them to defeat  the very end for which Government exists.’ For the word several which  I have italicized he ought to have substituted the majority. But he  ought also to have added that those principles of human nature which  make stable government at all possible may counteract the principles  he has described. As Maitland points out (Liberty, Collected Papers I)  no form of government can guarantee liberty, but only ‘an opinion of  right’. (The phrase is Hume’s.) But I know of little reason to think  that majority governments are less favourable to economic equality  and the resulting equal liberty than are plutocracies, aristocracies  or despotisms. Even Plato, idealizing aristocracy, thought it must  be precluded by communism from the temptation to oppression. And  I know of no reason to think that democracies are less tolerant of  free speech and propaganda than other kinds of government. Such  tolerance probably depends upon the amount of security a govern-  ment feels, which is apt to be in proportion to the equality of  political power.  éveoti pap nw¢ todto TH Tupavvidt  véonpa, toic pidowor uy nenobévan  (Aesch. Prom. Desm, 226).   The only way, then, in which I can rationalize the lamentations  of Sir James Fitzjames Stephen is to suppose’ him an unconscious  socialist, who so desired equality of liberty to be secured by economic  equality that he was willing to sacrifice equality of political power  and to institute some class or personal dictatorship.   Perhaps the most influential source for this vague antithesis of    130 E. F. CARRITT    freedom to equality is De Tocqueville’s L’Anaen Régime (1856). He  says that countries without an aristocracy are peculiarly liable to  despotic or ‘absolute’ government in its worst forms, and he quotes  Mirabeau: ‘Cette surface égale facilite exercise du pouvoir.’ He  does not seem to think that an aristocratic government could be  despotic. The elements of aristocracy he enumerates are: ‘parlement,  pays d’état, corps de clergé, privilégiés, noblesse’.   The reason he gives for his view is that ‘When men are no longer  united by bonds of caste, class, corporation, family, they tend to be  wholly preoccupied with their private interests’, and in particular  with money-making.   The inequality he thinks necessary for freedom is clearly not  economic; that existed in the France of his own day, which he con-  siders servile. He must mean political inequality.   He seems to have two points in mind:   (1) A highly centralized government is apt to be oppressive even  if unwillingly. But he shows no reason for supposing that democracy  (i.e. political equality) is especially favourable to centralization.  He emphasizes the high degree of centralization in the ancien régime.   (2) A ‘privileged nobility’ enjoys a good deal of freedom, though  the unprivileged masses may have none. This freedom, for instance  exemption from taxation, they will certainly try to defend against  aggression, and De Tocqueville seems to think that in doing so they  may incidentally defend or achieve some freedom for the unprivileged.  In England such defence happened to some extent in 1688, and such  achievement in 1832. In the ‘glorious revolution’ a land-owning  aristocracy resisted royal encroachment, with some advantage to the  liberty of the middle class. In the Reform Bill a new industrial  aristocracy by gaining liberty and political equality for itself made  them somewhat more accessible to the working classes; but it was  opposed by the landed and ecclesiastical aristocracy. Those who had  neither political nor economic privileges profited to some extent by  the battle between those who had both and those who had one and  coveted the other.   What evidence is there that democracy is more susceptible than  aristocracy to dictatorship? Napoleon, Hitler, Mussolini perhaps  all rose with the aid of something like democracy. Louis XIV, Lenin  and Franco did not. England, France, America, the Scandinavian  countries, Switzerland, Belgium, Holland are as democratic as any  countries, but have as yet no dictators. Whether economic equality  favours dictatorship is a question on which there can be little historical    LIBERTY AND EQUALITY 131    evidence, since no near approach to economic equality, unless in the  U.S.S.R., as a result of ‘the dictatorship of the proletariat’, has been  made. But most countries where the contrasts of wealth are com-  paratively small, such as the Scandinavian, have remained democratic.  Ancient tyrannies generally arose out of economic inequality, to  champion those who feared either exploitation or expropriation.  Modern dictatorships have found backing both from depressed  middle classes and from frightened capitalists. Among a people secure  in anything like economic equality such backing would be hard to  find. A dictator, thought to have established such equality, as in the  U.S.S.R., might indeed be enthusiastically retained, but that would  be in the belief, perhaps short-sighted, that he would maintain the  liberty he had thereby secured.   How unequal was the liberty which De Tocqueville thought equality  endangered can be shown by one or two quotations.In spite of corvée,  milice, arbitrary arrest, aristocratic exemption from taxation, he says:  “Il régnait dans l’ancien régime beaucoup plus de liberté que de nos  jours, mais une liberté toujours liée a Vidée d’exception et de  privilége, toujours contractée dans la limite des classes.’ ‘France dans  ses classes supérieures était libre.’ ‘Les nobles ne se préoccupaient  guere de la liberté générale des citoyens.”!   De Tocqueville then thinks political equality leads to governmental  oppression, and is not afraid of class or personal oppression. Stephen  thinks just the opposite; that political equality leads to class or  personal oppression. He feared the infraction of one subject’s liberty  by another more than its restriction by government. But his con-  temporaries who, like Mr. Cartland to-day, preached the incom-  patibility of liberty and equality, were mostly in the tradition of  laisser faire. Like De Tocqueville they were so frightened of any  governmental attempt to regulate or reform the economic system of  their day that they were careless how much that system itself allowed  of personal oppression. Lecky says (op. cit. pp. 212-215): “Equality  is only attained by a stringent repression of natural development.’  Nature is a familiar stalking horse for prejudice. It is equally true  that peace, order, security of life, limb, property are only attained  by a stringent repression of ‘natural’ development in one sense of  nature; and in that sense Hobbes thought the state of nature was one  of pretty equal fear and misery. The security of one man’s millions  or hundreds from crowds of overworked or unemployed is an    IMy italics.    132 E. F. CARRITT    inequality only maintained by a stringent repression of the ‘natural’  development called ‘helping oneself’ .! But there is another meaning of  the word nature, a meaning which, in the wilderness of the nineteenth  century, Matthew Arnold raised his voice to express. In his lecture to  the Royal Institution on Equality (Mixed Essays) he said:   ‘Property is created and maintained by law. It would disappear in  that state of private war and scramble which legal society supersedes.  That property should exist and that it should be held with a sense of  security” and with some power of disposal, may be taken, by us here  at any rate, as a settled maticr.* But that the power of disposal should  be practically unlimited, that the inequality should be enormous, or  that the degree of inequality admitted at one time should be admitted  always—this is by no means so certain. The right of bequest was in early  times .. . seldom recognized. In later times it has been limited in many  countries.... The cause of your being ill at ease is the profound  imperfectness of your social civilization .... The remedy is social  equality. Let me direct your attention to a reform of the law of bequest.  On the one side inequality harms by pampering; on the other by  vulgarizing and depressing.’   The characteristic Arnoldian refrain of the lecture is ‘Choose  Equality’. And equality is to be chosen because, far from repressing  ‘natural’ development, it liberates a ‘natural and vital instinct’ of  men, the instinct of ‘expansion’ or ‘humanization’. I think Arnold’s  diagnosis was right. The equality which his contemporaries thought  incompatible with liberty was mainly economic equality. They  thought that the promotion of this by law would somehow impair  the liberty of more people, or to a greater degree, than does the  maintenance of economic inequality (for of course they did not think  there should be xo laws of property). In order to make up our minds  how far, if at all, this is so, we must decide as nearly as we can  what we mean by liberty. Vaguely of course we all know; that is to  say we all agree on extreme cases: that a manacled man is not so free  as we are. But there are dubious instances where we might differ  from one another, or from ourselves at another time. For example:  how far is a man free who cannot throw up an ill-paid job without    'Cf. Dickinson, Justice and Liberty: ‘No regulation is more constant, more  crushing, more radical and severe, than that which is involved in property and the  police.’   ?This clause is otiose since Arnold clearly means by ‘property’ legal security  of possession.   *Consumption necessitates appropriation.’ Locke, Civ. Gout. I], 25-51.    LIBERTY AND EQUALITY 133    losing his house when there is an extreme shortage of housing?"   To make discussion profitable we must try to fix precisely the  sense in which we are now going to use the word. In this we must be  careful to depart as little as possible from normal usage, while avoid-  ing as far as possible its vagueness. Many of the confusions of theor-  izing on this topic have arisen from arbitrary definitions which went  against the very usages where all plain men would agree. But those  who adopt such arbitrary definitions seldom succeed in ridding  themselves or their readers of the ordinary associations of the word,  and so their procedure increases the very ambiguity it was meant to  avoid. Hegel, for instance identified freedom with obedience to the  laws of my State.? But the inescapable associations of the word  enabled him to suggest that therefore in obeying my State I am always  doing what I really want to do. Hardly less at variance with usage,  and. consequently hardly less confusing, has been the identification  of liberty with the unimpeded power to do what we ought? (or  perhaps what we think we ought) or to ‘contribute to the common  good’.t But it would go dead against ordinary usage to say that am  quite free if I am forbidden under penalties to smoke or to play  tennis on Sunday, though J never thought that either of these were  duties or ‘contributions to a common good’.   I think our definition of liberty should avoid the use of any moral  terms suchas ‘ought’, ‘right’, ‘good’. Only so can we avoid prejudicing  the subsequent questions whether liberty is incompatible with equality,  and if so, which we ought to promote, or which is ‘better’.   I offer a preliminary definition of liberty as ‘the power of doing  what one would choose without interference by other persons’ action’ .°    1*We should not be searching for the definition if we already knew precisely the  meaning of the term; but the fact that we accept a certain definition as correct shows  that we think the definition expresses more clearly the very thing we had in mind when  we used the term without knowing its definition. The correctness of a definition is  tested by two methods: by asking (i) whether the denotation of the term and that of  the proposed definition are the same; (ii) does the definition express explicitly what  we had implicitly in mind when we used the term?’ Ross, Foundations of Ethics, p. 259.   2Phil. d. Rechts. §§ 15, 140 (e), 206. Cf. Bradley, Ethical Studies. My Station  and its Duties; and Bosanquet, Philosophical Theory of the State, pp. 96, 181, 240,  and especially 107, 127. ct my Morals and Politics under ‘Hegel’ and “‘Bosanquet’  and ‘Liberty’.   3 Acton, op, cit. and, inconsistently enough, Hegel and Bosanquet.   4Green, Political Obligation, §§ 24, 120, etc.   5Maitland, Liberty (Collected Papers, 1) defines it as the absence of ‘External  restraints on human action which are themselves the results of human action’,  This is much the best discussion of Liberty which I know. I assume that by ‘restraint’    134 E. F. CARRITT    A maximum interference with liberty would be imprisonment with  manacles; a minimum, exclusion from one house or locked safe which  I wished to enter. Some of the elements of this definition need justifica-  tion.   (1) ‘Doing’. (a) Our thinking cannot be directly constrained (though  it can be influenced) by other persons. Thought is always free; (b) Our  feelings can be very painfully affected by others, for instance if they  smack us or whistle out of tune, or (when we love them) by their  indifference or neglect. And we do speak of ‘freedom from anxiety’ ;  but the qualification is necessary. A lover or anxious parent is not  thereby smpliciter unfree.   (2) ‘What one would choose’. (a) If I am ‘prevented’! from doing  what I should not choose to do, for instance, from cock-fighting or  stepping over a cliff in the dark, my freedom is not impaired. A  penal law against murder, then, limits the freedom of all who want to  murder but not of others. This is acceptable, but less welcome results  seem to follow. It might be plausibly argued that many, possibly most  people in this country, are so law-abiding that though they would  gladly be better fed, and though there is no just reason why they  should not be, they would not choose to help themselves, and that  therefore the laws of property and theft do not impair their freedom.  They refrain from stealing even where detection is impossible, and  therefore not through fear. The answer I think is that they act from an  inarticulate recognition of the admitted truth, that almost any  system of law giving some security of possession is better than none;  even bad laws secure more equal freedom than anarchy does, and so  it is our duty to support any system of security unless the contrary  behaviour will, with reasonable probability, contribute to substituting  a better system, which pilfering and swindling cannot do. If people  willingly conformed to a law forbidding access to mountains but  gladly profited by its repeal, I think usage compels us to say it had  inpaired their freedom. If, however, they actually voted or agitated  against its repeal, I think we must say it did not. If by long custom  people actually prefer and petition to remain slaves, or if they cease  even to wish to enjoy what is in the occupation of others, then their  slavery or exclusion does not diminish their freedom, unless they  change their mind. We may call them free fools and blame somebody    he means actual restraint. And potential restraint only becomes actual when I begin  to want to do what is forbidden. If freedom were power to do what I do not want, it  would be worthless.   ‘Or forbidden under penalty. See below.    LIBERTY AND EQUALITY 135    for their folly.’ (b) Bribes and promises do not impair freedom. The  man likes earning the bribe better than not earning it, whereas the  man deterred by threats would have preferred to act otherwise could  he have done so fearlessly.   (3) ‘Other persons’. (a) We can be prevented from doing what we  want by geographical conditions, weather, wild beasts or our own  bodily state. A swollen stream, a wolf, a broken leg do not impair our  freedom. If anybody thinks that usage is in favour of calling such  impediments a lack of freedom I would ask him to read in this essay  for ‘freedom (or liberty)’ ‘social freedom’. The same I think applies to  those who are prevented from doing what they like by belief about the  supernatural. They may be unfree from superstition or the fear of God  but not (socially or) sempliciter unfree. What influences their action  is not persons but the supposed nature of the universe. (b) But any  persons may impair my freedom: a neighbour, a dictator or a majority.  The fact that I have voted for the restriction makes no difference if  I should now like to break it. Ulysses’ sailors impaired his freedom by  his own orders when they prevented him from joining the Sirens. I  can even limit my own freedom by locking myself in an upper storey  and throwing the key out of the window; but not by vows or promises  without enforceable penalty. I am free to break them.   (4) ‘Interference’: so far as our action is impeded not by other  people’s action but merely by their failure to act, I think we should not  say our freedom was impaired. But the distinction here is clearly very  difficult to draw. To block my path limits my freedom. Not to clear  or repair it does not.   (5) ‘Action’. (a) I mean action here to include the credible threat  of action, since the most usual diminution of freedom is not by  physical constraint or violence but by the fear of it. (b) But I mean it,  though I am not sure I am right, to exclude deception. It seems clear  that our freedom is not impaired by the withholding of useful inform-  ation, and I am inclined to think not even by the giving of false  information. That wrongs us in some other way. I should say that  drugging a man, or (if that is possible) hypnotizing him against his    1Many women, e.g. resisted ‘emancipation’ from traditional domestic inferiority.  Their economic inferiority, which had only become conspicuous with newconditions,  was much more resented. Cf. Hume, Essays, H, xvi: ‘The bulk of mankind being  governed by authority, not reason, and never attributing authority to anything that has  not the recommendation of antiquity.” How dangerous this innate conservatism is he  shows two pages later: ‘Exorbitant power proceeds not, in any government, from new  laws, so much as from neglecting to remedy the abuses, which frequently rise from the  old ones.’ And cf. I, iv, ‘Antiquity always begets the opinion of right’.    136 E. F. CARRITT    will, impaired his freedom, but am inclined to think that propaganda  and excitement by rhetoric, music and similar tricks do not. To  prevent or forbid his access to contrary propaganda of course would  impair his freedom to obtain it if he wanted to do so.   If our definition, so explained, is accepted as the nearest we can  get to consistency without much violating common usage, two points  become clear.   (1) The first is that there are other good things, or other things  to which a man has claims and which it may be our duty to secure him,  besides liberty; for instance education, food, society, a good water-  supply.’ And it is possible that such claims might conflict with  the claim to liberty, and a compromise have to be struck. The writers  I have quoted seem to think equality is one of these things.   (2) The second point is that one man’s liberty is apt to be inimical  to his neighbour’s. I suppose the ideally free man outside a desert  island would be an irresponsible world-despot not even threatened  with assassination, but his freedom would almost certainly involve a  great deal of servitude for others. When therefore we say that men  have a right to freedom or that freedom is good (unless we mean  merely that each wants for himself all the freedom he can get) we can  only mean equal freedom. Indeed, if we use the language of natural  rights, the right to equality must be more fundamental than that to  liberty or life or anything else, since men cannot have absolute  rights to any of these things (for one man’s possession of them may  be incompatible with another’s) but only (ceteris paribus) equal clarms.?  Aristotle indeed identified justice (other than legal) with equality,  though an equality taking account of ‘desert’.? And justice (Recht)  is natural right. It is ‘the treatment of every man as an end’, ‘counting    ‘Maitland points out that Alexander Selkirk was completely free and very miserable.  1 may add that he might have a right to be rescued, but the moment he set foot on  ship his liberty would be diminished. He must obey the captain.   ?Equal claims to what is divisible (as liberty is) imply rights to equal shares,  e.g. to ten shillings in the pound where the assets are half the liabilities.   ‘An equal admission to the means of improvement and pleasure is a law  vigorously enjoined upon mankind by the voice of justice. All other changes in  society are good only as they are fragments of this or steps to its attainment. Godwin,  Political Justice, VIII, iv. Here the utilitarian joins hands with the adherent of natural  rights.   SEth. Nic. 129a 34, 130b 9, 131a 11, 158b 30, wat dfiav which I have rendered  ‘desert’ might I think include ‘need’ and be paraphrased ‘ceteris paribus’. Need is so  hard to assess that perhaps law should only attempt to assess it in relation to efficient  work. Does one man ‘need’ a first-rate execution of Bach but only a country-inn  parlour, and another a ‘luxury hotel’ but only a cinema-organ? : :    LIBERTY AND EQUALITY 137    every man for one’, an equality numerical till reason is shown to the  contrary. And, if for the moment we neglect other possibly conflicting  claims, the amount of freedom a man has a right to, the amount we  ought prima facie to secure him, is just so much as is compatible with  an equal amount for others. The maximum of freedom would be  obtained if men were never interfered with by others in doing anything  they chose except when what they chose to do interfered with others,  and if they then always were. One thing most people want to do which  can hardly affect the liberty of others is to express their opinions and  feelings. So, if we are merely considering maximum liberty, speech  should be free, even, I think, arguments for slavery or censorship.   It remains then to ask how far equal liberty is favoured or impaired  by equality in other respects.   (1) So far as ‘political equality’ goes, we have admitted that  majority government does not infallibly guarantee liberty any more  than any other form of government does. Empirically in most circum-  stances, certainly in modern civilization, it seems the most favourable  to it. At any rate our allegiance to a democratic or any other form of  government would seem to depend upon the degree in which it is likely  (or more likely than anything we could substitute) to secure men their  rights,’ prominent among which would be the right to equal liberty.   (2) ‘Equality before the law’, if, as I suppose, that only means law  effectively carried out and not arbitrarily perverted by caprice or  partiality, 1 is implied i in the very nature of law. And any system of law  giving some security of person and property is more favourable to  liberty than the anarchic state of ‘private war and scramble’. If    'Cf. Maitland, Liberty (Collected Papers, 1): ‘It is not possible to decide who  ought to govern until we know what a government ought to do.’  Cf. Hume, Essays, 1, v: ‘Government having ultimately no other object or purpose  i.e. justification? but the distribution of justice... obedience is a new duty  which must be invented to support that of justice.’ ‘No other’ is an exaggeration unless  justice is used in a very wide sense to include beneficence. A Sumerian king claims  fame as having given his people ‘equal justice and canals’ (Woolley, Abraham).    ?Known general laws, however bad, interfere less with freedom than decisions  based on no previously known rule.’ Maitland, op. cit. p. 81. In weighing the risks of  insecurity from innovation and of injustice from obsolescence, we may remember the  wise maxim of Hume that the breakdown of order would be the worse evil but the loss  of liberty is the more probable. Essays, I, v, vii. The best reasoned defence of anarchy  is perhaps Godwin, Political Justice, VII, viii, Of Law. Since ‘every case is a rule to  itself; it should be judged by pure equity, assessed by the unguided reason of the  judges. Presumably laws of conformity in indifferent matters, like the rule of the road,  would be allowed. Yet later (VIII, ii) Godwin says: ‘It is not easy to say whether  misery or absurdity would be most conspicuous in a plan which should invite every  man to seize upon everything he conceived himself to want’, and ‘Unless I can foresee,    138 E. F. CARRITT    “Equality before the law’ means not only that the rules are kept, but  also that they are made for the equal advantage of all whose needs or  deserts are equal (as it might be maintained most of our laws of murder  and assault are) then the question is raised how far this is also true of  our property laws. But I prefer to avoid the wider question whether  our property laws are just and confine myself to the question whether  the inequality which they protect and favour is, as has been suggested,  favourable to liberty. To decide the wider question we should have to  ascertain whether these now are (even if they once were) favourable to  the securing of all men’s other claims, such as those to improved  opportunities for health, education, enjoyment, as well as to the equal  distribution of the opportunities already available. And that might  involve us in economic considerations.   (3) What distribution of property then should be promoted and  protected by law if it is to secure men the maximum amount of doing  as they choose without interference? It is clear that all laws and all  taxes diminish, and are intended to diminish, somebody’s liberty,  frequently to the increase of general liberty, sometimes justifiably on  other grounds.’ A law which forbids me to appropriate what is in  my brother’s possession, if I want to do so, impairs my liberty as  truly as a law compelling me to give him half what is inmine. We may  not all covet our neighbour’s husband or wife, his ox or his ass. But  if we never covet his manservant or his maidservant, his leisure, job,  education, something that his money can buy, we are lucky.   What sort of property-distribution would produce the general  minimum of liberty as defined? Surely literal monopoly. Take an  extreme and simple case. If the total water supply of an island were  the legal property of one landlord and water-theft were a capital  offence, the rest of the population would desire more passionately  than anything else to do something which they were either prevented  from doing or could only do in fear of their lives. They would be  extremely unfree. No doubt the owner might be willing to sell at a  ‘reasonable’ rate, but so far as he had a monopoly also of other goods  he could not sell at a rate ‘reasonable’ to the purchasers. He would  have either to watch them perish, to sell water for labour, or to stand  drinks. Laws are not made good laws by being too absurd or inhuman  for enforcement.    in a considerable degree, the treatment I shall receive from my species . . . [can engage  in no valuable undertaking’.   “No law can be made that does not take something from liberty.’ Bentham;  Anarchical Fallacies. Preamble.    LIBERTY AND EQUALITY 139    Now just in proportion as the ownership of water were eq ualized,  the prohibition of water-theft would become less burdensome, less  obstructive of what each desired to do, even though still nobody  had so much as he would have liked. The only loss of liberty would be  in the original monopolist, and probably his loss would not be so  great as the gain of any one of his neighbours, since he could hardly  have desired to use his superfluity of water (say in watering orchids)  so passionately as the other had desired to moisten his tongue. To be  forcibly deprived of superabundance or even of conveniences impairs  liberty less than to be forcibly prevented from appropriating  necessities.!   If then we consider laws and institutions of property merely so  far as they directly affect liberty I think we must conclude that those  are most favourable to it which most favour equality in proportion to  need. Against such equality there may of course be other reasons.   There remains to notice the obvious relation of economic equality  with political equality and equality before the law. Clearly, without  freedom of speech, discussion, and information, the bare possession  of the vote is almost valueless, and great economic inequality gives  influence and power of propaganda which are as destructive of any  real equality of political power as a censorship itself. Even  ‘equality before the law’, that is legal justice itself, is endangered  by economic inequality in well-known ways. The expense of expert  legal advice and of protracted legislation heavily handicaps the poor.  From great economic inequalities rise class differences of education,  speech, standard of life, which may make it very difficult for judges to  sympathize with some of those who come before them.?   Those who contend that liberty and equality are incompatible  inherit Burke’s naive conservatism, the belief that the present  British social system is ideal and merely needs to be ideally admin-  istered:—the machine is perfect, if only we could eliminate friction,  but to plana machine with less friction is utopian. The liberty they  praise is a liberty within that system, with just the present institutions  of ownership, inheritance, taxation, combination, limited liability,  banking, all compulsory. The equality they condemn is any alteration  of that system which would secure a greater amount of liberty to a  greater number of persons; within the sacred system laisser-faire is    ‘Hume, Enquiry, Mil: ‘Whenever we depart from equality we rob the poor of  more satisfaction |and liberty] than we add to the rich’; and Essays, 1, vii: ‘Property  when united causes much greater dependence than the same property when dispersed.’   2And cf. Bentham, Book of Fallacies, 1: Appendix on ‘Sinister interest of Lawyers’.    140 E. F. CARRITT    divinely guided; but if we do not enforce just that system, providence  they think will lead us to ruin.   But ‘economics’ and ‘politics’ cannot be thus separated. We must  follow the same principle in judging what administration of the lawis  just and in judging what laws are just. If men have a right to liberty  and equality within the law, for the same reason they have a right to  laws that promote liberty and equality. Harrington in his Oceana  said that ‘Equality of estates causes equality of power, and equality  of power is liberty’, and Maitland (Equality, Collected Papers 11) adds  the rider that equality of political power tends to produce equality  of property. And that tends to produce liberty. Godwin! briefly  stated a position sometimes attributed to later writers: ‘It is only by  means of accumulation that one man obtains an unresisted sway over  multitudes of others. It is by means ofa certain distribution of income  that the present governments of the world are retained in existence.  Nothing more easy than to plunge nations, so organized, into war.’  Godwin really believed in both liberty and equality. His peculiarity  is that he could also believe in Jaisser-faire because he believed in  fraternity.    ' Political Justice, VIII, iii (1793).    Vil ~  TWO CONCEPTS OF LIBERTY’    SiR ISAIAH BERLIN    TO coerce a man is to deprive him of freedom—freedom from what?  Almost every moralist in human history has praised freedom. Like  happiness and goodness, like nature and reality, the meaning of this  term is so porous that there is little interpretation that it seems able  to resist. | do not propose to discuss either the history or the more  than two hundred senses of this protean word, recorded by historians  of ideas. I propose to examine no more than two of these senses—but  those central ones, with a great deal of human history behind them,  and, I dare say, still to come. The first of these political senses of  freedom or liberty (I shall use both words to mean the same), which  (following much precedent) I shall call the ‘negative’ sense, is involved  in the answer to the question “What is the area within which the  subject—a person or group of persons—is or should be left to do or be  what he is able to do or be, without interference by other persons?’  The second, which I shall call the positive sense, is involved in the  answer to the question ‘What, or who, is the source of control or  interference, that can determine someone to do, or be, one thing  rather than another?’ The two questions are clearly different, even  though the answers to them may overlap.    The notion of ‘negative’ freedom  Iam normally said to be free to the degree to which no human being  interferes with my activity. Political liberty in this sense is simply the  area within which a man can act unobstructed by others. If I am  prevented by other persons from doing what I could otherwise do, I    From the revised version of his Inaugural Lecture “Two Concepts of Liberty’  (Clarendon Press, 1958, pp. 6-19) in Four Essays on Liberty by Sir Isaiah Berlin  to be published as an Oxford University Press Paperback. Reprinted by permission  of the author and the Clarendon Press.    , ‘ine pages of the beginning and end of the original text have here been omitted.  Ed.    142 SIR ISAIAH BERLIN    am to that degree unfree; and if this area is contracted by other men  beyond a certain minimum, I can be described as being coerced, or, it  may be, enslaved. Coercion is not, however, a term that covers every  form of inability. If I say that I am unable to jump more than 10 feet  in the air, or cannot read because I am blind, or cannot understand  the darker pages of Hegel, it would be eccentric to say that lam to that  degree enslaved or coerced. Coercion. implies the deliberate inter-  ference of other human beings within the area in which I could other-  wise act. You lack political liberty or freedom only if you are prevented  from attaining a goal by human beings.' Mere incapacity to attaina  goal is not lack of political freedom. This is brought out by the use  of such modern expressions as ‘economic freedom’ and its counter-  part, ‘economic slavery’. It is argued, very plausibly, that ifa man is  too poor to afford something on which there is no legal ban—a loaf  of bread, a journey round the world, recourse to the law courts—he  is as little free to have it as he would be if it were forbidden him bylaw.  If my poverty were a kind of disease, which prevented me from buying  bread or paying for the journey round the world, or getting my case  heard, as lameness prevents me from running, this inability would not  naturally be described as a lack of freedom, least of all political  freedom. It is only because I believe that my inability to get a given  thing is due to the fact that other human beings have made arrange-  ments whereby I am, whereas others are not, prevented from having  enough money with which to pay for it, that I think myself a victim of  coercion or slavery. In other words, this use of the term depends on  a particular social and economic theory about the causes of my  poverty or weakness. If my lack of material means is due to my lack  of mental or physical capacity, then I begin to speak of being deprived  of freedom (and not simply of poverty) only if I accept the theory.  If, in addition, I believe that 1 am being kept in want by a specific  arrangement which | consider unjust or unfair, I speak of economic  slavery or oppression. ‘The nature of things does not maddenus, only  ill will does’, said Rousseau. The criterion of oppression is the part  that I believe to be played by other human beings, directly or    1T do not, of course, mean to imply the truth of the converse.   ?Helvétius made this point very clearly: ‘The free man is the man who is not in  irons, nor imprisoned in a gaol, nor terrorized like a slave by the fear of punish-  ment. .. it is not lack of freedom not to fly like an eagle or swim like a whale.’   3The Marxist conception of social laws is, of course, the best-known version  of this theory, but it forms a large element in some Christian and utilitarian, and  all socialist, doctrines.    TWO CONCEPTS OF LIBERTY 143    indirectly, with or without the intention of doing so, in frustrating  my wishes. By being free in this sense I mean not being interfered with  by others. The wider the area of non-interference the wider my  freedom.   This is what the classical English political philosophers meant  when they used this word.! They disagreed about how wide the area  could or should be. They supposed that it could not, as things were,  be unlimited, because if it were, it would entail a state in which all  men could boundlessly interfere with all other men; and this kind of  ‘natural’ freedom would lead to social chaos in which men’s minimum  needs would not be satisfied; or else the liberties of the weak would be  suppressed by the strong. Because they perceived that human  purposes and activities do not automatically harmonize with one  another; and, because (whatever their official doctrines) they put  high value on other goals, such as justice, or happiness, or culture,  or security, or varying degrees of equality, they were prepared to  curtail freedom in the interests of other values and, indeed, of freedom  itself. For, without this, it was impossible to create the kind of  association that they thought desirable. Consequently, it is assumed by  these thinkers that the area of men’s free action must be limited by  law. But equally it is assumed, especially by such libertarians as  Locke and Mill in England, and Constant and Tocqueville in France,  that there ought to exist a certain minimum area of personal freedom  which must on no account be violated; for if it is overstepped, the  individual will find himself in an area too narrow for even that  minimum development of his natural faculties which alone makes it  possible to pursue, and even to conceive, the various ends which men  hold good or right or sacred. It follows that a frontier must be drawn  between the area of private life and that of public authority. Where it  is to be drawn is a matter of argument, indeed of haggling. Men are  largely interdependent, and no man’s activity is so completely private  as never to obstruct the lives of others in any way. ‘Freedom for the  pike is death for the minnows’ ; the liberty of some must depend on the  restraint of others.? Still, a practical compromise has to be found.    1*A free man’, said Hobbes, ‘is he that...is not hindered to do what he hath  the will to do.’ Law is always a ‘fetter’, even if it protects you from being bound  in chains that are heavier than those of the law, say, arbitrary despotism or chaos.  Bentham says much the same.    2‘Freedom for an Oxford don’, others have been known to add, ‘is a very dif-  ferent thing from freedom for an Egyptian peasant.’   This proposition derives its force from something that is both true and important,  but the phrase itself remains a piece of political claptrap. It is true that to offer    144 SIR ISAIAH BERLIN    Philosophers with an optimistic view of human nature, and a beliet  in the possibility of harmonizing human interests, such as Locke or  Adam Smith and, in some moods, Mill, believed that social harmony  and progress were compatible with reserving a large area for private  life over which neither the state nor any other authority must be    olitical rights, or safeguards against intervention by the state, to men who are  palf-naked. illiterate, underfed, and diseased is to mock their condition; they need  medical help or education before they can understand, or make use of, an increase  in their freedom. What is freedom to those who cannot make use of it? Without  adequate conditions of freedom what is the value of freedom? First things come  first: there are situations, as a nineteenth-century Russian radical writer declared,  in which boots are superior to the works of Shakespeare; individual freedom is not  everyone’s primary need. For freedom is not the mere absence of frustration of what-  ever kind; this would inflate the meaning of the word until it meant too much or too  little. The Egyptian peasant needs clothes or medicine before, and more than,  personal liberty, but the minimum freedom that he needs today, and the greater  degree of freedom that he may need tomorrow, is not some species of freedom  peculiar to him, but identical with that of professors, artists, and millionaires.   What troubles the consciences of Western liberals is not, I think, the belief that  the freedom that men seek differs according to their social or economic conditions,    but that the minority who possess it have gained it by exploiting or, at least, averting  their gaze from the vast majority who do not. They believe, with good reason,  that if individual liberty is an ultimate end for human beings, none should be de-  prived of it by others; least of all that some should enjoy it at the expense of  others. Equality of liberty; not to treat others as I should not wish them to treat  me; repayment of my debt to those who alone have made possible my liberty or pros-  pertty or enlightenment; justice, in its simplest and most universal sense—these  are the foundations of liberal morality. Liberty is not the only goal of men. I can,  like the Russian critic Belinsky, say that if others are to be deprived of it—f my  brothers are to remain in poverty, squalor, and chains—then I do not want it for  myself, I reject it with both hands and infinitely prefer to share their fate. But  nothing is gained by a confusion of terms. To avoid glaring inequality or wide-  spread misery I am ready to sacrifice some, or all, of my freedom: I spe do so  willingly and freely: but it is freedom that I am giving up for the sake of justice  or equality or the love of my fellow men. I should be guilt-stricken, and rightly  so, if I were not, in some circumstances, ready to make this sacrifice. But a sacrifice  is not an increase in what is being sacrificed, namely freedom, however great the  moral need or the compensation for it. Everything is what it is: liberty is liberty,  not equality or fairness or justice or culture or human happiness or a quiet con-  science. If the liberty of myself or my class or nation depends on the misery of a  number of other human beings, the system which promotes this is unjust and  immoral. But if I curtail or lose my freedom, in order to lessen the a eels of  such inequality, and do not thereby materially increase the individual liberty of  others, an absolute loss of liberty occurs. This may be compensated for by a gain in  Justice or in happiness or in peace, but the loss remains, and it is a confusion of  values to say that although my ‘liberal’, individual freedom may go by the board, some  other kind of freedom—‘social’ or ‘economic’—is increased. Yet it remains true  that the freedom of some must at times be curtailed to secure the freedom of others.  Upon what principle should this be done? If freedom is a sacred, untouchable  value, there can be no such principle. One or other of these conflicting principles  must at any rate in practice yells not always for reasons which can be dearly  stated, let alone generalized into rules or universal maxims.    TWO CONCEPTS OF LIBERTY 145    allowed to trespass, Hobbes, and those who agreed with him,  especially conservative or reactionary thinkers, argued that if men  were to be prevented from destroying one another, and making social  life a jungle or a wilderness, greater safeguards must be instituted to  keep them in their places, and wished correspondingly to increase the  area of centralized control, and decrease that of the individual. But  both sides agreed that some portion of human existence must remain  independent of the sphere of social control. To invade that preserve,  however small, would be despotism. The most eloquent of all  defenders of freedom and privacy, Benjamin Constant, who had not  forgotten the Jacobin dictatorship, declared that at the very least  the liberty of religion, opinion, expression, property, must be guaran-  teed against arbitrary invasion. Jefferson, Burke, Paine, Mill,  compiled different catalogues of individual liberties, but the argument  for keeping authority at bay is always substantially the same. We must  preserve a minimum area of personal freedom if we are not to  ‘degrade or deny our nature’. We cannot remain absolutely free, and  must give up some of our liberty to preserve the rest. But total self-  surrender is self-defeating. What then must the minimum be? That  which a man cannot give up without offending against the essence of  his human nature. What is this essence? Whatare the standards which  it entails? This has been, and perhaps always will be, a matter of  infinite debate. But whatever the principle in terms of which the area  of non-interference is to be drawn, whether it is that of natural law  or natural rights, or of utility or the pronouncements of a categorical  imperative, or the sanctity of the social contract, or any other  concept with which men have sought to clarify and justify their  convictions, liberty in this sense means liberty from; absence of inter-  ference beyond the shifting, but always recognizable, frontier. “The  only freedom which deserves the name is that of pursuing our own  good in our own way ’, said the most celebrated of its champions. If this  is so, is compulsion ever justified? Mill had no doubt that it was. Since  justice demands that all individuals be entitled to a minimum of  freedom, all other individuals were of necessity to be restrained, if  need be by force, from depriving anyone of it. Indeed, the whole  function of law was the prevention of just such collisions: the state  was reduced to what Lassalle contemptuously described as the func-  tions of a nightwatchman or traffic policeman.   What made the protection of individual liberty so sacred to Mill?  In his famous essay he declares that unless men are left to live as  they wish ‘in the path which merely concerns themselves’, civilization    146 SIR ISAIAH BERLIN    cannot advance; the truth will not, for lack of a free market in ideas,  come to light; there will be no scope for spontaneity, originality,  genius, for mental energy, for moral courage. Society will be crushed  by the weight of ‘collective mediocrity’. Whatever is rich and diver-  sified will be crushed by the weight of custom, by men’s constant  tendency to conformity, which breeds only ‘withered capacities’,  ‘pinched and hidebound’, ‘cramped and warped’ human beings.  ‘Pagan self-assertion is as worthy as Christian self-denial.’ ‘All  the errors which a man is likely to commit against advice and warning  are far outweighed by the evil of allowing others to constrain him  to what they deem is good.’ The defence of liberty consists in the  ‘negative’ goal of warding off interference. To threaten a man with  persecution unless he submits to a life in which he exercises no choices  of his goals; to block before him every door but one, no matter how  noble the prospect upon which it opens, or how benevolent the  motives of those who arrange this, is to sin against the truth that he  is.a man, a being with a life of his own to live. This is liberty as it  has been conceived by liberals in the modern world from the days of  Erasmus (some would say of Occam) to our own. Every plea for civil  liberties and individual rights, every protest against exploitation and  humiliation, against the encroachment of public authority, or the  mass hypnosis of custom or organized propaganda, springs from this  individualistic, and much disputed, conception of man.   Three facts about this position may be noted. In the first place Mill  confuses two distinct notions. One is that all coercion is, in so far  as it frustrates human desires, bad as such, although it may have to be  applied to prevent other, greater evils; while non-interference,  which is the opposite of coercion, is good as such, although it is not  the only good. This is the ‘negative’ conception of liberty in its  classical form./The other is that men should seek to discover the  truth, or to develop ‘a certain type of character of which Mill  approved—fearless, original, imaginative, independent, non-  conforming to the point of etteiitiicity, and so on—and that truth  can be found, and such character can be bred, only in conditions of  freedom. Both these are liberal views, but they are not identical, and  the connexion between them is, at best, empirical. No one would  argue that truth or freedom of self-expression could flourish where  dogma crushes all thought. But the evidence of history tends to show  (as, indeed, was argued by James Stephen in his formidable attack on  Mill in his Liberty, Equality, Fraternity) that integrity,-love of truth  and fiery individualism grow at least as often in severely disciplined    fave    one se    om    TWO CONCEPTS OF LIBERTY 147    communities among, for example, the puritan Calvinists of Scotland  or New England, or under military discipline, as in more tolerant or  indifferent societies; and if this is so accepted, Mill’s argument for  liberty as a necessary condition for the growth of human genius falls  to the ground. If his two goals proved incompatible, Mill would be  faced with a cruel dilemma, quite apart from the further difficulties  created by the inconsistency of his doctrines with strict utilitarianism,  even in his own humane version of it.   