THE LIBRARY OF THE UNIVERSITY OF CALIFORNIA LOS ANGELES FROM THE LIBRARY OF ERNEST CARROLL MOORE THE SCOPE AND METHOD OF POLITICAL ECONOMY THE SCOPE AND METHOD OF POLITICAL ECONOMY BY JOHN NEVILLE KEYNES M.A. TJNIVEB8ITY LECTURER IN MORAL SCIENCE AND LATE FELLOW OF PEMBROKE COLLEGE IN THE UNIVERSITY OF CAMBRIDGE LATE EXAMINER IN POLITICAL ECONOMY IN THE UNIVERSITY OF LONDON Uonfcon MACMILLAN AND CO. AND NEW YORK 1891 [The Eight of Translation and Reproduction is reserved] Cambridge PRINTED BY C. J. CLAY M.A. AND SONS AT THE UNIVERSITY PRESS Library PREFACE. THE nature of the topics discussed in the following pages is sufficiently indicated in the introductory chapter, and a lengthy preface is therefore unnecessary. The abstract discussion of methods may appear to some to have mainly an academic interest, since it does not directly extend our knowledge of economic pheno- mena. Whilst, however, we ought to be upon our guard against allowing any such discussion to obscure the greater importance of actual economic investigations, the subject is one to which all students of economics must necessarily give some attention in the course of their reading, and its indirect bearing on the solution of practical economic questions is very far indeed from being without importance. Unfortunately almost every problem connected with the scope and method of poli- tical economy has given rise to conflict of opinion ; and the resulting controversies have sometimes been very VI PREFACE. bitter. Those readers, therefore, who already have any acquaintance with the literature of economic method, will be prepared to find that several of the chapters are more or less controversial in character. At the same time, I have endeavoured to avoid the tone of a partisan, and have sought, in the treatment of disputed questions, to represent both sides without prejudice. Whilst making no attempt to bring about a complete reconciliation between opposing views, I have been able to shew that the nature of the opposi- tion between them has sometimes been misunderstood, and its extent consequently exaggerated. Since the scope and method of a science can never be satisfactorily discussed at the commencement of its study, some knowledge of political economy in its general outlines is presupposed. As far as possible, however, illustrations of a fairly simple and familiar kind have been chosen. A good many illustrations that were included in the first draft of the book have been omitted, partly because they would have occupied too much space if given with any completeness, and partly in order to avoid points of controversy not essen- tially connected with the subject immediately under discussion. A certain amount of repetition has resulted from the frequent necessity of treating the same pro- blem from more than one point of view, and from the fact that the different questions to which the con- PREFACE. Vll sideration of economic method gives rise are in so many ways connected one with another. I have not hesitated to repeat the same thing several times in different connexions, if clearness seemed to be gained thereby. By means of quotations and references I have endeavoured to make clear my indebtedness to other writers ; and it is, therefore, for the most part unneces- sary to specify here the various sources from which I have derived assistance. To the works both of Pro- fessor Marshall and Professor Sidgwick, however, I am indebted in ways that it is impossible to identify and separately indicate. I am further under obligation to Professor Marshall, and also to Mrs Marshall, Mr W. E. Johnson, and Professor Nicholson, for their great kind- ness in reading the proof sheets of the book while it has been passing through the press. Their criticisms and suggestions have been most valuable, and have enabled me in many ways to improve my treatment of the subject. J. N. KEYNES. G, HARVEY ROAD, CAMBRIDGE, 12 December, 1890. CONTENTS. CHAPTER I. INTRODUCTORY. PAGE 1. Nature and importance of the enquiry into the scope and method of political economy .... 1 2. The conception of political economy as a theoretical, abstract, and deductive science .... 9 3. The conception of political economy as an ethical, realistic, and inductive science .... 20 4. The method of political economy cannot adequately be described by any single phrase ... 28 CHAPTER II. ON THE RELATION OF POLITICAL ECONOMY TO MORALITY AND PRACTICE. 1. Distinction between economic uniformities, economic ideals, and economic precepts . . . .31 X CONTENTS. PAGE 2. The possibility of studying economic laws or uni- formities without passing ethical judgments or formulating economic precepts .... 37 3. Grounds for recognising a distinct positive science of political economy, the sole province of which is to establish economic uniformities .... 46 4. Applied economics 53 5. Political economy and ethics 58 6. Methodological importance of the distinctions indi- cated in this chapter 61 NOTES TO CHAPTER II. A. ON POLITICAL ECONOMY AND Laisser Faire . . 65 B. ON THE SCOPE OF POLITICAL ECONOMY CONSIDERED AS AN ART 72 CHAPTER III. ON THE CHARACTER AND DEFINITION OF POLITICAL ECONOMY REGARDED AS A POSITIVE SCIENCE. 1. Political economy and physical science . . .81 2. Political economy and psychology .... 84 3. Political economy a social as distinguished from a political science 89 4. Definitions of Wealth and Economic Activity . . 90 5. Definition of Political Economy .... 97 NOTE TO CHAPTER III. ON THE INTERDEPENDENCE OF ECONOMIC PHENOMENA . 98 CONTENTS. XI CHAPTER IV. ON THE RELATION OF POLITICAL ECONOMY TO GENERAL SOCIOLOGY. PAGE 1. Conflicting views of the relation between economic science and the general science of society . . 108 2. The place of abstraction in economic reasoning . . Ill 3. Examples of economic problems requiring for their complete solution a realistic treatment. . .124 4. Distinction between political economy and other social enquiries 130 NOTES TO CHAPTER IV. A. ON THE DISTINCTION BETWEEN ABSTRACT AND CON- CRETE POLITICAL ECONOMY . . . .137 B. ON THE DISTINCTION BETWEEN THE STATICS AND THE DYNAMICS OF POLITICAL ECONOMY . .140 C. ON POLITICAL ECONOMY AND COMMON SENSE 142 CHAPTER V. ON DEFINITION IN POLITICAL ECONOMY. 1. The problem of definition in political economy . . 146 2. Conditions to be satisfied in framing economic defini- tions 152 3. Relativity of economic definitions . . . .159 Xll CONTENTS. CHAPTER VI. ON THE METHOD OF SPECIFIC EXPERIENCE IN POLITICAL ECONOMY. PAGE 1. Preliminary functions of observation in economic en- quiries 164 2. Limited scope for experiment in political economy . 169 3. The employment of the method of difference inde- pendently of deliberate experiment . . .178 4. The method of inductive generalization from a multi- plication of instances 190 5. Limitations of the empirical method . . .197 CHAPTER VII. ON THE DEDUCTIVE METHOD IN POLITICAL ECONOMY. 1. Nature of the deductive method .... 204 2. The application of the term " hypothetical " to eco- nomic science 205 3. Functions of observation in the employment of the deductive method 214 4. Kicardo's use of the deductive method . . . 222 5. The premisses of deductive political economy . .226 6. Special modifications of the deductive method . . 230 CHAPTER VIII. ON SYMBOLICAL AND DIAGRAMMATIC METHODS IN POLITICAL ECONOMY. 1. Mathematical character of political economy . . 236 CONTENTS. Xlll 2. The employment of arithmetical examples . . 239 3. Exact numerical premisses not essential to the em- ployment of mathematical methods . . . 240 4. Advantages resulting from the use of symbolical and diagrammatic methods in political economy . . 244 CHAPTER IX. ON POLITICAL ECONOMY AND ECONOMIC HISTORY. 1. Functions of economic history in theoretical investi- gations 252 2. Economic theories illustrated by history . . . 255 3. Economic theories criticized by history . . . 261 4. Economic theories established by history . . . 263 5. Functions of economic theory in historical investi- gations ". 269 6. Economic history and the history of economic theories 273 NOTES TO CHAPTER IX. A. ON THE LIMITS OF THE VALIDITY OF ECONOMIC DOCTRINES. 1. The relativity of concrete economic doctrines . . 281 2. Undesirability of limiting political economy to the theory of modern commerce .... 289 3. In what sense universality may be claimed for the principles of abstract economics .... 293 B. ON THE CONCEPTION OF POLITICAL ECONOMY AS A DISTINCTIVELY HISTORICAL SCIENCE. . . 296 XIV CONTENTS. CHAPTER X. ON POLITICAL ECONOMY AND STATISTICS. PAGE 1. The claims of statistics to be regarded as a distinct science 310 2. Statistics regarded as a method . . ... 320 3. The functions of statistics in economic enquiries . 324 NOTE TO CHAPTER X. ON SOME OP THE PRECAUTIONS REQUISITE IN THE USE OP STATISTICS IN ECONOMIC REASONINGS. 1. Conditions of the reliability of statistical data . . 332 2. The interpretation of simple statistics . . . 336 3. The range of statistics 341 | 4. The grouping of statistics 344 INDEX 349 THE SCOPE AND METHOD OF POLITICAL ECONOMY. CHAPTER I. INTRODUCTORY. 1. Nature and importance of the enquiry into the scope and method of political economy. In the terms economy and economic there is an ambiguity, that underlies much of the current confusion as to the nature of political economy. Any line of action is commonly termed economic, when it attains its end with the least possible expenditure of money, time, and effort ; and by economy is meant the employment of our resources with prudence and discretion, so that we may derive from them the maximum net return of utility. But the words are also used in a sense not implying any specially reasonable adaptation of means to ends ; f ~ K. 1 2 INTRODUCTORY. [CHAP. and in works on political economy, the term economic is generally employed simply as an adjective correspond- ing to the substantive wealth. By an economic fact, accordingly, is understood any fact relating to the phenomena of wealth. By economic activities are meant those human activities that direct themselves towards the creation, appropriation, and accumulation of wealth ; and by economic customs and institutions, the customs and institutions of human society in regard to wealth. Political economy or economics is a body of doctrine relating to economic phenomena in the above sense ; and the purpose of the following pages is to discuss the character and scope of this doctrine, and the logical method appropriate to its development. In seeking to define the scope of any department of study, the object in view is primarily to determine the distinguish- ing features of the phenomena with which it deals, and the kind of knowledge that it seeks concerning these phenomena. The enquiry also involves an ex- amination of the relations between the study in question and cognate branches of study. In passing to the consideration of method, we are dealing with a branch of applied logic, the object being to deter- mine the nature of the logical processes specially appropriate to the study that is, the methods of investigation and proof of which it can avail itself I.] LOGIC OF POLITICAL ECONOMY. 3 and the logical character of its conclusions as affected thereby. The discussion that follows belongs, then, to what may be called the philosophy or logic of political economy, and does not directly advance our knowledge of economic phenomena themselves. For this reason, a certain impatience is sometimes felt when any such discussion is proposed. What we want, it is said, is not any more talk about method, but rather useful applications of the right method; let us increase our actual stock of economic truths, instead of indulging in barren disputes about the way in which economic truths are to be attained. To this objection the logi- cian might reply that the enquiry has at any rate a logical, even if it has not an economic, significance. But it has also an economic significance. A moment's consideration will shew that from the point of view of political economy itself, it is of material importance that its scope and method should be rightly under- stood. There is, to begin with, a widely current confusion in regard to the nature of economic laws ; and for this reason, amongst others, it is imperative that the econo- mist should seek to define as accurately as possible the nature and limits of his sphere of enquiry. There should be no vagueness on the question whether political economy is concerned with the actual or the 12 4 INTRODUCTORY. [CHAP. ideal, whether it treats merely of what is, or asks further what ought to be, laying down rules for the attainment of those ends that it pronounces desirable. Even if theoretical and practical enquiries are both to be included within its scope, still the distinction between the two, and their mutual relations, need to be clearly and unambiguously set forth. Misunderstanding on these points has led to a misunderstanding of econo- mic truths themselves, and has consequently impaired the influence and authority of economic science. Next as to method, it is said that instead of arguing about what method of investigation is the right one, it is better to exemplify the right method by employ- ing it in the actual attainment of new economic truths. But are we then to beg the question of its lightness ? In the long run, time cannot but be saved by making a preliminary study of the instruments of investigation to be used, the proper way of using them, and the kind of results that they are capable of yielding. For in so far as methods of reasoning are employed without due regard to the conditions of their validity, the results gained must likewise be of uncertain valid- ity, and the progress of economic knowledge, instead of being advanced, will be retarded. The process, moreover, whereby a conclusion is reached affects its character and value, and the quali- fications and limitations as to range, &c., subject to- I.] LOGIC OF POLITICAL ECONOMY. 5 which it is to be accepted. If it is purely empirical, then it will be established only with a more or less high degree of probability, and it cannot be extended far beyond the range of space or time, over which the instances on which it is based were collected. If, on the other hand, it is obtained deductively, then it is hypothetical until it has been determined how far, and under what conditions, the assumptions on which it rests are realised in fact. It has been plausibly argued that Ricardo's chief weakness was that he did not clearly appreciate the true nature of his own method. At any rate he did not, in interpret- ing his results, take the precautions necessary to pro- vide against misconception on the part of many of his readers. It is true that it is one thing to establish the right method for building up a science, and quite another thing to succeed in building it up. It is also true, as the Austrian economist Menger has remarked, that sciences have been created and re- volutionized by those who have not stopped to analyse their own method of enquiry. Still their success must be attributed to their having employed the right method, even if they have employed it unconsciously or without going out of their way to characterize it. Their method must, moreover, be subjected to careful analysis before the value of 6 INTRODUCTORY. [CHAP. their contnbutions to the science can properly be estimated. Economics is not in any way peculiar in re- quiring that its method should be discussed. The logic of other sciences is, however, for the most part sufficiently dealt with in general works on logic or methodology. There are special reasons, partly to be found in the nature of the subject itself, and partly due to extrinsic causes, why the logic of political economy needs a more detailed consideration. In the first place, economic science deals with phenomena that are more complex and less uniform than those with which the natural sciences are con- cerned; and its conclusions, except in their most ab- stract form, lack both the certainty and the univer- sality that pertain to physical laws. There is a cor- responding difficulty in regard to the proper method of economic study ; and the problem of defining the conditions and limits of the validity of economic reason- ings becomes one of exceptional complexity. It is, moreover, impossible to establish the right of any one method to hold the field to the exclusion of others. Different methods are appropriate, according to the materials available, the stage of investigation reached, and the object in view ; and hence arises the special task of assigning to each its legitimate place and rela- tive importance. I.] LOGIC 1 OF POLITICAL ECONOMY. 7 Another reason for discussing the true principles of economic method in some detail is that fallacious reasonings are more common in political economy than in most other studies. This is due only in part to the difficulty and complexity of the subject- matter with which the science is concerned. It also deals with phenomena which, while encompassed with difficulties, are matters of every-day observation ; and it has few technical terms that are not also terms of every-day discourse. A not unnatural consequence is that people think themselves competent to reason about economic problems, however complex, without any such preparatory scientific training as would be universally considered essential in other departments of enquiry. This temptation to discuss economic questions without adequate scientific preparation is all the greater, because economic conditions exert so powerful an influence upon men's material interests. " Few men," says General Walker, " are presumptuous enough to dispute with the chemist or mechanician upon points connected with the studies and labours of his life ; but almost any man who can read and write feels himself at liberty to form and maintain opinions of his own upon trade and money. The economic literature of every succeeding year embraces works conceived in the true scientific spirit, and works ex- hibiting the most vulgar ignorance of economic history. 8 INTRODUCTORY. [CHAP. and the most flagrant contempt for the conditions of economic investigation. It is much as if astrology were being pursued side by side with astronomy, or alchemy with chemistry." Broadly speaking, the general tendency of popular economics is towards rash generalizations and fallacious arguments post hoc ergo propter hoc. This is frequently combined with an im- perfect analysis of fundamental conceptions, that leads to confusion of thought and the selection of false pro- positions as self-evident postulates ; and where deduc- tive reasoning is employed, its results are often applied without regard to the conditions requisite for their valid application. To this it must be added that the sharp distinctions drawn by opposing schools, and their narrow dogmat- ism, have unnecessarily complicated the whole problem. The subject has become involved in heated controver- sies, that have not only made it wearisome to unpre- judiced persons, but have also done injury to the credit of political economy itself. Outsiders are naturally suspicious of a science, in the treatment of which a new departure is so often and so loudly proclaimed essential. So far, it may be inferred, from economists having made progress in their science, they cannot even agree how to set about their work. The besetting fallacy of writers on economic method has been well said to be one of exclusiveness. A single I.] FALLACY OF EXCLUSIVENESS. 9 aspect or department of economic study is alone kept in view, and the method appropriate thereto aggran- dized, while other methods, of equal importance in their proper place, are neglected or even explicitly rejected. Hence the disputants on both sides, while right positively, are wrong negatively. Their criticisms on rejected methods are, moreover, too often based on misapprehension or misrepresentation. Methods are attacked for not doing what those who advocate their use have never imagined they could do ; and the quali- fications and limitations, with which each side ex- pounds its own method, are overlooked by the other side. Thus combined with the fallacy of exclusive- ness, or rather in consequence of it, there is in these controversies a remarkable prevalence of ignoratio elenchi. In the following pages, an attempt will be made to do justice to all the different instruments of investigation of which the economist can avail himself, while attention will also be drawn to the limitations to which each in turn is subject. 2. The conception of political economy as a theoretical, abstract, and deductive science. The main points involved in controversies about economic method may be indicated in outline by briefly contrasting two broadly distinguished schools, one of which de- scribes political economy as theoretical, abstract, and deductive, while the other describes it as ethical, 10 INTRODUCTORY. [CHAP. realistic, and inductive. It should be distinctly under- stood that this sharp contrast is not to be found in the actual economic writings of the best economists of either school. In the methods that they employ when they are really discussing the same problems there is to a great extent substantial agreement. They differ, how- ever, in the relative importance that they attach to different aspects of their work; and in their formal statements about method, these differences become ex- aggerated. The question of the right method of economic en- quiry was not as such discussed by Adam Smith ; and his views on the subject have, therefore, to be gathered from his way of dealing with actual economic problems. As a matter of fact, the support of his authority has been claimed on behalf of both the schools above re- ferred to. It has been said of him that he first raised political economy to the dignity of a deductive science. But he has also been regarded as the founder of the historical method in political economy. The reason for this apparent contradiction is not far to seek. It is to be found in Adam Smith's free- dom from excess on the side either of d priori or a posteriori reasoning. He rejected no method of en- quiry that could in any way assist him in investigating the phenomena of wealth. For argument or illustra- tion he had recourse, as the occasion might arise, either I.] ADAM SMITH, MALTHUS, AND RICARDO. 11 to elementary facts of human nature, or to complex facts of industrial life. He believed in a " natural " order of events, which might be deduced d priori from general considerations; but he constantly checked his results by appeals to the actual course of history. He worked up from abstractions to the complex realities of the economic world in which he lived. Thus, if on deductive grounds he lays down a doctrine of the ten- dency of wages to equality, he combines it with an inductive enquiry into the causes that check or restrict the operation of this tendency. If he sets forth the " natural " progress of opulence, he enters also upon an historical investigation of what the actual progress of opulence has been. If he condemns the doctrine of protection to native industry mainly on abstract grounds, he enforces his views with concrete illustra- tions and arguments in the greatest variety. As regards the inductive tendencies noticeable in Adam Smith, his successor is to be found in Malthus ; for the continuation and development of the abstract deductive tendencies we turn to Ricardo. Subsequent economists of the English school assimilated what was most characteristic in both these writers ; but it was Ricardo, rather than Malthus, who gave to their work a distinctive tone, particularly in their specific analysis of the method to be pursued. Senior and J. S. Mill were the earliest English 12 INTRODUCTORY. [CHAP. economists who definitely formulated principles of economic method. Senior's views are contained in his introductory lectures before the University of Oxford, and in his treatise on Political Economy ; Mill's views are to be found in his Essays on Some Unsettled Ques- tions of Political Economy, and in the sixth book of his Logic. The problem is discussed in more detail by Cairnes in his Character and Logical Method of Politi- cal Economy, a work of admirable lucidity, which has long been considered the authoritative text-book of English political economy, so far as concerns its logic. Bagehot's essays on the postulates of English political economy and on the preliminaries of political economy, published in his Economic Studies, have also in some respects a representative character. There are minor differences in the principles laid down by these four writers respectively, but funda- mentally they are in agreement in regarding political economy as a science that is in its scope theoretical, and in its method abstract and deductive. The follow- ing is a very brief summary of their characteristic doctrines. In the first place, a sharp line of distinction is drawn between political economy itself and its ap- plications to practice. The function of political eco- nomy is to investigate facts and discover truths about them, not to prescribe rules of life. Economic laws I.] DOCTRINES OF MILL AND CAIRNES. 13 are theorems of fact, not practical precepts. Political economy is, in other words, a science, not an art or a department of ethical enquiry. It is described as standing neutral between competing social schemes. It furnishes information as to the probable con- sequences of given lines of action, but does not itself pass moral judgments, or pronounce what ought or what ought not to be. At the same time, the greatest value is attached to the practical applications of economic science ; and it is agreed that the economist ought himself to turn his attention to them not, however, in his character as a pure economist, but rather as a social philosopher, who, because he is an economist, is in possession of the necessary theoretical knowledge. It is held that if this distinction is drawn, the social and ethical aspects of practical problems which may be of vital importance are less likely to be overlooked or subordinated. As to its position amongst the sciences, political economy is not regarded as inseparably bound up with social philosophy in general. Economic facts are, it is allowed, influenced by social facts of very various kinds, and in their turn influence them ; but it is neverthe- less held to be possible up to a certain point to isolate the study of the phenomena of wealth from the study of other phenomena of society. Such isolation is, in- deed, said to be necessitated by the requirements of 14 INTRODUCTORY. [CHAP. science, which always proceeds by analysing concrete phenomena, so as to deal separately with their differ- ent aspects and the different elements of which they are composed. Economic science constitutes, therefore, a distinct, though not entirely independent, depart- ment of sociological speculation. Passing to the means whereby the truths of the science are to be reached, it is held that on account of the variety and complexity of the influences to which economic phenomena are subject, the method of specific experience or direct induction is inadequate to yield more than empirical generalizations of uncertain validity. Experiment is, moreover, a resource from which the economist is debarred. It follows that we ought not to take as our starting point the analysis of concrete industrial facts. The right method of procedure is, on the contrary, deductive, or, as Mill puts it, a priori. The ultimate premisses upon which the deductive science is based are, moreover, limited in number, so that the more important of them admit of precise enunciation at the outset. For while the cir- cumstances helping in some degree to mould economic phenomena are indefinitely numerous, there are a few whose influence is predominant, far outweighing that of all the rest. These predominating circumstances con- sist of a few simple and indisputable facts of human nature as, for example, that in their economic deal- I.J DOCTRINES OF MILL AND CAIRNES. 15 ings men are influenced by the desire for wealth taken in connexion with the physical properties of the soil, and man's physiological constitution 1 . Political economy is accordingly spoken of as, in the main, an abstract science. For in basing its conclusions on a limited number of fundamental assumptions, it has to leave out of account many circumstances, which are of importance in individual cases, but are never- theless unimportant when instances are taken in the mass. That other motives besides the desire for wealth do operate on various occasions in determining men's economic activities is recognised. They are, however, to be neglected at any rate in the first instance since their influence is irregular, uncertain, and 1 There is, however, some difference of view as to the extent to which the application of the resulting doctrines needs limitation. Bagehot regards the doctrines of English political economy as not applicable to all states of society, but only to those in which com- merce has largely developed, and in particular taken the form of development which we find in England at the present time. The relativity of economic investigations is also indicated incidentally by Cairnes. Senior, on the other hand, remarks that those conclusions which relate to the nature and the production of wealth are universally true ; and although those which relate to the distribution of wealth are liable to be affected by the particular institutions of particular countries, still the natural state of things can be laid down as the general rule, and the anomalies produced by particular disturbing causes afterwards accounted for. In other words, while Bagehot regards the premisses of political economy as relating only to the economic habits and institutions of a particular age and country, Senior regards them as "natural," and, with slight qualifications, as independent of age and country. 16 INTRODUCTORY. [CHAP. capricious. On these grounds, it is argued that the abstraction, whereby the science takes as its principal subject-matter an " economic man," whose activities are determined solely by the desire for wealth, is both legitimate and necessary ; and, in further justification thereof, an analogy is drawn from mathematics and physics, which are said to be based on corresponding abstractions 1 . On similar grounds, the science is spoken of by Mill and Cairnes as hypothetical. For inasmuch as its premisses do not exhaust all the causes affecting the result, its laws are only true hypothetically, that is, in the absence of counteracting agencies. The same point is expressed by saying that political economy is a science of tendencies only, not of matters of fact, its object being to work out and ascertain the result of certain great forces, as if these alone operated, and nothing else exerted any modifying influence 2 . Senior sums up his views in the dictum that poli- tical economy "depends more on reasoning than on 1 Mill and Bagehot specially insist upon the high degree of ab- straction involved in economic reasonings. Bagehot more than once repeats that " English political economists are not speaking of real men, but of imaginary ones ; not of men as we see them, but of men as it is convenient to us to suppose they are" (Economic Studies, p. 5). 2 Senior, while affirming that the conclusions of political economy are true only in the absence of disturbing causes, still calls it a posi- tive, and not a hypothetical, science. By this he means that its pre- misses are not arbitrarily assumed. I.] DOCTRINES OF MILL AND CAIRNES. 17 observation." Mill. Cairnes, and Bagehot; however, all insist that the appeal to observation and experience must come in, before the hypothetical laws of the science can be applied to the interpretation and explan- ation of concrete industrial facts. For it then has to be ascertained how far, as regards the particular cases under consideration, allowance needs to be made for the operation of disturbing causes that is, for the peculiar modifications introduced by the minor in- fluences affecting economic phenomena. Comparison with observed facts provides a test for conclusions deductively obtained, and enables the limits of their application to be determined. Accordingly, while the method of specific experience is regarded as altogether inefficacious for the discovery of economic laws, and as incapable of affording independent proof of their validity, it is nevertheless considered to form an indis- pensable supplement to the deductive reasoning that constitutes the framework of the science. The above doctrines of economic method, which are those explicitly formulated by the writers referred to, need to be interpreted and in some respects qualified by reference to their actual economic writings. For if, from an examination of the latter, we seek to deduce their views on method, we find that their prac- tice does not precisely correspond with their theory ; and we are led to the conclusion that, judged by their K. 2 18 INTRODUCTORY. [CHAP. own writings, they state their doetrines on method in too absolute a manner, in particular exaggerating the abstractness of political economy taken as a whole. They also speak as if the science had reached the de- ductive stage in a more definitive manner than is ap- parent from their own way of dealing with economic problems. In treating of the production of wealth, for ex- ample, as is pointed out by Professor Sidgwick, Mill and other economists of his school have always em- ployed an inductive and analytical method, the deduc- tive element in their reasonings being in this part of their subject essentially subordinate. Mill is even more distinctly an inductive economist in his elaborate discussion of peasant proprietorship in its economic aspects. There is no doubt a deductive element, based on psychological data, in his argument as to the effect of ownership on the cultivator's industry and energy. But even on this point he brings a considerable amount of a posteriori evidence to bear ; and his general ar- gument depends mainly on an inductive and compara- tive investigation of the actual working of peasant pro- prietorship in France, Switzerland, and other countries, in which the operation of the system can be observed on a considerable scale. Cairnes, again, in his work on the Slave Power, where he analyses the general eco- nomic characteristics of slave labour, establishes some I.] J. S. MILL. 19 important economic doctrines by a careful inductive study of facts, little use being made of deductive reasoning. It is true that the general theory of distribution and exchange, expounded by the school of Mill, is based on reasoning of an abstract character ; but even here the writers, to whom reference has been made, tend to exaggerate the characteristics of their own method. They do not hold themselves aloof from the concrete realities of the actual economic world to anything like the extent that their description of the science would lead their readers to anticipate ; and it is very far from the truth to say that their doctrines are wholly constructed out of a few elementary laws of human nature. At all events, in order to establish their consistency, a large portion of their best economic work must be regarded as concerned with the practical modifications of the truths of political economy, rather than with those truths themselves. The contrast is specially marked between Mill's theory of method as contained in the Essays, and his practice as manifested in the Principles. In the former, the conception of the "economic man" occupies a position of central and all-pervading importance ; in the latter, it plays a much humbler part. Moreover, in his Principles of Political Economy, Mill avowedly treats, not merely of these principles themselves, but 22 20 INTRODUCTORY. [CHAP. also of "some of their applications to social philosophy." He states in his preface that while he desires to give an exposition of the abstract doctrines of political economy, he also desires to give something more than this ; his object is to include " a much wider range of ideas and of topics, than are included in political economy, considered as a branch of abstract specula- tion." Moral and social considerations, in the widest sense, receive accordingly their due share of attention ; and it would be difficult to find a better instance of an ethical treatment of economic problems than is con- tained in the chapter on "the probable future of the labouring classes." 