In the second place, the doctrine is comparatively modern. There   seems to be scarcely any discussion of individual liberty as a conscious  political ideal (as opposed to its actual existence) in the ancient world.  Condorcet has already remarked that the notion of individual rights is  absent from the legal conceptions of the Romans and Greeks; this  seems to hold equally of the Jewish, Chinese, and all other ancient  civilizations that have since come to light.? The domination of this  ideal has been the exception rather than the rule, even in the recent  history of the West. Nor has liberty in this sense often formed a  rallying cry for the great masses of mankind. The desire not to be  impinged upon, to be left to oneself, has been a mark of high civiliza-  tion both on the part of individuals and communities. The sense of  privacy itself, of the area of personal relationships as something  sacred in its own right, derives from a conception of freedom which,  for all its religious roots, is scarcely older, in its developed state,  than the Renaissance or the Reformation.’ Yet its decline would mark  the death of a civilization, of an entire moral outlook.   The third characteristic of this notion of liberty is of greater  importance. It is that liberty in this sense is not incompatible with  some kinds of autocracy, or at any rate with the absence of self-  government. Liberty in this sense is principally concerned with the  area of control, not with its source. Just as a democracy may, in fact,    ‘This is but another illustration of the natural tendency of all but a very few  thinkers to believe that all the things they hold good must be intimately connected, or  at least compatible, with one another. The history of thought, like the history of  nations, is strewn with examples of inconsistent, or at least disparate, elements  artificially yoked together in a despotic system, or held together by the danger  of some common enemy. In due course the danger passes, and conflicts between the  allies arise, which often disrupt the system, sometimes to the great benefit of mankind.    2See the valuable discussion of this in Michel Villey, Lecons d’Histotre de la Philosophie  du Droit, who traces the embryo of the notion of subjective rights to Occam.   Christian (and Jewish or Moslem) belief in the absohute authority of divine or  natural laws, or in the equality of all men in the sight of God, is very different from  belief in freedom to live as one prefers.    (ane!  (2-6)    148 SIR ISAIAH BERLIN    deprive the individual citizen of a great many liberties which he might  have in some other form of society, so it is perfectly conceivable that  a liberal-minded despot would allow his subjects a large measure of  personal freedom. The despot who leaves his subjects a wide area of  liberty may be unjust, or encourage the wildest inequalities, care  little for order, or virtue, or knowledge; but provided he does not  curb their liberty, or at least curbs it less than many other régimes, he  meets with Mill’s specification.'! Freedom in this sense is not, at any  rate logically, connected with democracy or self-government. Self-  government may, on the whole, provide a better guarantee of the  preservation of civil liberties than other régimes, and has been  defended as such by libertarians. But there is no necessary connexion  between individual liberty and democratic rule. The answer to the  question “Who governs me?’ is logically distinct from the question  ‘How far does government interfere with me?’ It is in this difference  that the great contrast between the two concepts of negative and  positive liberty, in the end, consists.! For the ‘positive’ sense of  liberty comes to light if we try to answer the question, not ‘Whatam I  free to do or be?’, but ‘By whom am I ruled?’ or ‘Whois to say what I  am, and what I am not, to be or do?’ The connexion between demo-  cracy and individual liberty is a good deal more tenuous than it seemed  to many advocates of both. The desire to be governed by myself, or at  any rate to participate in the process by which my life is to be control-  led, may be as deep a wish as that of a free area for action, and perhaps  historically older. But it is not a desire for the same thing. So different  is it, indeed, as to have led in the end to the great clash of ideologies  that dominates our world. For it is this—the ‘positive’ conception of  liberty: not freedom from, but freedom to—which the adherents of  the ‘negative’ notion represent as being, at times, no better than a  specious disguise for brutal tyranny.    ‘Indeed, it is arguable that in the Prussia of Frederick the Great or in the Austria  of Josef IJ, men of imagination, originality, and creative genius, and, indeed, minorities  of all kinds, were less persecuted and felt the pressure, both of institutions and  custom, less heavy upon them than in many an earlier or later democracy.    *‘Negative liberty’ is something the extent of which, in a given case, it is difficult  to estimate. It might, prima facie, seem to depend nape on the power to choose  between at any rate two alternatives. Nevertheless, not all choices are equally free,  or free at all. If in a totalitarian state I betray my friend under threat of torture,  perhaps even if I act from fear of losing my job, I can reasonably say that I did not  act freely. Nevertheless, I did, of course, make a choice, and could, at any rate in  theory, have chosen to be killed or tortured or imprisoned. The mere existence of  alternatives is not, therefore, enough to make my action free (although it may be    TWO CONCEPTS OF LIBERTY 149  II    The Notion of Positive Freedom   The ‘positive’ sense of the word ‘liberty’ derives from the wish on the  part of the individual to be his own master. I wish my life and decisions  to depend on myself, not on external forces of whatever kind. I wish to  be the instrument of my own, not of other men’s, acts of will. I wish to  be a subject, not an object; to be moved by reasons, by conscious  purposes which are my own, not by causes which affect me, as it were,  from outside. I wish to be somebody, not nobody; a doer—deciding,  not being decided for, self-directed and not acted upon by external  nature or by other men as if I were a thing, or an animal, or a slave  incapable of playing a human role, that is, of conceiving goals and  policies of my own and realizing them. This is at least part of what I  mean when I say that I am rational, and that it is my reason that  distinguishes me as a human being from the rest of the world. I wish,  above all, to be conscious of myself as a thinking, willing, active  being, bearing responsibility for his choices and able to explain them  by reference to his own ideas and purposes. I feel free to the degree that  I believe this to be true, and enslaved to the degree that I am made to  realize that it is not.   The freedom which consists in being one’s own master, and the  freedom which consists in not being prevented from choosing as I do    voluntary) in the normal sense of the word. The extent of my freedom seems to  depend on (a) how many possibilities are open to me (although the method of  counting these can never be more than impressionistic. Possibilities of action are  not discrete entities like apples, which can be exhaustively enumerated); (b) how  easy or difficult each of these possibilities is to actualize; (c) how important in my  plan of life, given my character and circumstances, these possibilities are when com-  pared with each other; (d) how far they are closed and opened by deliberate human  acts; (e) what value not merely the agent, but the general sentiment of the society in  which he lives, puts on the various possibilities. All these magnitudes must be ‘in-  tegrated’, and a conclusion, necessarily never precise, or indisputable, drawn from  this process. It may well be that there are many incommensurable degrees of freedom,  and that they cannot be drawn up on a single scale of magnitude, however conceived.  Moreover, in the case of societies, we are faced by such (logically absurd) questions  as ‘Would arrangement X increase the liberty of Mr. A more than it would that of  Messrs. B, C, and D between them, added together?’ The same difficulties arise in  applying utilitarian criteria. Nevertheless, provided we do not demand precise  measurement, we can give valid reasons for saying that the average subject of the  King of Sweden is, on the whole, a good deal freer today than the average citizen of  the Republic of Rumania. Total patterns of life must be compared directly as wholes,  although the method by which we make the comparison, and the truth of the con-  clusions, are difficult or impossible to demonstrate. But the vagueness of the concepts,  and the multiplicity of the criteria involved, is an attribute of the subject-matter  itself, not of our imperfect methods of measurement, or incapacity for precise thought.    150 SIR ISAIAH BERLIN    by other men, may, on the face of it, seem concepts at no great logical  distance from each other—no more than negative and positive ways  of saying the same thing. Yet the ‘positive’ and ‘negative’ notions of  freedom historically developed in divergent directions not always by  logically reputable steps, until, in the end, they came into direct  conflict with each other.   One way of making this clear is in terms of the independent  momentum which the, initially perhaps quite harmless, metaphor of  self-mastery acquired. ‘I am my own master’; ‘I am slave to noman’;  but may I not (as, for instance, T. H. Green is always saying) be a  slave to nature? Or to my own ‘unbridled’ passions? Are these not so  many species of the identical genus ‘slave’-—some political or legal,  others moral or spiritual? Have not men had the experience of  liberating themselves from spiritual slavery, or slavery to nature, and  do they not in the course of it become aware, on the one hand, ofa  self which dominates, and, on the other, of something in them which is  brought to heel? This dominant self is then variously identified with  reason, with my ‘higher nature’, with the self which calculates and aims  at what will satisfy it in the long run, with my ‘real’, or ‘ideal’, or  ‘autonomous’ self, or with my self ‘at its best’; which is then contrasted  with irrational impulse, uncontrolled desires, my ‘lower’ nature, the  pursuit of immediate pleasures, my ‘empirical’ or ‘heteronomous’  self, swept by every gust of desire and passion, needing to be rigidly  disciplined if it is ever to rise to the full height of its ‘real’ nature.  Presently the two selves may be represented as divided by an even  larger gap: the real self may be conceived as something wider than the  individual (as the term is normally understood), as a social ‘whole’  of which the individual is an element or aspect: a tribe, a race, a church,  a state, the great society of the living and the dead and the yet unborn.  This entity is then identified as being the ‘true’ self which, by imposing  its collective, or ‘organic’, single will upon its recalcitrant ‘members’,  achieves its own, and, therefore, their, ‘higher’ freedom. The perils  of using organic metaphors to justify the coercion of some men by  others in order to raise them to a ‘higher’ level of freedom have often  been pointed out. But what gives such plausibility as it has to this  kind of language is that we recognize that it is possible, and at times  justifiable, to coerce men in the name of some goal (let us say, justice  or public health) which they would, if they were more enlightened,  themselves pursue, but do not, because they are blind or ignorant or  corrupt. This renders it easy for me to conceive of myself as coercing  others for their own sake, in their, not my, interest. I am then    TWO CONCEPTS OF LIBERTY 151    claiming that I know what they truly need better than they know it  themselves. What, at most, this entails is that they would not resist  me if they were rational, andas wise as I, and understood their interests  as I do. But I may go on to claim a good deal more than this. I may  declare that they are actually aiming at what in their benighted state  they consciously resist, because there exists within them an occult  enuty—their latent rational will, or their ‘true’ purpose—and that  this entity, although it is belied by all that they overtly feel and do and  say, is their ‘real’ self, of which the poor empirical self in space and  time may know nothing or little; and that this inner spirit is the only  self that deserves to have its wishes taken into account.! Once I take  this view, I am in a position to ignore the actual wishes of men or  societies, to bully, oppress, torture them in the name, and on behalf,  of their ‘real’ selves, in the secure knowledge that whatever is the true  goal of man (happiness, fulfilment of duty, wisdom, a just society,  self-fulfilment) must be identical with his freedom—the free choice of  his ‘true’, albeit submerged and inarticulate, self.   This paradox has been often exposed. It is one thing to say that I  know what is good for X, while he himself does not; and even to ignore  his wishes for its—and his—sake; and a very different one to say that.  he has eo ipso chosen it, not indeed consciously, not as he seems in  everyday life, but in his role as a rational self which his empirical  self may not know—the ‘real’ self which discerns the good, and cannot  help choosing it once it is revealed. This monstrous impersonation,  which consists in equating what X would choose ifhe were something he  is not, or at least not yet, with what X actually seeks and chooses, is  at the heart of all political theories of self-realization. It is one thing  to say that I may be coerced for my own good which I am too blind  to see: this may, on occasion, be for my benefit; indeed it may enlarge  the scope of my liberty; it is another to say that if it is my good,  then I am not being coerced, for I have willed it, whether I know this  or not, and am free—or ‘truly’ free—even while my poor earthly body  and foolish mind bitterly reject it, and struggle against those who seek  however benevolently to impose it, with the greatest desperation.   This magical transformation, or sleight of hand (for which William    “The ideal of true freedom is the maximum of power for all the members of  human society alive to make the best of themselves’, said T. H. Green in 1881.  Apart from the confusion of freedom with equality, this entails that if a man chose  some immediate pleasure—which (in whose view?) would not enable him to make  the best of himself (what self?) what he is exercising is not ‘true’ freedom: and, if  deprived of it, he would not lose anything that mattered. Green was a genuine liberal:  but many a tyrant could use this formula to justify his worst oppression.    152 SIR ISAIAH BERLIN    James so justly mocked the Hegelians), can no doubt be perpetrated  just as easily with the ‘negative’ concept of freedom, where the self  that should not be interfered with is no longer the individual with  his actual wishes and needs as they are normally conceived, but the  ‘real’ man within, identified with the pursuit of some ideal purpose  not dreamed of by his empirical self. And, as in the case of the  ‘positively’ free self, this entity may be inflated into some super-  personal entity—a state, a class, a nation, or the march of history  itself, regarded as a more ‘real’ subject of attributes than the empirical  self. But the ‘positive’ conception of freedom as self-mastery, with its  suggestion of a man divided against himself, has, in fact, and as a  matter of the history of doctrines and of practice, lent itself more  easily to this splitting of personality into two: the transcendent,  dominant controller, and the: empirical bundle of desires and  passions to be disciplined and brought to heel. This demonstrates (if  demonstration of so obvious a truth is needed) that the conception of  freedom directly derives from the view that is taken of what constitutes  a self, a person, a man. Enough manipulation with the definition of  man, and freedom can be made to mean whatever the manipulator  wishes. Recent history has made it only too clear that the issue is not  merely academic.    IX  TWO CONCEPTS OF DEMOCRACY    JOSEPH SCHUMPETER  1-THE CLASSICAL DOCTRINE OF DEMOCRACY  1. THE COMMON GOOD AND THE WILL OF THE PEOPLE    THE eighteenth-century philosophy of democracy may be couched in  the following definition: the democratic method is that institutional  arrangement for arriving at political decisions which realizes the  common good by making the people itself decide issues through the  election of individuals who are to assemble in order to carry out its  will. Let us develop the implications of this.   It is held, then, that there exists a Common Good, the obvious  beacon light of policy, which is always simple to define and which every  normal person can be made to see by means of rational argument.  There is hence no excuse for not seeing it and in fact no explanation  for the presence of people who do not see it except ignorance—which  can be removed—stupidity and anti-social interest. Moreover, this  common good implies definite answers to all questions so that every  social fact and every measure taken or to be taken can unequivocally  be classed as ‘good’ or ‘bad’. All people having therefore to agree,  in principle at least, there is also a Common Will of the people  (= will of all reasonable individuals) that is exactly coterminous with  the common good or interest or welfare or happiness. The only thing,  barring stupidity and sinister interests, that can possibly bring in  disagreement and account for the presence of an opposition is a  difference of opinion as to the speed with which the goal, itself  common to nearly all, is to be approached. Thus every member of the  community, conscious of that goal, knowing his or her mind, dis-  cerning what is good and what is bad, takes part, actively and  responsibly, in furthering the former and fighting the latter and all  the members taken together control their public affairs.    From Capitalism, Socialism and Democracy by Joseph Schumpeter (grd edn., Allen  & Unwin, 1950), pp. 250-83. Copyright 1942, 1947, by Joseph A. Schumpeter.  Copyright 1950 by peel and Brothers. Reprinted by permission of George Allen and  Unwin and Harper and Row, Publishers.    154 JOSEPH SCHUMPETER    It is true that the management of some of these affairs requires  special aptitudes and techniques and will therefore have to be  entrusted to specialists who have them. This does not affect the  principle, however, because these specialists simply act in order to  carry out the will of the people exactly as a doctor acts in order to  carry out the will of the patient to get well. It is also true that ina  community of any size, especially if it displays the phenomenon of  division of labour, it would be highly inconvenient for every individual  citizen to have to get into contact with all the other citizens on every  issue in order to do his part in ruling or governing. It will be more  convenient to reserve only the most important decisions for the indivi-  dual citizens to pronounce upon—say by referendum—and to deal  with the rest through a committee appointed by them—an assembly or  parliament whose members will be elected by popular vote. This  committee or body of delegates, as we have seen, will not represent  the people in a legal sense but it will do so in a less technical one—it  will voice, reflect or represent the will of the electorate. Again as a  matter of convenience, this committee, being large, may resolve itself  into smaller ones for the various departments of public affairs. Finally,  among these smaller committees there will be a general-purpose  committee, mainly for dealing with current administration, called  cabinet or government, possibly with a general secretary or scapegoat  at its head, a so-called prime minister."   As soon as we accept all the assumptions that are being made by  this theory of the polity—or implied by it—democracy indeed acquires  a perfectly unambiguous meaning and there is no problem in conne-  xion with it except how to bring itabout. Moreover we need only forget  a few logical qualms in order to be able to add that in this case the  democratic arrangement would not only be the best ofall conceivable  ones, but that few people would care to consider any other. It is no  less obvious however that these assumptions are so many statements of  fact every one of which would have to be proved if we are to arrive at  that conclusion. And it is much easier to disprove them.   [There is, first, no such thing as a uniquely determined common  ood that all people could agree on or be made to agree on by the  force of rational argument] This is due not primarily to the fact that  some people may want thirigs other than the common good but to the  much more fundamental fact that to different individuals and groups    'The official theory of the functions of a cabinet minister holds in fact that he is  appointed in order to see to it that in his department the will of the people prevails.    Y    (3)    TWO CONCEPTS OF DEMOCRACY 155    the common good is bound to mean different things. This fact, hidden  from the utilitarian by the narrowness of his outlook on the world  of human valuations, will introduce rifts on questions of principle  which cannot be reconciled by rational argument because ultimate,    value ur. conceptions of what life and what society should be—are  beyond the range of mere logic. They may be bridged by compro-    mise in some cases but notin others. Americans who say, “We want this  country to arm to its teeth and then to fight for what we conceive  to be right all over the globe’ and Americans who say, ‘We want this  country to work outits own problems which is the only way it can serve  humanity’ are facing irreducible differences of ultimate values  which compromise could only maim and degrade.   Secondly, even if a sufficiently definite common good—such as for  instance the utilitarian’s maximum of economic satisfaction'—proved  acceptable to all, this would not imply equally definite answers to  individual issues. Opinions on these might differ to an extent  important enough to produce most of the effects of ‘fundamental’  dissension about ends themselves. The problems centring in the  evaluation of present versus future satisfactions, even the case of  socialism versus capitalism, would be left still open, for instance,  after the conversion of every individual citizen to utilitarianism.  ‘Health’ might be desired by all, yet people would still disagree on  vaccination and vasectomy. And so on.   The utilitarian fathers of democratic doctrine failed to see the  full importance of this simply because none of them seriously con-  sidered any substantial change in the economic framework and the  habits of bourgeois society. They saw little beyond the world of an  eighteenth-century ironmonger.   But, third, as a consequence of both preceding propositions, the  particular concept of the will of the poops or the volonté générale  that the utilitarians made their own vanishes into thin air. For that  concept presupposes the existence ofa uniquely determined common  good discernible to all. Unlike the romanticists the utilitarians had no  notion of that semi-mvystic entity endowed with a will of its own—that  ‘soul of the people’ which the historical school of jurisprudence made  so much of. They frankly derived their will of the people from the    ‘The very meaning of ‘greatest happiness’ is open to serious doubt. But even if this  doubt could be removed and definite meaning could be attached to the sum total of  economic satisfaction of a group of people, that maximum would still be relative to  geven situations and valuations which it may be impossible to alter, or compromise  on, in a democratic way.    156 JOSEPH SCHUMPETER    wills of individuals. And unless there is a centre, the common good,  toward which, in the long run at least, a// individual wills gravitate,  we shall not get that particular tvpe of ‘natural’ volonté generale.  The utilitarian centre of gravity, on the one hand, unifies individual  wills, tends to weld them by means of rational discussion into the  will of the people and, on the other hand, confers upon the latter the  exclusive ethical dignity claimed by the classic democratic creed.  This creed does not consist simply in worshipping the will of the people as such  but rests on certain assumptions about the ‘natural’ object of that will  which object is sanctioned by utilitarian reason. Both the existence and  the dignity of this kind of volonté générale are gone as soon as the  idea of the common good fails us. And both the pillars of the classical  doctrine inevitably crumble into dust.    IJ. THE WILL OF THE PEOPLE AND INDIVIDUAL VOLITION    Of course, however conclusively those arguments may tell against  this particular conception of the will of the people, they do not debar  us from trying to build up another and more realistic one. I do not  intend to question either the reality or the importance of the socio-  psychological facts we think of when speaking of the will of a nation.  Their analysis is certainly the prerequisite for making headway with  the problems of democracy. It would however be better not to retain  the term because this tends to obscure the fact that as soon as we  have severed the will of the people from its utilitarian connotation  we are building not merely a different theory of the same thing, but a  theory of a completely different thing. We have every reason to be on  our guard against the pitfalls that lie on the path of those defenders  of democracy who while accepting, under pressure of accumulating  evidence, more and more of the facts of the democratic process, yet  try to anoint the results that process turns out with oil taken from  eighteenth-century jars.   But though a common will or public opinion of some sort may still  be said to emerge from the infinitely complex jumble of individual and  group-wise situations, volitions, influences, actions and reactions of  the ‘democratic process’, the result lacks not only rational unity but  also rational sanction. The former means that, though from the stand-  point of analysis, the democratic pvocess is not simply chaotic—for  the analyst nothing is chaotic thai can be brought within the reach of  explanatory principles—yet the results would not, except by chance,  be meaningful in themselves—as for instance the realization of any    TWO CONCEPTS OF DEMOCRACY 157    definite end or ideal would be. The latter means, since that will is no  longer congruent with any ‘good’, that in order to claim ethical  dignity for the result it will now be necessary to fall back upon an  unqualified confidence in democratic forms of government as such—  a belief that in principle would have to be independent of the desir-  ability of results. As we have seen, it is not easy to place oneself on that  standpoint. But even if we do so, the dropping of the utilitarian  common good still leaves us with plenty of difficulties onour hands.   In particular, we still remain under the practical necessity of  attributing to the will of the individual an independence and a  rational quality that are altogether unrealistic. If we are to argue  that the will of the citizens per se is a political factor entitled to respect,  it must first exist. That is to say, it must be something more than an  indeterminate bundle of vague impulses loosely playing about given  slogans and mistaken impressions. Everyone would have to know  definitely what he wants to stand for. This definite will would have  to be implemented by the ability to observe and interpret correctly  the facts that are directly accessible to everyone and to sift critically  the information about the facts that are not. Finally, from that  definite will and from these ascertained facts a clear and prompt con-  clusion as to particular issues would have to be derived according to  the rules of logical inference—with so high a degree of general  efficiency moreover that one man’s opinion could be held, without  glaring absurdity, to be roughly as good as every other man’s." Andall  this the modal citizen would have to perform for himself and  independently of pressure groups and propaganda,” for volitions and   ‘This accounts for the strongly equalitarian character both of the classical doctrine  of democracy and of popular democratic beliefs. Ic will be pointed out later on how  Equality may acquire the status of an ethical postulate. As a factual statement about  human nature it cannot be true in any conceivable sense. In recognition of this the  postulate itself has often been reformulated so as to mean ‘equality of opportunity’.  But, disregarding even the difficulties inherent in the word opportunity, this reformu-  lation does not help us much because it is actual and not potential equality of  performance in matters of political behaviour that is required if each man’s vote 1s to  carry the same weight in the decision of issues.   It should be noted in passing that democratic phraseology has been instrumental  in fostering the association of inequality of any kind with ‘injustice’ which is so  impartant an element in the psychic pattern of the cnsucdiadlell and in the arsenal  of the politician who uses him. One of the most curious syenproms of this was the  Athenian institution of ostracism or rather the use to which it was sometimes put.  Ostracism consisted in banishing an individual by popular vote, not necessarily    for any ‘veaie? reason: it sometimes served as a method of eliminating an un-  comfortably prominent citizen who was felt to ‘count for more than one’.    2This term is here being used in its original sense and not in the sense which it is  rapidly acquiring at present and which suggests the definition: propaganda is any    158 JOSEPH SCHUMPETER    inferences that are imposed upon the electorate obviously do not  qualify for ultimate data of the democratic process. The question  whether these conditions are fulfilled to the extent required in order  to make democracy work should not be answered by reckless assertion  or equally reckless denial. It can be answered only by a laborious  appraisal of a maze of conflicting evidence.   Before embarking upon this, however, I want to make quite sure  that the reader fully appreciates another point that has been made  already. I will therefore repeat that even if the opinions and desires  of individual citizens were perfectly definite and independent data for  the democratic process to work with, and if everyone acted on them  with ideal rationality and promptitude, it would not necessarily  follow that the political decisions produced by that process from the  raw material of those individual volitions would represent anything  that could in any convincing sense be called the will of the people. It is    not only conceivable but, whenever individual wills are much divided, —  very likely that the political decisions produced will not conform te  ‘what people really want’. Nor can it be replied that, if not exactly  what de want, they will get a ‘fair compromise’. This may be so. The  chances for this to happen are greatest with those issues which are  quantitative in nature or admit of gradation, such as the question how  much is to be spent on unemployment relief provided everybody  favours some expenditure for that purpose. But with qualitative issues,  such as the question whether to persecute heretics or to enter upon  a war, the result attained may well, though for different reasons, be  equally distasteful to all the people whereas the decision imposed by  a non-democratic agency might prove much more acceptable to them.  An example will illustrate. I may, I take it, describe the rule of  Napoleon, when First Consul, as a military dictatorship. One of the  most pressing political needs of the moment wasa religious settlement  that would clear the chaos left by the revolution and the directorate  and bring peace to millions of hearts. This he achieved by a number  of master strokes, culminating in a concordat with the Pope (1801) and  the ‘organic articles’ (1802) that, reconciling the irreconcilable, gave  just the right amount of freedom to religious worship while strongly    statement emanating from a source that we do not like. I suppose that the term derives  from the name of the committee of cardinals which deals with matters concerning the  spreading of the Catholic faith, the congregatio de propaganda fide. In itself therefore it   es not carry any derogatory meaning and in particular it does not imply distortion  of facts. One can make propaganda, for instance, for a scientific method. It simply  means the presentation of facts and arguments with a view to influencing people’s  actions or opinions in a definite direction.    TWO CONCEPTS OF DEMOCRACY 159    upholding the authority of the state. He also reorganized and  refinanced the French Catholic church, solved the delicate question of  the ‘constitutional’ clergy, and most successfully launched the new  establishment with a minimum of friction. If ever there was any  justification at all for holding that the people actually want some-  thing definite, this arrangement affords one of the best instances in  history. This must be obvious to anyone who looks at the French class  structure of that time and it is amply borne out by the fact that this  ecclesiastical policy greatly contributed to the almost universal  popularity which the consular regime enjoyed. But it is difficult to see  how this result could have been achieved in a democratic way. Anti-  church sentiment had not died out and was by no means confined to  the vanquished Jacobins. People of that persuasion, or their leaders,  could not possibly have compromised to that extent.' On the other  end of the scale, a strong wave of wrathful Catholic sentiment was  steadily gaining momentum. People who shared that sentiment, or  leaders dependent on their good will, could not possibly have stopped  at the Napoleonic limit; in particular, they could not have dealt so  firmly with the Holy See for which moreover there would have been no  motive to give in, seeing which way things were moving. And the will  of the peasants who more than anything else wanted their priests, their  churches and processions would have been paralyzed by the very  natural fear that the revolutionary settlement of the land question  might be endangered once the clergy—the bishops especially—were  in the saddle again. Deadlock or interminable struggle, engendering  increasing irritation, would have been the most probable outcome of  any attempt to settle the question democratically. But Napoleon was  able to settle it reasonably, precisely because all those groups which  could not yield their points of their own accord were at the same time  able and willing to accept the arrangement if imposed.   This instance of course is not an isolated one.” Ifresults that prove in  the long run satisfactory to the people at large are made the test of  government for the people, than government by the people, as  conceived by the classical doctrine of democracy, would often fail to  meet it.   'The legislative bodies, cowed though they were, completely failed in fact to support  Napoleon in this policy. And some of his most trusted paladins opposed it.    2Other instances could in fact be adduced from Napoleon’s practice. He was an  autocrat who, whenever his dynastic interests and his foreign policy were not con-  cerned, simply strove to do what he conceived the people wanted or needed. This is  what the advice amounted to which he gave to Eugéne Beauharnais concerning the  latter’s administration of northern Italy.    160 JOSEPH SCHUMPETER    III. HUMAN NATURE IN POLITICS    It remains to answer our question about the definiteness and  independence of the voter’s will, his powers of observation and  interpretation of facts, and his ability to draw, clearly and promptly,  rational inferences from both. This subject belongs to a chapter  of social psychology that might be enutled Human Nature in Politics.’   During the second half of the last century, the idea of the human  personality that is a homogeneous unit and the idea of a definite  will that is the prime mover of action have been steadily fading—even  before the times of Théodule Ribot and of Sigmund Freud. In  particular, these ideas have been increasingly discounted in the  field of social sciences where the importance of the extra-rational and  irrational element in our behaviour has been receiving more and  more attention, witness Pareto’s Mind and Society. Of the many sources  of the evidence that accumulated against the hypothesis of rationality,  I shall mention only two.   The one—in spite of much more careful later work—may still be  associated with the name of Gustave Le Bon, the founder or, at an  rate, the first effective exponent of the psychology of crowds (psychologie  des foules).? By showing up, though overstressing, the realities of  human behaviour when under the influence of agglomeration—in  particular the sudden disappearance, in a state of excitement, of  moral restraints and civilized modes of thinking and feeling, the  sudden eruption of primitive impulses, infantilisms and criminal  propensities—he made us face gruesome facts that everybody knew  but nobody wished to see and he thereby dealt a serious blow to the    ‘This is the title of the frank and charming book by one of the most lovable English  radicals who ever lived, Graham Wallas. In spite of all that has since been written on the  subject and especially in spite of all the‘detailed case studies that now make it possible  to see so much more clearly, that book may still be recommended as the best introduc-  tion to political psychology. Yet, after having stated with admirable honesty the case  against the uncritical acceptance of the classical doctrine, the author fails to draw the  obvious conclusion. This is all the more remarkable because he rightly insists on the  necessity of a scientific attitude of mind and because he does not fail to take Lord  Bryce to task for having, in his book on the American commonwealth, professed him-  self ‘grimly’ resolved to see some blue sky in the midst of clouds of disillusioning facts.  Why, so Graham Wallas seems to exclaim, what should we say of a meteorologist who  insisted from the outset that he saw some blue sky? Nevertheless in the constructive  part of his book he takes much the same ground. |   *The German term, Massenpsychologie, suggests a warning: the psychology of crowds  must not be confused with the psychology of the masses. The former does not neces-  sarily carry any class connotation and in itself has nothing to do with a study of the  ways of thinking and feeling of, say, the working class.    TWO CONCEPTS OF DEMOCRACY 161    picture of man’s nature which underlies the classical doctrine of  democracy and democratic folklore about revolutions. No doubt  there is much to be said about the narrowness of the factual basis of  Le Bon’s inferences which, for instance, do not fit at all well the  normal behaviour of an English or Anglo-American crowd. Critics,  especially those to whom the implications of this branch of social  psychology were uncongenial, did not fail to make the most of its  vulnerable points. But on the other hand it must not be forgotten  that the phenomena of crowd psychology are by no means confined to  mobs rioting in the narrow streets of a Latin town. Every parliament,  every committee, every council of war composed ofa dozen generals in  their sixties, displays, in however mild a form, some of those features  that stand out so glaringly in the case of the rabble, in particular a re-  duced sense of responsibility, a lower level of energy of thought and  greater sensitiveness to non-logical influences. Moreover, those  phenomena are not confined to a crowd in the sense of a physical  agglomeration of many people. Newspaper readers, radio audiences,  members of a party even if not physically gathered together are  terribly easy to work up into a psychological crowd and into a state  of frenzy in which attempt at rational argument only spurs the animal  spirits.   The other source of disillusioning evidence that I am going to  mention is a much humbler one—no blood flows from it, only  nonsense. Economists, learning to observe their facts more closely,  have begun to discover that, even in the most ordinary currents of  daily life, their consumers do not quite live up to the idea that the  economic text-book used to convey. On the one hand their wants are  nothing like as definite and their actions upon those wants nothing  like as rational and prompt. On the other hand they are so amenable  to the influence of advertising and other methods of persuasion that  producers often seem to dictate to them instead of being directed by  them. The technique of successful advertising is particularly instruc-  tive. There is indeed nearly always some appeal to reason. But mere  assertion, often repeated, counts more than rational argument and so  does the direct attack upon the subconscious which takes the form of  attempts to evoke and crystallize pleasant associations of an entirely  extra-rational, very frequently of a sexual, nature.   The conclusion, while obvious, must be drawn with care. In the  ordinary run of often repeated decisions the individual is subject to  the salutary and rationalizing influence of favourable and unfavour-  able experience. He is also under the influence of relatively simple and    162 JOSEPH SCHUMPETER    unproblematical motives and interests which are but occasionally  interfered with by excitement. Historically, the consumers’ desire for  shoes may, at least in part, have been shaped by the action of pro-  ducers offering attractive footgear and campaigning for it; yet at any  given time it is a genuine want, the definiteness of which extends  beyond ‘shoes in general’ and which prolonged experimenting clears  of much of the irrationalities that may originally have surrounded  it.! Moreover, under the stimulus of those simple motives consumers  learn to act upon unbiased expert advice about some things (houses,  motor-cars) and themselves become experts in others. Itis simply not  true that housewives are easily fooled in the matter of foods, familiar  household articles, wearing apparel. And, as every salesman knows to  his cost, most of them have a way of insisting on the exact article they  want.   This of course holds true still more obviously on the producers’  side of the picture. No doubt, a manufacturer may be indolent, a bad  judge of opportunities or otherwise incompetent; but there is an  effective mechanism that will reform or eliminate him. Again  Taylorism rests on the fact that man may perform simple handicraft  operations for thousands of years and yet perform them inefficiently.  But neither the intention to act as rationally as possible nor a steady  pressure toward rationality can seriously be called into question at  whatever level of industrial or commercial activity we choose to look.?   And so it is with most of the decisions of daily life that lie within the  little field which the individual citizen’s mind encompasses with a full  sense of its reality. Roughly, it consists of the things that directly  concern himself, his family, his business dealings, his hobbies, his  friends and enemies, his township or ward, his class, church, trade  union or any other social group of which he is an active member—  the things under his personal observation, the things which are  familiar to him independently of what his newspaper tells him, which  he can directly influence or manage and for which he develops the kind    ‘In the above passage irrationality means failure to act rationally upon a given wish.  It does not refer to the reasonableness of the wish itself in the opinion of the observer.  This is important to note because economists in appraising the extent of consumers’  irrationality sometimes exaggerate it by confusing the two things. Thus, a ee Es  finery may seem to a professor an indication of irrational behaviour for which there is  no other explanation but the advertiser’s arts. Actually, it may be all she craves for.  If so her expenditure on it may be ideally rational in the above sense.   2This level differs of course not only as between epochs and places but also, ata given  time and place, as between different industrial sectors and classes. There is no such  thing as a universal pattern of rationality.    TWO CONCEPTS OF DEMOCRACY 163    of responsibility that is induced by a direct relation to the favourable  or unfavourable effects of a course of action.   Once more: definiteness and rationality in thought and action’ are  not guaranteed by this familiarity with men and things or by that  sense of reality or responsibility. Quite a few other conditions which  often fail to be fulfilled would be necessary for that. For instance,  generation after generation may suffer from irrational behaviour in  matters of hygiene and yet fail to link their sufferings with their  noxious habits. As long as this is not done, objective consequences,  however regular, of course do not produce subjective experience.  Thus it proved unbelievably hard for humanity to realize the relation  between infection and epidemics: the facts pointed to it with what to  us seems unmistakable clearness; yet to the end of the eighteenth  century doctors did next to nothing to keep people afflicted with  infectious disease, such as measles or smallpox, from mixing with  other people. And things must be expected to be still worse whenever  there is not only inability but reluctance to recognize causal relations  or when some interest fights against recognizing them.   Nevertheless and in spite of all the qualifications that impose them-  selves, there is for everyone, within a much wider horizon, a narrower  field—widely differing in extent as between different groups and  individuals and bounded by a broad zone rather than a sharp line—  which is distinguished by a sense of reality or familiarity or respon-  sibility. And this field harbours relatively definite individual volitions.  These may often strike us as unintelligent, narrow, egotistical; and it  may not be obvious to everyone why, when it comes to political  decisions, we should worship at their shrine, still less why we should  feel bound to count each of them for one and none of them for more  than one. If, however, we do choose to worship we shall at least not  find the shrine empty.?    ‘Rationality of thought and rationality of action are two different things. Rationality  of thought does not always guarantee rationality of action. And the latter may be  present without any conscious deliberation and irrespective of any ability to formulate  the rationale of one’s action correctly. The observer, particularly the observer who  uses interview and questionnaire methods, often overlooks this and hence acquires an  exaggerated idea of the importance of irrationality in behaviour. This is another source  of those overstatements which we meet so often.    21t should be observed that in speaking of definite and genuine volitions I do not  mean to exalt them into ultimate data for all kinds of social analysis Of course they are  themselves the product of the social process and the social environment. All Imean  is that they may serve as data for the kind of special-purpose analysis which the  economist has in mind when he derives prices from tastes or wants that are ‘given’ at  any moment and need not be further analysed each time. Similarly we may for our pur-    164 JOSEPH SCHUMPETER    Now this comparative definiteness of volition and rationality of  behaviour does not suddenly vanish as we move away from those  _concerns of daily life in the home and in business which educate  and discipline us. In the realm of public affairs there are sectors  that are more within the reach of the citizen’s mind than others.  This is true, first, of local affairs. Even there we find a reduced power  of discerning facts, areduced preparedness toact upon them, areduced  sense of responsibility. We all know the man—and a very good speci-  men he frequently is—who says that the local administration is not his  business and callously shrugs his shoulders at practices which he  would rather die than suffer in his own office. High-minded citizens  in a hortatory mood who preach the responsibility of the individual  voter or taxpayer invariably discover the fact that this voter does  not feel responsible for what the local politicians do. Still, especially  in communities not too big for personal contacts, local patriotism  may be a very important factor in ‘making democracy work’, Also, the  problems of a town are in many respects akin to the problems ofa  manufacturing concern. The man who understands the latter also  understands, to some extent, the former. The manufacturer, grocer or  workman need not step out of his world to havea rationally defensible  view (that may of course be right or wrong) on street cleaning or town  halls.   