3. The conception of political economy as an ethical, realistic, and inductive science. The emphasis with which the earlier systematic writers on economic method, especially in England, dwelt upon the abstract side of political economy led to a reaction, which took its rise in Germany, and is especially connected with the names of Roscher, Hildebrand, and Knies. The two schools, thus broadly distinguished, are sometimes spoken of as the English and the German respectively. These designations have the merit of brevity ; and, taking into account what was actually written about method by English and German economists respectively during the middle part of this century, they are not without some justification. Interpreted literally, how- I.] ENGLISH AND GERMAN SCHOOLS. 21 ever, they are misleading. The doctrine of method set forth in the preceding section does not fairly represent the many-sidedness of English work in eco- nomics. In particular, it fails to assign a sufficiently important place to the mass of historical and statistical material that the labour of English economists has provided. The doctrine would, moreover, be accepted only in a modified and broadened form by those con- temporary economists., who avowedly carry on the traditions of the English school. Again, the so-called German doctrines, whatever may have been their origin, are no longer the peculiar possession of any one country. They are, for example, represented by a rising school of economists in the United States, who expressly repudiate the assertion that the new move- ment is exclusively a German movement. Even in England the spirit of the reaction was manifested long ago by Richard Jones, and in recent years very forcible expression has been given to it by Cliffe Leslie and others. On the other hand, amongst distinguished economists, who have employed a highly abstract method of treating economic problems, several Germans, e.g., von Thiinen, are to be included ; and recently there has sprung up in Austria a new school, which insists very emphatically on the necessity of an abstract treat- ment of the science 1 . 1 Professor Carl Menger of Vienna is one of the principal leaders 22 INTRODUCTORY. [CHAP. Subject to the foregoing explanation, it is conve- nient to speak of the school of Roscher and Knies as the German school. The explicit teaching of this school in regard to the scope and method of economics is briefly indicated in the following paragraphs 1 . In the first place, a more extended scope is given to the science than is usual with English economists ; for it is avowedly made to treat of what ought to be as well as of what is. The possibility of drawing in this later development of German opinion. Compare his Unter- suchiuigen iiber die Methode der Socialwissenschaften und der Politischen Oekonomie insbesondere. He specially insists on the necessity of dis- tinguishing theoretical political economy from economic history and statistics on the one hand, and from the practical sciences of political economy on the other ; and he accuses the dominant German school both of misunderstanding the point of view of the abstract method, and of attributing an exaggerated importance to the historical. He further charges them with error in attempting to give an ethical direction to theoretical political economy. He speaks still more strongly in a very controversial series of letters specially directed against Schmoller, and published under the title Die Irrthiimer des Historismus in der Deutschen Nationalokonomie. Professor Emil Sax of Prague is in agreement with Menger on fundamental points, but presents his views in a less controversial form. He insists strongly on the importance of pure theory. Compare his Wesen und Anfgaben der Nationalokonomie. 1 On the points that follow compare Eoscher, Geschichte der National-Oekonomik in Deutschland, especially pp. 1032 1036 ; Knies, Die Politische Oekonomie vom Standpunkte der geschichtliclien Methode ; Schonberg's article on Die Volkswirthschaft (in his Hand- buck), 1 13; and Wagner, Systematische Nationalokonomie (Jalir- biicher fur Nationalokonomie und Statistik, March, 1886). Wagner's article is translated in the Quarterly Journal of Economics (published for Harvard University), vol. i. p. 113. I.] THE GERMAN HISTORICAL SCHOOL. 23 any clear line of separation between these enquiries is, indeed, practically denied. It is held that there can be no purely theoretical science of political economy, such as was contemplated by Cairnes. The school explicitly calls itself ethical ; it regards political economy as having a high ethical task, and as concerned with the most important problems of human life. The science is not merely to classify the motives that prompt to economic activity ; it must also weigh and compare their moral merit. It must determine a standard of the right production and distribution of wealth, such that the demands of justice and morality may be satisfied. It must set forth an ideal of eco- nomic development, having in view the intellectual and moral, as well as the merely material, life ; and it must discuss the ways and means such as the strengthening of right motives, and the spread of sound customs and habits in industrial life, as well as the direct intervention of the State by which that ideal is to be sought after 1 . Another characteristic of the German historical school is the manner in which its adherents insist 1 It should be observed that differences in regard to the scope of a science may be to a considerable extent merely verbal. One writer may include within the science itself enquiries which another writer regards as belonging only to its applications ; but it does not follow that the latter neglects these enquiries, or even that he in the slightest degree attaches less importance to them. 24 INTRODUCTORY. [CHAP. upon the social side of political economy, and the in- terdependence of economic and other social phenomena. It is held that because of this interdependence, political economy cannot be adequately treated except in close connexion with other branches of social science. The treatment adopted ought, accordingly, to be realistic. It is maintained that the economist should only very sparingly, if at all, abstract from the complex realities of actual economic life ; and should consequently in most of his reasonings deal, not with an abstract "eco- nomic man," subject only to a single motive, the desire for wealth, but directly with men as they really are, moved by diverse motives, and influenced by the actual conditions of the age and society in which they live. Closely connected with this characteristic is the insist- ence upon the relativity of economic doctrines. The economic conditions of life are subject to variation ; and subject to like variation are the laws by which men's economic activities are regulated. As to the method of reasoning by which economic knowledge is to be extended, great stress is laid on the necessity of appealing constantly to specific observation of the actual economic world, and generalizing there- from. Hence the school is spoken of as inductive and statistical. It is still more distinctively designated historical, from its special insistence on the importance of historical material in building up the science. Only I.] THE GERMAN HISTORICAL SCHOOL. 25 by reference to the past, it is held, can the present be properly understood ; and only by a comparison of the economic conditions of different periods and different countries can the limitations of economic doctrines be adequately realised, and economists saved from one- sided and narrow dogmatism. The importance of studying the course of economic evolution is, accordingly, emphasized. It should be added that, independently of differences in regard to the scope and method of political economy, the .dominant German school is distinguished from the older English economists by a difference of attitude towards laisser faire and government interference. This is, however, a point of contrast with which we are not directly concerned in the present treatise. It will be observed that the above-mentioned cha- racteristics are by no means independent of one another. In some cases, the connexion is very close indeed. The more realistic our standpoint, for example, the more obvious becomes the necessity of direct appeals to history and statistics ; the historical method leads per- force to the recognition of the relativity of economic doctrines ; and the realistic and the social standpoints are also closely connected. In its turn, the ethical con- ception of the science emphasizes all the other points ; and in fact, if it be granted that political economy is directly concerned with what ought to be, then most 26 INTRODUCTORY. [CHAP. of the rest may be said logically to follow 1 . It results from this dependence of characteristics that in discuss- ing the various questions at issue a certain repetition is unavoidable. Accordingly, in the following pages, even when we are treating apparently distinct prob- lems, there will not unfrequently be found a recurrence of the same fundamental points, viewed in different aspects. Within the new school itself very important differ- ences of tone and attitude are to be observed. The more extreme members of the school are arrogant and exclusive in their pretensions. Not content with em- phasizing the importance of the historical method, they reject the aid of any other method except in extreme subordination to it. They are not simple re- formers, but revolutionaries ; for they advocate a com- plete reconstruction and transformation of political economy. In their view, the science in the past has been barren of valuable results; only by a radical change of method can it hope to be fruitful in the future. The old doctrines, and the old ways of reaching them, are to be put on one side and seen no more. Professor Schmoller and Dr Ingram may be taken as examples of this advanced wing of the new school. 1 It will be shewn, however, that the converse does not hold good ; that it is, in other words, possible to adopt a realistic treatment of economic problems without passing ethical judgments. I.] DOCTRINES OF ROSCHER AND WAGNER. 27 The former would practically identify political economy and economic history, or at any rate resolve political economy into the philosophy of economic history. The latter, whose aim is somewhat different, though he is equally revolutionary in his tendency, would absorb political economy into general sociology. The position taken by the more moderate adherents of the German school, including Roscher himself, is in marked contrast to the above. They adopt a tone of moderation, and an attitude of compromise. While in- sisting on the importance of historical investigation in political economy, they admit the necessity of em- ploying other methods in conjunction therewith ; and while taking a realistic view of the science as a whole, they recognise the value of abstraction, at any rate in certain preparatory stages. They accept many of the most characteristic of the old conclusions, and on the old grounds. According to Professor Adolph Wagner, who may be taken as a leading representative of the more moderate section of the new school, the inductive and deductive methods both have their place in econo- mics. " These, then," he says, " are the two methods : on the one hand, deduction from psychological mo- tives first and foremost, deduction from the motive of individual advantage, then from other motives ; on the other hand, induction from history, from statistics, and from the less exact and less certain, yet indis- 28 INTRODUCTORY. [CHAP. pensable, process of common observation and expe- rience. With both methods we are to approach the various problems of political economy, and to solve them so far as we can. Which method is most to be used depends on the nature of the particular problems ; but it depends also on the turn of mind, very likely on the accident of training and education, of the individual investigator." 1 4. The method of political economy cannot ade- quately be described by any single phrase. We must not then exaggerate the opposition between what may be called the classical English school and the new school. The former realise more vividly the abstract problems of the science, and in writing on method 1 Quarterly Journal of Economics, vol. I. p. 124. Dr von Scheel expresses himself similarly. Different methods, he says, are ser- viceable for the solution of economic problems. " We must use both inductive and deductive methods. The most suitable method will continually vary with the particular nature of the problem to be solved" (Schonberg's Handbuch, Die politische Oekonomic ah Wissen- schaft, 3). Professor Gustav Cohn may be quoted to the same effect. The idea, he says, that mere collections of historical or statistical material can be made available for science, without de- ductive aids, is just as much an extravagance, as the opposite idea that out of deductions from elementary hypotheses the whole science can be constructed (Grundlegung der Nationalokonomie, p. 35). Dr E. R. A. Seligman, again, writing on behalf of the American supporters of the new movement, remarks that the more extreme of the Germans "have themselves overshot the mark, have unduly undervalued the work of the English school, and have in their zeal too dogmatically denied the possibility of formulating any general laws" (Science Economic Discussion, p. 21). I.] DIFFERENT METHODS SERVICEABLE. 29 keep these problems mainly in view. The latter realise more vividly the concrete problems, and hence lay stress on all the points which the English school have tended to overlook. But the difference is strictly speaking one of degree only; and we find the oppo- sition reduced to a minimum, when we compare the actual procedure in the solution of given problems adopted by the best contemporary economists, whether they profess to belong to the new school or are content to be classed with the old 1 . As to the doctrine to be expounded in the following pages, it will suffice here to say that while great im- portance will be attached to the place of the deductive method in economic enquiry, and while a protest will be entered against the unhistorical spirit evinced by those adherents of the new movement who proclaim the necessity for a complete reorganization of the science, still no attempt will be made to justify the doctrines of the older school in the precise form in which they were laid down by Mill and Cairnes. The method of political economy cannot adequately be de- scribed by any single phrase ; and accordingly no one method will be advocated to the entire exclusion of other methods. It will, on the contrary, be shewn 1 On this point, compare Professor Dunbar's very valuable essay on the Beaction in Political Economy, in the Quarterly Journal of Economics, October, 1886. 30 INTRODUCTORY. [CHAP. I. that, according to the special department or aspect of the science under investigation, the appropriate method may be either abstract or realistic, deductive or inductive, mathematical or statistical, hypothetical or historical. CHAPTER II. ON THE RELATION OF POLITICAL ECONOMY TO MORALITY AND PRACTICE. 1. Distinction between economic uniformities, economic ideals, and economic precepts. In regard to the scope of political economy, no question is more important, or in a way more difficult, than its true relation to practical problems. Does it treat of the actual or of the ideal ? Is it a positive science con- cerned exclusively with the investigation of uni- formities, or is it an art having for its object the determination of practical rules of action ? What, for example, is the true problem of political economy in regard to the influence of competition on wages ? Is it to investigate the precise nature of that influence, and to enquire how far and in what ways the operation of competition is or may be modified by other agencies? Or is it rather to determine how far the effects of competition can be morally approved, and to what extent it is desirable that its operation should be supplemented or superseded by combination or direct governmental interference ? 32 POLITICAL ECONOMY AND PRACTICE. [CHAP. The distinction here indicated is indeed threefold rather than twofold as is usually implied. For when we leave the enquiry into the veritable order of eco- nomic phenomena, their coexistences and sequences, under existing or assumed conditions, we still have to take account of a further subdivision of some signifi- cance. There is, on the one hand, the investigation of economic ideals, and the determination of a standard by reference to which the social worth of economic activities and conditions may be judged ; and there is also the investigation of economic rules, i.e., the de- termination of maxims or precepts by obedience to which given ends may best be attained 1 . Thus, in 1 There is still another distinction, which need not, however, be dwelt upon namely, the distinction between economic maxims as formulated by the student, and their practical outcome in the actual legislation of different countries. Even this distinction is not always clearly recognised. It seems, for instance, to be obscured in the following passage in Lord BramwelTs recent address as President of Section F of the British Association: "What will be the best way to add to the wealth of a society must be a subject of study by that society, which will lay down rules that is to say, make laws for the purpose ; and this is political economy. Adam Smith was not the first political economist, though well called the father of those rules which now prevail. But rules for the purpose existed before him, the great objection to them being that most of them were wrong. There was a law that the dead should be buried in woollen. Laws were made for fixing wages; laws were made against regrating and forestalling. Then think of the usury laws. You cannot deny that these were economical laws because you think them wrong." The laws here referred to cannot in any proper sense be called laws of political economy. Even if political economy is regarded as an art, the precepts of that art must be distinguished from the actual practice II.] UNIFORMITIES, IDEALS, AND PRECEPTS. 33 regard to the payment of interest, we have, first, the positive enquiries why, under certain conditions of in- dustry, interest is paid at all, and what determines the rate paid. We have, secondly, the enquiries whether interest ought to be paid, and, if it ought, what constitutes a fair rate of interest. We have, thirdly, the enquiries whether any interference in regard to the payment of interest is desirable, and, if so, what are the best means whereby such payment may either be abolished or at least approximated to a fan- standard. Another illustration may be taken from the depart- ment of taxation. The investigation of the incidence of- taxation is in itself a positive enquiry ; so is the problem of the influence of different forms of taxation on relative values. These are, in other words, enquiries as to matters of fact. Passing to problems that belong to a different category, we may distinguish the determination of the ideal of taxation from that of rules of taxation in the narrower and stricter sense. It is one thing to ask in what sense, if any, and why, equality of taxation should be our aim ; it is another thing to enquire by what rules, e.g., the adoption of a system of progressive taxation or the judicious combination of direct and indirect taxation, such of politicians and finance ministers, however much their acts may be the direct legislative embodiment of the precepts. K. 3 34 POLITICAL ECONOMY AND PRACTICE. [CHAP. equality can with the nearest possible approximation be attained. Intimate as are the connexions between the above kinds of enquiry, they are in themselves distinct in character, and belong to different departments in a classification of knowledge. The first belongs to posi- tive science, the second to normative or regulative science (along with ethics, if indeed it be not a branch of ethics or of what may be called applied ethics), and the third not to science at all in the more modern use of the term, but to art as distinguished therefrom. As the terms are here used, a positive science may be defined as a body of systematized knowledge con- cerning what is 1 ; a normative or regulative science as a body of systematized knowledge discussing criteria of what ought to be, and concerned therefore with the ideal 1 The use of the term positive to mark this kind of enquiry is not altogether satisfactory ; for the same term is used by Cairnes and others in contrast to hypothetical, which is not the antithesis here intended. It is difficult, however, to find any word that is quite free from ambiguity. Theoretical is in some respects a good term, and we shall sometimes use it. In certain connexions, however, it is to be avoided, inasmuch as it may be understood to imply an antithesis with actual, as when theory and fact are contrasted; it may also suggest that the enquiries referred to have little or no bearing on practical questions, which is of course far from being the case. Pro- fessor Sidgwick in his Methods of Ethics employs the term speculative; but this term, even more than theoretical, suggests something very much in the air, something remote from the common events of every- day life. It seems best, therefore, not to use it in the present connexion. II.] UNIFORMITIES, IDEALS, AND PRECEPTS. 35 as distinguished from the actual 1 ; an art as a system of rules for the attainment of a given end 8 . The object of a positive science is the investigation of uni- formities, of a normative science the determination of ideals, of an art the formulation of precepts. The problem whether political economy is to be re- garded as a positive science, or as a normative science, or as an art, or as a combination of these, is to a certain extent a question merely of nomenclature and classifi- cation. It is, nevertheless, most important to distin- guish economic enquiries according as they belong to the three departments respectively; and it is also im- portant to make clear their mutual relations. Con- fusion between them is common and has been the source of many mischievous errors. An endeavour will be made in the following pages 1 It should be particularly observed that a department of know- ledge does not necessarily belong to the category of art, as distinguished from science, simply because it is concerned with what ought to be. Logic and ethics are both of them sciences, although they are con- cerned with right reasoning and right conduct respectively. In the following pages, however, whenever science is contrasted with art without further qualification, positive science, and not normative science, is had in view. 2 To avoid misunderstanding, it should be added that Adam Smith and his contemporaries, as well as some modern economists, use the term science without any reference to the distinction between science and art as above indicated. They mean by a science any systematic body of knowledge, whether consisting of theoretical propositions, or of practical rules of action. The best recent authorities, however, at any rate in this country, use the term in the narrower sense. 32 36 POLITICAL ECONOMY AND PRACTICE. [CHAP. to shew that it is both possible and desirable to discuss economic uniformities independently of economic ideals, and without formulating economic precepts, although the converse proposition cannot be affirmed; and it follows that, if this view be correct, we ought at least to recognise as fundamental a positive science of political economy which is concerned purely with what is, and which seeks to determine economic laws 1 . It is a further question whether or not we should also recognise, as included under political economy in the widest sense but distinct from the positive science (a) a branch of ethics which may be called the ethics of political economy, and which seeks to determine economic ideals', and (6) an art of political economy, which seeks to formulate economic precepts. 1 We here use the term law, as it will consistently be used in the following pages, in its scientific, and not in its jurisprudential, sense. We mean by a laic a theorem, the statement of a uniformity, not a command enforced by sanctions. The law of supply and demand, the Bicardian law of rent, Gresham's law, and the like, may be given as examples of economic laws in the above sense. The validity of such laws is a purely theoretical question, and our attitude towards them is not, or at any rate should not be, affected by our ethical or political views. It is otherwise as soon as we begin to lay down rules for the guidance of statesmen and legislators. When we argue for fair trade or for free trade, when we advocate the legislative re- striction of hours of labour or the nationalisation of the land, or when we contend for a general policy of laisser faire, we have advanced a stage further. Considerations based on political economy, conceived as a positive science, may still form the foundation of our argument, but such data have to be controlled by ethical and political considerations. II.] THE SCOPE OF POLITICAL ECONOMY. 37 2. The possibility of studying economic laws or uniformities without passing ethical judgments or formu- lating economic precepts. It has been pointed out in the preceding chapter that the prevailing tendency amongst a certain school of economists is to widen the scope of political economy, by giving it a distinctly ethical character, and making little attempt to sepa- rate its treatment as a practical science from its treat- ment as a theoretical science. It is even maintained that such a separation is impossible. Thus Professor Wagner, while clearly distinguishing the positive and the ethical problems, explicitly denies that either of them admits of being treated apart from the other, although taken together their treatment may be sepa- rated from that of the dependent art. He gives the five following problems (of which the first two belong to positive science as above defined, the third and fourth to normative or ethical science, and the fifth to art) as between them constituting the great general problem of political economy: (1) the description of economic phenomena; (2) the explanation of the causes upon which they depend ; (3) the determination of a standard by which their social merit may be mea- sured ; (4) the setting up of an aim for economic pro- gress ; (5) the examination of the ways and means for attaining this aim. Of these problems he regards the first four as too closely connected to permit a sepa- 38 POLITICAL ECONOMY AND PRACTICE. [CHAP. ration. Only the fifth, he considers, where we have to deal with the practical questions of an art, can be clearly distinguished from the rest 1 . Others would not even admit the degree of separation that Wagner al- lows. Dr von Scheel, for instance, remarks that the history, theory, and art of political economy form one indivisible whole 8 . Yet on reflection it seems clear that there can be no inherent reason, why we should not both describe and explain economic phenomena without either pass- ing a judgment on their moral worth, or setting up an aim for economic progress ; although of course the converse does not hold good. It is true, for example, that we cannot determine how nearly the results of free competition approximate to our economic ideal until we know what those results are. Nor can we say how far it is desirable that the effects which would be brought about by unimpeded competition should be modified by governmental interference or voluntary combination, until we have also ascertained what kind of modification would ensue, and what would be the collateral effects of such interfering agencies. We can, however, successfully investigate the nature of eco- nomic phenomena under the regime of competition, without comparing them with any ideal standard ; and 1 Quarterly Journal of Economics, vol. i. pp. 124 128. * Schonberg's Handbuch, vol. I. pp. 71, 72. II.] THE CONNEXION BETWEEN SCIENCE AND ART. 39 we can also correctly ascertain the effects exerted, or capable of being exerted, by agencies other than com- petitive such as law, public opinion, voluntary com- bination, and the like without expressing an opinion on the practical question how far it is desirable that the operation of agencies such as these should be specially encouraged 1 . 1 In discussing the connexion between science and art, it is necessary to distinguish the logical order from what may be called the historical order. It has often been pointed out that while in the logical order science precedes art, the historical order is the reverse of this. In other words, the demand for guidance arising from men's practical needs is recognised, and attempts are made to satisfy it, before bodies of speculative truth are systematically formulated. Thus an empirical art of medicine exists before there is any distinct science of physiology. Indeed, as is remarked by Sir George Cornewall Lewis, " the purely scientific treatment of any subject, without an attempt to lay down precepts or rules of practice, is in general one of the latest stages in the journey of knowledge " (Methods of Observation and Reasoning in Politics, Chapter 19, 5). But, as we have said, in the logical order science precedes art, that is, we cannot satisfactorily lay down rules for practical guidance except on the basis of knowledge of facts. When, therefore, this knowledge is not to be found elsewhere, the art must seek it as best it can for itself, thus becoming at the same time both a science and an art, the two enquiries, however, not being definitely distinguished. Strictly speaking, instead of saying that historically art precedes science, it would for the above reason be more accurate to say that at the outset there is no clearly marked line of distinction between them. Accordingly in early treatises on any art we expect to find, and we do find, theorems of science more or less explicitly set forth, justifying the rules which it is the authors' main purpose to expound. Herein is the explanation of the fact that, while Adam Smith conceives political economy as an art, the Wealth of Nations assumes for the most part the form of a science. The system of political economy there advocated, being " the obvious and simple system of natural 40 POLITICAL ECONOMY AND PRACTICE. [CHAP. The proposition that it is possible to study economic uniformities without passing ethical judgments or for- mulating economic precepts seems in fact so little to need proof, when the point at issue is clearly grasped, that it is difficult to say anything in support of it that shall go beyond mere truism. We may, however, seek to explain away certain difficulties, based on misap- prehension, that have tended to prevent its truth from being universally recognised. The idea probably is that any attempt to treat economic laws, without pass- ing ethical judgments, and without reference to an ideal to be aimed at, is certain to result in a practical denial that moral considerations have any bearing on economic phenomena at all. It has indeed been made a specific charge against English economics of thirty years ago that, seeking to be purely theoretical, it became in the worst sense unmoral, its tendency being to claim for economic action a sphere altogether independent of moral laws. Whilst it would be difficult to substantiate this charge from the actual writings of English economists liberty," does not in itself consist of any elaborate code of rules. Adam Smith is chiefly concerned to confute on scientific grounds other systems, and to establish the scientific basis of his own. His first three books, to a large extent his fourth, and to some extent his fifth, are thus taken up with discussions in which the actual relations of phenomena are discussed and expounded. On this point, compare Sidgwick, Principles of Political Economy, Introduction, Chapter 2. II.] ECONOMIC ACTIVITIES SUBJECT TO MORAL LAWS. 41 of the first rank at any period, a certain justification for it may be found in the tone and attitude of some popular interpreters of economic science at the time referred to ; and it will be useful to seek to discover the source of the error into which they fell. That it was an error hardly needs to be insisted upon. Nothing can be more deplorable than that the econo- mist should be understood to imply that, in his in- dustrial dealings, a man is freed from the ordinary obligations of justice and humanity. To refer an in- justice in the economic world to demand and supply may possibly account for it ; but it cannot seriously be maintained that from the point of view of the moralist or the social reformer this settles the matter. It needs no proof that neither economic activities nor any other class of human activities can rightly be made inde- pendent of moral laws. But it is far from being the case that the fallacious attitude of mind here combated is a necessary con- sequence of the attempt to construct a purely positive science of economics. On the other hand, it is rather the failure to recognise the fundamentally distinct character of enquiries into what is, and enquiries into what ought to be, that is really responsible for attempts to solve practical economic questions without reference to their ethical aspects. And this danger will cer- tainly not be diminished by endeavouring systemati- 42 POLITICAL ECONOMY AND PRACTICE. [CHAP. cally to fuse the two classes of problems. There is, however, a further source of confusion, to which it is necessary at this point specially to call attention, due to the non-recognition of the fact that from the purely positive standpoint the operation of moral forces may need to be taken into account. It has too often been implied, though it may not often have been expressly stated, that at any rate in regard to what can actually happen, as distinguished from what one might desire to see happen the last word has been said when the effects of competition have been correctly ascertained and set forth 1 . As a matter of fact, although the 1 In Kingsley's Alton Locke, the tailor Crossthwaite speaks as follows "But you can recollect as well as I can, when a deputation of us went up to a member of Parliament one that was reputed a philosopher, and a political economist, and a Liberal and set before him the ever-increasing penury and misery of our trade and of those connected with it; you recollect his answer that however glad he would be to help us, it was impossible he could not alter the laws of nature that wages were regulated by the amount of competition among the men themselves, and that it was no business of Govern- ment or anyone else, to interfere in contracts between the employer and employed, that those things regulated themselves by the laws of political economy, which it was madness and suicide to oppose " (ch. 10). It cannot be denied that there have been those who have spoken in this strain in the name of political economy, though such is not the teaching of leading English exponents of the science. "The distribution of wealth," says J. S. Mill em- phatically, "depends on the laws and customs of society. The rules by which it is determined are what the opinions and feelings of the ruling portion of the community make them, and are very differ- ent in different ages and countries ; and might be still more different, if mankind so chose" (Political Economy, ii. 1, 1). Compare also II.] OPERATION OF MORAL FORCES. 