Second, there are many national issues that concern individuals and  groups so directly and unmistakably as to evoke volitions that are  genuine and definite enough. The most important instance is afforded  by issues involving immediate and personal pecuniary profit to  individual voters and groups of voters, such as direct payments, pro-  tective duties, silver policies and so on. Experience that goes back to  antiquity shows that by and large voters react promptly and rationally  to any such chance. But the classical doctrine of democracy evidently  stands to gain little from displays of rationality of this kind. Voters  thereby prove themselves bad and indeed corrupt judges of such  issues,’ and often they even prove themselves bad judges of their own    ps speak of genuine and definite volitions that at any moment are given independent-  of attempts to manufacture them, although we recognize that these genuine volitions  themselves are the result of environmental influences in the past, propagandist  influences included. This distinction between genuine and mukeldared will (see  below) is a difficult one and cannot be applied in all cases and for all purposes. For  our purpose however it is sufficient to point to the obvious common-sense case which  can be made for it.   'The reason why the Benthamites so completely overlooked this is that they did not    consider the possibilities of mass corruption in modern capitalism. Committing in  their political theory the same error which they committed in their economic theory,    TWO CONCEPTS OF DEMOCRACY 165    long-run interests, for it is only the short-run promise that tells  politically and only short-run rationality that asserts itself effectively.   However, when we move still farther away from the private con-  cerns of the family and the business office into those regions of  national and international affairs that lack a direct and unmistakable  link with those private concerns, individual volition, command of facts  and method of inference soon cease to fulfil the requirements of the  classical doctrine. What strikes me most of all and seems to me to be  the core of the trouble is the fact that the sense of reality’ is so  completely lost. Normally, the great political questions take their  place in the psychic economy of the typical citizen with those leisure-  hour interests that have not attained the rank of hobbies, and with  the subjects of irresponsible conversation. These things seem so far  off; they are not at all like a business proposition; dangers may not  materialize at all and if they should they may not prove so very  serious; one feels oneself to be moving ina fictitious world.   This reduced sense of reality accounts not only for a reduced  sense of responsibility but also for the absence of effective volition.  One has one’s phrases, of course, and one’s wishes and daydreams  and grumbles; especially, one has one’s likes and dislikes. But  ordinarily they do not amount to what we call a will—the psychic  counterpart of purposeful responsible action. In fact, for the  private citizen musing over national affairs there is no scope for such  a will and no task at which it could develop. He is a member of an  unworkable committee, the committee of the whole nation, and this  is why he expends less disciplined effort on mastering a political  problem than he expends on a game of bridge.”    they felt no compunction about postulating that, ‘the people’ were the best judges  of their own individual interests and that these must necessarily coincide with the  interests of all the people taken together. Of course this was made easier for them  because actually though not intentionally they philosophized in terms of bourgeois  interests which had more to gain from a parsimonious state than from any direct  bribes.    ‘William James’ ‘pungent sense of reality’. The relevance of this point has been  particularly emphasized by Graham Wallas.    2It will help to clarify the point if we ask ourselves why so much more intelligence  and clear-headedness show up at a bridge table than in, say, political discussion among  non-politicians. At the bridge table we have a definite task; we have rules that  discipline us; success and failure are clearly defined; and we are prevented from  behaving irresponsibly because every mistake we make will not only immediately tell  but also be immediately allocated to us. These conditions, by iheit failure to be  fulfilled for the political behaviour of the ordinary citizen, show why it is that in  politics he lacks a 1 the alertness and the judgement he may display in his profession.    166 JOSEPH SCHUMPETER    The reduced sense of responsibility and the absence of effective  volition in turn explain the ordinary citizen’s ignorance and lack of  judgement in matters of domestic and foreign policy which are ifany-  thing more shocking in the case of educated people and of people who  are successfully active in non-political walks of life than it is with  uneducated people in humble stations. Information is plentiful and  readily available. But this does not seem to make any difference. Nor  should we wonder at it. We need only compare a lawyer’s attitude to  his brief and the same lawyer’s attitude to the statements of political  fact presented in his newspaper in order to see what is the matter. In  the one case the lawyer has qualified for appreciating the relevance of  his facts by years of purposeful labour done under the definite  stimulus of interest in his professional competence; and under a  stimulus that is no less powerful he then bends his acquirements, his  intellect, his will to the contents of the brief. In the other case, he  has not taken the trouble to qualify; he does not care to absorb the  information or to apply to it the canons of criticism he knows so well  how to handle; and he is impatient of long or complicated argument.  All of this goes to show that without the initiative that comes from  immediate responsibility, ignorance will persist in the face of masses  of information however complete and correct. It persists even in the  face of the meritorious efforts that are being made to go beyond  presenting information and to teach the use ofit by means of lectures,  classes, discussion groups. Results are not zero. But they are small.  People cannot be carried up the ladder.   Thus the typical citizen drops down to a lower level of mental  performance as soon as he enters the political field. He argues and  analyses in a way which he would readily recognize as infantile within  the sphere of his real interests. He becomes a primitive again. His  thinking becomes associative and affective.’ And this entails two  further consequences of ominous significance.   First, even if there were no political groups trying to influence  him, the typical citizen would in political matters tend to yield to  extra-rational or irrational prejudice and impulse. The weakness of  the rational processes he applies to politics and the absence of  effective logical control over the results he arrives at would in  themselves suffice to account for that. Moreover, simply because he  is not ‘all there’, he will relax his usual moral standards as well and  occasionally give in to dark urges which the conditions of private    1See ch. xii.    TWO CONCEPTS OF DEMOCRACY 167    life help him to repress. But as to the wisdom or rationality of his  inferences and conclusions, it may be just as bad if he gives in to a  burst of generous indignation. This will make it still more difficult  for him to see things in their correct proportions or even to see more  than one aspect of one thing at a time. Hence, if for once he-does  emerge from his usual vagueness and does display the definite will  postulated by the classical doctrine of democracy, he is as likely as  not to become still more unintelligent and irresponsible than he  usually is. At certain junctures, this may prove fatal to his nation.’   Second, however, the weaker the logical element in the processes of  the public mind and the more complete the absence of rational criti-  cism and of the rationalizing influence of personal experience and  responsibility, the greater are the opportunities for groups with an  axe to grind. These groups may consist of professional politicians or of  exponents of an economic interest or of idealists of one kind or  another or of people simply interested in staging and managing  political shows. The sociology of such groups is immaterial to the  argument in hand. The only point that matters here is that, Human  Nature in Politics being what it is, they are able to fashion and,  within very wide limits, even to create the will of the people. What we  are confronted with in the analysis of political processes is largely  not a genuine but a manufactured will. And often this artefact is all  that in reality corresponds to the volonté générale of the classical  doctrine. So far as this is so, the will of the people is the product and  not the motive power of the political process.   The ways in which issues and the popular will on any issue are being  manufactured is exactly analogous to the ways ot commercial advertis-  ing. We find the same attempts to contact the subconscious. We find  the same technique of creating favourable and unfavourable associa-  tions which are the more effective the less rational they are. We find  the same evasions and reticences and the same trick of producing  opinion by reiterated assertion that is successful precisely to the  extent to which it avoids rational argument and the danger of  awakening the critical faculties of the people. And so on. Only, all    ‘The importance of such bursts cannot be doubted. But it is possible to doubt  their genuineness. Analysis will show in many instances that they are induced by  the action of some group and do not spontaneously arise from the people. In this  case they enter into a (second) class of phenomena which we are about to deal with.  Persona! ly, I do believe that genuine instances exist. But I cannot be sure that more  thorough analysis would not reveal some psycho-technical effort at the bottom of  them.    168 JOSEPH SCHUMPETER    these arts have infinitely more scope in the sphere of public affairs  than they have in the sphere of private and professional life. The  picture of the prettiest girl that ever lived will in the long run prove  powerless to maintain the sales of a bad cigarette. There is no equally  effective safeguard in the case of political decisions. Many decisions of  fateful importance are of a nature that makes it impossible for the  public to experiment with them at its leisure and at moderate cost.  Even if that is possible, however, judgement is as a rule not so easy to  arrive at as it is in the case of the cigarette, because effects are less easy  to interpret.   But such arts also vitiate, to an extent quite unknown in the field of  commercial advertising, those forms of political advertising that  profess to address themselves to reason. To the observer, the antira-  tional or, at all events, the extra-rational appeal and the defenceless-  ness of the victim stand out more and not less clearly when cloaked in  facts and arguments. We have seen above why it is so difficult to impart  to the public unbiased information about political problems and  logically correct inferences from it and why it is that information and  arguments in political matters will ‘register’ only if they link up  with the citizen’s preconceived ideas. As a rule, however, these  ideas are not definite enough to determine particular conclusions.  Since they can themselves be manufactured, effective political argu-  ment almost inevitably implies the attempt to twist existing volitional  premises into a particular shape and not merely the attempt to im-  plement them or to help the citizen to make up his mind.   Thus information and arguments that are really driven home are  likely to be the servants of political intent. Since the first thing man  will do for his ideal or interest is to lie, we shall expect, and as a  matter of fact we find, that effective information is almost always  adulterated or selective! and that effective reasoning in politics  consists mainly in trying to exalt certain propositions into axioms  and to put others out of court; it thus reduces to the psycho-technics  mentioned before. The reader who thinks me unduly pessimistic need  only ask himself whether he has never heard—or said himself—that  this or that awkward fact must not be told publicly, or that a certain  line of reasoning, though valid, is undesirable. If men who according  to any current standard are perfectly honourable or even high-  minded reconcile themselves to the implications of this, do they not  thereby show what they think about the merits or even the existence  of the will of the people?    Selective information, if in itself correct, is an attempt to lie by speaking the truth.    TWO CONCEPTS OF DEMOCRACY 169    There are of course limits to all this.! And there is truth in Jefferson’s  dictum that in the end the people are wiser than any single individual    can be, or in Lincoln’s about the impossibility of ‘fooling all the    peaple all the time’. But both dicta stress the long-run aspect ina  highly significant way. It is no doubt possible to argue that given time  us_as highly reasonable and even shrewd. History however consists  “of a succession of short-run situations that may alter the course of  events for good. If all the people can in the short run be ‘fooled’ step  by step into something they do not really want, and if this is not an  exceptional case which we could afford to neglect, then no amount  of retrospective common sense will alter the fact that in reality they  neither raise nor decide issues but that the issues that shape their fate  are normally raised and decided for them. More than anyone else the  lover of democracy has every reason to accept this fact and to clear his  creed from the aspersion that it rests upon make-believe.    IV. REASONS FOR THE SURVIVAL OF THE CLASSICAL DOCTRINE    But how is it possible that a doctrine so patently contrary to fact  should have survived to this day and continued to hold its place in the  hearts of the people and in the official language of governments ? The  refuting facts are known to all; everybody admits them with perfect,  frequently with cynical, frankness. The theoretical basis, utilitarian  rationalism, is dead; nobody accepts it as a correct theory of the body  politic. Nevertheless that question is not difficult to answer.   First of all, though the classical doctrine of collective action may  not be supported by the results of empirical analysis, it is powerfully  supported by that association with religious belief to which I have  adverted already. This may not be obvious at first sight. The utilitarian  leaders were anything but religious in the ordinary sense of the term.  In fact they believed themselves to be anti-religious and they were so  considered almost universally. They took pride in what they thought  was precisely an unmetaphysical attitude and they were quite out of  sympathy with the religious institutions and the religious move-  ments of their time. But we need only cast another glance at the picture  they drew of the social process in order to discover that it embodied  essential features of the faith of protestant Christianity and was in    ‘Possibly they might show more clearly if issues were more frequently decided  by referertdum. Politicians presumably know why they are almost invariably hostile  to that institution, ;    170 JOSEPH SCHUMPETER    fact derived from that faith. For the intellectual who had cast off his  religion the utilitarian creed provided a substitute for it. For many of  those who had retained their religious belief the classical doctrine  became the political complement of it.!   Thus transposed into the categories of religion, this doctrine—and  in consequence the kind of democratic persuasion which is based  upon it—changes its very nature. There is no longer any need for  logical scruples about the Common Good and Ultimate Values. All  this is settled for us by the plan of the Creator whose purpose defines  and sanctions everything. What seemed indefinite or unmotivated  before is suddenly quite definite and convincing. The voice of the  people that is the voice of God for instance. Or take Equality. Its  very meaning is in doubt, and there is hardly any rational warrant for  exalting it into a postulate, so long as we move in the sphere of  empirical analysis. But Christianity harbours a strong equalitarian  element. The Redeemer died for all: He did not differentiate between  individuals of different social status. In doing so, He testified to the  intrinsic value of the individual soul, a value that admits of no grada-  tions. Is not this a sanction—and, as it seems to me, the only possible  sanction’—of ‘everyone to count for one, no one to count for more  than one’—a sanction that pours super-mundane meaning into  articles of the democratic creed for which it is not easy to find any  other? To be sure this interpretation does not cover the whole ground.  However, so far as it goes, it seems to explain many things that other-  wise would be unexplainable and in fact meaningless. In particular,  it explains the believer’s attitude toward criticism: again, as in the  case of socialism, fundamental dissent is looked upon not merely as  error but as sin; it elicits not merely logical counterargument but  also moral indignation.   We may put our problem differently and say that democracy, when  motivated in this way, ceases to be a mere method that can be dis-    ‘Observe the analogy with socialist belief which also is a substitute for Christian  belief to some and a complement of it to others.    It might be objected that, however difficult it may be to attach a general meaning  to the word Equality, such meaning can be unravelled from its context in most if not  all cases. For instance, it may be permissible to infer from the circumstances in which  the ge pi address was delivered that by the ‘proposition that all men are created  free and equal’, Lincoln simply meant equality ae legal status versus the kind of  inequality that is implied in the recognition of slavery. This meaning would be  definite enough. But if we ask why that proposition should be morally and politically  binding and if we refuse to answer ‘Because every man is by nature exactly like every  other man’, then we can only fall back upon the divine sanction supplied by Christian  belief. This solution is conceivably implied in the word ‘created’.    TWO CONCEPTS OF DEMOCRACY 171    cussed rationally like a steam engine or a disinfectant. It actually  becomes what from another standpoint I have held it incapable of  becoming, viz., an ideal or rather a part of an ideal schema of things.  The very word may become a flag, a symbol of alla man holds dear, of  everything that he loves about his nation whether rationally con-  tingent to it or not. On the one hand, the question how the various  propositions implied in the democratic belief are related to the facts  of politics will then become as irrelevant to him as is, to the believing  Catholic, the question how the doings of Alexander VI tally with the  supernatural halo surrounding the papal office. On the other hand,  the democrat of this type, while accepting postulates carrying large  implications about equality and brotherliness, will be in a position  also to accept, in all sincerity, almost any amount of deviations from  them that his own behaviour or position may involve. That is not even  illogical. Mere distance from fact is no argument against an ethical  maxim or a mystical hope.   Second, there is the fact that the forms and phrases of classical  democracy are for many nations associated with events and develop-  ments in their history which are enthusiastically approved by large  majorities. Any opposition to an established regime is likely to use  these forms and phrases whatever its meaning and social roots may  be. If it prevails and if subsequent developments prove satisfactory,  then these forms will take root in the national ideology.   The United States is the outstanding example. Its very existence  as a sovereign state is associated with a struggle against a monarchial  and aristocratic England. A minority of loyalists excepted, Americans  had, at the time of the Grenville administration, probably ceased to  look upon the English monarch as their king and the English aristo-  cracy as their aristocracy. In the War of Independence they fought  what in fact as well as in their feeling had become a foreign monarch  and a foreign aristocracy who interfered with their political and  economic interests. Yet from an early stage of the troubles they  presented their case, which really was a national one, as a case of the  ‘people’ versus its ‘ruler’, in terms of inalienable Rights of Man and  in the light of the general principles of classical democracy. The  wording of the Declaration of Independence and of the Constitution    ‘It might seem that an exception should be made for oppositions that issue into  frankly autocratic regimes, But even most of these rose, as a matter of history, in  democratic ways and based their rule on the approval of the people. Caesar was  not killed by plebeians. But the aristocratic oligarchs who did kil him also used  democratic phrases.    172 JOSEPH SCHUMPETER    adopted these principles. A prodigious development followed that  absorbed and satisfied most people and thereby seemed to verify the  doctrine embalmed in the sacred documents of the nation.  Oppositions rarely conquer when the groups in possession are in  the prime of their power and success. In the first half of the nineteenth  century, the oppositions that professed the classical creed of democracy  rose and eventually prevailed against governments some of which—  especially in Italy—were obviously in a state of decay and had become  bywords of incompetence, brutality and corruption. Naturally though  not quite logically, this redounded to the credit of that creed which  moreover showed up to advantage when compared with the benighted  superstitions sponsored by those governments. Under these cir-  cumstances, democratic revolution meant the advent of freedom and  decency, and the democratic creed meant a gospel of reason and  betterment. To be sure, this advantage was bound to be lost and the  gulf between the doctrine and the practice of democracy was bound  to be discovered. But the glamour of the dawn was slow to fade.  Third, it must not be forgotten that there are social patterns in  which the classical doctrine will actually fit facts with a sufficient  degree of approximation. As has been pointed out, this is the case  with many small and primitive societies which as a matter of fact  served as a prototype to the authors of that doctrine. It may be the  case also with societies that are not primitive provided they are not  too differentiated and do not harbour any serious problems.  Switzerland is the best example. There is so little to quarrel about in  a world of peasants which, excepting hotels and banks, contains no  great capitalist industry, and the problems of public policy are so  simple and so stable that an overwhelming majority can be expected  to understand them and to agree about them. But if we can conclude  that in such cases the classical doctrine approximates reality we  have to add immediately that it does so not because it describes an  effective mechanism of political decision but only because there are  no great decisions to be made. Finally, the case of the United States  may again be invoked in order to show that the classical doctrine  sometimes appears to fit facts even in a society that is big and highly  differentiated and in which there are great issues to decide provided  the sting is taken out of them by favourable conditions. Until this  country’s entry into the First World War, the public mind was con-  cerned mainly with the business of exploiting the economic possi-  bilities of the environment. So long as this business was not seriously  interfered with nothing mattered fundamentally to the average citizen    TWO CONCEPTS OF DEMOCRACY 173    who looked on the antics of politicians with good-natured contempt.  Sections might get excited over the tariff, over silver, over local  misgovernment, or over an occasional squabble with England. The  people at large did not care much, except in the one case of serious  disagreement which in fact produced national disaster, the Civil War.   And fourth, of course, politicians appreciate a phraseology that  flatters the masses and offers an excellent opportunity not only for  evading responsibility but also for crushing opponents in the name of  the people.    