43 forces of competition may usually exert a prepon- derating influence in the economic world, they have not the universality and necessity which is here as- cribed to them, nor are they incapable of being, as we may say, moralized. Economic phenomena depend upon the activity of free agents, whose customary behaviour may be modified not merely by legislative interference, but also by changes in their own moral standard, or in the social pressure brought to bear upon them by public opinion ; and it follows that, in general, we are not justified in assuming finality in regard to concrete industrial facts, or in affirming that, in the economic world, what is must be. It is true that extra-regarding motives are not in economic affairs as powerful or as constant in their operation as motives of a self-regarding character. Still they none the less do exercise an appreciable influence, and as the sense of social responsibility grows stronger and becomes more diffused their importance is likely to be increased 1 . Mill's Autobiography, p. 246, where he speaks of the modes of the distribution of wealth as dependent on human will, and capable there- fore of being modified by human effort. By "the ruling portion of the community" in the above extract we ought to understand not merely those who have a voice in framing a country's laws, but also those who mould public opinion and exert an influence on the moral tone of a people. 1 When people talk about supply and demand, they sometimes forget that these are themselves phenomena depending upon human will, and that among the changes which may lead to modifications in supply or demand are changes in moral conditions. This may 44 POLITICAL ECONOMY AND PRACTICE. [CHAP. It involves confusion of thought, however, to sup- pose that economic phenomena are for the above reason incapable of being studied positively, or that in our in- vestigation of them we are necessarily bound to pass a judgment upon their moral worth. To recognise the influence, actual or potential, exerted by the economic ideals that men may frame for themselves is not the same thing as to discuss the objective validity of those ideals ; and our treatment of economic science may re- main strictly positive (in the sense in which we are now using that term), while at the same time we enquire in detail in what ways economic phenomena are or may be affected by the pressure of public opinion, or by motives of justice, and kindliness, and concern for the general well-being. It has been argued that the science cannot be sepa- rated from the art of economics, because of the in- be the case, for instance, if, because the public conscience has been touched, people will not purchase commodities which they believe to have been produced under what they regard as immoral con- ditions; or if they will not deal with shops where the employers have the reputation of treating their employees meanly or harshly. The fact that, at any rate in the estimation of traders themselves, causes of this kind may operate to a very appreciable extent is shewn by the anxious indignation with which some large London firms re- pudiated certain statements made before the Select Committee of the House of Lords on the Sweating System (1888) in regard to their manner of paying their workpeople. More than one firm specially called the attention of the Committee to the fact that they "were being damnified and injured in their business by reason of the state- ments which were being made before their lordships." II.] INFLUENCE OF ECONOMIC IDEALS. 45 fluence exerted by the latter upon the actual course of economic development 1 . There is an element of truth in this argument, which has perhaps been sometimes overlooked ; but it does not establish the desired conclusion. Men are influenced in what they actually do by what they think they ought to do ; and economic precepts, when enforced by the agency of the law or public opinion, lead to modifications of economic facts But all this may be taken into account without leaving the positive for the practical standpoint. Consider, for example, the influence exerted upon mediaeval trade by doctrines of the illegitimacy of usury and of what constitutes a reasonable price. It is one thing to study the nature and extent of this influence. It is another thing to enquire into the validity of the doctrines themselves. And although the historian may more or less combine the two dis- cussions, they clearly admit of logical separation. We may conclude the argument contained in this section by the remark that, just as the science of psy- chology recognises the existence and operation of moral motives, yet does not pass ethical judgments, so po- litical economy may recognise the operation of moral motives in the economic world, and yet not become an ethical science. 1 Compare Professor H. C. Adams in Science Economic Discussion, p. 102. 46 POLITICAL ECONOMY AND PRACTICE. [CHAP. 3. Grounds for recognising a distinct positive science of political economy, the sole province of which is to establish economic uniformities. Granting that it is logically possible to separate the positive from the ethical and practical study of economic phenomena, there is still no absolute inconsistency in holding that such a separation is undesirable. It may be pointed out how enormous is the influence exerted upon the well-being of mankind by the modes in which wealth is produced and distributed; and stress may be laid upon the fact that those human activities, which constitute the subject-matter of the economist's investigations, have an ethical significance, which is at least as worthy of consideration as their economic significance. It is indeed not strange that the idea of an essentially ethical treatment of political economy should have a strong fascination for earnest minds. Nor is it strange that as our social sympathies grow broader and stronger, the notion of stopping short at the purely positive enquiry should be viewed with an increasing degree of impatience. But in all this the point really at issue is ob- scured. No one desires to stop short at the purely theoretical enquiry. It is universally agreed that in economics the positive investigation of facts is not an end in itself, but is to be used as the basis of a prac- tical enquiry, in which ethical considerations are allowed II.] POLITICAL ECONOMY AS A POSITIVE SCIENCE. 47 their due weight. The question is not whether the posi- tive enquiry shall complete as well as form the founda- tion of all economic discussion, but whether it shall be systematically combined with ethical and practical enquiries, or pursued in the first instance independently. The latter of these alternatives is to be preferred on grounds of scientific expediency. Our work will be done more thoroughly, and both our theoretical and our practical conclusions will be the more trustworthy, if we are content to do one thing at a time. The following are, in rather more detail, the reasons that may be given for explicitly recognising the independence of the positive enquiry 1 . (1) The attempt to fuse together the enquiries into what is, and what ought to be, is likely to stand in the way of our giving a clear and unbiassed answer to either question. Our investigation, for instance, of the laws that determine competitive wages cannot but be seriously hampered, if the very same discussion is to serve for a solution of the problem whether wages so determined are fair wages. The value of economic 1 The question of combining positive and ethical enquiries is a somewhat different one from that of combining enquiries that belong respectively to the departments of science and art. The two questions have, however, a good deal in common, and we shall, therefore, in order to avoid unnecessary repetition, treat them together. The reader will observe that in the following arguments the undesirability sometimes of the former, and sometimes of the latter, combination is chiefly had in view. 48 POLITICAL ECONOMY AND PRACTICE. [CHAP. theories is, indeed, rightly measured by their ultimate bearing on practical questions; and the economist should always seek to direct his theoretical investiga- tions into the channels that will eventually prove most useful from the practical standpoint. But while the ultimate aim may be to guide human conduct, the immediate object to be kept in view is knowledge of positive facts. Such knowledge is not likely to be accurate and thorough, if, instead of pursuing his theoretical enquiries systematically, the economist works them out piecemeal, as they happen to rise into importance in connexion with particular practical issues. It may require an effort to keep the practical problems in the background even temporarily, but in the long run the guidance afforded will be the more trust- worthy, if its scientific foundations are first made secure 1 . 1 Bacon, in an often quoted passage, comments on the hasty and untimely eagerness with which men are apt to turn aside from pure science to its practical applications. "Whence it comes that, like Atalanta, they go aside to take up the golden apple, so meanwhile interrupting their course and letting victory slip out of their hands. But in the true course of experiment, and the carrying it on to new effects, the Divine Wisdom and Order are entirely to be taken as our examples. Now God on the first day of Creation, created only Light ; and gave a whole day for that work, and on that day created no material object. Similarly, in experience of every kind, first the discovery of causes and true Axioms is to be made ; and light-bringing not fruit-bringing experiments to be sought for. But Axioms rightly discovered and established supply practical uses not scantily but in crowds ; and draw after themselves bands and troops of effects " (Novum Organum, Book i. Aph. 70). II.] CONFUSION BETWEEN LAWS AND PRECEPTS. 49 It may be added that since purely economic data rarely by themselves suffice for the complete solution of practical problems, either our solution of the latter will be incomplete, or else the discussion that belongs to the positive science of economics will not improbably be overlaid by the introduction of considerations which, so far as it is concerned, are extraneous. (2) The attempt to combine theoretical and prac- tical enquiries tends to confirm the popular confusion as to the nature of many economic truths. What are laid down as theorems of pure science are constantly interpreted as if they were maxims for practical guid- ance. In spite of repeated protests from economists themselves, there is an inveterate disposition on the part of the public to regard the principles of political economy as essentially rules of conduct, even when the sole intention of those who formulate them is to de- termine what is, and not to prescribe what ought to be. Thus, because in economic theory men's action in buy- ing and selling is commonly assumed to be governed by self-interest, political economy is supposed to in- culcate selfishness ; because many economic truths are based on the postulate of competition, trades-unions are spoken of as violating economic laws ; and because it is laid down that, in a perfect market, price is de- termined by supply and demand, the science is repre- sented as teaching that price ought so to be determined. K. 4 50 POLITICAL ECONOMY AND PRACTICE. [CHAP. This kind of confusion is perhaps particularly common in England where, for reasons that are to be found in the historical development of the science, political economy has to a large extent become identified in the public mind with the policy of laisser faire 1 . In order to remove this prejudice, it is essential that economic precepts should be carefully distin- guished from the theorems of the positive science upon which they are based. But if theoretical and practical enquiries are merged together, the distinction can never be made thoroughly clear. Moreover, if we profess that our treatment of the subject is throughout ethical, then where we do not blame we shall naturally be understood to approve, or at least to excuse. Actually to express on every occasion an ethical judgment, and in so doing to strike always the true note, is an im- possibility. It may be added that the moral character of economic phenomena varies even when their scientific character is the same. (3) There is a further reason why a positive science of political economy should receive distinct and in- dependent recognition. With the advance of knowledge, it may be possible to come to a general agreement in regard to what is or what may be in the economic 1 The connexion between political economy and the doctrine of laisser faire is touched upon further in a note at the conclusion of this chapter. II.] CONFLICTING IDEALS. 51 world, sooner than any similar agreement is attainable in regard to the rules by which the economic activities of individuals and communities should be guided. The former requires only that there shall be unanimity as to facts ; the latter may be prevented by conflicting ideals, as well as by divergent views as to the actual or the possible. Take, for instance, the problem of socialism versus individualism. Even if philosophers are agreed as to facts, they may still arrive at contrary solutions of this problem, because they differ as to the true ideal of human society, and as to the comparative importance to be assigned, say, to the realisation of individual freedom. If political economy regarded from the theoretical standpoint is to make good progress, it is essential that all extrinsic or premature sources of controversy should be eliminated ; and we may be sure that the more its principles are discussed independently of ethical and practical considerations, the sooner will the science emerge from the controversial stage. The intrusion of ethics into economics cannot but multiply and per- petuate sources of disagreement. For if an ethical treatment of economic problems is to be systematic and thorough, and not merely sentimental and super- ficial, fundamental ethical questions that have long been the subject of controversy cannot be excluded such questions, for instance, as the determination of a 42 52 POLITICAL ECONOMY AND PRACTICE. [CHAP. standard of justice, and the relation of this standard to the ordinary utilitarian standard. However necessary it may be to face these questions at a later stage, there is no reason why we should not have a positive science of economics that is independent of them. In the following pages, then, it is this positive science that is meant by political economy when that term is used without further qualification 1 . At the same time, it does not follow that in pursuing his theoretical investigations, the economist need con- sider himself altogether precluded from indicating the ethical or practical significance of the theorems of fact that it is his primary object to formulate. An iso- lation of this kind is generally speaking impracticable. There are, indeed, some practical questions, especially in currency and banking, in which economic considera- 1 As a designation for the positive science, economics or economic science may be preferred to political economy, as being less likely to be nsed ambiguously. The name political economy is, however, too well established to be altogether discarded ; and we, therefore, use all three of the names more or less indiscriminately. It is to be added that whatever ambiguity attaches to political economy is beginning also to attach to the other terms. Dr Cunningham, for instance, in his Economics and Politics in the main understands by economics, and by economic science, a system of maxims. "Economic science," he says, " is wholly practical, it has no raison d'etre except as directing conduct towards a given end," namely, the pursuit of wealth; and the principles of economics are accordingly described as practical principles (such as are embodied in the mercantile system or in the system of laisser faire) which state the means to be pursued with reference to this end. II.] APPLICATIONS OF ECONOMIC THEOREMS. 53 tions are of such paramount importance, and the con- nexion between theory and practice is so immediate and obvious, that the refusal to consider at once the prac- tical bearing of the theoretical discussion might seem to be unnecessarily pedantic, and to involve needless repetition. All that is meant is that if moral judg- ments are expressed, or practical applications pointed out, they should be regarded as digressions, not as economic dicta, constituting integral and essential portions of economic science itself. In other words, the theoretical and practical enquiries should not be systematically combined, or merged in one another, as is maintained by those who declare that political eco- nomy is an indivisible whole of theoretical and practical investigations. 4. Applied economics. It is unnecessary to insist upon the enormous practical importance of the theoretical knowledge that economic science affords. Industrial and financial policy can be rightly directed, only if based upon such knowledge ; and whether we seek to construct social ideals, or to decide upon ade- quate steps towards their attainment, an indispensable preliminary is a study of the economic consequences likely to result from varying economic conditions. The question may, therefore, be raised whether, granting the existence of an independent economic science, economists should not supplement their treat- 54 POLITICAL ECONOMY AND PRACTICE. [CHAP. ment of this science by constructing a definite art of political economy, in which maxims for practical guid- ance are explicitly formulated. In support of this view, it may be argued that if the economist is led to regard certain practical questions as definitely within his province, he is the more likely to direct his theoretical investigations into the most useful channels, so that they may ultimately become not only light-giving but also fruit-bearing. It may further be urged that by explicitly recog- nising the twofold character of political economy, while at the same time carefully distinguishing the stand- points from which it becomes a science and an art respectively, we shall best remedy the popular miscon- ception as to the true nature of economic laws '. Granting, however, the desirability of treating systematically the practical applications of economic science, it may be doubted whether the phrase art of political economy does not suggest a body of doctrine, definite in scope and at the same time complete in itself, such as is practically unattainable. The practical applications of economic theories are many and various ; and the precepts based upon a study of the science may vary, according as we take the individual, or the 1 This is one of the main grounds given by Dr Sidgwick for recognising a distinct art as well as a science of political economy (Principles of Political Economy, 1887, p. 395). II.] SCOPE OF POLITICAL ECONOMY AS AN ART. 55 national, or the cosmopolitan point of view. Leaving this point on one side, however, a more serious difficulty results from the universally recognised fact that but few practical problems admit of complete solution on economic grounds alone. It is true, as suggested in the preceding section, that in a few departments such as those of currency and banking we meet with cases where, having determined the economic consequences of a given proposal, we practically have before us all the data, requisite for a wise decision in regard to its adoption or rejection. But more usually when we pass, for instance, to problems of taxation, or to problems that concern the relations of the State with trade and industry, or to the general discussion of communistic and socialistic schemes it is far from being the case that economic considerations hold the field exclusively. Account must also be taken of ethi- cal, social, and political considerations, that lie outside the sphere of political economy regarded as a science. If, therefore, the art confines itself to the practical applications of the science, pure and simple, its precepts will necessarily lack finality. They cannot be more than conditional. At the same time, there is danger of their hypothetical character being forgotten, and of the idea consequently gaming currency, that the economist desires to subordinate all considerations that are not purely economic. 56 POLITICAL ECONOMY AND PRACTICE. [CHAP. If, on the other hand, the art attempts a complete solution of practical problems, it must of necessity be to a large extent non-economic in its character, and its scope becomes vague and ill-defined. It may, accordingly, be objected that in attempting to formulate an economic art, that shall lay down absolute rules for the regulation of human conduct, economists are claiming to occupy too wide a range, and to frame a body of so-called economic doctrine, that is really much more than economic, and that cannot with any advantage be separated from general political and social philo- sophy 1 . The question of recognising a definite art of political economy is to a certain extent a verbal one ; and if all possible misunderstanding as to the scope of such an art and its relation to positive economic science can be removed, the way in which this question is decided is comparatively unimportant. On the whole, however, it seems likely to conduce to clearness of thought to regard the branch of enquiry under consideration as forming the economic side of political philosophy, or of the art of legislation, or of social philosophy, as the case may be, rather than as constituting a distinct art of political economy. In lieu of such an art, we should 1 The difficulty of assigning a definite scope to political economy, considered as an art, is discussed in further detail in a note at the conclusion of the chapter. II.] APPLIED ECONOMICS. 57 then recognise special departments of political and social philosophy, dealing with practical questions, in which economic considerations are of material importance, for the discussion of which, therefore, economic knowledge is essential, and to the treatment of which economists will naturally turn their attention. Adopting this alternative, we may still sum up the more important practical applications of economic science under the name applied economics. This term has the special merit that it does not suggest a definite body of prin- ciples with scientifically demarcated limits 1 . 1 It must be pointed out, however, that the name applied economic* is also not altogether free from ambiguity. For a science may be applied in two ways : first, to the explanation of particular facts ; secondly, to afford guidance in matters of conduct. The term applied economics or applied political economy has indeed been employed in three different senses: (a) in the sense suggested in the text; (6) to designate the application of economic theory to the interpretation and explanation of particular economic phenomena, without any necessary reference, however, to the solution of practical questions ; (c) to mark off the more concrete and specialized portions of economic doctrine from those more abstract doctrines that are held to pervade all economic reasoning. Compare the following : (a) " Applied political economy studies economic phenomena with the immediate aim of providing safe rules for administration, or of directing economic institutions so that they may conduce to the general welfare. Its aim is therefore immediately practical, since it does not investigate the Iww or why of certain facts, but seeks rules for doing certain things well" Cossa, Guide to the Study of Political Economy, Part i, Chapter 2. Compare also Cornewall Lewis's division of politics into pure and applied (Methods of Observation and Reasoning in Politics, Chapter 3, 5). (6) "The following essays consist in part of attempts to apply the principles of economic science to the solution of actual problems, 58 POLITICAL ECONOMY AND PRACTICE. [CHAP. We may also, if we please, speak of certain of the practical enquiries under consideration, as the art of industrial legislation, the art of taxation or of State finance, and so on. To these phrases there seems to be no objection. In each case we have a distinct and fairly compact body of doctrine, and there is no im- plication that our data are exclusively economic. For the notion of one supreme art of political economy, we should thus substitute a series of arts, in each of which there is a limitation to some particular sphere of economic activity. 5. Political economy and ethics. The relation between political economy and ethics may now be stated more explicitly, although this will involve little more than a repetition of what has been already in- dicated. We have seen that since men's economic of which those presented by the California!! and Australian gold discoveries, and by the state of land tenure in Ireland, are the most important. So much of the volume may not improperly be described as essays in applied political economy. The remaining essays deal mostly with topics of a theoretical kind." Cairnes, Essays in Political Economy theoretical and applied, Preface. Under the name applied political economy, both kinds of applications noted above are here had in view. (c) " Currency, banking, the relations of labour and capital, those of landlord and tenant, pauperism, taxation, and finance are some of the principal portions of applied political economy, all involving the same ultimate laws, manifested in most different circumstances." Jevons on The Future of Political Economy. Applied economics in this last sense constitutes what may be called the concrete, as dis- tinguished from the abstract, portion of economic science itself. II.] ETHICS OF POLITICAL ECONOMY. 59 activities are partly determined by moral considerations, it may be necessary in positive economic science to take account of the operation of moral motives. It is not, however, the function of the science to pass ethical judgments ; and political economy, regarded as a posi- tive science, may, therefore, be said to be independent of ethics. But it is a different matter when we turn to the applications of economic science to practice, that is, to applied economics; for no solution of a practical problem, relating to human conduct, can be regarded as complete, until its ethical aspects have been considered. It is clear, accordingly, that practical discussions of an economic character cannot be isolated from ethics, except in so far as the aim is merely to point out the practical bearing of economic facts, without any attempt to lay down absolute rules of conduct. It may be added that although in the past there may have been a tendency with a certain school of economists to attempt the solution of practical economic questions without adequate recognition of their ethical aspects, there is, at the present time, no such tendency discernible amongst economists who have any claim to speak with authority. Here, then, is the third of those subdivisions of economic enquiry in the widest sense, which we began by distinguishing from one another. In logical order, this division stands intermediate between the two 60 POLITICAL ECONOMY AND PRACTICE. [CHAP. others between the positive science, that is to say, and the so-called art. It may be regarded as a branch of applied ethics, and may perhaps be called the ethics of political economy. In it the functions of the eco- nomist and the moralist are combined, the general principles of social morality being considered in their special bearing on economic activities 1 . In pursuing this enquiry, our object is scientifically to define men's duties in their economic relations one with another, and, above all, the duties of society, in so far as by its action it can control or modify economic con- ditions. In other words, we seek to determine standards, whereby judgment may be passed on those economic activities, whose character and consequences have been established by our previous investigation of economic facts. We seek, moreover, to determine ideals in regard to the production and the distribution of wealth, so as best to satisfy the demands of justice and morality. It is subsequently the function of applied economics, or of the so-called art of political economy, to enquire how nearly the ideal is capable of being attained, and by what means; and to determine how, subject to the above condition, the greatest aggregate happiness may be made to result from the least expenditure of effort. 1 A distinction may be drawn between the public and the private ethics of political economy. AS an illustration of doctrines belonging to the latter category, attention may be called to a volume of sermons by Mr W. Kichmond entitled Christian Economics. II.] ETHICS OF POLITICAL ECONOMY. 61 As an illustration, it may be pointed out that the many problems raised by mediaeval moralists, in con- nexion with the question as to what constitutes a just price, belong to the ethics of political economy. For instance Is it right to sell a thing for more than it is worth ? Is it right to sell a thing which is not of the substance or measure or quality it professes to be ? Is the seller bound to reveal a fault in an article ? Is it right in trade to buy cheap and sell dear? The modern doctrine that, under a system of thoroughgoing competition, normal value is determined by cost of pro- duction is, on the other hand, a doctrine that belongs to positive science. The true solution of the ethical ques- tion, as to what constitutes a just price, may of course be held to be that competitive price will be a just price, if only it can be guaranteed that the competition is really free and effective on the part of all concerned. Or it may be held that, while this is not an ideally just price, no juster price is practically attainable. But these doctrines are in no way implied in the ordinary doctrine of cost of production as the regulator of normal value. 6. Methodological importance of the distinctions indicated in this chapter. We may conclude the present chapter by briefly pointing out the methodological importance of the distinctions that have been indicated. The main point to notice is that the endeavour to merge 62 POLITICAL ECONOMY AND PRACTICE. [CHAP. questions of what ought to be with questions of what is tends to confuse, not only economic discussions them- selves, but also discussions about economic method. The relative value to be attached to different methods of in- vestigation is very different, according as we take the ethical and practical standpoint, or the purely scientific standpoint. Thus it would be generally agreed that, in dealing with practical questions, an abstract method of treatment avails less and carries us much less far than when we are dealing with theoretical questions. On the other hand, we are more directly dependent upon history and inductive generalization. Again, while economic uniformities and economic precepts are both, in many cases, relative to particular states of society, the general relativity of the latter may be affirmed with less qualification than that of the former. " Political economy," says Sir James Steuart, and by this he means the art of political economy, "in each country must necessarily be different"; and this is hardly too strong a statement. On practical questions there is nearly always something to be said on both sides, so that practical decisions can be arrived at only by weighing counter-arguments one against another. But the relative force of these arguments is almost certain to vary with varying conditions. Hence, in general, a given economic policy can be definitely formulated only for nations having particular economic II.] RELATIVITY OF ECONOMIC PRECEPTS. 63 surroundings, and having reached a certain stage of economic development. Applied to nations, not simi- larly situated, the policy is likely at least to require modification. It is even possible that what is excellent for a given nation at a given time may be actively mis- chievous and injurious for another nation, or for the same nation at a different period of its economic history. It follows, similarly, that the value of the economic institutions of the past cannot adequately be judged by reference to existing conditions alone. We are not here denying the relativity of economic the- orems, but merely affirming the greater relativity of economic precepts. Unless the distinction between theorems and precepts is carefully borne in mind, the relativity of the former is likely to be over-stated. It is through differences of this kind being over- looked that divergences of view on questions of method become exaggerated. In the controversies that ensue, one set of disputants is thinking mainly of theoretical problems, while the other set is thinking mainly of practical problems ; and hence each in turn is liable to commit the fallacy of ignoratio elenchi. Again, because political economy is regarded as wholly practical, some writers have been led erroneously to deny that any economic doctrines admit of definite or exact expression. It is implied that the study cannot yield anything more than a useful collection of rules, 64 POLITICAL ECONOMY AND PRACTICE. [CHAP. II. having a restricted validity and application, and subject to numerous limitations and exceptions. Even if it be granted that this description is not altogether inappli- cable to political economy conceived as an art, it is obviously a fallacy to assume, without having made any clear distinction between them, that what is true of economic precepts is equally true of economic theorems. The frequency of errors such as these must be our excuse for treating in so much detail the distinctions indicated in the present chapter. The chapters that follow relate almost exclusively to the scope of political economy conceived as a positive science, and to the methods whereby the theorems of this science are to be established. NOTES TO CHAPTER II. A. ON POLITICAL ECONOMY AND Laisser Faire. THE connexion between political economy and laisser faire may be discussed from two different points of view, which are not always as clearly distinguished as they should be. There is, first, the connexion between political eco- nomy and laisser faire considered as an assumption or basis of reasoning ; there is, secondly, the connexion be- tween political economy and laisser faire considered as a maxim or rule of conduct. (1) Abstract economic doctrines are for the most part based upon the assumption of free competition and absence of government interference. This assumption may indeed be said to have occupied a central position in the develop- ment of economic theories during the last hundred years. There are two reasons why this has been so. One is to be found in the general principle of reasoning that it is best to take the simplest cases first. If we can accurately determine what will ensue under conditions of economic freedom, we shall be the better able to deal subsequently with more complicated cases, and to estimate the influence exerted by various interfering agencies. The second reason is that in modern economic societies laisser faire has been K. 5 66 POLITICAL ECONOMY [CHAP. II. as a matter of fact the general rule. Conclusions, there- fore, based on the assumption of non-interference have more nearly corresponded with the actual facts of modern industry than those that could have been obtained from any other equally simple hypothesis. Beyond this, however, there is no essentially necessary connexion between political economy and laisser faire regarded as a basis of reasoning. Economists recognise that in certain states of society, actual or possible, the conditions are so different from those of modern industry that conclusions depending on the hypothesis of non- interference are not even approximately applicable. More- over, in relation to modern economic phenomena themselves, it becomes necessary ultimately to deal with more complex problems in which various interferences with thorough- going competition have to be taken into account. The assumption of laisser faire represents, therefore, only a preliminary stage, and by its aid we traverse only a portion of the ground that has to be covered in the course of our economic reasonings. A little reflection will shew that it is far from being the case that political economy always presupposes the absence of government interference. It investigates the effects of export and import duties, of bounties, and of state-created monopolies such as the opium monopoly in Bengal. It seeks to determine the influence exerted on wages by the existence of poor relief guaranteed by the State. In nearly all modern currency discussions as, for instance, those relating to bimetallism or the regulation of convertible paper currencies the whole argument is so far from being based on the assumption of laisser faire, that everything turns on NOTE A.] AND LAISfiER FAIRS. 67 the supposition that some control over the currency is exercised by governments. In short, wherever government intervention becomes a prominent factor, the economist recognises and discusses the influence exerted by it ; and if in the future the part played by the State in economic affairs is extended, account will have to be taken of the fact in current political economy. A contrast is sometimes drawn between a socialistic and an economic state of society ; but when the distinction is thus expressed, the term economic is not used in the sense given to it by economists themselves. It is true that in a purely communistic society a good deal of ordinary economic theory relating to distribution and exchange would be irrelevant or inapplicable. But although in such a society men's economic activities would be in cer- tain directions controlled, they would clearly not be anni- hilated; and a scientific discussion of economic phenomena would, therefore, be by no means unnecessary. The func- tions of capital and the manner of its co-operation with labour would, for example, still require elucidation. Cost of production would still admit of analysis, and we should still have the phenomena of increasing and diminishing returns. "VVe should indeed still have the phenomenon of rent, meaning thereby the difference between cost of pro- duction under more favourable and under less favourable circumstances. Schemes of socialism, moreover, as distinguished from pure communism, do not necessarily involve the entire abolition of free exchange. Under such schemes, therefore, a theory of exchange would still be required. And unless our socialistic community were isolated from all others, 52 68 POLITICAL ECONOMY [CHAP. II. there would still remain for discussion extremely complicated questions of foreign trade and international exchanges. We may be sure, finally, that if some of the old eco- nomic phenomena were to become obsolete, new ones would arise demanding scientific treatment. While then a contrast may be drawn between our current political economy in so far as it is specially designed to elucidate the existing economic order, and the form that the science would be likely to assume in so far as specially designed to elucidate the phenomena of a socialistic society, it is clearly erroneous to imagine that the triumph of socialism would mean the extinction of political economy as a science. (2) We may pass to a consideration of the connexion between political economy and laisser faire regarded as a maxim of conduct. The question is quite distinct from that which we have just been discussing. For it is clear that we may on the one hand work out the consequences of laisser faire with the very object of discrediting it as a practical principle; or that we may on the other hand recognise the necessity of investigating the economic effects of government interference, while deploring the fact that instances of such interference are ever to be met with. Nevertheless the questions are not unfrequently con- fused together, and because laisser faire is a common economic postulate it is supposed to be a necessary economic precept. This confusion of thought has been encouraged by the circumstance that, until comparatively recently, the leading modern writers on the science have in their practical teaching expressly advocated a general policy of non-interference with trade and industry. Hence, NOTE A.] AND LAISSER FAIRS. 69 starting with the conception of political economy as the general art of legislation in matters relating to wealth, the public have come to identify it with the particular system of reducing government interference to a minimum ; and the maxim of natural liberty that everyone should be left free to use his mind, his body, and his property in the manner he deems best for himself is often regarded as the fundamental economic axiom. Political economy being thus transformed into a dog- matic creed, the worth of the study itself is measured by the degree of acceptance accorded to this creed. So common is the identification of political economy with the principle of non-interference that rarely do we find any professed attack on the former that does not on analysis resolve itself mainly into an attack on the latter. Com- pare, for instance, Dr Hutchison Stirling's vigorous dia- tribe in his Secret of Hegel. 1 Similarly, when people talk about political economy being exploded and becoming a thing of the past, all they mean is that laisser faire is ceasing to be an accepted maxim. On reflection it is clear that there is an inherent absurdity in attacking political economy as distinct from any particular system of political economy. For if particular systems are exploded, that only necessitates their being replaced by some other system. Regarding political economy as a positive science, it is of course clear that neither laisser faire nor any other maxim of conduct can form an integral portion of its teaching. Hence the advocacy of a policy of laisser faire 1 Volume 2, pp. 569, ff. Compare, also, Carlyle's various attacks upon political economy. 70 POLITICAL ECONOMY [CHAP. II. by individual economists, based though it may be on their interpretation of economic truths, belongs at any rate to applied economics. It has been said above that the leading English economists have in their practical writings been as a rule in favour of laisser faire. Looking a little closer, however, we find that their advocacy of the principle is at any rate accompanied by numerous qualifications and excep- tions. They do not regard it as an axiomatic and inexorable formula by which all particular proposals may be finally tested, but as a practical conclusion whose validity in every case depends on particular circumstances. 1 Adam Smith, for example, holds that besides maintain- ing such public institutions as are necessary for defence and the administration of justice, it is the duty of govern- ments to maintain certain institutions for facilitating com- merce and promoting education. " The third and last duty of the sovereign or commonwealth," he remarks, "is that of erecting and maintaining those public institutions and those public works, which though they may be in the highest degree advantageous to a great society, are, how- ever, of such a nature, that the profit could never repay the expense to any individual, or small number of indi- viduals ; and which it, therefore, cannot be expected that any individual, or small number of individuals, should erect or maintain." He moreover admits exceptions to a 1 "Let us remember," says Cairnes, "that laisser faire is a prac- tical ruU, and not a doctrine of science ; a rule in the main sound, but like most other sound practical rules, liable to numerous excep- tions ; above all, a rule which must never for a moment be allowed to stand in the way of the candid consideration of any promising proposal of social or industrial reform" (Essays in Political Economy, p. 251). NOTE A.] AND LAISSER FAIEE. 71 policy of free trade; for he explicitly recognises certain cases in which protection to native industry is desirable, and other cases in which it may rightly be a matter of deliberation "how far it is proper to continue the free importation of certain foreign goods," or " how far, or in what manner, it may be proper to restore that free im- portation, after it has been for some time interrupted." There are other instances in which he justifies interference. "The law," he says, "which obliges the masters in several different trades to pay their workmen in money, and not in goods, is quite just and equitable." Again, while allowing that any regulations affecting the note issues of a country may be regarded as " in some respect a violation of natural liberty," he nevertheless considers that certain regulations of the kind may be justified on the ground that "those exertions of the natural liberty of a few individuals, which might endanger the security of the whole society, are, and ought to be, restrained by the laws of all governments; of the most free, as well as of the most despotical." 1 Turning to Malthus, we find in the ranks of leading English economists a defender of the corn laws. Mill's long list of exceptions to the rule of laisser faire is well known. Coming to quite recent writers there is still less justification for the notion of an essential and necessary 1 On Adam Smith's attitude towards laisser faire, compare Sidgwick, Scope and Method of Economic Science, pp. 5 7. " To attribute to Adam Smith," says Dr Sidgwick, " a dogmatic theory of the natural right of the individual to absolute industrial independence as some recent German writers are disposed to do is to construct the history of economic doctrines from one's inner consciousness." Compare, also, Professor Nicholson's edition of the Wealth of Nations, Intro- ductory Essay, pp. 14 18. 72 POLITICAL ECONOMY AND LAISSKR FAIRS. [CHAP. II. connexion between political economy and the principle of non-interference. One of the distinguishing marks of a powerful school of economists at the present time is the definite repudiation of this principle ; and even those, who find in the study of economics very strong arguments against protective and against socialist legislation, would still never think of setting up the acceptance of a policy of unrestricted freedom of trade and industry as a test of economic orthodoxy. B. ON THE SCOPE OF POLITICAL ECONOMY CONSIDERED AS AN ART. SOME of the difficulties, which arise in the endeavour to determine the scope of political economy considered as an art, have been briefly indicated in the preceding chapter. In this note they will be discussed in somewhat further detail. Questions may be raised in regard to, first, the range of well-being contemplated by the art; and, secondly, the precise nature of the ideal at which it aims. (1) Under the first of the above heads it may be asked whether the aim of the economic art is individual or social ; and whether it is national or cosmopolitan. (a) It is clear that individuals as well as societies may in their own interests turn to account their study of economic science. The monopolist may derive practical guidance from the treatment of monopoly -value; the manufacturer from the discussion of over-production and industrial depression; the banker from the enquiry into NOTE B.] SCOPE OF POLITICAL ECONOMY AS AN ART. 73 the conditions under which crises tend to become periodic ; the trades-unionist from the analysis of the conditions favourable to the success of a strike. We might, accord- ingly, recognise a branch of the economic art, concerned with the principles according to which private persons should be guided in the pursuit of their own economic interests. There are, moreover, technical arts, such as the art of banking, which are to some extent based on economic science, but whose aim cannot be described as social. It is, however, generally agreed by those who advocate the recognition of an art of political economy that it aims at some result, that is desirable, not merely from the point of view of any given individual, but from the point of view of society taken as a whole. It is not regarded as an art of getting rich, or as an art of speculation, or as an art of investment, or as professing to indicate how producers should organize and carry on their business, in order to make their profits as great as possible. The art of political economy is, in other words, not identified with the whole of the practical applications of economic science '. 1 A minor question may here be raised, namely, whether in so far as the aim of the economic art is social, it is concerned wholly with legislation. M. de Laveleye defines political economy as determining "what laws men ought to adopt in order that they may, with the least possible exertion, procure the greatest abundance of things useful for the satisfaction of their wants; may distribute them justly, and con- sume them rationally" (Elements of Political Economy, 2). The departments of economic practice here had in view are clearly of the utmost importance; and there may be good grounds for giving a separate recognition to what may be called the art of State finance, and the art of industrial legislation. The former of these would in- clude a discussion of the general principles of taxation and of national debts from the practical standpoint. The latter would enquire how 74 SCOPE OF POLITICAL ECONOMY [CHAP. II. (b) Assuming that the aim of the economic art is social, not individual, the further question may be asked, whether it aims merely at national prosperity and national greatness, or at some result that is desirable for the -whole human race. This point is suggested by List's distinction between political economy and cosmopolitical economy. He regards the former as limiting its teaching to " the enquiry how a given nation can obtain (under the existing conditions of the world) prosperity, civilization, and power, by means of agriculture, industry, and commerce " ; while the latter "teaches how the entire human race may attain pros- perity". 1 It is true that cases of conflict between the precepts of the two arts here indicated are not likely to be frequent. But if it is maintained that no real conflict can ever occur, attention may be called to certain problems connected with emigration and immigration, and with export and import far and in what manner any State regulation of trade and industry is to be recommended. The art of political economy is, however, more usually regarded as having a wider scope than either of these. It may, for instance, in the matter of private almsgiving claim in the interests of society to formulate maxims for individual guidance ; or, with a view to the more equitable distribution of wealth, it may advo- cate the voluntary adoption of the co-operative principle or of profit- sharing ; or, accepting as its function a high moral task, it may seek in various ways to influence the economic activities of individuals so as to bring them into harmony with sound economic morality and secure the supremacy of right habits and customs in industrial life. 1 The National System of Political Economy (Sampson Lloyd's translation), p. 119. It is recognised by other economists besides List including some distinctly free trade economists that the general problem of free trade and protection needs to be handled somewhat differently according as the national or the cosmopolitan standpoint is adopted. NOTE B.] AS AN ART. 75 duties. Mill points out cases in which it is possible for a country to gain at the expense of other countries by the imposition of export duties. The opium trade between India and China affords an actual instance of a country raising a large revenue from foreigners by means of what is practically a tax on exports. "It is certain, however," Mill adds, taking the cosmopolitan standpoint, "that what- ever we gain is lost by somebody else, and there is the expense of the collection besides : if international morality, therefore, were rightly understood and acted upon, such taxes, as being contrary to the universal weal, would not exist". 1 Further, in reference to the question of restricting the exportation of machinery, Mill observes that even if by such means a country might individually gain, the policy would still in his opinion be unjustifiable on the score of international morality. "It is evidently", he says, "the common interest of all nations that each of them should abstain from every measure by which the aggregate wealth of the commercial world would be diminished, although of this smaller sum total it might thereby be enabled to attract to itself a larger share". 2 Since then a conflict is sometimes possible, it behoves the exponent of the economic art to make clear his view as to what the aim of the art really is. Perhaps the most ob- vious solution is to recognise, as List suggests, two distinct arts an art of cosmopolitan economy, and an art of iiational economy. The precepts of the former might often require modification to suit the special circumstances of different nations, but it would be cosmopolitan in the sense that it 1 Unsettled Questions of Political Economy, p. 25. 2 Of. cit., p. 31. 76 SCOPE OF POLITICAL ECONOMY [CHAP. II. would have regard to the well-being of the greatest number, irrespective of nationality. The latter would deliberately sacrifice the interests of other nations, if they happened to be in conflict with those of the nation specially under consideration. (2) A more fundamental question in regard to the scope of political economy considered as an art relates to the nature of the ideal at which it aims. () Does it seek merely to point out the laws, and institutions, and eco- nomic habits, that are most favourable to the production and accumulation of wealth ? (b) Or does it enquire further by what means an ideally just distribution of wealth may be attained ? (c) Or does it widen its range still further, and ask how all economic activities both of the State and of individuals should be moulded, with a view to the general well-being in the fullest and broadest sense 1 This last alternative represents the prevailing view amongst German economists. (a) If the end at which the art of political economy aims is simply the increased production of wealth, its scope is certainly definite, and the data upon which its conclu- sions are based belong exclusively to economic science. Since, however, the production of wealth is not the sole or supreme end that a society will have in view in framing its laws and shaping its institutions, the economic art so conceived can lay down no absolute or final rules. It can only speak conditionally, and say that in so far as the in- creased production or accumulation of wealth is concerned, such and such a line of action should be adopted. Hence, before deciding to act upon the hypothetical precepts of political economy (so interpreted), it is necessary to enquire NOTE B.] AS AN ART. 77 how far they are consistent with other social aims, and how far they satisfy the claims of justice. Wherever there is conflict, an appeal must be made to some other and higher authority. This authority must determine to what extent each set of considerations shall be subordinated to the others. It seems a doubtful gain to construct a definite art of political economy in this sense. For inasmuch as the science of economics itself contains all the information that is requisite, it will suffice to call attention to the practical bearing of its theorems, without systematically converting them into precepts. To frame a definite system of precepts, having regard entirely to the increased production of wealth, can indeed hardly fail to give rise to misapprehension. As a matter of fact, political economy has not unfrequently been subjected to startling misrepresentations, because it has first been identified with the art of making wealth a maximum, and then the necessarily hypothetical character of such an art has been forgotten. It is of little use to protest that economic precepts are not necessarily to be acted upon. If we have once formulated maxims of policy, and proclaimed that economic principles are directly prac- tical, the impression that economists desire to subordinate all considerations to the increase of wealth will certainly be encouraged. If, however, it can be made clear that eco- nomic principles are in themselves positive, and that, whilst economics shews, amongst other things, how laws and insti- tutions influence the production and accumulation of wealth, still it does not itself base any rules of action upon such knowledge, but merely places the results of its investiga- tions at the service of the legislator and social reformer, to be by them duly weighed and considered, then the chance 78 SCOPE OF POLITICAL ECONOMY [CHAP. II. of such misapprehension will at any rate be reduced to a minimum. (b) According to Professor Sidgwick, "we may take the subject of political economy considered as an art to in- clude, besides the theory of provision for governmental ex- penditure, (1) the art of making the proportion of produce to population a maximum,... and (2) the art of rightly distributing produce among members of the community, whether on any principle of equity or justice, or on the economic principle of making the whole produce as useful as possible". 1 This conception of the economic art is broader than that discussed above. But it seems to go either too far or else not quite far enough. For we pass outside the boundary of economic considerations in the narrowest sense, taking account also of considerations of justice ; and yet our maxims will still be, in some cases, only conditional. They cannot claim to be absolute, until we have taken into account all classes of considerations that may in any way be pertinent. In framing maxims of taxation and State finance, for example, political and social aims have to be borne in mind as well as equitable and strictly economic aims. The same may be said of free trade or protectionist maxims. Again, in seeking to deter- mine what is the ideal distribution of wealth, we ought to consider not merely the relation of distribution to desert, but also the manner in which methods of distribution affect the various other elements of social well-being. The indi- vidualistic organization of industry is by some writers con- demned on the ground of the anti-social spirit, engendered 1 Political Economy, 1887, p. 397. NOTE B.] AS AN ART. 79 by the competitive struggle. On the other hand, the so- cialistic organization of industry is by a different set of writers condemned on the ground that it hinders the reali- sation of individual freedom, and the development of indi- vidual character. Both these arguments are independent of the effects of socialism and individualism on the produc- tion and distribution of wealth. (c) According to the third conception of the economic art, its aim is to direct the economic activities of the State and of individuals, with a view to the completest realization of social well-being. "Political economy", says Professor Schonberg, representing the view of the dominant German school, as well as his own view, "does not ask primarily whether the greatest possible amount of wealth is produced, but rather how men live, how far through their economic activity the moral aims of life are fulfilled, and how far the demands of justice, humanity, and morality are satisfied". 1 Professor Ely, taking a similar view, and writing on behalf of the so-called "new school" of economists in the United States, describes the ideal of political economy as "the most perfect development of all human faculties in each individual, which can be attained". The aim, he goes on to say, is "such a production, and such a distribution, of economic goods as must in the highest practicable degree subserve the end and purpose of human existence for all members of society".* The end had in view is now the supreme end for which a society exists, and every question that arises is to be considered from all sides, and not from a single point of view. The rules laid down will accord- 1 Haiulbuch, Die Volksioirthschaft, 9. 2 Science Economic Discussion, p. 50. 80 SCOPE OF POLITICAL ECONOMY AS AN ART. [CHAP. II. ingly be no longer conditional, but absolute, at any rate in relation to the particular country or state of civilization under discussion. The above corresponds with the attitude that the great majority of economists of all schools have at least desired to take, so far as they have attempted a complete solution of practical problems for social purposes. The conception seems, moreover, to raise the economist to a position of greater importance than he can occupy, so long as he limits himself to purely theoretical investigations or merely con- ditional precepts. But does he not herein become a good deal more than an economist ? He will certainly need for his scientific basis very much more than economic science can by itself afford, for he must be a -student of political and social science in the widest sense. He must also solve fundamental problems of social morality. We have, in fact, no exception to the general rule that arts, claiming to lay down absolute rules, cannot be based exclusively on single theoretical sciences. We are, accordingly, led to the conclusion, indicated in the preceding chapter, that a definitive art of political economy, which attempts to lay down absolute rules for the regulation of human conduct, will have vaguely defined limits, and be largely non-economic in character. CHAPTER III. ON THE CHARACTER AND DEFINITION OF POLITICAL ECONOMY REGARDED AS A POSITIVE SCIENCE. 1. Political economy and physical science. In- asmuch as the production of material wealth is de- pendent largely upon physical conditions, it may be asked whether political economy does not partake to some extent of the nature of physical science. This question is, however, to be answered in the negative, on the ground that while the science has to take ac- count of the operation of physical laws, it is still con- cerned with them only indirectly; such laws do not constitute its subject-matter. It does not, for ex- ample, seek to establish or explain the physical laws that are involved in agriculture or mining or manu- facture. This is the function of such sciences as mechanics, chemistry, geology, and the science of agri- culture. The only concern of political economy with these laws is that it assumes certain of them as pre- misses or data, and makes them the basis of its own K. 6 82 POLITICAL: 7JJ(TJP,4I- i ? "AND PHYSICAL SCIENCE. [CHAP. reasonings, tracing the influence which they exert in moulding and modifying men's economic activities. Thus even the law of diminishing return from land, regarded as a bare physical fact, is hardly to be con- sidered a true economic law, although it no doubt occupies a unique position amongst the physical pro- legomena of economic science. Its economic import- ance consists in its relation to the productiveness of human labour as applied to land, and in its consequent influence upon the distribution and exchange of wealth. If economists are led into giving fuller details about agriculture than they usually give about processes of manufacture, it is because, from this point of view, the importance of the above law is exceptionally great. The relation of political economy to the physical sciences is then simply this, that it presupposes them ; it is sometimes concerned with physical laws as pre- misses, but never as conclusions. Accordingly when the production of wealth is said to be one of the great departments of economic science, reference is made primarily to what may be called the social laws of the production of wealth (i.e., to the various influences exerted on production by division of labour, foreign trade, methods of distribution, and so forth), rather than to the physical processes by whose aid production is carried on. The physical requisites III.] DIFFERENTIA OF ECONOMIC LAWS. 83 of the production of wealth need to be summarised in their broadest outlines; but the science is not directly concerned with the technique of different trades and occupations. Again, whilst economists re- cognise the physical conditions affecting men's eco- nomic efficiency, the immediate effects of these con- ditions are accepted as facts from physiology and other sciences ; it is only in so far as they indirectly affect or are affected by the social facts of wealth that economic science itself investigates them. The differentia of economic laws, as contrasted with purely physical laws, consists in the fact that the former imply voluntary human action 1 . The forces of competition are, indeed, sometimes spoken of as if they were themselves mechanical and automatic in their operation. But, as we have already had occasion to remark, this is not the case. When, for instance, we speak of the price of a commodity as determined by supply and demand, we mean by supply not the total amount in existence, but the amount offered for sale by holders of the commodity; and it is clear that in 1 " Various laws of nature have to be considered in connexion with human economy, but these are not economic laws. By the latter we understand laws of economic facts. An economic fact is not a pheno- menon of the natural, material world. It originates when in some way man, as an intelligent being with free will, enters actively into co-operation with natural phenomena, for the purpose of satisfying human needs." Schonberg, Handbuch der Politischen Ofkonomie, Die Volkswirtluschaft, 13. 62 84 POLITICAL ECONOMY AND PSYCHOLOGY. [CHAP. this sense supply, equally with demand, is dependent upon human judgment and will. 2. Political economy and psychology. In order to mark off political economy from the physical sciences, it is spoken of sometimes as a moral science, sometimes as a social science. Of these descriptions, the latter is to be preferred. The term moral science is, to begin with, not free from ambiguity. This term is no doubt sometimes used in a broad sense as including all the separate sciences that treat of man in his subjective capacity, that is, as a being who feels, thinks, and wills. But more frequently it is used as a synonym for ethics ; and hence to speak of economic science as a moral science is likely to obscure its positive character 1 . But the above is not the only reason why it is better not to describe political economy simply as a moral science. The sciences that relate to man fall into two subdivisions those that are concerned with man in his purely individual capacity, and those that are concerned with him principally as a member of society. Political economy belongs to the latter of these subdivisions. It is true that some of the pro- blems discussed by the science those relating, for example, to the functions of capital would arise in a more or less rudimentary form in relation to an isolated 1 Compare what has been said in the preceding chapter in regard to the relation of political economy to ethics. III.] POLITICAL ECONOMY A SOCIAL SCIENCE. 