2-ANOTHER THEORY OF DEMOCRACY  I. COMPETITION FOR POLITICAL LEADERSHIP    I THINK that most students of politics have by now come to accept the  criticisms levelled at the classical doctrine of democracy in the preced-  ing chapter. I also think that most of them agree, or will agree before  long, in accepting another theory which is much truer to life and at  the same time salvages much of what sponsors of the democratic  method really mean by this term. Like the classical theory, it may be  put into the nutshell of a definition.   It will be remembered that our chief troubles about the classical  theory centred in the proposition that ‘the people’ hold a definite  and rational opinion about every individual question and that they  give effect to this opinion—in a democracy—by choosing ‘represent-  atives’ who will see to it that that opinion is carried out. Thus the  selection of the representatives is made secondary to the primary  purpose of the democratic arrangement which is to vest the power of  deciding political issues in the electorate. Suppose we reverse the  roles of these two elements and make the deciding of issues by the  electorate secondary to the election of the men who are to do the  deciding. To put it differently, we now take the view that the role of  the people is to produce a government, or else an intermediate body  which in turn will produce a national executive!or government. And  we define: the democratic method is that institutional arrangement  for arriving at political decisions in which individuals acquire the    1The insincere word ‘executive’ really points in the wrong direction. It ceases  however to do so if we use it in the sense in which we speak of the ‘executives’ of a  business corporation who also do a great deal more than ‘execute’ the will of stock-  holders.    174 JOSEPH SCHUMPETER    power to decide by means of a competitive struggle for the people’s  vote.   Defence and explanation of this idea will speedily show that, as  to both plausibility of assumptions and tenability of propositions,  it greatly improves the theory of the democratic process.   First of all, we are provided with a reasonably efficient criterion  by which to distinguish democratic governments from others. We  have seen that the classical theory meets with difficulties on that score  because both the will and the good of the people may be, and inmany  historical instances have been, served just as well or better by govern-  ments that cannot be described as democratic according to any  accepted usage of the term. Now we are ina somewhat better position  partly because we are resolved to stress a modus procedendi the presence  or absence of which it is in most cases easy to verify.   For instance, a parliamentary monarchy like the English one ful-  fils the requirements of the democratic method because the monarch  is practically constrained to appoint to cabinet office the same people  as parliament would elect. A ‘constitutional’ monarchy does not  qualify to be called democratic because electorates and parliaments,  while having all the other rights that electorates and parliaments have  in parliamentary monarchies, lack the power to impose their choice  as to the governing committee: the cabinet ministers are in this case  servants of the monarch, in substance as well as in name, and can in  principle be dismissed as well as appointed by him. Such an arrange-  ment may satisfy the people. The electorate may reaffirm this fact by  voting against any proposal for change. The monarch may be so  popular as to be able to defeat any competition for the supreme  office. But since no machinery is provided for making this competition  effective the case does not come within our definition.   Second, the theory embodied in this definition leaves all the room  we may wish to have for a proper recognition of the vital fact of  leadership. The classical theory did not do this but, as we have seen,  attributed to the electorate an altogether unrealistic degree of in-  itiative which practically amounted to ignoring leadership. But  collectives act almost exclusively by accepting leadership—this is  the dominant mechanism of practically any collective action which is  more than a reflex. Propositions about the working and the results of  the democratic method that take account of this are bound to be  infinitely more realistic than propositions which do not. They will not    ‘See however the fourth point below.    TWO CONCEPTS OF DEMOCRACY 175    stop at the execution of a volonté générale but will go some way toward  showing how it emerges or how it is substituted or faked. What we have  termed Manufactured Will is no longer outside the theory, an aberra-  tion for the absence of which we piously pray; it enters on the ground  floor as it should.   Third, however, so far as there are genuine group-wise volitions  at all—for instance the will of the unemployed to receive unemploy-  ment benefit or the will of other groups to help—our theory does not  neglect them. On the contrary we are now able to insert them in  exactly the role they actually play. Such volitions do not as a rule  assert themselves directly. Even if strong and definite they remain  latent, often for decades, until they are called to life by some political  leader who turns them into political factors. This he does, or else his  agents do it for him, by organizing these volitions, by working them up  and by including eventually appropriate items in his competitive  offering. The interaction between sectional interests and public  opinion and the way in which they produce the pattern we call the  political situation appear from this angle in a new and much clearer  light.   Fourth, our theory is of course no more definite than is the concept  of competition for leadership. This concept presents similar difficulties  as the concept of competition in the economic sphere, with which it  may be usefully compared. In economic life competition is never com-  pletely lacking, but hardly ever is it perfect.! Similarly, in political  life there is always some competition, though perhaps only a potential  one, for the allegiance of the people. To simplify matters we have  restricted the kind of competition for leadership which is to define  democracy, to free competition for a free vote. The justification for  this is that democracy seems to imply a recognized method by which  to conduct the competitive struggle, and that the electoral method is  practically the only one available for communities of any size. But  though this excludes many ways of securing leadership which should  be excluded,? such as competition by military insurrection, it does  not exclude the cases that are strikingly analogous to the economic    ‘In Part II we had examples of the problems which arise out of this.    21It also excludes methods which should not be excluded, for instance, the acquisition  of political leadership by the people’s tacit acceptance of it or by election guast per  inspirationem. The latter differs ee election by voting only by a technicality. But the  former is not quite without importance even in modern politics; the swa held by a  party boss within his party is often based on nothing but tacit acceptance of his leader-  ship. Comparatively speaking however these are details which may, I think, be  neglected in a sketch like this.    176 JOSEPH SCHUMPETER    phenomena we label ‘unfair’ or ‘fraudulent’ competition or restraint  of competition. And we cannot exclude them because if we did we  should be left with a completely unrealistic ideal.! Between this ideal  case which does not exist and the cases in which all competition with  the established leader is prevented by force, there is a continuous  range of variation within which the democratic method of government  shades off into the autocratic one by imperceptible steps. But if we  wish to understand and not to philosophize, this is as it should be.  The value of our criterion is not seriously impaired thereby.   Fifth, our theory seems to clarify the relation that subsists between  democracy and individual freedom. If by the latter we mean the  existence of a sphere of individual self-government the boundaries of  which are historically variable—no society tolerates absolute freedom  even of conscience and of speech, no society reduces that sphere to  zero—the question clearly becomes a matter of degree. We have seen  that the democratic method does not necessarily guarantee a greater  amount of individual freedom than another political method would  permit in similar circumstances. It may well be the other way round.  But there is still a relation between the two. If, on principle at least,  everyone is free to compete for political leadership? by presenting  himself to the electorate, this will in most cases though not in all  mean a considerable amount of freedom of discussion for all. In  particular it will normally mean a considerable amount of freedom of  the press. This relation between democracy and freedom is notabsolute-  ly stringent and can be tampered with. But, from the standpoint of  the intellectual, it is nevertheless very important. At the same ume,  it is all there is to that relation.   Sixth, it should be observed that in making it the primary function  of the electorate to produce a government (directly or through an  intermediate body) I intended to include in this phrase also the func-  tion of evicting it. The one means simply the acceptance of a leader  or a group of leaders, the other means simply the withdrawal of this  acceptance. This takes care of an element the reader may have missed.  He may have thought that the electorate controls as well as installs.  But since electorates normally do not control their political leaders  in any way except by refusing to reelect them or the parliamentary  majorities that support them, it seems well to reduce our ideas about  this control in the way indicated by our definition. Occasionally,   1As in the economic field, some restrictions are implicit in the legal and moral  principles of the community.   ? Free, that is, in the same sense in which everyone is free to start another textile mill.    TWO CONCEPTS OF DEMOCRACY 177    spontaneous revulsions occur which upset a government or an in-  dividual minister directly or else enforce a certain course of action.  But they are not only exceptional, they are, as we shall see, contrary  to the spirit of the democratic method.   Seventh, our theory sheds much-needed light onan old controversy.  Whoever accepts the classical doctrine of democracy and in con-  sequence believes that the democratic method is to guarantee that  issues be decided and policies framed according to the will of the  people must be struck by the fact that, even if that will were undeniably  real and definite, decision by simple majorities would in many cases  distort it rather than give effect to it. Evidently the will of the majority  is the will of the majority and not the will of ‘the people’. The latter  is a mosaic that the former completely fails to ‘represent’. To equate  both by definition is not to solve the problem. Attempts at real  solutions have however been made by the authors of the various plans  for Proportional Representation.   These plans have met with adverse criticism on practical grounds.  It is in fact obvious not only that proportional representation will  offer opportunities for all sorts of idiosyncrasies to assert themselves  but also that it may prevent democracy from producing efficient  governments and thus prove a danger in times of stress.’ But before  concluding that democracy becomes unworkable if its principle is  carried out consistently, it is just as well to ask ourselves whether  this principle really implies proportional representation. As a matter  of fact it does not. If acceptance of leadership is the true function  of the electorate’s vote, the case for proportional representation  collapses because its premises are no longer binding. The principle  of democracy then merely means that the reins of government should  be handed to those who command more support than do any of the  competing individuals or teams. And this in turn seems to assure the  standing of the majority system within the logic of the democratic  method, although we might still condemn it on grounds that lie  outside of that logic.    Il. THE PRINCIPLE APPLIED    The theory outlined in the preceding section we are now going to  try out on some of the more important features of the structure and  working of the political engine in democratic countries.   'The argument against proportional representation has been ably stated by    Professor F. A. Hermens in ‘The Trojan Horse of Democracy’, Social Research,  November 1938.    178 JOSEPH SCHUMPETER    1. In a democracy, as I have said, the primary function of the  elector’s vote is to produce government. This may mean the election  of a complete set of individual officers. This practice however is in  the main a feature of local government and will be neglected hence-  forth.! Considering national government only, we may say that  producing government practically amounts to deciding who the  leading man shall be.? As before, we shall call him Prime Minister.   There is only one democracy in which the electorate’s vote does  this directly, viz., the United States.* In all other cases the electorate’s    'This we shall do for simplicity’s sake only. The phenomenon fits perfectly into our  schema.    *This is only eerie! true. The elector’s vote does indeed put into power  a group that in all normal cases acknowledges an individual leader but there are as  a rule leaders of second and third rank who carry political guns in their own right and  whom the leader has no choice but to put into appropriate offices. This fact will be  recognized presently.  Another point must be kept in mind. Although there is reason to expect that a man  who rises to a position of supreme command will in general be a man of considerable  ersonal force, whatever else he may be—to this we shall return later on—it does not  ollow that this will always be the case. Therefore the term ‘leader’ or ‘leading man’  is not to imply that the individuals thus designated are necessarily endowed with  qualities of leadership or that they always do give any personal leads. There are political  situations favourable to the rise of men deficient in leadership (and other qualities)  and unfavourable to the establishment of strong individual positions. A party or a  combination of parties hence may occasionally be acephalous. But everyone recognizes  that this is a paidloged state and one of the typical causes of defeat.    3We may, I take it, disregard the electoral college. In calling the President of the  United States a prime minister I wish to stress the fundamental similarity of his  position to that of prime ministers in other democracies. But I do not wish to minimize  the differences, although some of them are more formal than real. The least important  of them is that the President also fulfils those largely ceremonial functions of, say,  the French presidents. Much more important is it that he cannot dissolve Congress—  but neither could the French Prime Minister do so. On the other hand, his position  is stronger than that of the English Prime Minister by virtue of the fact that his  leadership is independent of his having a majority in Congress—at least legally; for  as a matter of fact he is checkmated if he has none. Also, he can appoint and dismiss  cabinet officers (almost) at will. The latter can hardly be called ministers in the  English sense of the word and are really no more than the word ‘secretary’ conveys  in common parlance. We might say, therefore, that in a sense the pens Me is not  only prime minister but sole minister, unless we find an analogy between the functions  of an English Cabinet minister and the functions of the managers of the administra-  tion’s forces in Congress.   There is no difficulty about interpreting and explaining these and many other  peculiarities in this or any other country that uses the democratic method. But in  order to save space we shall mainly think of the English pattern and consider all  other cases as more or less important ‘deviations’ on the theory that thus far the logic  of democratic government has worked itself out most completely in the English practice  though not in its legal forms.    TWO CONCEPTS OF DEMOCRACY 179    vote does not directly produce government but an intermediate organ,  henceforth called parliament,! upon which the government-producing  function devolves. It might seem to account for the adoption or  rather the evolution of this arrangement, both on historical grounds  and on grounds of expediency, and for the various forms it took in  different social patterns. But it is nota logical construct; it isa natural  growth the subtle meanings and results of which completely escape  the official, let alone legal, doctrines.   How does a parliament produce government? The most obvious  method is to elect it or, more realistically, to elect the prime minister  and then to vote the list of ministers he presents. This method is  rarely used.? But it brings out the nature of the procedure better than  any of the others. Moreover, these can all be reduced to it, because  the man who becomes prime minister is in all normal cases the one  whom parliament would elect. The wayinwhich heis actually appoint-  ed to office, by a monarch as in England, bya President as in France  or by a special agency or committee as in the Prussian Free State of  the Weimar period, is merely a matter of form.   The classical English practice is this. After a general election the  victorious party normally commands a majority of seats in Parliament  and thus is in a position to carry a vote of want of confidence against  everyone except its own leader who in this negative way is designated  ‘by Parliament’ for national leadership. He receives his commission  from the monarch—kisses hands’—and presents to him his list of  ministers of which the list of cabinet ministers is a part. In this he  includes, first, some party veterans who receive what might be called  complimentary office; secondly, the leaders of the second rank, those  men on whom he counts for the current fighting in Parliament and  who owe their preferment partly to their positive political value and  partly to their value as potential nuisances; third, the rising men  whom he invites to the charmed circle of office in order to “extract  the brains from below the gangway’; and sometimes, fourth, a few  men whom he thinks particularly well qualified to fill certain offices.*    ‘It will be recalled that I have defined parliament as an organ of the state. Although  that was done simply for reasons of formal (legal) logic, this definition fits in par-  ticularly well with our conception of the democratic method: Membership in parlia-  ment is hence an office.   For example, it was adopted in Austria after the breakdown in 1918.   ’To lament, as some people do, how little fimess for office counts in these arrange-  ments is beside the point where description is concerned; it is of the essence of  democratic government that political valles should count primarily and fitness only  incidentally.    180 JOSEPH SCHUMPETER    But again, in all normal cases this practice will tend to produce the  same result as election by Parliament would. The reader will also  see that where, as in England, the prime minister has the actual power  to dissolve (‘to go to the country’), the result will to some extent  approximate the result we should expect trom direct election of the  cabinet by the electorate so long as the latter supports him.! This may  be illustrated by a famous instance.   2. In 1879, when the Beaconsfield (Disraeli) government, after  almost six years of prosperous tenure of power culminating in the  spectacular success of the Congress of Berlin,? was on all ordinary  counts entitled to expect a success at the polls, Gladstone suddenly  roused the country by a series of addresses of unsurpassable force  (Midlothian campaign) which played up Turkish atrocities so success-  fully as to place him on the crest of a wave of popular enthusiasm  for him personally. The official party had nothing to do with it. Several  of its leaders in fact disapproved. Gladstone had resigned the leader-  ship years before and tackled the country single-handed. But when  the liberal party under this impetus had wona smashing victory, it was  obvious to everyone that he had to be again accepted as the party  leader—nay, that he had become the party leader by virtue of his  national leadership and that there simply was no room for any other.  He came into power in a halo of glory.    1If, as was the case in France, the prime minister has no such power, parliamentary  cotenes acquire so much independence that this parallelism between acceptance of a  man by parliament and acceptance of the same man by the electorate is weakened or  destroyed. This is the situation in which the parlour game of parliamentary politics  runs riot. From our standpoint this is a deviation from the design of the machine.  Raymond Poincaré was of the same opinion.   Of course, such situations also occur in England. For the Prime Minister’s power  to dissolve—strictly, his power to ‘advise’ the monarch to dissolve the House of  Commons—is inoperative either if his party’s inner circle sets its face against it or  if there is no chance that elections will strengthen his hold upon Parliament. That is  to say, he may be stronger (though pase still weak) in Parliament than he is in the  country. Such a state of things tends to develop with some regularity after a govern-  ment has been in power for some years. But under the English system this deviation  from design cannot last very long.    21 do not mean that the temporary settlement of the questions raised by the Russo-  Turkish War and the acquisition of the perfectly useless island of Cyprus were in  themselves such masterpieces of statesmanship. But I do mean that from the stand-  pan of domestic politics they were just the kind of showy success that would normally   atter the average citizen’s vanity and would greatly enhance the government’s  prospects in an atmosphere of jingo patriotism. In fact it was the general opinion that  Disraeli would have won if he had dissolved immediately on returning from Berlin.    TWO CONCEPTS OF DEMOCRACY 181    Now this instance teaches us a lot about the working of the demo-  cratic method. To begin with, it must be realized that it is unique only  in its dramatic quality, but in nothing else. It is the oversized specimen  of a normal genus. The cases of both Pitts, Peel, Palmerston, Disraeli,  Campbell Bannerman and others differ from it only in degree.   First, as to the Prime Minister’s political leadership.! Our example  shows that it is composed of three different elements which must not  be confused and which in every case mix in different proportions, the  mixture then determining the nature of every individual Prime  Minister’s rule. On the face of it, he comes into office as the leading  man of his party in Parliament. As soon as installed however, he  becomes in a sense the leader of Parliament, directly of the house of  which he is a member, indirectly also of the other. This is more than  an official euphemism, more also than is implied in his hold upon  his own party. He acquires influence on, or excites the antipathy of,    ‘It is characteristic of the English way of doing things that official recognition  of the existence of the Prime Minister’s office was deferred until 1907, when it was  allowed to appear in the official order of precedence at court. But it is as old as  democratic government. However, since democratic government was never introduced  by a distinct act but slowly evolved as part of a comprehensive social process, it is not  easy to indicate even an approximate birthday or birth eae There is a long stretch  that presents embryonic cases. It is tempting to date the institution from the reign  of William III, whose position, so much weaker than that of the native rulers had been,  seems to give colour to the idea. The objection to this however is not so much that  England was no ‘democracy’ then—the reader will recall that we do not define demo-  cracy by the extent of eects ae that, on the one hand, the embryonic case of  aay haa occurred under Charles II and that, on the other hand, William III never  reconciled himself to the arrangement and kept certain matters successfully in his own  hands. We must not of course confuse prime ministers with mere advisers, however  powerful with their sovereign and however firmly entrenched in the very centre of the  public power plant they may be—such men as Richelieu, Mazarin or Strafford for  instance. Godolphin and Harley under Queen Anne were clearly transitional cases.  