85 individual; and it is accordingly possible to illustrate certain elementary economic principles by reference to the conduct of a Robinson Crusoe. As soon, however, as we advance beyond the threshold of the science, it becomes necessary to regard human beings, not in isolation, but as members of associated communities consisting of others besides themselves. The most prominent characteristic of actual economic life is the relation of mutual dependence that subsists between different individuals; and political economy may be said to be essentially concerned with economic life as a special aspect of social life. Political economy should then be described as a social, rather than as a moral or psychological, science. It presupposes psychology just as it presupposes the physical sciences, and the natural starting point for the economist in his more abstract enquiries is a considera- tion of the motives by which individuals are usually influenced in their economic relations ; but the science is not therefore a branch of psychology. The bare facts that other things being equal men prefer a greater to a smaller gain, that under certain conditions they will forego present for the sake of future gratifications, and the like, are psychological facts of great economic im- portance. But they are assumed by the economist, not established by him. He does not seek to explain or analyse them ; nor does he investigate all the con- 86 POLITICAL ECONOMY AND PSYCHOLOGY. [CHAP. sequences to which they lead. Economic laws in the strict sense are different from the above. They are not simple laws of human nature, but laws of complex social facts resulting from simple laws of human nature. An illustration may be quoted from Cairnes. "Rent," he observes, "is a complex phenomenon, arising from the play of human interests when brought into con- tact with the actual physical conditions of the soil in relation to the physiological character of vegetable productions. The political economist does not attempt to explain the physical laws on which the qualities .of the soil depend; and no more does he undertake to analyse the nature of those feelings of self-interest in the minds of the landlord and tenant which regulate the terms of the bargain. He regards them both as facts, not to be analysed and explained, but to be ascertained and taken account of; not as the subject-matter, but as the basis of his reasonings. If further information be desired, recourse must be had to other sciences: the physical fact he hands over to the chemist or the physiologist; the mental to the psychological or the ethical scholar". 1 The fact that social, rather than purely psychological, phenomena constitute the subject-matter of political economy is clear if we take any recognised work on the science, such as the Wealth of Nations. Adam 1 Logical Method of Political Economy, pp. 37, 8. III.] CHARACTER OF ECONOMIC LAWS. 87 Smith traces many of the phenomena of wealth to man's mental constitution, but it is not man's mental constitution that it is his purpose to analyse. This is not always sufficiently borne in mind when the Wealth of Nations is contrasted with the Theory of Moral Sentiments, and spoken of as forming its sup- plement. Mill, indeed, in speaking of political economy uses the phrase "moral or psychological science"; and he goes on to define political economy as "the science relating to the moral or psychological laws of the production and distribution of wealth". 1 Take, how- ever, the following laws as formulated by Mill himself in his Principles of Political Economy: rent does not enter into the cost of production of agricultural pro- duce ; the value of money depends, cceteris paribus, on its quantity together with the rapidity of circulation; a tax on all commodities would fall on profits. Such laws as these ought certainly not to be described as moral or psychological, even though they rest ulti- mately upon a psychological basis. Moreover, as will be shewn in the following chapter, notwithstanding the importance of psychological pre- misses in certain departments of economic enquiry, the phenomena of the industrial world cannot be explained in their entirety simply by deductive reasoning from 1 Unsettled Questions of Political Economy, pp. 129, 133. 88 POLITICAL ECONOMY AND PSYCHOLOGY. [CHAP. a few elementary laws of human nature. To what purpose, and subject to what conditions, political eco- nomy uses its psychological data will be considered later on; it need only be said at this point that reason- ing from such data requires to be in various ways sup- plemented by direct observation of the complex social facts which constitute economic life 1 . 1 It may be remarked in passing that the description of economics adopted by Jevons in his Theory of Political Economy seems to give the science too much of a psychological, and too little of a social, character. The theory of economics is described as "the mechanics of utility and self-interest" (p. 23) ; it is "entirely based on a calculus of pleasure and pain; and the object of economics is to maximise happiness by purchasing pleasure, as it were, at the lowest cost of pain" (p. 25). A few pages further on the same idea is expanded. " Pleasure and pain are undoubtedly the ultimate objects of the cal- culus of economics. To satisfy our wants to the utmost with the least effort to procure the greatest amount of what is desirable at the expense of the least that is undesirable in other words, to maximise pleasure, is the problem of economics" (p. 40). The outcome of Jevons's conception of a calculus of pleasure and pain is a theory of utility, whose economic importance it would be difficult to exaggerate. Still this theory does not itself constitute the central theory of eco- nomics. It should indeed be regarded as an essential datum or basis of economic reasonings, rather than as itself an integral portion of the science at all. It seems more properly to belong to a branch of applied psychology, to which the name hedonics may be given. At the same tune, because of its economic importance, the economist must work out the theory for himself, if he does not find it worked out independently. Thus accidentally as it were, it may occupy an important place in economic writings, and yet be a premiss rather than an ultimate conclusion of the science. Jevons himself, after laying down his theory of utility, goes on to consider its applications to what are economic phenomena in the strictest sense. He thus throws so much light upon these phenomena that his Theory of III.] POLITICAL ECONOMY AND POLITICS. 89 3. Political economy a social as distinguished from a political science. From whatever point of view we look at it, political economy is best described as a social science; and if a distinction is drawn between social and political sciences, it must, notwithstanding its name, be regarded as belonging to the former, and not to the latter, category 1 . For while the science has sometimes to take account of political and legal condi- tions, it is essentially concerned with man in his social as distinguished from his political relations. It is, in other words, only in certain departments of political economy that we are concerned with men in their special character as members of a State. As remarked by Knies, "a large preliminary division of political eco- nomy has to investigate only the social economic life of man independently of all political influences". 8 The laws of distribution and exchange under conditions of free contract may be taken as an example. These laws do not exhaust political economy, but at any rate they fill a large and fundamentally important place in the science. Again, whilst economic doctrines may Political Economy, taken as a whole, may rightly be regarded as one of the most suggestive and valuable contributions to the science that have ever been made. 1 Hence a further reason, besides the one given in the note on p. 52, why some recent writers prefer to speak of the science as economics rather than as political economy. 2 Die Politische Oekonomie vom geschichtlichen Standpuncte, 1883, p. 3. 90 POLITICAL ECONOMY AND POLITICS. [CHAP. be in some cases relative to particular political con- ditions, they are more frequently relative to particular stages of industrial organization that are to a consider- able extent, if not altogether, independent of political influences. The above remarks relate primarily to the positive science of economics. Regarding political economy in its practical aspect, the connexion with politics is more intimate. Applied economics may indeed be said to be mainly concerned with the economic activities of the State in its corporate capacity, or of individuals as controlled by the State. Still, as we have already had occasion to shew, economic maxims having for their object the interests of society as a whole may also be formulated for the guidance of individuals act- ing independently of external constraint 1 . 4. Definitions of Wealth and Economic Acti- vity. The point has now been reached at which it seems desirable to give a formal definition of political economy, regarded as a positive science ; but before doing this, it is necessary briefly to discuss the meaning of certain terms that we have already had occasion frequently to make use of, namely, the terms wealth and economic activity. Wealth is one of those words that may without disadvantage be defined somewhat differently from different points of view; and it must 1 Compare Chapter 2, Note B. III.] DEFINITION OF WEALTH. 91 be borne in mind that our present object is merely to give a definition that shall suffice broadly to distinguish economic enquiries from those relating to other human interests. No attempt need, therefore, be made to deal with the difficulties that arise in connexion with the measurement of wealth. Utility may be defined as the power of satisfying, directly or indirectly, human needs and desires; and the possession of utility is the one characteristic that all writers are agreed in ascribing to wealth. It seems clear, however, that we cannot from our present stand- point identify wealth with all sources of utility what- soever, since there are many means of satisfying human needs, such as family affection, the esteem of acquaint- ances, a good conscience, a cultured taste, which have never been included within the scope of political eco- nomy, and the laws of whose production and distribu- tion have hardly anything in common with the laws that are as a matter of fact discussed by economists. Some characteristic besides the possession of utility must therefore be added, whereby such sources of utility as consist in a man's own nature, or in the subjective attitude of others towards him, may be ex- cluded from the wealth category. This further charac- teristic may be found in the quality of being potentially exchangeable. It is not meant that nothing is wealth unless it is actually bought and sold. For a thing may 92 DEFINITION OF WEALTH. [CHAP. be potentially exchangeable without being actually made the subject of exchange. Even in a communistic society the criterion would be applicable. It is true that with special reference to such a society, it might more naturally be expressed in another form, the essen- tial characteristic of wealth being described as the capability of being distributed by fiat of government. The sources of utility capable of being thus distributed would, however, be identical with those that could in a state of economic freedom be acquired by purchase. In either case personal qualities, and such objects of desire as affection and esteem, would be excluded. Wealth may then be defined as consisting of all potentially exchangeable means of satisfying human needs 1 . This definition brings within the category of wealth, in the first place, desirable material commodities that are capable of appropriation, such as food, books, build- ings, machines ; in the second place, rights and oppor- tunities to use or receive or in any way derive benefit 1 Whilst exchangeability may be selected as the most convenient differentia of wealth for purposes of definition, it should be pointed out that other characteristics become more prominent in some of the departments of economic science. For example, in the department of production, the primary notion is that wealth is the result of labour and sacrifice. Again, in the department of distribution, the fundamental idea is the right of appropriation. Compare Professor Nicholson's article on Wealth in the ninth edition of the Encyclopaedia Britannica. III.] SERVICES. 93 from material commodities, such as mortgages and other debts, shares in public and private companies, patents and copyrights, access to libraries and picture- galleries, and the like ; in the third place, personal services not resulting in any material product, as, for example, those rendered by actors, soldiers, domestic servants, lawyers, physicians; and, lastly, the right to command or control the services of any person over a given period. In regard to services it is to be observed that al- though the benefits they confer may be more or less permanent, they are in themselves merely transient phenomena. They are, however, the produce of labour ; they admit of being made the subject of exchange ; and they may possess exchange- value 1 . They give rise, 1 When a material commodity is sold or given away, its ownership is transferred from one person to another. When, however, a service is rendered, nothing passes into the possession of one person that was previously in the possession of another ; and hence it has been denied that services can possibly be exchanged. But it is a mistake to suppose that change of ownership either constitutes exchange or is essential to it. On the one hand, material commodities may change ownership gratuitously ; and on the other hand, we have all that is really essential to exchange, when A confers a benefit upon B on the understanding that B confers some other specified benefit upon him, and vice versa. Either benefit thus conferred may consist in the possession of some material commodity ; but it may also consist in a service rendered, that is, in an expenditure of effort on the one side accompanied by the satisfaction of some actual or supposed need on the other. It is, therefore, quite correct to say that a service may be rendered in exchange for another service, or in exchange for a material commodity; and it is quite correct to speak of services as having exchange-value. 94 DEFINITION OF WEALTH. [CHAP. therefore, to problems analogous to those which present themselves in connexion with material wealth; and they are accordingly rightly included under wealth from our present standpoint 1 . It will be noticed that the above definition excludes 1 It is sometimes considered essential to the idea of wealth to be susceptible of accumulation; and in speaking of services as wealth there is no doubt some departure from the ordinary usage of language. But, on the other hand, as Dr Sidgwick observes, "in ordinary estimates of the aggregate income of the inhabitants of a country, directly useful or, as we might say, 'consumable' services are commonly included : for as such services are reckoned as paid out of income, if we add the nominal incomes, estimated in money, of those who render such services as well as those who receive them, the result will only represent the aggregate real income of the country, if this latter notion is extended so as to include services" (Principles of Political Economy, 1887, p. 78). The following passage, from a journal of high standing specially devoted to economic questions, serves as an example of the kind of inaccuracy to which the omission of services from the category of consumable wealth may possibly give rise. "What comes out most strongly upon a review of the distribution of wealth is the smallness of the portion which is even theoretically available for redressing apparent inequality. It is only a small part of a wealthy man's riches which can actually be taken from himself, because it is after all but a small part which he actually con- sumes himself. The greater part is only his in so far as he directs the mode of spending or employing it. He directly maintains a number of persons who might conceivably be more usefully employed than in lounging in his hall, attending to his horses, or cultivating his flowers, but who are maintained, nevertheless, out of his wealth. Nearly the whole of his income goes in paying directly or indirectly for labour, and to take it from him means a general dislocation of the whole apparatus." This passage, if not actually erroneous, is at least very misleading. In addition to the material wealth that the rich man consumes, he enjoys a multiplicity of services which, under other conditions, might be distributed more equally through the com- munity. III.] PERSONAL ATTAINMENTS. 95 from the category of wealth personal abilities and attain- ments of all kinds. For abilities and attainments are not in themselves capable of being made the subject of exchange. We sometimes speak indeed of paying for the use of a man's skill ; but in reality the payment is for services rendered by aid of the skill. The right to command the services of any one over a given period is included under wealth, as pointed out above. The exclusion of human qualities and capacities from the wealth category is on the whole in accordance with scientific convenience, as well as with popular thought and speech, which as Professor J. B. Clark observes broadly distinguishes between the able man and the wealthy man, between what a man is and what he has 1 . At the same time, it is important to recognise that labour expended in the acquirement of skill is indirectly productive of wealth, in so far as it ulti- mately results either in material commodities or useful services. Thus, labour expended in acquiring the skill of the actor or the doctor is productive, as well as that expended in acquiring the skill of the carpenter or the shoemaker. Moreover, in any correct estimate of the productive resources of a country, the natural and ac- quired abilities of its inhabitants may occupy a position of the greatest importance. It may be added that the scope of political economy is practically not affected by the question whether 1 Philosophy of Wealth, pp. 5, 6. 96 DEFINITION OF ECONOMIC ACTIVITY. [CHAP. wealth is limited to exchangeable sources of utility, as in the above definition, or is used in a broader sense so as to include under the title of personal wealth all those capacities that enable men to be efficient producers of exchangeable wealth. For although so-called personal wealth does not, under the former alternative, directly constitute part of the subject-matter of the science, it still comes in for discussion as a source of wealth, and as such has still to be recognised as an economic factor of vital consequence. Wealth being defined as above, economic activity may be correspondingly defined as human activity which directs itself towards the production and appro- priation of such means of satisfying human needs as are capable of being made the subject of exchange. The economic life of a community is constituted by the eco- nomic activities of the members of which it is composed, acting either in their individual or in their corporate capacity. The term economy is sometimes used as equivalent to economic life, and by national economy is meant accordingly the economic life of a nation. It is to be observed that as civilisation advances, each individual becomes more and more dependent on others for the satisfaction of his needs; and hence economic life increases in complexity. In other words, with the progress of society, the organization of industry and the distribution of industrial functions grow increasingly complicated, and the phenomena resulting from men's III.] DEFINITION OF POLITICAL ECONOMY. 97 economic activities become more and more varied in character. 5. Definition of Political Economy. Political economy, regarded as a positive science, is often briefly denned as the science of the phenomena of wealth; and this definition has the merit of directness and simplicity. There seems, however, some advantage in attempting so to define the study as explicitly to indicate that it is not primarily concerned either with purely physical phenomena or with psychological or political phenomena as such, but with phenomena that originate in the activity of human beings in their social relations one with another. With this object, political economy may be defined as the science which treats of the phenomena arising out of the economic activities of mankind in society. It is not pretended that this or any other definition can by itself suffice unambiguously to express the nature of economics. The proposed formula must, therefore, be taken subject to the various explana- tions that have been already given, or that may subse- quently be given, in regard to the province of the science, and its relations to other branches of enquiry. It may be said of the definition of political economy, as of most other definitions, that the discussion leading up to it is of more importance than the particular formula ultimately selected. K. 7 NOTE TO CHAPTER III. ON THE INTERDEPENDENCE OF ECONOMIC PHENOMENA. THE phenomena with which political economy is con- cerned are usually classified under the heads of the pro- duction, distribution, exchange, and consumption of wealth. This separation of the science into distinct departments should not, however, be regarded as absolute or essential. The object of the classi6 cation is convenience of exposition; but such is the action and reaction between the phenomena in question, that it is impossible satisfactorily and com- pletely to discuss any one of the departments without having regard to the others also. Taking, for instance, production and consumption, it is obvious that men's habitual consumption determines what kinds of wealth shall be produced ; and, as indicated in the distinction between productive and unproductive con- sumption, the form in which wealth is consumed materially affects the amount produced. It is not quite so obvious, but it is equally true, that the production of wealth, both in kind and amount, is influenced by its distribution. The very rich consume luxuries, which, if wealth were more equally distributed, would in all probability not be produced at all or at any rate not to the same extent. Again, if CHAP. III. NOTE.] PRODUCTION OF WEALTH. 99 wealth were more equally distributed, there would in all probability be an increase in the average efficiency of the previously worst paid classes of the community, either in consequence of their being better fed, housed, and clothed, or in consequence of a better education and training having been provided for them by their parents; on the other hand, the number of hours during which they would be willing to work might be diminished. That the amount of wealth produced would in some such ways as these be affected by changes in distribution seems pretty certain ; although it is of course impossible to say a priori in what direction the effect would predominate. Turning to the connexion between production and exchange, it is to be observed that, as soon as division of labour is carried at all far, the former involves in some form or other the latter. Those whose function may priind facie appear to be simply and entirely to facilitate the exchange of wealth for example, bankers and bill-brokers, whole- sale merchants and retail dealers all play their part, and sometimes an important part, in assisting in the production of wealth. For without exchange in some form or other it is obvious that production could be carried but a very little way ; and, strictly speaking, the work of production ought not to be considered complete, until commodities have found their way into the possession of those persons whose intention it is to consume them '. In the case of exchange 1 J. S. Mill in his treatment of production introduces, under the head of capital, questions of distribution that might perhaps have been avoided. In his fundamental propositions respecting capital, and especially in the much criticized proposition that demand for commodities is not demand for labour, the truth of his conclusions 72 100 DISTRIBUTION AND EXCHANGE. [CHAP. III. and consumption, there is a still more intimate connexion ; for rates of exchange depend mainly upon laws of demand, and these in their turn depend directly upon laws of con- sumption. The connexion between distribution and exchange may be discussed from more than one point of view. If it is asked how the distribution of wealth is effected under modern industrial conditions, the answer clearly is by means of exchange. As it has been well expressed, "the adjustment of rates of exchange constitutes, in the aggregate, the process of distribution". 1 We may go even further, and say that in an individualistic society the theory of distribution resolves itself immediately into a theory of exchange-value. Each share into which the net produce of a community is divided represents the price paid for a certain service or utility afforded by the recipient of that share. Wages may thus be regarded as the ex- change-value of labour, interest as the exchange-value of the use of capital, rent as the exchange-value of the use of land 2 . From another point of view, the theory of the exchange- value of material commodities depends upon the theory of distribution. At any rate, as Cliffe Leslie insists, the theory of cost of production involves the whole theory of depends partly upon the assumption of the perfect mobility of capital and labour. But that is a subject that comes in for explicit discussion only in connexion with distribution, and much later in Mill's work. Here is perhaps one reason why Mill's chapters on capital are by some readers found so difficult, and, it may be added, unsatisfactory. 1 J. B. Clark, Philosophy of Wealth, p. 64. 2 Compare Sidgwick, Principles of Political Economy, 1887, p. 168. NOTE.] CONSUMPTION OF WEALTH. 101 wages and profits ; in other words, unless we have already determined a law of normal wages and a law of normal profits, the doctrine of cost of production is meaningless. It is, therefore, clear that theories of distribution and exchange cannot be divorced from one another, or discussed to any purpose in isolation. In connexion with the interdependence of economic phenomena, we may briefly touch upon a controversy that has been raised as to whether the consumption of wealth should or should not be regarded as constituting a distinct department of political economy 1 . The question is to a large extent one of convenience of arrangement, rather than of actual divergence of view in regard to the scope of the science. The following are among the topics, in addition to an analysis of the nature of economic consumption, that have been treated by different economists under the head of the consumption of wealth : the theory of utility, and the relation between utility and value 8 ; the distinction between 1 It may be observed that the consumption of wealth, in the sense in which the term is used by the economist, does not of necessity in- volve its destruction. We may say that by the consumption of wealth in political economy we mean its utilization, to which its destruction may or may not be incidental. Thus, in the economic sense, jewels are in process of consumption when they are being worn as ornaments ; so are the houses in which we live, and the pictures that hang on our walls. A house in which no one lives, or a picture that is stowed away in a lumber room, lasts at any rate no longer than one that is being rationally "consumed." Compare Senior, Political Economy, p. 54 ; and Walker, Political Economy, 328. Senior remarks that " it would be an improvement in the language of political economy if the expression 'to use' could be substituted for that 'to consume.'" 2 Compare Jevons, Theory of Political Economy. General Walker is of opinion that what has led to the practical excision of the whole 102 CONSUMPTION OF WEALTH. [CHAP. III. different kinds of consumption, and in particular the distinction between productive and unproductive con- sumption 1 ; the effects of different kinds of consumption, and in particular the effects of luxury 2 ; the policy of department of consumption from so many recent works is " the fascination of the mathematical treatment of economical questions, and the ambition to make political economy an exact science" (Poli- tical Economy, p. 298). We can hardly regard this view as correct ; for, if we take Jevons, who has insisted more strongly than any other English economist on the mathematical nature of the science, we find his most characteristic doctrines distinctly based on a theory of utility, which theory of utility he himself rightly regards as a theory of consumption. Indeed he lays it down explicitly that " the theory of economics must begin with a correct theory of consumption " (Theory of Political Economy, 1879, p. 43). Elsewhere he declares that the doctrine of consumption is the most important branch of the science, and he regards it as unaccountable and quite paradoxical that English economists should, with few exceptions, have ignored that doctrine. See Fortnightly Review, vol. 26, p. 625. Professor Walras of Lausanne, who is another representative mathematical economist, takes up practically the same position as Jevons. A con- sideration of the satisfaction of needs, which must involve a theory of consumption, is the basis of his doctrine of exchange-value. See his Elements d'Economie Politique Pure. 1 Compare J.-B. Say, Traite d'Economie Politique; James Mill, Elements of Political Economy; M c Culloch, Principles of Political Economy; Boscher, Grundlagen der NationaKkonomie; E. de Lave- leye, Elements de VEconomie Politique ; Leroy-Beaulieu, Precis d'Economie Politique; Lexis, Die volkswirthschaftliche Comumtion, in Schonberg's Handbuch der Politischen Oekonomie. 2 Compare Say, M c Culloch, Eoscher, de Laveleye, Leroy-Beaulieu, Lexis. M c Culloch, in his treatment of the consumption of wealth, brings out incidentally, but clearly, a point often supposed to be over- looked by economists, namely, that a taste for luxuries tends to in- crease, not to diminish, the amount of wealth produced. " The mere necessaries of life may, in favourable situations, be obtained with but little labour ; and the uncivilised tribes that have no desire to possess its comforts are proverbially indolent and poor, and are NOTE.] CONSUMPTION OF WEALTH. 103 sumptuary laws, and of other laws attempting to regulate consumption 1 ; the causes of commercial depression, and the impossibility of general over-production 8 ; insurance exposed in bad years to the greatest privations. To make men industrious to make them shake off that lethargy which benumbs their faculties when in a rude or degraded condition, they must be inspired with a taste for comforts, luxuries, and enjoyments" (Principles of Political Economy, p. 493). Cliffe Leslie goes a little too far in the same direction, when he remarks that unproductive expenditure and consumption " are the ultimate incentives to all production, and without habits of considerable superfluous expendi- ture a nation would be reduced to destitution" (Essays, 1888, p. 170). Men produce in the first place in order that they may live, and con- sumption which sustains a worker in efficient working condition is not usually spoken of as unproductive. Cournot (Principes de la Theorie des Richesses, 31) points out that we can conceive a state of society in which there would be no such thing as strictly unpro- ductive consumption ; for every satisfaction given to animal appe- tites might tend either to the preservation of health, the increase of strength, the prolongation of existence, or the propagation of the species. General Walker under the head of consumption exposes the fallacy that the mere destruction of wealth in some way increases production. This fallacy may be regarded as in a way the comple- mentary error of the true theory that a taste for luxuries and a high standard of comfort do, under certain conditions, tend to increase productive efficiency. 1 Compare M c 'Culloch, Eoscher, and Lexis. Professor Lexis dis- cusses the danger of the supply of certain commodities becoming exhausted (e.g., coal, petroleum, quicksilver), and touches on the policy of restraining in some way their consumption with the view of protecting the interests of future generations. He touches also upon interferences with consumption based on climatic, sanitary, and moral considerations; e.g., restrictions upon the destruction of forests, regulations in regard to house accommodation for the labour- ing classes, restrictions upon the consumption of alcohol. 8 Compare James Mill, Eoscher, Lexis, and Walker. This topic is treated under the head of consumption, because it relates to the equilibrium between production and consumption. Professor Lexis 104 CONSUMPTION OF WEALTH. [CHAP. III. and its economic advantages 1 ; government expenditure and the theory of taxation 2 ; the doctrine of population, and in particular the existence of economic wants and a standard of comfort as affecting the increase of population 8 . It is easy to shew that most of the above topics may quite naturally be dealt with in other departments of the science, as in fact they are by those economists who do not profess to treat explicitly of the consumption of wealth. The distinction, for instance, between productive and unpro- ductive consumption, and the effects generally of different forms of consumption on production, are not inappropriately discussed under the head of production itself; while the phenomena of (actual or apparent) over-production may be taken in connexion with the theory of exchange, since only goes on to touch upon possible remedies or palliatives for trade crises, such as State undertakings whose main object is to provide work for the unemployed, and emigration. 1 Compare Roscher, and de Laveleye. 2 This topic is brought under the head of the consumption of wealth by Say, James Mill, de Laveleye, and others, on the ground that taxation is the means whereby the consumption of government is provided for. 3 Compare Part v. of General Walker's Political Economy; also Part iv, Chapter 3, of Leroy-Beaulieu's Precis d' Economic Politique. It will be observed that even among those economists who recognise a distinct doctrine of consumption, there is far from being complete agreement as to what problems should be included within this depart- ment. The theory of population, for example, which General Walker introduces in this connexion, is treated by James Mill under distribu- tion, and by M c Culloch under production, while Eoscher makes it a fifth department of political economy, distinct from production, cir- culation, distribution, and consumption. There is, again, a divergence of view in regard to the place of commercial depression, and the theory of taxation ; and the theory of consumption of Jevons and Walras is quite different from that of any of the other writers above referred to. NOTE.] CONSUMPTION OF WEALTH. 105 under a system of exchange can these phenomena arise. Again, the incidence of taxation is directly connected with the phenomena and laws of the distribution of wealth ; and the remainder of the theory of taxation, except in so far as it relates to the effect of different forms of taxation on production, belongs to applied economics, rather than to the positive science of economics, with which alone we are here concerned. This last remark applies also to the dis- cussion of sumptuary laws, and to all enquiries how far and in what directions the increase of consumption should be encouraged or discouraged. Insurance may fairly be re- garded as a question of distribution. As to the theory of population, since labour is one of the requisites of produc- tion, the law of its increase may be discussed in connexion with production; or it may be included in the theory of distribution, in connexion with the laws regulating, through the supply of labour, the normal rate of wages. The theory of utility occupies, as we go on to shew, a unique position. It is, however, intimately connected with the determination of laws of exchange-value. The truth is that the phenomena of production, dis- tribution, exchange, and consumption, respectively, all so act and react upon one another, that if any one of these classes of facts is given no independent treatment, it must nevertheless come in for a large share of discussion in con- nexion with the others. Whether all propositions relating to consumption should be arranged by themselves or discussed as they arise in relation to other topics is, therefore, to a cer- tain extent a mere question of convenience of exposition 1 . 1 Thus, Dr Sidgwick, while explicitly admitting the fundamental importance of certain propositions relating to consumption, thinks 106 CONSUMPTION OF WEALTH. [CHAP. III. On the whole, it appears that the distinction between productive and unproductive consumption, the phenomena of over-production, the principles of taxation, &c., discussed under the head of consumption by James Mill and others, fall quite naturally and conveniently into other departments of economic science. The theory of utility, as discussed by Jevons, stands on a different footing. For unlike theories of taxation, population, and so on, it relates in itself purely to the consumption of wealth, and hence has a much stronger claim to be considered a distinct theory of consump- tion. At the same time, for reasons that have been already briefly indicated in connexion with Jevons's description of economics, this theory may be regarded as constituting part of the necessary prolegomena of economic science, rather than one division of completed economic doctrine. The consumption of wealth is not so much an economic that in such a treatise as his own, it is more convenient to introduce these propositions in dealing with the questions of production, distri- bution, and exchange, which they help to elucidate, rather than to bring them together under a separate head (Principles of Political Economy, 1887, p. 27). J. S. Mill (Unsettled Questions, p. 132, note) and Cherbuliez (Precis de la Science Economique, p. 5) are somewhat less guarded in their rejection of consumption as a special topic for dis- cussion. "Political economy," says Mill, "has nothing to do with the consumption of wealth, further than as the consideration of it is inseparable from that of production, or from that of distribution. We know not of any laws of the consumption of wealth as the subject of a distinct science : they can be no other than the laws of human enjoyment." The consumption of wealth, says Cherbuliez, is in its only important form a phenomenon which cannot be separated from the production of wealth. Unproductive consumption is the appli- cation of wealth to the needs for which it was produced. It requires no further discussion. When wealth gets into the hands of the unproductive consumer, economic activity is at an end. NOTE.] CONSUMPTION OF WEALTH. 107 activity in the sense in which that term has been above defined, as itself the end and aim of all economic activity. Wealth is produced, distributed, exchanged, in order that it may be consumed. The satisfaction of human needs is the motive power throughout. Thus, a true theory of con- sumption is the keystone of political economy ; but it may, nevertheless, be regarded as occupying the position of a fundamental datum or premiss of the science, rather than as constituting in itself an economic law or laws on a par with the laws of production, distribution, and exchange. CHAPTER IV. ON THE RELATION OF POLITICAL ECONOMY TO GENERAL SOCIOLOGY.' 1. Conflicting views of the relation between economic science and the general science of society. Before pro- ceeding further, it is necessary explicitly to enquire whether political economy is really entitled to rank as a distinct department of study at all. It is main- tained by Comte and his followers that on account of the extremely intimate connexion between the phenomena of wealth and other aspects of social life, any attempt to separate economic science from social philosophy in general must necessarily end in failure. The phe- nomena of society, it is said, being the most compli- cated of all phenomena, and the various general aspects of the subject being scientifically one and inseparable, it is irrational to attempt the economic or industrial analysis of society, apart from its intellectual, moral, and political analysis, past and present. It is admitted that certain of the facts of wealth may by a scientific CHAP. IV.] POLITICAL ECONOMY AND SOCIOLOGY. 109 artifice be studied separately, but it is denied that their investigation can constitute a distinct science 1 . In striking contrast to the above is the view of those economists, who regard political economy as an independent abstract science, dealing with the phe- nomena of wealth in isolation, and having no concern with other social phenomena. While on the former view the relation of economics to sociology is properly one of entire subordination or rather inclusion, on this view it is one of absolute independence ; the facts of wealth are to be studied in and by themselves ; they are to be treated as having no relation to other social facts; man is to be considered as a being who is oc- cupied solely in acquiring and consuming wealth. The truth lies between these two extreme views. What may be called the extreme separatist doctrine affirms of political economy as a whole what is true 1 Compare Miss Martineau's Positive Philosophy of Auguste Comte, vol. 2, pp. 51 54; also Mr Frederic Harrison's essay entitled Pro- fessor Cairnes on M. Comte and Political Economy in the Fortnightly Review for July, 1870. Comte's views have recently been given fresh prominence in this country by Dr Ingram's article on Political Economy in the ninth edition of the Encyclopaedia Britannica (republished as a History of Political Economy). Dr Ingram combines his history with an elaborate attack on the manner in which the science has been for the most part studied in England ; and arrives at the conclusion that political economy cannot any longer command attention as a fruitful branch of speculation unless it is subsumed under and absorbed into general sociology. "The one thing needful," he says, "is not merely a reform of political economy, but its fusion in a complete science of society." 110 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. only of a certain portion or aspect of it, and hence would leave the science incomplete. Comte's view, on the other hand, overlooks the fact that only by special- ization within proper limits can scientific thoroughness and exactness be attained in any department of know- ledge. Students of economics may, moreover, naturally and fairly ask to have the province of sociology itself more explicitly denned, and to see its own fundamental doctrines more clearly formulated, before they can be expected to shew a willingness to have political economy subsumed under and absorbed into it. It will be our endeavour to shew that whilst the study of economic phenomena cannot be completed without taking account of the influence exerted on the industrial world by social facts of very various kinds, it is nevertheless both practicable and desirable to recognise a distinct systematized body of knowledge, which is primarily and directly concerned with economic phe- nomena alone. On this view, economics is regarded as constituting one division of the general philosophy of society, of which other divisions are jurisprudence, the science of political organization, and the philosophy of religious, moral, and intellectual development ; but it is allowed its own set of specialists, and the necessity of systematically combining the study of economic phe- nomena with that of other aspects of human existence is denied. It is, in other words, held to be possible for IV.] POLITICAL ECONOMY AND SOCIOLOGY. Ill the economist to steer a middle course, neither as- suming throughout the whole range of his investigations an entirely unreal simplicity, nor, through the neglect of that specialization which has been found indispen- sable in the physical sciences, allowing himself to be hopelessly baffled by the complexity of the actual phe- nomena. It should be carefully borne in mind that through- out this chapter, as in the chapters that follow, political economy is regarded as a positive science. Similarly by sociology is understood a body of theoretical truth, not a system of practical maxims. The separate existence of economic theory is not imperilled, when it is admitted that practical arguments based on economic grounds alone are rarely in themselves decisive. The two questions are often not clearly distinguished from one another. It is, however, important to recognise that those, who stand out most strongly for the recognition of a separate economic science, may hold equally strongly that no true guidance in matters of conduct is to be obtained by appealing simply to economic considerations, all social consequences of a non-economic character being disregarded. 2. The place of abstraction in economic reasoning. According to what has been above called the extreme separatist view, political economy takes one aspect of human society and action, and considers it absolutely 112 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. alone and apart, the science being concerned with man solely as a being who desires to possess wealth. Just as the geometer considers the dimensions of bodies apart from their physical properties, and the physicist their physical properties apart from their chemical consti- tution, so the economist is said to regard man simply as a being who, in all his economic relations, is actuated by an enlightened self-interest, and who is also free to act accordingly, so far as he does not interfere with a like freedom on the part of others. This view of political economy is taken by Mill in his Essays, although, as we have previously had occasion to remark, his constructive treatise on the science is worked out on different and much broader lines 1 . He describes economics as treating of the laws of the pro- duction and distribution of wealth, not so far as these laws depend upon all the phenomena of human nature, 1 Professor Marshall puts the contrast very clearly and forcibly. " In 1830," he says, "John Mill wrote an essay on economic method, in which he proposed to give increased sharpness of outline to the abstractions of the science. He faced Bicardo's tacit assumption that no motive of action except the desire for wealth need be much considered by the economist ; he held that it was dangerous so long as it was not distinctly stated, but no longer ; and he half promised a treatise, which should be deliberately and openly based on it. But he did not redeem the promise. A change had come over his tone of thought and of feeling before he published in 1848 his great economic work. He called it Principles of Political Economy with some of their Applications to Social Philosophy ; and he made in it no attempt to mark off by a rigid line those reasonings which assume that man's sole motive is the pursuit of wealth from those which do not." IV.] ECONOMICS DEFINED AS AN ABSTRACT SCIENCE. 113 but only so far as they depend upon the pursuit of wealth, or upon the perpetually antagonizing principles to this pursuit, namely, aversion to labour, and desire of the present enjoyment of costly indulgences. Entire abstraction is to be made from every other human pas- sion or motive. In other words, the economist is sup- posed to take as his subject of study, not the entire real man, as we know him in all the complexity of actual life, but an abstraction usually spoken of as the economic man a being, who, in the pursuit of wealth, moves along the lines of least resistance, and does not turn aside towards other ends. In accordance with this view, political economy is defined as " the science which traces the laws of such of the phenomena of society as arise from the combined operations of mankind for the production of wealth, in so far as those phenomena are not modified by the pursuit of any other object". 1 It is admitted that the 1 Unsettled Questions of Political Economy, p. 140. Bagehot, in his Economic Studies, expresses himself similarly. "Political eco- nomy in its complete form, and as we now have it, is an abstract science, just as statics and dynamics are deductive sciences. And, in consequence, it deals with an unreal and imaginary subject" (p. 73). It "deals not with the entire real man as we know him in fact, but with a simpler, imaginary man a man answering to a pure defini- tion from which all impairing and conflicting elements have been fined away. The abstract man of this science is engrossed with one desire only the desire of possessing wealth" (p. 74). This view of political economy is justified by Bagehot on the ground that "the maxim of science is simply that of common sense simple cases first; K. 8 114 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. economist must allow for the interference of other impulses in applying his results, and that he ought, even in the formal exposition of his doctrines, to introduce many practical modifications ; but the recog- nition of these other impulses is excluded from the science itself. The one error involved in the above view is that of mistaking a part for the whole, and imagining political economy to end as well as begin with mere abstrac- tions. The practical modifications, of which Mill speaks, themselves demand a scientific treatment, and should, therefore, have a place accorded to them within the science itself. For in many cases they are not mere isolated modifications, admitting of application to in- dividual instances only. It is often possible to generalize on other foundations than that of the economic man ; and, at any rate, the various interferences with free competition admit of scientific enumeration and classi- fication. begin with seeing how the main force acts when there is as little as possible to impede it, and when you thoroughly comprehend that, add to it in succession the separate effects of each of the encumbering and interfering agencies" (p. 74). The maxim here cited may be accepted without hesitation ; but it seems hardly consistent with the previous statement that political economy in its complete form is an abstract science, <&c. Our contention is that while economics rightly begins with abstractions, in its complete form it has as good a claim to be called a realistic as to be called an abstract science. What appears to be the correct doctrine is clearly laid down by Cairnes in his Logical Method of Political Economy, pp. 42 45. IV.] CONCEPTION OF THE "ECONOMIC MAN". 115 The abstraction by which men are supposed to act solely with a view to their own material advantage and the advantage of those immediately dependent upon them has, nevertheless, its place in political economy *. For while it is true that our economic activities are subject to the influence of a variety of motives, which sometimes strengthen and sometimes counteract one another, it is also true that in economic affairs the desire for wealth exerts a more uniform and an in- definitely stronger influence amongst men taken in the mass than any other single motive. Hence, in order to introduce the simplicity that is requisite in a scientifically exact treatment of the subject, it is legitimate and even indispensable to begin by tracing the results of this motive under the sup- position that it operates without check. By thus ignoring at the outset all other motives and circum- stances, except those implied in the notion of free and thoroughgoing competition, we may, at any rate in certain departments of enquiry, determine the more constant and permanent tendencies in operation, and hence reach a first approximation towards the truth. As a matter of fact, this approximation is in some cases a very near approximation indeed. In dealing, 1 The "economic man" ought not to be described as a pure egoist. In making a choice of occupations for his children, for example, he has always been regarded as acting for them as he would for himself, and as desiring wealth on their account. 82 116 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. for instance, with prices on the Stock Exchange, or in the great wholesale markets, under modern in- dustrial conditions, we are for the most part con- cerned with the economic activities of persons who practically realise in actual life the notion of the economic man. This by no means implies that the persons referred to are what we should ordinarily call selfish. For men of the most unselfish character are, in many of their commercial dealings, influenced mainly by what are sometimes termed strictly com- mercial motives subject only to the restraints of law and of ordinary commercial custom and morality. They may desire wealth in order to educate and bring up their children with a view to their children's best interests, or in order that they may devote their wealth to particular philanthropic objects, or to the general well-being of the community to which they belong. Still the desire for wealth is in its imme- diate economic effects the same whatever its ulterior object may be 1 . 1 Cliffe Leslie criticizes the conception of the desire of wealth as a barren abstraction, in which are confounded together many different desires whose actual consequences are indefinitely various. "No other branch of philosophy," he remarks, " is still so deeply tinctured with the realism of the schools as economic science. A host of different things resemble each other in a single aspect, and a common name is given to them in reference to the single feature which they have in common. It is, properly speaking, only an indication of this common feature, but it puts their essential differences out of mind, and they come to be thought of in the lump as one sort of thing. The IV.] THE DESIRE FOR WEALTH. 117 There are cases, then, where motives other than pecuniary exert so little active influence that they may, desire of wealth is a general name for a great variety of wants, desires, and sentiments, widely differing in their economical character and effect, undergoing fundamental changes in some respects, while preserving an historical continuity in others. Moralists have fallen into a similar error, though from an opposite point of view, and, in their horror of an abstraction, have denounced, under the common name of love of wealth, the love of life, health, cleanliness, decency, knowledge, and art, along with sensuality, avarice, and vanity. So all the needs, appetites, passions, tastes, aims, and ideas, which the various things comprehended in the word ' wealth ' satisfy, are lumped together in political economy as a principle of human nature, which is the source of industry and the moving principle of the economic world... The division of labour, the process of exchange, and the intervention of money, have made abstract wealth or money appear to be the motive to production, and veiled the truth that the real motives are the wants and desires of consumers; the demands of consumers determining the commodities supplied by producers. After all the reproach cast on the Mercantile School, modern economists have themselves lapsed into the error they have imputed to it. If every man produced for himself what he desires to use or possess, it would be patent and palpable how diverse are the motives summed up in the phrase ' desire for wealth ' motives which vary in different individuals, different classes, different nations, different sexes, and different states of society... The desire for wealth is by no means necessarily an incentive to industry, and still less to abstinence. War, conquest, plunder, piracy, theft, fraud, are all modes of acquisi- tion to which it leads. The robber baron in the reign of Stephen, and the merchant and the Jew whom he tortured, may have been influenced by the same motives. The prodigal son who wastes his substance in riotous living is influenced by the same motives the love of sport, sensual pleasure, luxury, and ostentatious display which impel many other men to strenuous exertion in business" (Essays, pp. 166 170). The whole of the above argument is very persuasively put, but it does not establish the conclusion that Cliffe Leslie desires to establish. By the desire of wealth is meant the desire of general purchasing potcer, that is, the desire to increase one's 118 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. without serious risk of error, be neglected even in the concrete applications of economic doctrines. More usually, however, this abstraction from other influences yields only an approximation towards the actual truth, which approximation needs subsequently to be de- veloped and corrected. Even the above degree of validity is denied to the command over the necessaries and conveniences of life in general ; and nothing that Cliffe Leslie says proves it to be either an illegiti- mate or a barren assumption that in their ordinary economic dealings men are in the main influenced by this desire, and that, in consequence, a greater gain is preferred to a smaller. That there are enormous variations in men's ideas, as to the particular things that constitute the necessaries and conveniences of life, is nothing to the point. For, as observed in the text, the immediate effects of the desire of wealth may be the same, although the ulterior objects had in view may be very different. A man may desire general purchasing power, in order that he may be assisted towards the attainment of the noblest and most unselfish ends, but it does not follow that he will therefore sell his services or his goods at less than their market value. Granting that the object of men's desires are very various, still, as Dr Sidgwick puts it, so far as they are exchangeable and commensurable in value, they "admit of being regarded as definite quantities of one thing wealth; and it is just because the 'desire of wealth' may, for this reason, be made to include ' all the needs, appetites, passions, tastes, aims, and ideas, which the various things comprehended under the word wealth satisfy', that we are able to assume, to the extent required in deductive political economy, its practical universality and unlimitedness" (Political Economy, 1887, p. 36). We may add that it is also not to the point that, under different conditions, the desire of wealth may lead to very different lines of conduct. The assumption that men are actuated by this desire is, in economic reasonings, com- bined with other assumptions as, for example, the absence of force and fraud which circumscribe within certain limits the modes in which the motive can operate. IV.] CRITICISM OF THE METHOD OF ABSTRACTION. 119 postulate in question by some economists. They hold that if the abstraction whereby we suppose men to act solely with a view to their own advantage is not con- demned as leading to positive error, it should at least be rejected as practically useless. It is in manifest contradiction, they say, to the facts of life. Knies, for example, rejects it on the ground that a society of men actuated solely and continuously by self-interest, and allowed absolute freedom of action, has never actually existed. He admits the possibility of hypothetically working out the laws of price, &c., that would arise in such a society were it to exist, but he denies that such a hypothetical enquiry has any utility or practical justi- fication. One might just as well, he says, base an enquiry on the hypothesis that all men are inspired by altruism, or that they all have an equally strong im- pulse towards charity ; and he implies that such enqui- ries as these would be in all respects as serviceable, in enabling the economist to understand and explain the phenomena of the actual economic world 1 . The first point in the above argument that no society of pure egoists has ever actually existed is, strictly speaking, irrelevant. For the economists, whom Knies is criticizing, have always insisted that they are dealing with abstractions, with imaginary beings of a 1 Die Politische Oekotwmie vom geschichtlichen Standpunete, 1883, p. 504. 120 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. simpler type than are to be met with in real life. They have never posited the actual existence of a society of men, guided in all their actions by pure self-interest. The gist of their argument is not even that, in the one sphere of life with which they as economists are con- cerned, the desire for wealth operates by itself and subject to no interference from the operation of other motives. All they affirm is that, taking a broad survey of the economic sphere, the desire of each man to promote the material interests of himself and his family is far more powerful, and far more uniform in its operation, than the other motives, which sometimes act as a drag upon it. They hence infer that by calculating the consequences of this desire, they will be materially assisted towards determining what is most likely to happen on the average or in the long run 1 . The latter part of Knies's criticism is strictly rele- vant, if it is valid. It amounts to this :that a doctrine, based on the hypothesis of pure altruism, would be just as near the concrete reality, as one based on the hypo- thesis of pure egoism ; in other words, that, in their eco- nomic dealings with one another, men are as uniformly 1 Menger specially insists that the use of the so-called dogma of self-interest is misunderstood by the historical school of German economists, when they regard it as forming such a disturbing contrast to "full empirical actuality." See his Hethode der Sociahcissen- schaften, p. 79. IV.] DEFENCE OF THE METHOD OF ABSTRACTION. 121 and as powerfully influenced by an immediate desire to augment their neighbour's wealth as by a desire to augment their own. But this argument is certainly contradicted by all the facts of actual economic life. Look where we will in the industrial world, do we not find self-interest controlled though it may be by moral, legal, and social considerations the main force determining men's actions? Is it not a patent fact that in buying and selling, in agreeing to pay or to accept a certain rate of wages, in letting and hiring, in lending and borrowing, the average man aims at making as good a bargain for himself as he can? He may be restrained within certain limits by law, morality, and public opinion ; and the influence exerted by restraining forces such as these must be ultimately taken into account. Still the desire for wealth is, under normal conditions, the active impelling force ; and the immediate economic consequences of this desire are the same, however unselfishly the wealth gained may ultimately be expended. It is this fact of common experience that justifies economists in starting from the conception of the economic man, as approximately typical of actual men con- sidered in their economic relations. Conclusions based on this conception contain a hypothetical element ; but they are nevertheless, at any rate in certain departments of the science, within measurable dis- 122 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. tance of the concrete realities of the actual economic world 1 . 1 The above is freely allowed by some writers who on other grounds criticize the English school of political economy. "No economist," says Professor E. M. Smith, "would venture on the solution of an economic problem without taking into consideration the fact that men are ordinarily moved by self-interest" (Science Economic Discussion, p. 113). "Hypothetically," says Wagner, "the use of the theory of self-interest is always proper; and, for the isolation of causes, it has proved the best of methodological tools. For we have here an element common to all men. We have an element founded on a law which is in truth a 'natural' and universal law. It is based on the physical nature of man, on his mental nature (which depends primarily on his physical nature), and on his relations to the external world. As it affects the individual, so, also, it represents the interest of the species, since the species exists and is continued only through the individual. The objections of historical economists are obscure, and are carried too far, when, instead of admitting the hypothetical value of de- duction from self-interest, they deny that it has any value what- ever. They make a mistake which is the reverse of the mistake of the advocates of pure deduction ; and their mistake is the greater. In considering the modifications of industrial self-interest in dif- ferent individuals, different peoples, at different times, its various combinations with other motives, they forget that there is, after all, a universal element of humanity in this selfishness " (Jahr- biicher fur Nationattkottomie und Statistik, March, 1886, p. 231; Quarterly Journal of Economics, October, 1886, p. 118). Wagner goes on to point out, what we have already insisted upon, that "self- interest, when spoken of as the motive of industrial action, often does not mean one's individual interest alone, but includes the interest of others; to be sure, of others in whose welfare the person who acts takes an interest. Consider the family, the acquisition of property for transmission to descendants. Here the egoistic action passes over into altruistic action. But it may nevertheless be said that, although there is a widening of the egoistic motive beyond the individual, it still remains egoistic." There may seem contradiction here ; but there is no real contradiction. An individual action may form one link in a series of actions which, considered in their totality, IV.] LIMITS OF THE METHOD OF ABSTRACTION. 123 At the same time, in so far as it can be shewn that in certain spheres of economic action men are normally moved by altruistic motives, this can be more or less recognised in the abstractions upon which the economist's more general reasonings are based. It is, however, doubtful whether it would be possible, in any case, to base upon the hypothesis of general altruism an exact science corresponding to English political economy. For the desire for the general wel- fare is not a motive capable of being measured in the same way as the desire for wealth can be measured 1 . But it is time now to turn to the other side of the picture. Whilst the process of abstraction from the full empirical actuality is an instrument of the greatest possible utility in economic investigations, the eco- nomist cannot by this means alone explain all indus- trial facts. Neither the conception of the economic man nor any other abstraction can suffice as an adequate basis upon which to construct the whole science of economics. In completing our investigations we have generally speaking to deal with something far more complex. As Roscher puts it, we must in our finished theory "turn to the infinite variety of real life". 2 are altruistic; and yet considered by itself, and in relation to its immediate consequences, it may be indistinguishable from an action that is purely egoistic. 1 Compare Marshall, Present Petition of Economics, 8 11. 2 "The abstraction according to which all men are by nature the 124 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. 3. Examples of economic problems requiring for their complete solution a realistic treatment. It is in attempting the final solution of problems, relating to the distribution of wealth, that it is most obviously insufficient to regard mankind as simply and entirely concerned in the pursuit of gain, irrespective of social surroundings, and the operation of other than pecuniary- motives 1 . The love of a certain country or a certain same, different only in consequence of a difference of education, position in life, &c., all equally well equipped, skilful, and free in the matter of economic production and consumption, is one which, as Bicardo and von Thiinen have shewn, must pass as an indispens- able stage in the preparatory labours of political economists. It would be especially well, when an economic fact is produced by the co-operation of many different factors, for the investigator to mentally isolate the factor of which, for the tune being, he wishes to examine the peculiar nature. All other factors should, for a time, be con- sidered as not operating, and as unchangeable, and then the question asked, What would be the effect of a change in the factor to be examined, whether the change be occasioned by enlarging or dimi- nishing it? But it never should be lost sight of, that such a one is only an abstraction after all, from which, not only in the transition to practice, but even in finished theory, we must turn to the infinite variety of real life " (Principles of Political Economy, 22). 1 If by an economic motive is meant any motive that influences men's economic activities, it is clear that the desire of wealth is not the only economic motive. Wagner gives a five-fold classification of economic motives in the above sense: four egoistic, and one non- egoistic. (1) The wish for gain, and the fear of want. (2) The hope of reward of a non-economic kind (e.g., approval), and the fear of disadvantages of a non-economic kind (e.g., punishment). The operation of such motives as these is important in connexion with slave labour. (3) The sense of honour, and the fear of disgrace. The guild system under ideal conditions is an example of the operation of these motives. Another example is to be found in the pride which IV.] ECONOMIC MOTIVES. 125 locality 1 , inertia, habit, the desire for personal esteem, the love of independence or power, a preference for every good workman takes in the quality of his work. (4) The impulse to activity and to the exercise of power, and the fear of the results of inactivity. "Sometimes, in the restless activity of men who carry on industry on a great scale, the wish to accumulate property is the immediate aim but not so much for the sake of material advantage, as for the sake of the power which a fortune confers." The motive of rivalry, which may also under certain conditions exert an appreciable influence, is closely akin to the love of power. (5) The non-egoistic motive is the sense of duty and the fear of conscience. " Because of it, competition is not pressed to the utmost, prices do not reach the highest or lowest limits which the pursuit of individual advantage would fix, and would fix without encountering an effective check in the mere sense of honour and propriety. Under this head, we are to class not only all charitable action, but the cases where an industrial or social superior purposely refrains from making his own interest the exclusive ground of his economic conduct " (Quarterly Journal of Economics, vol. 1, pp. 118121). 