The first man to be universally recognized at the time and by political historians  was Sir Robert Walpole. But he as well as the Duke of Newcastle (or his brother Henry  Pelham or both jointly) and in fact all the leading men down to Lord Shelburne  (including the elder Pitt who even as foreign secretary came very near to fulfilling our  requirements in substance) lack one or another of Me characteristics. The first full-  fledged specimen was the younger Pitt.   It is interesting to note that what his own time recognized in the case of Sir Robert  Walpole (and later in that of Lord Carteret Earl of Granville]) was not that here was an  organ essential to democratic government that was breaking through atrophic tissues.  On the contrary, public opinion felt it to be a most vicious cancer the growth of which  was a menace to the national welfare ahd to democracy—‘sole minister’ or ‘first  minister’ was a term of opprobrium hurled at Walpole by his enemies. This fact is  significant. It not only indicates the resistance new institutions usually meet with. It  also indicates that this institution was felt to be incompatible with the classic doctrine  of democracy which in fact has no place for political leadership in our sense, hence  no place for the realities of the position of a prime minister.    182 JOSEPH SCHUMPETER    the other parties and individual members of the other parties as  well, and this makes a lot of difference in his chances of success. In  the limiting case, best exemplified by the practice of Sir Robert Peel,  he may coerce his own party by means of another. Finally, though in  all normal cases he will also be the head of his party in the country,  the well-developed specimen of the prime ministerial genus will have  a position in the country distinct from what he automatically acquires  by heading the party organization. He willlead party opinion creative-  ly—shape it—and eventually rise toward a formative leadership of  public opinion beyond the lines of party, toward national leadership  that may to some extent become independent of mere party opinion.  It is needless to say how very personal such an achievement is and  how great the importance of such a foothold outside of both party and  Parliament. It puts a whip into the hand of the leader the crack of  which may bring unwilling and conspiring followers to heel, though  its thong will sharply hit the hand that uses it unsuccessfully.   This suggests an important qualification to our proposition that  in a parliamentary system the function of producing a government  devolves upon parliament. Parliament does normally decide who  will be Prime Minister, but in doing so it is not completely free. It  decides by acceptance rather than by initiative. Excepting patho-  logical cases like the French chambre, the wishes of members are not  as a rule the ultimate data of the process from which government  emerges. Members are not only handcuffed by party obligations. They  also are driven by the man whom they ‘elect’—driven to the act of  the ‘election’ itself exactly as they are driven by him once they have  ‘elected’ him. Every horse is of course free to kick over the traces and  it does not always run up to its bit. But revolt or passive resistance  against the leader’s lead only shows up the normal relation. And this  normal relation is of the essence of the democratic method. Gladstone’s  personal victory in 1880 is the answer to the official theory that  Parliament creates and cashiers government.!   ‘Gladstone himself upheld that theory strongly. In 1874, when defeated at the polls,  he still argued for meeting Parliament because it was up to Parliament to pass the  sentence of dismissal. This of course means nothing at all. In the same way he studiously  professed unbounded deference to the crown. One biographer after another has  marvelled at this courtly attitude of the great democratic leader. But surely Queen  Victoria showed better discernment than did those biographers if we may judge trom  the strong dislike which she displayed for Gladstone from 1879 on aid which the  biographers attribute simply to the baleful influence of Disraeli. Is it really necessary  to point out that professions of deference may mean two different things? The man  who treats his wife with elaborate courtliness is not as a rule the one to accept comrade-    ship between the sexes on terms of equality. As a matter of fact, the courtly attitude  is precisely a method to evade this.    TWO CONCEPTS OF DEMOCRACY 183    3. Next, as to the nature and role of the cabinet.’ It is a curiously  double-faced thing, the joint product of Parliament and Prime  Minister. The latter designates its members for appointment, as we  have seen, and the former accepts but also influences his choice.  Looked at from the party’s standpointitisanassemblage of subleaders  more or less reflecting its own structure. Looked at from the Prime  Minister’s standpoint it is an assemblage not only of comrades in  arms but of party men who have their own interests and prospects to  consider—a miniature Parliament. For the combination to come  about and to work it is necessary for prospective cabinet ministers  to make up their minds—not necessarily from enthusiastic love—to  serve under Mr. X and for Mr. X to shape his programme so that his  colleagues in the cabinet will not too often feel like ‘reconsidering  their position’, as official phraseology has it, or like going ona sit-  down strike. Thus the cabinet—and the same applies to the wider  ministry that comprises also the political officers not in the cabinet—  has a distinct function in the democratic process as against Prime  Minister, party, Parliament and electorate. This function of inter-  mediate leadership is associated with, but by no means based upon,  the current business transacted by the individual cabinet officers in the  several departments to which they are appointed in order to keep  the leading group’s hands on the bureaucratic engine. And it has only  a distant relation, if any, with ‘seeing to it that the will of the people  is carried out in each of them’. Precisely in the best instances, the  people are presented with results they never thought of and would  not have approved of in advance.   4. Again, as to Parliament. I have both defined what seems to me  to be its primary function and qualified that definition. But itmight  be objected that my definition fails to do justice to its other functions.  Parliament obviously does a lot of other things besides setting up  and pulling down governments. It legislates. And it even administers.  For although every act of a parliament, except resolutions and    1§till more than the evolution of the prime minister’s office, that of the cabinet is  blurred by the historical continuity that covers changes in the nature of an institution.  To this day the English cabinet is legally the operative part of the Privy Council, which  of course was an instrument of government in decidedly predemocratic times. But  below this surface an entirely different organ has evolved. As soon as we realize this  we find the task of dating its emergence somewhat easier than we found the analogous  task in the case of the prime minister. Though embryonic cabinets existed in the time  of Charles II (the ‘cabal’ ministry was one, and the committee of four that was formed  in connexion with Temple’s experiment was another), the Whig ‘junto’ under  William UI is a fair candidate for first place. From the reign of Anne on only minor  points of membership or functioning remain to disagree on.    184 JOSEPH SCHUMPETER    declarations of policy, makes ‘law’ in a formal sense, there are man  acts which must be considered as administrative measures. The budget  is the most important instance. To make it is an administrative  function. Yet in this country it is drawn up by Congress. Even where  it is drawn up by the minister of finance with the approval of the  cabinet, as it is in England, Parliament has to vote on it and by this  vote it becomes an act of Parliament. Does not this refute our theory?  When two armies operate against each other, their individual moves  are always centred upon particular objects that are determined by  their strategical or tactical situations. They may contend for a par-  ticular stretch of country or for a particular hill. But the desirability  of conquering that stretch or hill must be derived from the strategical  or tactical purpose, which is to beat the enemy. It would be obviously  absurd to attempt to derive it from any extra-military properties the  stretch or hill may have. Similarly, the first and foremost aim of  each political party is to prevail over the others in order to get into  ower or to stay in it. Like the conquest of the stretch of country or  the hill, the decision of the political issues is, from the standpoint of  the politician, not the end but only the material of parliamentary  activity. Since politicians fire off words instead of bullets and since  those words are unavoidably supplied by the issues under debate, this  may not always be as clear as it is in the ~ilitary case. But victory over  the opponent is nevertheless the essence of both games.!  Fundamentally, then, the current production of parliamentary  decisions on national questions is the very method by which Parlia-  ment keeps or refuses to keep a government in power or by which  Parliament accepts or refuses to accept the Prime Minister’s leader-  ship.? With the exceptions to be noticed presently, every vote is a vote    ‘Sometimes politicians do emerge from phraseological mists. To cite an example  to which no objection can be raised on the score of frivolity: no lesser politician  than Sir Robert Peel characterized the nature of his craft when he said after his   arliamentary victory over the Whig government on the issue of the latter’s policy  in Jamaica: ‘Jamaica was a good horse to start’. The reader should ponder over this.    2This of course pee to the pre-Vichy French and pre-Fascist Italian practice  just as much as to the English practice. It may however be called in question in the  case of the United States where defeat of the administration on a major issue does  not entail resignation of the President. But this is merely due to the fact that the  Constitution, which embodies a different political theory, did not permit parlia-  mentary practice to develop according to its logic. In actual fact this logic did not  entirely fail to assert itself. Defeats on major issues, though they cannot displace the  President, will in general so weaken his prestige as to oust him from a position of  leadership. For the time being this creates an abnormal situation. But whether he wins  or loses the subsequent presidential election; the conflict is then settled in a way that  does not fundamentally differ from the way in which an English Prime Minister deals  with a similar situation when he dissolves Parliament.    TWO CONCEPTS OF DEMOCRACY 185    of confidence or want of confidence, and the votes that are technically  so called merely bring out in abstracto the essential element that is  common to all. Of this we can satisfy ourselves by observing that the  initiative in bringing up matters for parliamentary decision as a rule  lies with the government or else with the opposition’s shadow cabinet  and not with private members.   It is the Prime Minister who selects from the incessant stream of  current problems those which he is going to make parliamentary  issues, that is to say, those on which his government proposes to  introduce bills or, if he is not sure of his ground, at least resolutions.  Of course every government receives from its predecessor a legacy  of open questions which it may be unable to shelve; others are taken  up as a matter of routine politics; it is only in the case of the most  brilliant achievement that a Prime Minister is in a position to impose  measures about a political issue which he has created himself. In any  case however the government’s choice or lead, whether free or not,  is the factor that dominates parliamentary activity. If a bill is brought  in by the opposition, this means that it is offering battle: such a move  is an attack which the government must either thwart by purloining  the issue or else defeat. If a major bill that is not on the governmental  menu is brought in by a group of the governmental party, this spells  revolt and it is from this angle and not from the extra-tactical merits  of the case that it is looked upon by the ministers. This even extends  to the raising of a debate. Unless suggested or sanctioned by the  government, these are symptoms of the government forces’ getting  out of hand. Finally, if a measure is carried by inter-party agreement,  this means a drawn battle or a battle avoided on strategical grounds.!   5. The exceptions to this principle of governmental leadership in  ‘representative’ assemblies only serve to show how realistic it is.  They are of two kinds.   ‘Another highly significant piece of English technique may be mentioned in this  connexion. A major bill is or was usually not proceeded with if the majority for it fell  to a very low figure on the second reading. This practice first of all recognized an  important limitation of the majority principle as actually applied in well-managed  democracies: it would not be correct to say that in a democracy the minority is always  compelled to surrender. But there is a second point. While the minority is not always  compelled to yield to the majority on the particular issue under debate, it is practically  eae ante oa were exceptions even to this—compelled to yield to it on the question  whether the cabinet is to stay in power. Such a vote on the second reading of a major  government measure may be said to combine a vote of confidence with a vote for shelv-  ing a bill. If the contents of the bill were all that mattered there would hardly be any  sense in voting for it if it is not to make the statute book. But if Parliament is primarily    concerned with keeping the cabinet in office, then such tactics become at once under-  standable.    186 JOSEPH SCHUMPETER    First, no leadership is absolute. Political leadership exerted accord-  ing to the democratic method is even less so than are others because  of that competitive element which is of the essence of democracy.  Since theoretically every follower has the right of displacing his  leader and since there are nearly always some followers who have a  real chance of doing so, the private member and—f he feels that he  could do with a bigger hat—the minister within and without the inner  circle steers a middle course between an unconditional allegiance to  the leader’s standard and an unconditional raising of a standard of  his own, balancing risks and chances with a nicety that is sometimes  truly admirable.! The leader in turn responds by steering a middle  course between insisting on discipline and allowing himself to be  thwarted. He tempers pressure with more or less judicious con-  cessions, frowns with compliments, punishments with benefits. This  game results, according to the relative strength of individuals and their  positions, in a very variable but in most cases considerable amount  of freedom. In particular, groups that are strong enough to make  their resentment felt yet not strong enough to make it profitable to in-  clude their protagonists and their programmes in the governmental  arrangement will in general be allowed to have their way in minor  questions or, at any rate, in questions which the Prime Minister can  be induced to consider as of minor or only sectional importance.  Thus, groups of followers or even individual members may occa-  sionally have the opportunity of carrying bills of their own and still  more indulgence will of course be extended to mere criticism or to  failure to vote mechanically for every government measure. But we  need only look at this in a practical spirit in order to realize, from the  limits that are set to the use of this freedom, that it embodies not  the principle of the working of a parliament but deviations from it.   Second, there are cases in which the political engine fails to absorb  certain issues either because the high commands of the government’s  and the opposition’s forces do not appreciate their political values  or because these values are in fact doubtful.? Such issues may then    ‘One of the most instructive examples by which the above can be illustrated is  afforded by the course taken by Joseph Chamberlain with respect to the Irish question  in the 1880’s. He finally outmanoeuvered Gladstone, but he started the campaign while  officially an ardent adherent. And the case is exceptional only in the force and  brilliance of the man. As every political captain knows, only mediocrities can be  counted on for loyalty. That is why some of the greatest of those captains, Disraeli  for instance, surrounded themselves by thoroughly second-rate men.    *An issue that has never been tried out is the typical instance of the first class. The  typical reasons why a government and the shadow cabinet of the opposition may tacitly  agree to leave an issue alone in spite of their realizing its potentialities are technical  difficulty of handling it and the fear that it will cause sectional difficulties.    TWO CONCEPTS OF DEMOCRACY 187    be taken up by outsiders who prefer making an independent bid for  power to serving in the ranks of one of the existing parties. This of  course is perfectly normal politics. But there is another possibility.  A man may feel so strongly about a particular question that he may  enter the political arena merely in order to have it solved in his way  and without harbouring any wish to start in on a normal political  career. This however is so unusual that it is difficult to find instances  of first-rank importance of it. Perhaps Richard Cobden was one. It  is true that instances of second-rank importance are more frequent,  especially instances of the crusader type. But nobody will hold that  they are anything but deviations from standard practice.  ~ We may sum up as follows. In observing human societies we do not  as a rule find it difficult to specify, as least in a rough common-sense  manner, the various ends that the societies under study struggle to  attain. These ends may be said to provide the rationale or meaning  of corresponding individual activities. But it does not follow that  the social meaning of a type of activity will necessarily provide the  motive power, hence the explanation of the latter. If it does not, a  theory that contents itself with an analysis of the social end or need  to be served cannot be accepted as an adequate account of the activi-  ties that serve it. For instance, the reason why there is such a thing as  economic activity is of course that people want to eat, to clothe them-  selves and so on. To provide the means to satisfy those wants is the  social end or meaning of production. Nevertheless we all agree that  this proposition would make a most unrealistic starting point fora  theory of economic activity in commercial society and that we shall  do much better if we start from propositions about profits. Similarly,  the social meaning or function of parliamentary activity is no doubt  to turn out legislation and, in part, administrative measures. But  in order to understand how democratic politics serve this social end,  we must start from the competitive struggle for power and office and  realize that the social function is fulfilled, as it were, incidentally—  in the same sense as production is incidental to the making of profits.  6. Finally, as to the role of the electorate, only one additional  point need be mentioned. We have seen that the wishes of the  members of a parliament are not the ultimate data of the process  that produces government. A similar statement must be made con-  cerning the electorate. Its choice—ideologically glorified into the  Call from the People—does not flow from its initiative but is being  shaped, and the shaping of it is an essential part of the democratic  process. Voters do not decide issues. But neither do they pick their  members of parliament from the eligible population with a perfectly    188 JOSEPH SCHUMPETER    open mind. In all normal cases the initiative lies with the candidate  who makes a bid for the office of member of parliament and such local  leadership as that may imply. Voters confine themselves to accepting  this bid in preference to others or refusing to accept it. Even most of  those exceptional cases in which a man is genuinely drafted by the  electors come into the same category for either of two reasons:  naturally a man need not bid for leadership if he has acquired leader-  ship already; or it may happen that a local leader who can control  or influence the vote but is unable or unwilling to compete for election  himself designates another man who then may seem to have been  sought out by the voters acting on their own initiative.   But even as much of electoral initiative as acceptance of one of the  competing candidates would in itself imply is further restricted by  the existence of parties. A party is not, as classical doctrine (or  Edmund Burke) would have us believe, a group of men who intend to  promote public welfare ‘upon some principle on which they are all  agreed’. This rationalization is so dangerous because it is so tempting.  For all parties will of course, at any given time, provide themselves  with a stock of principles or planks and these principles or planks  may be as characteristic of the party that adopts them and as im-  portant for its success as the brands of goods a department store sells  are characteristic of it and important for its success. But the depart-  ment store cannot be defined in terms ofits brands and a party cannot  be defined in terms of its principles. A party is a group whose members  propose to act in concert in the competitive struggle for political  power. If that were not so it would be impossible for different parties  to adopt exactly or almost exactly the same programme. Yet this  happens as everyone knows. Party and machine politicians are simply  the response to the fact that the electoral mass is incapable of action  other than a stampede, and they constitute an attempt to regulate  political competition exactly similar to the corresponding practices    of a trade association. The psycho-technics of party management  and party advertising, slogans and marching tunes, are not accessor-  ies. They are of the essence of politics. So is the political boss.    Xx  JUSTICE AND THE COMMON GOOD    BRIAN BARRY  I    Social Principles and the Democratic State, by S. 1. Benn and R. S. Peters  (George Allen & Unwin, London, 1959) is far more than a textbook;  for the authors’ object is not merely to say, ‘Justice means this,  equality means that, freedom means the other. Where they conflict you  take your pick’. Their thesis is the daring one that all political  arguments fit into a single pattern and that this pattern is identical  with morality.   The authors reject any sociological definition of ‘moral rule’.  According to them, a rule is moral if and only if it is:    critically accepted by the individual in the light of certain criteria. The  criteria can be summarized by saying that a rule should be considered in the  light of the needs and interests of people likely to be affected by it with no  partiality towards the claims of any of those whose needs and interests are at  stake. (p. 56)   People have interests and needs which they put forward as claims. The  criteria of impersonality and respect for persons are satisfied when claims  are assessed on relevant grounds, and privileges excluded as a basis for  allowing a claim. (p. 51)    I think the inconvenience of this definition is fairly clear: it follows from  it, for example, that one can never talk about the moral rules of a  group unless one has first ascertained whether everyone who acts on  the rules and expects others to act on them has accepted them after an  impartial consideration of their effects. The theoretical disadvantages  of making stipulative restrictions on the expression ‘moral rules’ are  very similar to the disadvantages (which the authors recognize) of  calling only good laws ‘laws’.   My main objection, however, is that no single scheme can be suf-  ficient to cover all the arguments which even in liberal-democratic    From Analysis, Vol. 21 (Blackwell, 1960-61), pp. 86-g0. Reprinted by permission  of the author, Analysis, and Basil Blackwell.    fy    190 BRIAN BARRY    communities would ordinarily be thought ‘moral’. If Benn and Peters  do succeed in interpreting their schema so as to fitany argument pre-  sented into it, this is ata cost of both misrepresenting most arguments  and making the schema vacuous.    