1 "Political economists," said Mr Chamberlain in a speech at Inverness in September, 1885, "find it difficult, perhaps, to under- stand how such unpractical considerations as the traditions and the history of a race, the love of home and of country, religious en- thusiasm, and political sentiments, should absolutely prevent a High- lander from accepting with complacency a proposal to exile himself from the land, which his forefathers have possessed and cultivated, for which they have shed their blood, and in which they lie buried. But human nature is a greater force even than the laws of political economy, and the Almighty Himself has implanted in the human breast that passionate love of country, which rivets with irresistible attraction the Esquimaux to his eternal snows, the Arab to his sandy desert, and the Highlander to his rugged mountains." None of our leading economists are really open to this reproach. In the appli- cations of political economy, if not as an integral portion of the science itself, they recognise the necessity of investigating inductively the operation of all the forces that affect the movement of labour from country to country, and from place to place within the same 126 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP country life, class prejudice, obstinacy and feelings of personal ill-will *, public spirit, sympathy, public opinion, the feeling of duty with regard to accepted obligations, current notions as to what is just and fair, are amongst the forces exerting an influence upon the distribution of wealth, which the economist may find it necessary to recognise, though the precise weight to be attached to them varies enormously under different conditions. The special influence that may be exerted by ethical motives has been referred to in rather more detail in an earlier chapter. It is to be remarked that even in the abstract theory the economist assumes that the rules of conventional morality in matters of business are generally accepted and obeyed. The standard of such conventional morality is, however, subject to vari- ations, as also the extent to which departures from it are common ; and variations of this kind should not be overlooked by the economist in his more concrete in- vestigations. As a special case, attention may be called to the extent to which the conventional morality country; and amongst such forces they do not overlook those here referred to by Mr Chamberlain. The above quotation may serve to illustrate how narrow and one-sided is the political economist in the general estimation ; and how common it is to substitute for the views of economists themselves, the opinions of superficial readers who have separated fragments of economic doctrine from their proper context. Because of misunderstandings of this sort it becomes the more necessary to insist that the abstract theory does not exhaust the whole of the science. 1 These forces may exert an important influence in trade disputes. IV.] CIRCUMSTANCES AFFECTING WAGES. 127 of the market pushes the rule of caveat emptor. Even in the same society at the same time this varies in different classes of transactions. Amongst important circumstances affecting wages are qualities of co-operativeness and habits of combi- nation amongst the labouring classes, as well as the social forces and legal regulations which determine how far these qualities and habits shall have free play. Again, in discussing the labour question, it is obvious that differences of enterprise and knowledge as affect- ing a man's willingness or ability to change his con- dition must not be overlooked. Regard must also be had to legislation of every kind in so far as it directly or indirectly affects the mobility of labour. To take a special case the economist has to discuss the cir- cumstances determining the wages of women, and to enquire whether these are in any way different from those determining the wages of men; but only by investigating the operation of various social influences can he obtain anything like an adequate solution 1 . Turning to the more general problem of the increase of the supply of labour, we find it to be one that depends materially on the social, intellectual, and moral causes, that determine men's standard of comfort, as 1 Compare the treatment of this problem in Walker's Wages Question, pp. 372 384 ; and in Mr and Mrs Marshall's Economics of Industry, pp. 175 177. 128 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. well as on such conditions as the price of food and other necessaries. This point may be illustrated by reference to Mill's treatment of the argument that under socialism, " the prudential restraint on the multi- plication of mankind would be at an end, and popula- tion would start forward at a rate which would reduce the community, through successive stages of increasing discomfort, to actual starvation ". 1 He says that "there would certainly be much ground for this apprehension", if socialism "provided no motives to restraint, equi- valent to those which it would take away"; but he then goes on to speak of the force of public opinion as possibly supplying a new motive, to which, in ac- cordance with the general tenour of his remarks, might be added that of public spirit and care for the general well-being. It is easy to exaggerate the probable efficacy of these forces ; but the illustration will at least serve to shew how, in arguing from one state of society to another, there is need to investigate and allow for the effects which different surroundings may have on human action. When we pass to the production and accumulation of wealth, we find that the motives determining them vary in different instances 8 . Work, for instance, that is 1 Principles of Political Economy, ii. 1 3. 2 "In vast permanent societies, in long ages of history, populations such as the Egyptian and the Indian, under a strict caste system, have shewn an astonishing degree of industry, directly stimulated by IV.] INFLUENCE OF VARIOUS MOTIVES. 129 inspired by mere love of routine, and saving that has become a mere habit, are not so uncommon as CL priori we might be inclined to imagine. One consequence of the latter fact is that we cannot discover the laws determining the accumulation of capital, or the manner in which a fall in the rate of interest will affect saving, by considering exclusively the effect of the desire of wealth. Even in the case of a purely monetary question, such as the circumstances determining the amount of the depreciation of an inconvertible currency, an im- portant consideration may be the extent to which a people's distrust is aroused, and this in its turn may depend partly on their political sympathies, or on their knowledge and intelligence, or on the extent to which their power of moral restraint prevents them from giving way to unreasoning panic. This last point is still more clearly important in connexion with the phenomena that constitute a financial crisis. The habit, social feeling, and religious duty, and, in a very slight degree, by personal desire of gain. In religious societies under very different kinds of faith, very active industry, on a scale quite decisive as an experiment, has been stimulated by purely religious motives. Some of the most splendid results of industry ever recorded the clearing of wildernesses ; vast public works, such as bridges, monuments, and temples ; the training of whole races of savages into habits of toil have been accomplished by purely religious bodies on purely religious motives, by monks, missionaries, and priests." Frederic Harrison on the Limite of Political Economy in the Fortnightly Review, June 15, 1865. K. 9 130 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. theory, for example, of the recurrence of such crises at regular intervals, so far as it does not involve the operation of physical causes (as in Jevons's sun-spot theory), may require to be modified according to the stage of a nation's intellectual and moral development. Further illustrations might be added, bringing out in particular the influence exerted on industrial pheno- mena by legal conditions 1 , and by political and social institutions; but enough has been said to shew that, while the pursuit of wealth may be the main force of which account has to be taken, still if economic science is to succeed in affording an adequate explana- tion and elucidation of the facts of economic life it is necessary also to have regard to social surroundings, and the operation of diverse other motives. 4. Distinction between political economy and other social enquiries. Since a realistic treatment of eco- nomic problems is usually essential to their complete solution, it is necessary that economists should keep in view all the various aspects of social life ; and it is clearly mischievous to aim at an entire isolation of economics from other social sciences. But political economy does not, therefore, lose its individuality. For the recognition that the various forms of social activity 1 It is pointed out by Schonberg that the legal factor is one that must always operate, however much the maxim of laisser faire is allowed to exert an influence. There must be laws relating to property, Ac. IV.] SCIENTIFIC DIVISION OF LABOUR. 131 are in many ways interdependent does not destroy the significance of the differences between them ; and to do away with the boundaries, that now separate the differ- ent social sciences, would be to sacrifice all the gain resulting from scientific division of labour. Political economy has necessarily to take account of facts that belong primarily to other subdivisions of social enquiry ; but its study of them is confined to one particular point of view; it is concerned with them only in so far as they have a direct economic bearing. Hence the eco- nomist rightly neglects, or passes over very lightly, many phenomena and relations between phenomena, that are of central importance from the standpoint of other social philosophers, e.g., the jurist, the moralist, or the student of political science. " The tendency of scientific progress;' as Cherbuliez has well remarked, " has always been to separate the sciences, not to con- fuse them ; to divide and subdivide the domain of their investigations, not to make of them a single field, cultivated by the same hands, following the same methods 'V It may be observed in passing that physical, as well as social, conditions have to be taken into account in political economy. The possibility of a rise in wages being maintained may depend on the effect of better food upon the recipient's efficiency as a workman. The 1 Precis de la Scie)ice Economique, vol. i, p. 9. 92 132 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. law of diminishing return has a direct physical basis. It is impossible properly to investigate the effects of free trade apart from the assumption of physical differ- ences between different countries. But no one therefore regards political economy as having no existence indepen- dently of the physical sciences. We may argue further from the analogy of the physical sciences themselves. For it is also true that phenomena in the physical world are in various ways interdependent. Geological phenomena, for instance, are dependent upon physical and chemical phenomena. But no one therefore denies the right of geology to be recognised as a distinct science. It has been truly said that in a sense everything includes everything else, and no doubt on the problem of the rent of land it might be possible to build up an ency- clopaedia of the sciences. Nevertheless, subdivision and specialization are necessary, if we are to advance in accurate knowledge. Granting, then, that political economy is not a wholly independent and isolated science, it is still to be regarded as a distinct division of speculative truth ^ and it may rightly take rank as a social science marked off by special characteristics from other social sciences. It is, in other words, a science in which a form of social activity of a distinctive character is singled out for distinctive treatment 1 . 1 It is to be observed that Dr Schonberg who may be taken as IV.] DISTINCTIVE CHARACTER OF ECONOMICS. 133 Even if we recognise in the most unreserved way that political economy is only one division or depart- ment of social science, there is a special reason why, in the present state of sociological knowledge, we should not seek to give economic doctrine an entirely new representing the prevailing view of German economists whilst fully recognising, and indeed insisting upon, the interdependence of economic and other social phenomena, still speaks of political economy as a special independent science ("erne eigene selbsttindige Wissenschaft"). He gives the following as the great fundamental life spheres of every people, constituting, in their totality, national life : justice, art, science and education, family life, social life and morality, religious life, political life, and economic life. The last of these spheres, he remarks, stands in the closest causal connexion with the others ; it influences them, and is influenced by them. In studying it, there- fore, we must recognise these causal relations. But at the same time economic life is a distinct sphere of national life ; in it men pursue peculiar ends ; in it peculiar forces are developed ; it depends on special institutions ; into it enter peculiar problems ; and it is consequently the subject of an independent science. See Schonberg's Handbuch, vol. i, pp. 3, 16, 17. Compare, also, Dr von Scheel in the same volume of the Handbuch, p. 69. "It has been proposed," he says, "that political economy should be enlarged into the science of society, while it ought only to be said of it that it should be enlarged into one of the social sciences in opposition to the too narrow conceptions which were held formerly." Knies expresses himself similarly " One may take into consideration a science which, under the name of sociology or any other name, would have to set forth the underlying universal theory for all state and social sciences. But with all that, the just claim and unavoidable requirement of a special care for political and social economy, in the sense in which it is known to us, is still not in the least affected. If this branch of science, which has grown up with the development of the scientific division of labour for the special investigation of a very large and important sphere of human social life, were not there already, then it would have to be immediately invented" (Die Politische Oekonomie vom geschichtlichen Standpuncte, 1883, p. 9). 134 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. form with a view to its absorption into general so- ciology. Comte charged political economy with being radically sterile as regards results. But what results has sociology, conceived as a master-science dealing with man's social life as a whole, yet to shew ? It has been well said by Lord Sherbrooke that sociology, as distinct from the special social sciences, has yet its spurs to win. The time may come when in the domain of social science wide generalizations are established, from which each special science that deals with man in society may learn. There may thus be constituted a body of general sociological doctrine, to which political economy is subordinated. But economics cannot wait for sociology in this sense to be built up. " It is vain," says Professor Marshall, " to speak of the higher autho- rity of a unified social science. No doubt if that ex- isted, economics would gladly find shelter under its wing. But it does not exist; it shews no signs of coming into existence. There is no use in waiting idly for it; we must do what we can with our present resources 'V 1 Present Position of Economics, p. 35. Dr Sidgwick in his Scope and MetJiod of Economic Science expresses himself to a similar effect. He discusses in some detail the claims of sociology to be regarded as a positive or established science, taking Comte's own tests of (1) Con- sensus or Continuity, and (2) Prevision (p. 46) ; and on both grounds he decides the question in the negative. "There is no reason," he says in conclusion, "to despair of the progress of general sociology; but I do not think that its development can be really promoted IV.] THEORY OF ECONOMIC PROGRESS. 135 It may be added, partly by way of qualification, that there is one particular department of economic enquiry, in which the connexion with the general philosophy of society is closer than in other departments. It seems clear that in seeking a theory of economic progress, the method of specialization can be carried less far than when we are discussing laws that presuppose given conditions of industry. It is indeed possible to trace historically the actual course of progress from the specifically economic standpoint ; and generalizations relating to points of detail in economic development are attainable apart from any general theory of social progress. But since it is admitted that the economic conditions of any given stage in the progress of society are determined not merely by the economic conditions, but by the general social characteristics, of the pre- ceding stage, no theory of the tendencies of economic evolution as a whole seems likely to be reached inde- pendently of some theory of the general tendencies of social development. by shutting our eyes to its present very rudimentary condition. When the general science of society has solved the problems which it has as yet only managed to define more or less clearly when for positive knowledge it can offer us something better than a mixture of vague and variously applied physiological analogies, imperfectly verified historical generalizations, and unwarranted political pre- dictions when it has succeeded in establishing on the basis of a really scientific induction its forecasts of social evolution its existence will be irresistibly felt throughout the range of the more special enquiries into different departments of social fact to which we have hitherto restricted ourselves" (pp. o- r >, SCO. 136 POLITICAL ECONOMY AND SOCIOLOGY. [CHAP. IV. It is from the department of economic progress that those who attack the old political economy draw their most forcible illustrations ; and it is further to be ob- served that as general sociology is frequently con- ceived, its one fundamental problem is "to find the laws according to which any state of society produces the state which succeeds it and takes its place". 1 When sociology, as thus interpreted, can lay down pro- positions that are definitely formulated and clearly established, then the theory of economic progress may with advantage be specially subordinated to it. 1 J. S. Mill, Logic, vol. 2, p. 510. NOTES TO CHAPTER IV. A. ON THE DISTINCTION BETWEEN ABSTRACT AND CONCRETE POLITICAL ECONOMY. THE discussion in the preceding chapter naturally leads to the recognition of two stages in economic doctrine, which may be called the abstract and the concrete stage respec- tively '. In the abstract or pure theory of political economy we concern ourselves entirely with certain broad general principles irrespective of particular economic conditions; or, as Jevons puts it, with " those general laws which are so simple in nature, and so deeply grounded in the con- stitution of man and the outer world, that they remain the same throughout all those ages which are within our consideration." The method of the abstract theory is almost wholly deductive and hypothetical ; for though based ultimately on observation, it works from artificially simplified data. The results obtained are in one sense of universal application, since they are ready to be modified to suit particular circumstances as the occasion may arise ; but they are in themselves always incomplete, since we cannot by their aid alone adequately understand the eco- nomic phenomena of actual life. 1 Compare Jevons on the Future of Political Economy in the Fortnightly Review for November, 1876, p. 625. 138 ABSTRACT AND CONCRETE ECONOMICS. [CHAP. IV. Concrete economics comes in to supplement the pure theory, and is not content with merely hypothetical results. Its laws are obtained either by direct generalization from experience, or by the aid of the deductive method. In the latter case, however, the premisses are adapted to suit special circumstances, and both premisses and conclusions are constantly tested by direct appeals to experience. In formulating concrete economic doctrines we seek to lay down laws that are operative over a given period or in a given state of society. Such laws are for the most part relative, not universal, in their application 1 . We have the pure theory par excellence, when we concern ourselves with economic men, supposed to deal exclusively with one another in a state of economic free- dom. On this basis the laws of competitive values, wages, rent, interest, &c. are worked out in their most general and abstract forms. Jevons's Theory of Political Economy may be given as a typical example of an abstract treat- ment of the subject, as contrasted with such a work as Walker's Wages Question, where the treatment is in the main concrete. The line between abstract and concrete political eco- nomy hardly admits, however, of being quite rigidly deter- mined ; for the extent to which we have in view special circumstances and conditions of society may sometimes be a matter of degree. Even the same doctrine (e.g., the doctrine of cost of production as the regulator of value) 1 What is here spoken of as concrete economics has sometimes been called applied economics. The latter designation, however, as already pointed out, is ambiguous ; and on the whole it seems best to keep it for what is also called the art of political economy. NOTE A.] NATURE OF THE DISTINCTION. 139 may be regarded as having an abstract or a concrete character according to the mode in which it is treated ; and in some cases the concrete doctrines are just the abstract doctrines plus something more, namely, an enquiry into the special conditions under which alone the latter can be applied to existing facts, and an investigation of the modifications of doctrine needed in consequence thereof. Instead, therefore, of attempting to draw any hard and fast line between the two sets of doctrines, it may be better to say simply that political economy is abstract in so far as it neglects special conditions of time, place, and circumstance ; while it becomes more and more con- crete as it takes such conditions into account. This rela- tivity does not detract from the importance of the distinc- tion, which is of special utility in relation to problems of method. It should be added that the manner in which the dis- tinction has been above expressed, and even the distinction itself, would not be universally accepted. For, as we have seen, some economists practically deny the possibility, or at any rate the utility, of any abstract or hypothetical treatment of economics at all, while others seem to regard the pure theory as exhausting economic science. An en- deavour has been made in the preceding chapter to con- trovert both the above views. The pure theory may rightly be regarded as of great and even indispensable value as the general basis of economic reasoning, while it is at the same time held to be only part of a larger whole. It may, indeed, sometimes be possible to pass imme- diately from the pure theory to the interpretation of indi- vidual phenomena of the actual economic world ; but more 140 ABSTRACT AND CONCRETE ECONOMICS. [CHAP. IV. usually there is required the intervention of a body of doctrine, which, while possessing a certain generality of form, is still not purely abstract in character, or capable of being worked out merely by the aid of those simple and general data, which alone are recognised by the abstract theory. It is this body of doctrine that constitutes con- crete political economy, as broadly distinguished from the pure theory of the science. The reasoning of the abstract theory has a logical precision which concrete economics for the most part lacks. Being hypothetical it can be made demonstrative and ne- cessary, so that among properly trained persons there should be no room for differences of opinion in regard to its conclusions. Concrete economic doctrines are in com- parison contingent and indeterminate. But it does not follow that they therefore foraa no part of the science, or that they are essentially unscientific and untheoretical. It ought frankly to be recognised that not all science is of the demonstrative type ; and it would be a great mistake to narrow our conception of political economy to the pure theory alone, simply in order to attain perfection of logical form. B. ON THE DISTINCTION BETWEEN THE STATICS AND THE DYNAMICS OF POLITICAL ECONOMY. IN speaking in the preceding chapter of the laws of economic progress, another subdivision of political economy has been indicated, about which a few more words may here be added. Economic doctrines may treat (a) of the NOTE B.] ECONOMIC STATICS AND DYNAMICS. 141 phenomena of wealth as they present themselves under given economic conditions ; or (6) of the manner in which these conditions themselves vary over long periods of time, together with the economic changes that ensue thereupon. The former of these branches of doctrine constitutes the main body of economic science. To it belongs, for example, the investigation of the laws which in any given society regulate the division of what is produced into the shares of rent, interest, and earnings. The latter enquiry may be distinguished as the study of economic progress; and the resulting doctrines constitute in their totality a general theory of economic development or evolution. The laws of the movement from status to contract, and of the transition from collective to individual property, may be given as examples of special doctrines belonging to this branch of enquiry. Using terms which Comte introduced into the nomen- clature of social science, Mill and some other economists speak of these two branches as the statics and dynamics of political economy respectively. These terms are not, as a matter of fact, specially appropriate; they may even be misleading. In so-called economic statics we are fre- quently engaged in examining the effects of particular changes, e.g., changes in demand, in cost of production, in the amount of currency in circulation, and the like. The economic world, even in a given state of society, is in perpetual movement; prices, wages, profits, systems of cur- rency, tariffs, )conomie, p. 3. 202 308 HISTORICAL CONCEPTION OF ECONOMICS. [CHAP. IX- It has been already shewn that economic history itself needs to be interpreted by theory. Moderate advocates of the historical method, such as Arnold Toynbee, have clearly recognised that "without the help of deduction, this method can serve only to accumulate a mass of unconnected and unserviceable facts". 1 It follows that to postpone con- siderations of theory until an indefinite number of facts have already been collected is, even from the historical point of view, a mistake. The case against the supremacy of the historical method in economics is all the stronger, if we regard the method as literally confining itself to the facts of past times. For the purely historical method is obviously much narrower than the inductive ; and it will hardly be denied that the facts which are essential to the economist are to a very great extent obtained from contemporary observations or from records so recent that they have hardly yet passed into what we understand by economic history. Moreover, inferences based on historical research, as distinguished from observation of the present order of events, labour under special disadvantages. Often there is more or less uncertainty concerning the facts themselves. "History has suffered to drop from her pages, perhaps has never recorded, much of the information which would now be most precious to us" 2 ; and an incomplete record may be even worse than no information at all, so far as affording a basis for theoretical conclusions is concerned. We see the past as it were through a mist; and we cannot " cross- 1 F. C. Montague, Arnold Toynbee (Johns Hopkins University Studies in Historical and Political Science, Seventh Series, i. p. 33). 2 Richard Jones, Literary Remains, p. 570. NOTE B.] HISTORICAL CONCEPTION OF ECONOMICS. 309 examine its facts" as we often can the facts of the present time '. It is still more important to observe that by very reason of the evolution of industrial systems, and the shifting character of economic conditions, upon which the historical school of economists so much insist, the study of the past is rendered the less serviceable for the solution of present-day problems. Upon many of these problems extremely little light is thrown by economic history that is more than a hundred years old. How indeed can gene- ralizations based upon one set of circumstances be safely applied to quite another set of circumstances? Not only may the problems calling for solution be novel in their character; there may even arise new industrial classes. With what classes in the fourteenth century, for example, are we to compare the modern factory operative and the modern capitalist employer? If, therefore, for no other reason than that institutions and habits arid conditions change, another method of investigation than the historical must for very much of our economic work be essential. Political economy can never become a specifically historical science. 1 Compare Marshall, Present Position of Economics, 17. CHAPTER X. ON POLITICAL ECONOMY AND STATISTICS. 1. The claims of statistics to be regarded as a distinct science. A leading German statistician has gone so far as to say that there are almost as many different views of the nature and province of statistics, as there are writers who have occupied themselves with the subject. Many different definitions of the term statistics have also been proposed ; a list of 180, more or less differing from one another, was drawn up by Quetelet as long ago as 1869. Even the etymology of the term, or at all events the mode of its derivation from the Latin status, has been a matter of dispute. 1 1 The correct account of the derivation of statistics seems to be that it came through the Italian stato, which was in the fifteenth century first used in the signification of territory, or " state " in the political sense. See Riimelin, article Statistik in Schonberg's Hand- buch, 2 ; and Wappaus, Einleitung in das Studium der Statistik, p. 7. " Achenwall," says Professor Wappaus, " never in his writings explained the origin of the name statistics, but it comes out in his lecture-notes. His explanation is as follows. The Italians were the first to form a science of the State, and called it Ragione di Stato. From this in Latin writings, and lectures given in Latin arose CHAP. X.] ACHENWALL. 311 Only two or three of those views, however, that are most broadly distinguished from one another need be noticed here. Gottfried Achenwall, professor of law and politics at Gottingen about the middle of the last century, though not the originator of the Latin adjective statisticus, appears to have been the first to use the German substantive Statistik, and he is usually regarded as the founder of statistics considered as a special branch of knowledge 1 . He meant by statistics a collection of noteworthy facts concerning States the historical and descriptive material upon which political science, as we now understand it, is largely dependent. It is to be added that as treated by the school of statisticians to which Achenwall belonged the so-called "descrip- tive" school statistics was not essentially numerical or quantitative. Verbal description took the first place, and figures were used merely as accessory thereto. Since the time of Achenwall the term, as ordinarily used, has changed its meaning, and the distinguishing Ratio status, or Disciplina de ratione status, or Disciplina de statu. There being in classical Latin no simple expression for ' state ' in our sense, the word status was used with this meaning. The Italians at the same time gave to any one learned in the above science or art the name statista. German scholars adopted the word statiata into Latin, and formed the adjective statisticus." 1 This view is not strictly correct, as Achenwall had predecessors at the German universities (Conring, Schmeitzel, and others) whose subject-matter and method resembled his own. His treatment was, however, more thorough than theirs, and attracted more attention. 312 POLITICAL ECONOMY AND STATISTICS. [CHAP. mark of statistics is considered to be the employment of numerical data 1 . Moreover, regarded as a science, statistics is not content to be merely descriptive, but claims to be theoretical and speculative. The principal question at issue is whether statistics can legitimately be regarded as constituting a distinct science at all. This question is complicated by the fact that, in ad- dition to Achenvvall's view, two very different concep- tions of statistics as a science have been formed. Statistical science is, according to Dr Mouat, "a special science of methods from which, and by which 1 Statistics in this sense may be traced back to the " Political Arithmetick " of Sir William Petty, and other English writers of the seventeenth and eighteenth centuries ; and the influence of Quetelet and Knies was important about the middle of the present century in favour of the arithmetical, as opposed to the descriptive, school. The latter still has some, though not many, adherents. The late Pro- fessor Wappaus of Gottingen, for example, defends Achenwall's con- ception in his Einleitung in das Studium der Statistik, published after his death in 1881. He recognises that quantitative data have assumed a greater relative importance than formerly, but he ascribes this simply to the fact that they are now more easily procurable. "The increased facilities for obtaining numerical facts have," he says, "necessarily affected the method of statistics. We now have two equally important resources : description, and numerical expres- sion. The two methods supplement one another, and it is a mistake to try to make two separate branches of science out of them. Mis- understandings have arisen about statistics, because demands have been made upon it, which cannot be fulfilled demands only to be satisfied by a purely philosophical branch of study, which statistics is not. Statistics is a positive science, an aggregate of knowledge brought together for a practical end, namely, the knowledge of the concrete State. This is a very simple definition, and justified by the genesis and history of the science " (pp. 32 4). X.] STATISTICS REGARDED AS A SCIENCE. 