II    This must be so, I suggest, because when we are dealing with  interests there are two conflicting principles at work: aggregative and  distributive. They are both, it seems to me, independently operative  in most men’s minds; and where they give conflicting answers there is  no higher principle to which the conflict can be referred.   Suppose, for example, that one can see no reason in terms of desert  or need why science teachers should be paid more than others; but  that one also believes that unless all the money available for raising  pay goes to them the standard of science teaching will decline with  grave results. Or again, suppose that one believes on the one hand  that dropping two atomic bombs on Japan will cause less suffering  than continuing the war with ‘conventional’ weapons, but that it is  unjust to use weapons which rely on devastating civilian populations  rather than military targets. In both cases we have a conflict between  an aggregative result and a distributive principle. If these were all  the relevant considerations (which in fact they are not) then how one  decided would depend on how highly one ranked the two kinds of  principle (moral philosophers can be divided according to which side  they exaggerate at the expense of the other).   The point I wish to make is that ‘impartiality’ is no help here in  providing a schema for decision. If it means ‘everyone to count for one  and nobody for more than one’, this is satisfied by the result of aggre-  gation, which may still be morally unacceptable to many people. Ifon  the other hand one is to say, ‘The arts teachers aren’t given enough con-  sideration by straight aggregation; they must be given some more’,  my objection is that the formula is now useless as a guide to decision.  It merely gives us a thoroughly misleading way of justifying ex post  facto whatever decision we may in fact reach.    Ill    There is still, however, one possible escape route for the authors’  formula. This lies in emphasizing the bit about ‘relevant grounds for  treating people differently’. But this manoeuvre fails in one of two    JUSTICE AND THE COMMON GOOD 191    ways, depending on the construction of ‘relevant’. If you say that what  subject someone teaches is not a relevant criterion in determining  his pay, you certainly get the answer that scientists ought not to be  paid more. But this is too good a demonstration, for according to the  authors, anyone who admits this and still says that science teachers  ought to be paid more is not arguing morally; yet such a position  seems to me perfectly reasonable. Some one who takes this view has  chosen in favour of the education of children and against professional  equity—surely a perfectly reputable thing to do. If on the other hand  you say that what subject someone teaches is made relevant simply by  the fact that science teachers are in shorter supply than others and  paying them more will keep up educational standards, you are right  back at a simple aggregationist position.   In fact the authors give some support to each interpretation. Thus,  on p. 112, they say‘... it is up to whoever should make distinctions  to justify the criteria in terms . . . ultimately of a balance ofadvantage  to all concerned’. Phrases such as ‘beneficent consequences’ and  ‘beneficial results’ occur in a similar context on pp. 169 and 170.  This sounds like a straight aggregationist position: you show that  something is just, or for the common good, by showing that it makes  relevant distinctions, and these distinctions are relevant if they  provide ‘a balance of advantage to all concerned’. This form of  argument is certainly used by the authors and I shall discuss an  example in section IV.   Against this, they say on pp. 272-3:    Two politicians may each say, with perfect sincerity, that he is seeking  the public interest, or the common good, though one proposes to ex-  propriate private capital and the other to defend it for the death. Does one  of them have to be wrong... ? Is the disagreement about fact at all? It may  be...but the probability is that this is less important than a disagree-  ment on moral principle. One holds private capital to be an immoral thing  in itself, the other that it represents the legitimate fruits of thrift, industry,  and other economic virtues..,. What then have the two politicians in  common that enables them to appeal, with equal sincerity to ‘the common  good’? ... (Each) is saying, in effect, that having considered the claims of  all sections in a spirit of impartiality, the balance of advantage lies in the  course he recommends.    Here it is quite clear that ‘the balance of advantage’ is not something  obtainable by aggregation; it is simply a repetition of the procedural  point that one must have considered all claims impartially. To say on  thus analysis that relevant distinctions are those based on a balance    192 BRIAN BARRY    of advantage is to add nothing. ‘We have to decide what is relevant’ (p.  113).   My object in pointing out this inconsistency is not to score a  cheap debating point but to substantiate my view that it is impossible  to fit into one theory questions of distributive principle (the sort of  thing the two politicians in the argument are disagreeing about) and  aggregative ones (which phrases such as ‘balance of advantage’ would  naturally be thought to refer to).    IV    In this section I shall apply the above analysis to two of the  discussions of particular questions in the book to illustrate how the  authors’ insistence on trying to show all concepts as aspects of a single  criterion leads them to distort characteristic forms of argument. In  the first example, we see a straight question of distribution obscured  by reference to an aggregative concept. On p. 272, the authors say:    ...the government would have resented being told in 1957 that decon-  trolling rents was not for the common good. But the government clearly  had to choose between the interests of the landlords and the interest of the  tenants. Whether or not it chose rightly, it did little good to the tenants.    Now if it is correct that the issue is one where one side’s losses are  the other side’s gains, I would suggest that the ‘common good’ is out  of place. Of course the government would resent being told its action  was not for the common good; but neither would it justify the measure  by saying it was for the common good. The concept is out of place. If  the government is willing to admit that it is simply transferring money  from one set of pockets to another, it will say, for example, that  control was unfair between owners of different forms of property, or  between owner-occupiers and tenants, or that it was unjust between  landlords and tenants; i.e., it would support a distributive change  by distributive arguments. (More likely, of course, it would deny that  only one side would be benefited, pointing to the benefits of a free  market in producing a rational allocation of resources, and then it  could talk about the common good; but I follow the authors’ assump-  tion for the sake of the argument.)   What I am trying to show is that it is not an accident that we have  different concepts; they really do have different jobs. We have one set  which point out various distributive comparisons, such as justice,  fairness, equity, equality (that these all differ is not hard to show but    JUSTICE AND THE COMMON GOOD 193    it would take me out of my way here. One example: a lottery is fair if  honestly run, but a lottery which distributed prizes justly, i.e.,  according to desert or need, would no longer be fair). And we have  another set which point out the results of various methods of aggrega-  tion, such as ‘public interest’, ‘common good’ and ‘general welfare’  (‘good’ and ‘interest’ for example require one to include different  ways in which people are affected). To say, as Benn and Peters do,  that ‘to seek the common good’ means ‘to try to act justly’, is to make  nonsense of the subtle and complex way in which we go about criti-  cizing political programmes, in the pursuit of a tidy but barren  theory.   My second example shows the same error operating in reverse. In  chapters 4 and 5, the authors have an excellent study of the grounds  on which one might justify various claims to income. Unfortunately,  however, they are again hampered by their theoretical apparatus, for  all claims, according to their general theory, must be established  by being shown to be just. This works excellently for claims based on  personal desert and need; but it does not work at all for arguments  for property based on the advantage of having a group of politically  independent or cultured citizens (pp. 167-70):    If it were true that fortunes based on inheritable property were indis-  pensable for an elite of this sort, and if such an elite were really so valuable,  the property system would be justified by its beneficient consequences.  (p. 169) ; .    Now, this is fine; and the situation should in my view be summed  up by saying that in this case justice would have to be qualified by  utilitarian considerations.   But this course is not open to Benn and Peters. They have to say  that the general advantage of property makes the amount of money  your father had a relevant and therefore just ground for differences  in income (pp. 169-70). This seems to me highly misleading.  Although Hume used the expression ‘rules of justice’ to cover  precisely such things as property rules, ‘justice’ is nowadays  analytically tied to ‘desert’ and ‘need’, so that one could quite  properly say that some of what Hume called ‘rules of justice’ were  unjust. Again, we see how the attempt to reduce all arguments to  one pattern forces Benn and Peters to assimilate quite different  kinds of arguments to one another.    NOTES ON THE CONTRIBUTORS    JOHN PLAMENATzZ is Chichele Professor of Social and Political Theory  at Oxford and a Fellow of All Souls College. He was formerly a Fellow  of Nuffield College. His German Marxism and Russian Communism was  published in 1961, and Man and Society, in two volumes, in 1963.    P. H. PARTRIDGE teaches at the Australian National University,  Canberra.    H. L. A. HART has been Professor of Jurisprudence in Oxford since  1952. Among his publications are Causation in the Law (1961), and Law,  Liberty, and Morality, (1963).    STANLEY I. BENN, until recently Lecturer in Government at the  University of Southampton, is now at the Australian National  University, Canberra. He is the author, with R. S. Peters, of Social  Principles and the Democratic State (1959).    R. S. PETERS is Professor of the Philosophy of Education in the  University of London Institute of Education. Among his publications  are The Concept of Motivation (1958), Soctal Principles and the Democratic  State (with S. I. Benn, 1959), and Ethics and Education (1966).    PETER WINCH, formerly of the University College of Swansea, is now  Professor of Philosophy at King’s College, London; his The Idea of a  Social Science was published in 1958.    BRIAN BARRY has been a Fellow of Nuffield College, Oxford, since  1966. His book Political Argument was published in 1965.    E. F. CARRITT, who’ died in 1963, was for many years a Fellow of  University College, Oxford. His book Ethical and Political Thinking was  published in 1947, and he wrote also on the philosophical problems of  aesthetics.    Sir IsataH BERLIN is President of Wolfson College, Oxford. He was  Chichele Professor of Social and Political Theory at Oxford from  1957 to 1967. Among his publications are The Hedgehog and the Fox  (1953), a study of Tolstoy, and Karl Marx (2nd. edn., 1956).    JOSEPH A. SCHUMPETER went to Harvard from his native Austria in  1932, and taught there for many years before his death in 1950. He  published a number of widely influential works in the field of econo-  mics.    BIBLIOGRAPHY    (not including material in this volume)    I. General and Methodological   The best elementary introduction to political theory is J. D.  Mabbott’s The State and the Citizen (Hutchinson, London, 1948). ©  Charles Vereker’s The Development of Political Theory (Hutchinson,  London, 1957) gives an overall survey of the historical development  of the subject. S. I. Benn and R. S. Peters, in Social Principles and the  Democratic State (Allen and Unwin, London, 1959), cover the main  topics of political theory from the point of view of analytic philosophy.  Arnold Brecht’s massive Political Theory (Princeton U.P., Princeton,  1959) does the same thing in a more Teutonic and elaborately scholarly  way.   G. H. Sabine’s History of Political Theory (Harrap, London, 1937)  deserves its reputation as a model text-book, being lucid, thorough  and reliable to a very high degree. J. P. Plamenatz’s Man and Society  (Longmans, London, 1963) is confined to the major figures in the  history of political thought but subjects their ideas to a full critical  examination. Sheldon Wolin’s Politics and Vision (Allen and Unwin,  London, 1961) is more interpretative again and is as much concerned  with the historical setting as with the logical cogency of political  theories; a most impressively intelligent and original book.   A useful, if somewhat mechanically written, survey of methodolo-  gical issues is The Study of Political Theory by Thomas P. Jenkin (Random  House, New York, 1955). The relations between political philosophy  and political science are thoroughly and penetratingly examined with  a wealth of references in David Easton’s The Political System (Knopf,  New York, 1953). T. D. Weldon’s The Vocabulary of Politics (Penguin  Books, London, 1953) expresses the hostility of some analytic philo-  sophers to political theory with the artless enthusiasm ofaconvert. At  the opposite extreme is the defence of the traditional attitude, which  takes political theory to be concerned to demonstrate the timeless,  essential nature of the state, in Leo Strauss’s What is Political Philosophy  (Free Press, Glencoe, 1959). G. H. Sabine’s ‘What is political theory ?  in the Jowmal of Politics (1939), and Isaiah Berlin’s ‘Does political theory  still exist?’, in Philosophy, Politics and Society, second series, ed. Laslett  and Runciman (Blackwell, Oxford, 1962) are noteworthy brief general    196 BIBLIOGRAPHY    statements. Cf. also J. D. Mabbott, ‘Political Concepts’ in Philosophy,  (1938); J. M. CameronandT. D. Weldon, ‘The Justification of Political  Attitudes’ in the Proceedings of the Aristotelian Society, Supplementary  Volume (1955); H. B. Action ‘Political Justification’ in Contemporary  British Philosophy, Third Series, ed. H. D. Lewis (Allen and Unwin,  London, 1956).   R. M. Maclver’s The Web of Government (Macmillan, New York, 1947)  is a large, reflective survey of the field of political science. Robert A.  Dahl’s Modern Political Analysis (Prentice-Hall, New Jersey, 1963) is an  elementary introduction to political science with a strong methodo-  logical emphasis. S. M. Lipset’s Political Man (Doubleday, New York,  1960) is a notable example of political sociology.    II. State, law and morality   A definition of the political is developed in chapter 2 of Robert  A. Dahl’s Modern Political Analysis (Prentice-Hall, New Jersey, 1963).  For sovereignty and related issues see S. 1. Bennand R. S. Peters, Social  Prinaples and the Democratic State (Allen and Unwin, London, 1959),  chapters 11, 12 andiz- A.D. Lindsay’s ‘Sovereignty’ in Proceedings of  the Aristotelian Society (1923-4); W. J. Rees’s ‘The Theory of Sovereignty  Restated’ in Mind (1950) and in Philosophy, Politics and Society, first series,  ed. Laslett (Blackwell, Oxford, 1956).   A classical and much-discussed definition of law is to be found in  H. Kelsen’s General Theory of Law and the State (Harvard 1945). There isa  most delicate and penetrating discussion of the subject in H. L. A.  Hart’s The Concept of Law (Clarendon Press, Oxford, 1961). See also  Benn and Peters, Op. cit., chapter 3.   On rights in general and natural rights in particular see E. F.  Carritt, Morals and Politics (Clarendon Press, Oxford, 1935), chapter  13, and also his Ethical and Political Thinking (Clarendon Press,  Oxford, 1947); J. P. Plamenatz, Consent, Freedom and Political Obliga-  tion (Clarendon Press, Oxford, 1938); A. C. Ewing, ‘The Rights of the  Individual against the State’ in Revue Internationale de Philosophie (1948),  and in chapter 2 of his Individual, State and World Government (Mac-  millan, New York, 1947); M. Macdonald, ‘Natural Rights’, in Proceed-  ings of the Aristotelian Society (1946-7), and in Philosophy, Politics and  Society, first series, ed. Laslett (Blackwell, Oxford, 1956); J. D. Mabbott  The State and the Citizen (Hutchinson, London, 1948), part B; A. P.  d’Entréves Natural Law (Hutchinson, London, 1951); J. P. Plamenatz  and H. B. Acton, ‘Rights’, in Proceedings of the Aristotelian Society,  Supplementary Volume (1950); A. I. Melden and W. K. Frankena,    BIBLIOGRAPHY 197    ‘Human Rights’, in Proceedings of the American Philosophical Association,  Eastern Division, volume 1 (1952); S$. M. Brown and W. K. Frankena,  in Philosophical Review (1955) (commenting on chapter 3 of this antho-  logy, Hart’s ‘Are There Any Natural Rights ?’); Benn and Peters, op.  cit., chapter 4; R. B. Brandt, Ethical Theory (Prentice-Hall, New Jersey,  1959) chapter 17; J. Hospers, Human Conduct (Harcourt Brace, New  York, 1961), chapter 19.    III. Political Obligation   The first chapter of J. P. Plamenatz’s Consent, Freedom and Polttical  Obligation (Clarendon Press, Oxford, 1938) on consent is an excellent  example of the clarificatory power of philosphical analysis in the field  of political ideas. Chapter 7 of the same book deals with political  obligation. Theories of obligation are discussed in E. F. Carritt’s  Morals and Politics (Clarendon Press, Oxford, 1935), chapter 14, and  in his Ethical and Political Thinking (Clarendon Press, Oxford, 1947),  part II, chapter 14. Benn and Peters, op. cit., chapter 14, give a brief  and convenient survey of different grounds of political obligation.   The social contract theory is reinterpreted in a somewhat far-  fetched way in H. D. Lewis’s ‘Is there a Social Contract?’ in Philosophy  (1940). A survey of the history of the contract theory from its first  beginnings in Greek thought is given in J. W. Gough, The Social  Contract (Clarendon Press, Oxford, 1936).   Discussions of the general will theory are more numerous. See  H. D. Lewis ‘Natural Rights and the General Will’ in Mind (1937);  J. P. Plamenatz, Constant, Freedom and Political Obligation (Clarendon  Press, Oxford, 1938) chapters 2 and 3; J. D. Mabbott, The State and The  Citizen (Hutchinson, London, 1948), part D; G. R. G. Mure, ‘The  Organic State’ in Philosophy (1949); B. Mayo, ‘Is there a case for the  general will?’, in Philosophy (1951), and in Philosophy, Politics and  Society, first series, ed. Laslett (Blackwell, Oxford, 1956); B. Blanshard,  Reason and Goodness (Allen and Unwin, London, 1961), chapter 14.   Traditionalism is examined in H. B. Action, ‘Tradition and some  other forms of order’, in Proceedings of the Aristotelian Soctety (1952-3),  and supported in M. Oakeshott, Rationalism in Politics (Methuen,  London, 1962).    IV. The ends of government   J. R. Pennock’s Liberal Democracy (Rinehart, New York, 1950) is a  good general survey of liberal ideals, from the point of view of both  meaning and justification, and contains a most valuable bibliography.    198 BIBLIOGRAPHY    As a counterpoise mention may be made of a piece of writing much  older than any other included here, J. F. Stephen’s Liberty, Equality,  Fraternity (Smith Elder, London, 1873), both on account of its analytic  clarity and the trenchancy of its opinions. B. M. Barry, Political  Argument (Routledge, London, 1965) is a recent work of comparable  scope, much greater philosophical technicality and utterly different  opinions.   Liberty is defined in chapter 5 and justified in chapter 6 of J. P.  Plamenatz, Consent, Freedom and Political Obligation (Clarendon Press,  Oxford, 1938) with the author’s characteristic clarity and definiteness.  C. I. Lewis’s ‘The Meaning of Liberty’, in Revue Internationale de  Philosophie (1948), is brief but all-embracing. See also Benn and  Peters, op. cit., chapter 10.   On democracy see R. A. Dahl, Preface to Democratic Theory (Chicago  U. P., 1961); Benn and Peters, op. cit., chapter 15; J. Hospers, Human  Conduct (Harcourt Brace, New York, 1961) chapter 18; R. Wollheim,  ‘A Paradox in the Theory of Democracy’, in Philosophy, Politics and  Society, second series, ed. Laslett and Runciman (Blackwell, Oxford,  1962); C. B. Macpherson, The Real World of Democracy (Clarendon  Press, Oxford, 1966).   On justice and equality see H. Spiegelberg in Philosophical Review  (1944); D. D. Raphael, ‘Equality and Equity’, in Philosophy (1946),  and his ‘Justice and Liberty’ in Proceedings of the Aristotelian Society  (1950-51). I. Berlin and R. Wollheim, ‘Equality’, in Proceedings of the  Anstotelian Society, Supplementary Volume, (1955); R. B. Brandt,  Ethical Theory (Prentice-Hall, New Jersey, 1959), chapter 16; J. Rawls,  ‘Justice as fairness’, in Philosophical Review (19 58); J. Hospers, Human  Conduct (Harcourt Brace, New York, 1961), chapters 20-22; Bennand  Peters, op. cit., chapters 5 and 6; B. A. O. Williams, ‘The idea of  equality’ in Philosophy, Politics and Society, second series, ed. Laslett  and Runciman (Blackwell, Oxford, 1962).    INDEX OF NAMES  (not including authors mentioned only in the Bibliography)    Acton, Lord, 128   Aristotle, 1, 54n., 136   Arnold, Matthew, 132   Austin, John, 6, 67, 95   Bagehot, W., 34, 128   Barker, Sir Ernest, 68n., 6gn., 71n.  Barry, B. M., 3, 13, 18   Benn, S. I., 3, 9, 115, 123, 189-93  Bentham, Jeremy, 7, 138n., 139n.  Bentley, A. F., 113   Berlin, Sir Isaiah, 17, 52n.   Black, D., 122   Bodin, J., 67   Bosanquet, B., 34, 133n.   Bradley, F. H., 133n.   Braybrooke, D., 32n.   Brecht, A., 32n.   Bryce, Lord, 67, 79n., 160n.  Burke, Edmund, 24, 25, 46, 127, 139, 145, 188    Campbell, A. H., 58n.  Carritt, E. F., 18   Cartland, R., 127, 131  Catlin, G. E. G., 32n.  Cobban, A., 32n.  Collingwood, R. G., 100, 101  Condorcet, 147   Constant, Benjamin, 143, 145  Crosland, A., 39    Dahl, R. A., 35, 41, 43, 46n., 50-52  Descartes, 109  Dicey, A. V., 67  Dodds, E. R., 88  Downs, E., 46n, 47N., 49n.  Duguit, L., 71n.  Durkheim, E., 21  Engels, F., 30  199    200 INDEX    Erasmus, 146    Field, G. C., 32n.  Freud, S., 29n., 160    Gallie, W. B., 4   Gellner, E., 88n.   Godwin, W., 140   Gray, J. C., 6gn.   Greaves, H. B., 32n., 36n.   Green, T. H., 34, 38n., 133n., 150, 151n.    Hampshire, S., 53n.   Hare, R. M., 123   Harrison, W., 32n.   Hart, H. L.A., 9   Hayek, F. A., 2   Hegel, 1, 7, 12, 13, 22, 33, 34, 52M., 133, 142   Helvetius, 142n.   Hermens, F. A., 177n.   Hobbes, 1, 12, 15, 33, 34, 83-6, 91, 95, 97-8, 102, 109, 110, 119,  131, 143N., 145   Hohfeld, W. N., 55n., 64n.   Homans, G. C., 106n.   Horsburgh, H., 45n.   Hume, 10, 11, 110, 129, 135N., 137N., 139N., 193    Jaffa, H. V., 32n.   James, William, 152, 165n.   Jefferson, 145, 169   Jouvenel, B. de, 83-4, 86, 98, 101-2, 104, 106  Kant, 56   Kelsen, H., 69n., 73n., 75~7   Lamont, W. D., 55n.   Laslett, P., 32n.   Lassalle, F., 145   Le Bon, G., 160, 161   Lecky, W. E., 128, 131   Lincoln, Abraham, 169, 170n.   Lindsay, Lord, 68n.   Locke, 1, 5, 7, 12, 33, 110, 115, 143, 144  Macdonald, M., 38n.   Machiavelli, 1   Maitland, F. W., 129, 133n., 136n., 137N., 140  Marshall, G., 70   Marx, 2, 23, 24, 30, 40n.    INDEX 201    May, Erskine, 128   Mill, James, 129   Mill, J. S., 1, 17, 50, 143-8  Miller, J. D. B., 113-5  Morris Jones, W. H., 41n.  Murdoch, Iris, 39   Myrdal, G., 39   Niebuhr, R., 43    Paine, Thomas, 145   Pareto, V., 160   Peters, R. S., 3, 13, 97-9, 101, 103, 105-10, 123, 189-93  Plamenaw, J., 3, 5, 32N., 36n.   Plato, 1, 33, 54n., 88, 129   Polanyi, M., 44   Popper, Sir Karl, 4, 41, 44    Rees, J. C., gan.   Rees, W. J., 67, 68n., 6gn., 71n., 74n., 77-8, 80  Ribot, T., 160   Ross, Sir David, 133n.   Rostow, W. W., 48n.   Rousseau, 1, 12, 13, 15, 22, 34, 119-24, 142    Salmond, J. W., 6gn.   Schelling, T., 120   Schubert, G., 112   Schumpeter, J. A., 17, 46-50, 51  Shils, E., 41, 497.   Shkiar, J., 36n.   Smith, Adam, 144   Sorauf, F. J., 112   Stephen, Sir J. F., 128, 129, 146  Strachey, J., 39   Strauss, R., 32n.   Tocqueville, A. de, 34, 130, 131, 143  Truman, D. B., 113    Villey, M., 147n.    Wallas, G., 34, 160n., 165n.   Warrender, H., 85, 93   Weber, M., 6, 86-8, 100n., 101, 106, 107  Weldon, T. D., 2, 22, 23, 33, 92   Winch, P. G., 13   Wittgenstein, L., 99   Wollheim, R. A., 72n.