313 alone, the natural laws can be deduced which govern most of the conditions of man, and many of those of the animal and vegetable kingdoms." " There is not," Dr Mouat adds, " a branch of human knowledge to which the science of statistics is not closely allied, and for the correct understanding of which the scien- tific marshalling of figures, and observation of aggre- gate facts, is not more or less necessary. That the laws deduced from them fall into the ranks of the branches of knowledge to which they belong when they are fairly established does not, in my humble judg- ment, invalidate the scientific claim of the agency to which they owe their existence". 1 The concluding portion of the above statement would probably meet with universal acceptance. A method or an agency may, however, be scientific with- out thereby becoming itself a science. Statistics, or statistical method, as understood by Dr Mouat, is a very important means whereby human knowledge is extended ; but as such it is to be regarded as a scien- tific instrument, rather than as an independent body of doctrine constituting a distinct science. It is, indeed, necessary to recognise a theory of statistics, dealing with what may be called the tech- nique of the statistical method, that is to say, the con- 1 History of the Statistical Society of London (Jubilee volume of the Statistical Journal), p. 47. 314 POLITICAL ECONOMY AND STATISTICS. [CHAP. ditions that statistical data must fulfil, the modes in which they are to be ascertained and collected, the manner of their arrangement and employment for pur- poses of reasoning, the criteria determining the validity of arguments based upon them, and the logical charac- ter of the conclusions established by their aid. But all this is really antecedent to the actual use of statistics for any particular purpose. The whole discussion con- stitutes, not a separate science, but a special branch or department of inductive logic or methodology that is, of the science or art which treats of scientific method in general. It is, however, in quite a different sense from the above that the existence of an independent science of statistics is affirmed by the majority of continental statisticians, and also by some English writers 1 . Sta- tistical science is regarded, not as an abstract science of methods dealing with phenomena of very various kinds under a distinctive aspect, but as a concrete science with a distinctive subject-matter. A distinction is clearly drawn between statistics as a method and sta- 1 Compare Mr Wynnard Hooper's paper on the Method of Statisti- cal Analysis in the Statistical Journal for March, 1881 ; and the same writer's article on Statistics in the ninth edition of the Encyclopedia Britannica. Mr Hooper himself takes the view advocated in the text, namely, that there is no independent science of statistics. The two views are sometimes spoken of as the English and the Continental respectively. There is not, however, universal agreement amongst Continental, any more than amongst English, writers on the subject. X.] STATISTICS REGARDED AS A SCIENCE. 315 tistics as a science. It is recognised that the method has a very wide application ; but the science is de- scribed as studying exclusively man's social life. As thus interpreted, statistical science becomes prac- tically equivalent to sociology, with the implication that the sole means whereby sociological, including economic, knowledge can be attained is the systematic collection and inductive interpretation of social phenomena. There is the further implication that the data are mainly, if not exclusively, numerical. Dr Mayr, taking this view, defines the science of statistics as "the systematic statement and explanation of actual events, and of the laws of man's social life that may be deduced from these, on the basis of the quantitative observation of aggregates". 1 If it is asked why the quantitative observation of social aggregates should constitute a distinct science, while no similar claim is made in regard to the obser- vation of purely physical aggregates, the reply given is that in the determination of the laws of social life statisti- cal enquiry is " the only possible mode of investigation," and not as in the case of the physical sciences a merely secondary or supplementary method. Social 1 This definition is given in Dr Mayr's Die Gesctzmiissigkeit im Gesellncliaftsleben, an abridged translation of which will be found in the Statistical Journal for September, 1883. For definitions given by English statisticians, who take a similar view, see Statistical Journal, December, 1865, p. 492 ; and Jubilee volume, p. 8. 316 POLITICAL ECONOMY AND STATISTICS. [CHAP. science and political economy are spoken of as branches or departments of the science of statistics, a science which studies social and economic phenomena in the only satisfactory way, namely, by the accumulation of facts and generalization from them. It will be observed that the doctrine here set forth is even narrower than that which regards induction as the sole valid method of economic enquiry. For we are now limited to quan- titative induction ; qualitative induction, whether his- torical or comparative, is out of place as well as the deductive method. Our grounds for rejecting this view have been given in a previous chapter, and to pursue the discussion here would merely carry us back to a class of considera- tions that have been already sufficiently insisted upon. It may, however, be added that for the general science of society we have at any rate another name sociology or social science which does not beg the question as to method, and which is free from the ambiguity that at best must attach to the term statistics. For it has to be allowed that by this term is also meant a method of analysis having an indefinitely wide range of applica- tion outside the science of man in society. Thus we speak of moral and intellectual statistics, of vital and medical statistics, of astronomical and meteorological statistics, of physical and physiological statistics, as well as of economic and political statistics. It can hardly be X.] STATISTICS REGARDED AS A SCIENCE. 317 said that there is any concrete department of enquiry, in which statistics as a method may not find a place ; and it is upon this character of universality that the claim of statistical research to public recognition and encouragement is frequently based. If a less extreme view than that above described is taken, and statistics is considered to be a distinct science, but nevertheless not to include the whole of social science or of economics, then it becomes a part which is only differentiated from the remainder by the employment of a particular method. Professor R. M. Smith explicitly recognises statistical science as "a branch of social science employing a specific method, and devoting itself to those problems of life in society which can best be solved by that method". 1 It seems, however, both unusual and undesirable to differentiate sciences by their method as distinguished from their subject-matter. We might equally well identify other scientific methods with those particular sciences in the development of which they happen to be of special importance. At any rate the question now becomes little more than a verbal one. There need be no fun- damental disagreement between those who take the view just indicated, and those who prefer to treat 1 Statistics and Economics (Publications of the American Eco- nomic Association, vol. 3), p. 118. Prof. Smith divides social statis- tics into population statistics, economic statistics, and statistics of vice and crime. 318 POLITICAL ECONOMY AND STATISTICS. [CHAP. statistics simply as a particular method or instrument of scientific enquiry, which is not peculiar to the study of social facts, although it may be of much greater relative importance in connexion with that study than it is in other departments of knowledge. There seems, however, to be an idea that if the claim of statistics to recognition as a science be not admitted, then the statistician becomes a mere drudge, who is denied the luxury of opinions, and whose sole function is to collect materials for others to reason about and base theories upon. Professor Smith's main ground for calling statistics a science is that this is the only way of rescuing the study from " the barrenness which results from viewing its object as simply the collection of masses of figures, with which the statistician has nothing further to do." He regards the question whether statistics is or is not a science as not a merely verbal one, because the answer to it determines "the position of the statistician and the authority with which he speaks." By another writer it is said that if statistics is not a science, then the statistician is merely as one who binds up sheaves of wheat for others to thresh out 1 . But this by no means follows. There is, un- doubtedly, special risk of error when statistics are used by others than those who have prepared them. For, 1 Dr Guy in the Statistical Journal, 1865, p. 483. X.] STATISTICS REGARDED AS A SCIENCE. 319 as Mr Hooper remarks, "there are usually 'pitfalls' even in the simplest statistical statement, the position and nature of which are known only to the persons who have actually handled what may be called the 'raw material' of the statistics in question." Hence the statistician rightly, and even necessarily, per- forms the function of interpreting results. But in so doing he becomes the economic statistician, the politi- cal statistician, the medical statistician, the physical statistician, as the case may be. He applies his statis- tics, that is to say, within the domain of some parti- cular science ; and it may be added that unless he has an adequate knowledge of that science, not only will he probably go astray in his interpretation, but the very facts themselves are not likely to be suitably selected or arranged. But all this, it is clear, applies to medical and physical and other statistics, just as much as to social statistics; and no one would maintain that we have a distinct science, wherever we have a branch of knowledge in which statistics may be usefully em- ployed. Hence if we go out of our way to recognise a science of statistics which is concerned with social phenomena alone, we seem thereby to cast an un- deserved slur upon statistics falling into other de- partments of enquiry. On the other hand, we may refuse to recognise any distinct statistical science, with- 320 POLITICAL ECONOMY AND STATISTICS. [CHAP. out lowering the standard of what is required from statisticians, or in any way underrating the importance of the functions which they perform. 2. Statistics regarded as a method. In seeking to define statistics regarded as a method, it is con- venient to adopt the somewhat clumsy phrase already quoted from Dr Mayr, and say that it is a scientific me- thod based on the quantitative observation of aggregates. It is, in the first place, a method based on observation. It goes direct to facts, which it collects and systemati- cally arranges. It is, in the second place, based on an observation of quantities. It deals with phenomena that are measurable, and hence capable of numerical expression. It is, in the third place, concerned with aggregates, as distinguished from individuals or units. Series of isolated numerical facts are popularly called statistics, and they may be of use simply as information or as means of description or illustration, but they are of little or no value as a scientific instrument. In the scientific use of statistics the observations must be made in the mass, they must involve a certain degree of continuity, and the resulting figures must be care- fully and systematically grouped 1 . 1 The term statistics, when used as a singular noun, signifies the method above described, or if we recognise such a science the science of statistics (German, die Statistik; French, la statistique). In English, however, the term is generally used as a plural noun, X.] THE METHOD OF STATISTICS. 321 By the aid of statistics, thus understood, we are enabled to employ the method of concomitant varia- tions. In this way, quantitative inductions are estab- lished, and the laws of the variations of phenomena determined. There is a close connexion between the statistical method and the doctrine of chances. On the basis of the quantitative observation of aggregates, the influ- ence exerted in individual cases by accidental causes may be eliminated. For when instances are taken in the mass, it is often a fair assumption that accidental causes will operate so as to neutralise one another. The effects of permanent agencies may thus be calculated, even though in individual cases their influence is but slight. The manner in which, when a sufficient number of instances are taken, aggregate regularity is found to emerge out of individual irregularity has been one of the most striking results of statistical research. It was this fact that excited the enthusiasm of the Belgian mathematician Quetelet; and to his influence may be ascribed the impetus given to the study of statistics in the second quarter of the present century. In the use of statistics, considerable assistance may often be derived from the employment of diagrams. The graphic method is not only useful for the popular and it then signifies the numerical data which constitute the basis of the statistical method. K. 21 322 POLITICAL ECONOMY AND STATISTICS. [CHAP. exposition of statistics, enabling the mind more accu- rately to realise numerical comparisons ; but it has also a genuine scientific value. Thus by means of graphic representation we may employ the special method of quantitative induction called by Whewell the method of curves*. The relative positions of curves can be more easily compared than columns of figures ; and corres- 1 The graphic method takes different forms. Straight lines of different lengths, for example, are sometimes used, and rectangles, or triangles, or other geometrical figures, the relative size of which can easily be compared. Maps are also very serviceable as a popular method of illustration, where the statistics relate to different geo- graphical divisions. Statistical maps are sometimes called carto- grams, cartography being denned as " the employment of geography for the graphic illustration of statistics." The above forms have not, however, the scientific value that belongs to the use of curves, so drawn as to represent the manner in which the variations of some given quantity are related to the variations of some other quantity. Whewell defines the method of curves as follows : " The Method of Curves consists in drawing a curve, of which the observed quantities are the Ordinates, the quantity on which the change of these quantities depends being the Abscissa. The efficacy of this method depends upon the faculty which the eye possesses of readily detecting regularity and irregularity in forms. The method may be used to detect the Laws which the observed quantities follow ; and also, when the observations are inexact, it may be used to correct these observations, so as to obtain data more true than the observed facts themselves" (Novum Organon Renovatum, Aphorism xliv). On the graphic method of statistics in general see a paper by Professor Marshall in the Jubilee volume of the Statistical Journal ; also Dr D. B. Dewey's Elementary Notes on Graphic Statistics. Practical examples of the employment of this method will be found in Jevons's Investigations in Currency and Finance. The use of diagrams for statistical purposes should be clearly distinguished from their em- ployment in economic theory as discussed in a previous chapter. X.] GRAPHIC STATISTICS. 323 pondences may thus be observed, and empirical laws suggested, that would otherwise have escaped attention. This is especially true when there are more than two series of phenomena whose mutual relations are made the subject of investigation. The mere saving of space may be a matter of importance. Several distinct curves can be placed in a single chart, and it thus becomes possible to grasp at one and the same moment a greater multiplicity of detail 1 . The use of curves also renders us less liable to be distracted by movements of a partial or temporary cha- racter. There are indeed cases, as pointed out both by Whewell and Jevons, where diagrams to a certain ex- tent supersede the taking of averages ; for we may apprehend intuitively the general course of a curve, neglecting individual irregularities. A similar appre- hension would generally speaking not be possible, or at any rate not equally reliable, were we limited to mere columns of figures 2 . 1 Care must be taken to render it easy to follow the course of the different curves without confusing them one with another. They may be distinguished by form as well as by colour. Thus one curve may consist of an unbroken line, another of a broken line, another of a succession of dots, another of dots and dashes alternately, and so on. 2 It should be added that whilst the graphic method of statistics is for the reasons above stated scientifically important, certain precautions are necessary in order to prevent the comparison of curves from proving deceptive. This is especially the case when we take curves, representing the progress of phenomena, with a view to com- 21 2 324 POLITICAL ECONOMY AND STATISTICS. [CHAP. 3. The functions of statistics in economic en- quiries. Cairnes lays it down that "the relation of statistics to political economy is in no respect different from that in which they stand to other sciences which have reached the deductive stage". 1 But this summary dismissal of the question cannot be accepted. In the first place, notwithstanding the importance of deduction in economics, the science cannot be regarded as having reached the deductive stage in the same definitive manner as those sciences with which an analogy is here suggested for instance, physics and astronomy. Its premisses are less determinate than theirs, and greater prominence needs to be assigned to empirical confirmation and criticism. In the second place, al- though statistics ought not to be identified with soci- ology, the quantitative observation of aggregates is certainly of far greater relative importance in the social sciences than it is in the great majority of the physical sciences. In the latter for instance, in optics or in electricity so far as conclusions rest on an inductive basis, it is usually a basis of experiment. Individual cases can be treated as typical; and where a repetition paring their proportional rates of progress. Professor Marshall indi- cates the nature of the needful precautions in the paper referred to in a preceding note. 1 Logical Method of Political Economy, p. 86. "While the above is given as summing up Cairnes's view of the relation between statis- tics and political economy, some of his incidental remarks on the subject seem to indicate a less extreme position. X.] FUNCTIONS OF STATISTICS. 325 of trials is necessary, it is only in order to guard against error. The statistical method is important in special instances, e.g., in meteorology, but generally speaking it occupies a subordinate position. In the social sci- ences, on the other hand, there is little room for experi- ment, while statistics play a part for which no substitute can be found. Political economy, in particular being concerned pre-eminently with quantities has a special tendency to become on its inductive side statistical, just as on its deductive side it tends to become mathe- matical. To begin with, statistics are of importance in eco- nomic enquiries in respect of their merely descriptive functions. For example, statistics of production and of wages and prices are essential elements in any com- plete description of the social condition of a community; statistics of exports and imports in the description of its foreign trade and intercourse with other nations ; statistics of taxation and of national indebtedness in the description of its financial condition. This point is, however, so obvious that we need not dwell upon it, but may pass on at once to the more strictly scientific uses of statistics. The functions of statistics in economic theory are, first, to suggest empirical laws, which may or may not be capable of subsequent deductive explanation; and secondly, to supplement deductive reasoning by check- 326 POLITICAL ECONOMY AND STATISTICS. [CHAP. ing its results, and submitting them to the test of experience. Statistics play a still more important part in the applications of economic science to the elu- cidation and interpretation of particular concrete phe- nomena. We have seen that there are certain departments of economics in which we are compelled to content our- selves with empirical laws. In such cases we are usually concerned with aggregates, and can make little or nothing of individual phenomena taken by themselves. Our main reliance must, therefore, be placed upon sta- tistics, which may be either historical or contemporary. An illustration is once more afforded by the Malthusian doctrine of population. Mai thus himself made elaborate statistical enquiries concerning the proportion of yearly marriages to population ; the fruitfulness of marriages indifferent countries; the effects of epidemics on births, deaths, and marriages; and so forth. He hence inferred that under favourable circumstances population tends to double itself in twenty-five years; and on a similar basis he estimated the effects of the various checks to population operating in the less civilised parts of the world, and in past times, as well as in the different States of modern Europe. In this connexion it is necessary once more to point out the characteristic weakness of empirical generaliza- tions. They may be true of a given state of society, X.] GENERALIZATIONS BASED ON STATISTICS. 327 but with the changes incident to the progress of time may become false. It is necessary, therefore, to keep as it were a watch upon them, and from time to time bring up to date the statistics upon which they are based. It is no exaggeration to say, as Dr Giffen has said, that Malthus's statistical enquiries remain as valu- able as ever. At the same time, further experience gained during the present century suggests certain qualifications in the statement of the Malthusian doc- trine, and in some degree modifies the practical conclu- sions drawn from it. Empirical laws need not, however, always remain so. Statistical investigations may suggest laws which can subsequently be established on a more or less satisfac- tory deductive basis. In other words, the observed uniformities may be referred to causes which it is shewn will account for them, and which are also shewn to be in operation. Thus the tendency of financial crises to recur at periodical intervals was not first worked out theoretically ; it was disclosed by statistical observations, and theories to account for the cyclical movement were afterwards propounded. Another simple illustration is to be found in the autumnal drain on the Money Market. Besides affording absolute additions to economic knowledge, statistics are of great value in enabling the deductive economist on the one hand to test and where 328 POLITICAL ECONOMY AND STATISTICS. [CHAP. necessary modify his premisses, and on the other hand to check and verify his conclusions. By means of statistics, also, he may sometimes roughly measure the force exerted by disturbing agencies. Thus, Mr Bagehot appeals to statistics to test the legitimacy of the postulate that in modern industrial communities there tends to be a movement of labour from the worse paid to the better paid localities. He holds that patent statistical facts shew what may be called " the tides " of the people, the set of labour being steadily and rapidly from the counties where there is only agriculture and little to be made of new labour, towards those where there are many employments and where much is to be made of it 1 . Statistics may, again, be called in to determine how far in a given state of society this tendency does actually result in an equality of wages, or how far there are other strong forces in operation which succeed in more or less coun- teracting it. Another illustration is afforded by the functions of statistics in the controversy between free traders and protectionists. Statistics cannot by themselves decide this controversy, but they are of assistance in supple- menting more abstract reasoning. The free trader, whilst basing his conclusions mainly on a deductive process, is bound to deal with all available statistics, 1 Economic Studies, p. 22. X.] PROBLEMS INVOLVING STATISTICS. 329 shewing in what respects they bear out his theory, and explaining away any apparent inconsistencies 1 . Passing from economic science in the strict sense to its applications, the necessity for statistics is found to be relatively even greater. There are many important problems of fact especially where a comparison is in- stituted between different times or places, though not in this case alone which imply the use of statistics in their very statement: for example, the enquiry whether during the last twenty years there has or has not been an appreciation in the value of gold ; the comparison between the position of the labouring classes now and fifty years ago; the analysis and explanation of the recent depression of trade ; the investigation of the relative pressure of taxation, under existing conditions, upon different classes of the community. A sound knowledge of theory is requisite for a satisfactory treatment of problems of this kind. Theory guides us in our selection of statistics, and teaches us how to 1 Compare Sir T. Farrer, Free Trade versus Fair Trade, and Dr Giffen on the Use of Import and Export Statistics (Essays in Finance, Second Series). "Statistics," Dr Giffen remarks, "though they cannot logically prove the affirmative in the direct issue between free trade and protection, from the difficulty of rinding exactly parallel cases and eliminating other causes, may be used to prove negatively that there is nothing in the apparent facts to help the protectionist" (p. 223). In some other of Dr Giffen's essays there are excellent instances of the mutual bearing on one another of statistics and deductive reasoning. 330 POLITICAL ECONOMY AND STATISTICS. [CHAP. turn them to the best account. But the data for the solution of the problems must necessarily be numerical. In the majority of cases, moreover, aggregate regu- larity has to be evolved out of individual irregularity, and hence a special reason why we must deal with phenomena in the mass, and not individually. Thus again taking as our examples the enquiries above re- ferred to if we compare prices now with prices twenty years ago, some will be found to have fallen, some to have risen; similarly if we compare wages now with wages fifty years ago; even in a year of depression some trades are found to be flourishing ; the pressure of taxation varies in the case of different individuals in the same class. Averages must, therefore, be taken ; and it is clear that the essential conditions for the right solution of the problems are reliable statistics, and ability to use statistics correctly. The right use of statistics is, indeed, far from being a simple matter. Statistics, it is often said, can be made to prove anything. And if they are used without special knowledge, or grouped simply with the object of establishing a foregone conclusion, the charge is well founded. As against ignorant or prejudiced statisti- cians, or against the casual employment of a few figures picked up at random and regardless of what may be called their context, it is not difficult to defend the paradox that there is nothing more misleading than X.] IMPORTANCE OF THE STATISTICAL METHOD. 331 facts except figures. For reasoning from statistics, in addition to the dangers which it has in common with all empirical reasoning, is subject to difficulties and dangers peculiar to itself 1 . If, however, the limitations of statistics are clearly recognised, if they are accurately collected over an adequate range, if they are employed without prejudice and after full enquiry into their true significance, and if they are fairly and properly grouped, then their value is unique, and the statistical method easily makes good its claim to rank as a thoroughly effective and reliable instrument of science. 1 The nature of these difficulties and dangers is briefly indicated in the note following this chapter. NOTE TO CHAPTER X. ON SOME OF THE PRECAUTIONS REQUISITE IN THE USE OF STATISTICS IN ECONOMIC REASONINGS. 1. Conditions of the reliability of statistical data. If arguments based on statistics are to be of any value, particular attention must be paid to the following points : (a) the sources from which the statistics are obtained, with special reference to their reliability ; (b) their true meaning and significance ; (c) their completeness or incom- pleteness as covering the whole range of the phenomena to which they relate ; (d) the manner of their grouping, with special reference to the taking of averages. Each of these points may be briefly considered in turn 1 . The initial difficulty in the use of statistics is the pos- sible inaccuracy of the original data. Statistics may be obtained and published officially, or they may be collected through private channels. Under the former of these con- ditions, the accuracy of the figures is sometimes practically unquestionable ; as, for example, in the case of railway 1 In what follows I am specially indebted to Dr Giffen's Essays in Finance, First and Second Series ; to papers on Statistics by Professor II. M. Smith in the Political Science Quarterly, the Quarterly Journal of Economics, and the Publications of the American Economic Associa- tion ; and to various papers in the Journal of the Statistical Society. NOTE.] THE RELIABILITY OF STATISTICS. 333 traffic receipts. But this is by no means the universal rule, even when official statistics are forthcoming. Thus up to 1854 the values of imports into this country were calculated at the prices of the early part of the eighteenth century. From 1854 to 1870 they were officially computed according to the best information obtainable. At the present time the values of both imports and exports are declared by the importers and exporters. The returns are of course checked by the officers compiling the statistics; but still their accuracy depends to a considerable extent upon the good faith and carefulness of the consignees and exporting agents, Avho it is said are often insufficiently instructed by their principals. The chance of error is greater in some cases than in others. For example, when goods are sent into the country to be sold on commission, there is no available invoice of the same definite character as when they are sent to order. Again, the check exercised by the customs' officers is likely to be more effective in the case of goods that are subject to a duty than in the case of those that are non-dutiable. The uncertainty attaching to the accuracy of statistics is still greater when they are collected through private channels. Wages statistics may be taken as a special example. If they are obtained by the simple process of writing to some individual in each district under investiga- tion, and asking him to give the best information in his power, they can hardly be of much practical value, unless the sources of the informant's own knowledge are fully set forth and are themselves capable of being tested. Botli employers and workmen are in danger of being more or less unconsciously biassed by class prejudice or by conside- 334 THE RELIABILITY OF STATISTICS. [CHAP. X. rations of the uses to which the information they afford may be put. Probably the best sources of information are the actual ledgers and pay-rolls of large establishments if recourse can be had to them, or records kept by trade societies primarily with a view to the enlightenment of their own members. Where two or more sources of infor- mation are available, they will serve mutually to check one another. It is of importance that access should be had to the ultimate data themselves, and not merely to calculations already based upon them. For instance, knowledge of the actual wages paid to individuals, and the number of men employed at each rate, is of much greater value to the statistician than ready made averages provided either by employers or workmen. This of course applies not only to wage-statistics, but to all statistics. A distinction may be drawn between the absolute accu- racy of statistical data, and their relative accuracy. Abso- lute accuracy may in some cases be unattainable ; and yet it may be possible to institute very valuable comparisons. But in such cases relative accuracy is essential ; the sta- tistics should be collected on the same method, and under similar circumstances. If these conditions are fulfilled, error within certain limits need not be seriously misleading; for in accordance with the doctrine of chances, it may be a legitimate assumption that the error will exert approxi- mately the same proportional influence on each side of the comparison. It follows that in comparing statistics of any kind for consecutive years, what is specially important is that no change shall have been made in the mode or circumstances NOTE.] THE RELIABILITY OF STATISTICS. 335 of their collection. For example, in comparing income tax returns at different periods, it may be necessary to make allowance for improvements in the means adopted for preventing false returns. It should also be remembered that in consequence of changes in the rate of the tax, in- ducements to falsify statements of income may be increased or diminished. In the case of official statistics, it is most desirable that on the occurrence of any change in methods of collection, the old method should for a few years be carried on along- side of the new. It can then be approximately calculated what allowance must be made for the change in compari- sons involving periods before and after it. A similar difficulty frequently arises when statistics of different nations are compared. There is, for instance, the greatest variety in the methods by which the values of exports and imports are calculated in different countries. It has already been mentioned that in England the practice is for values to be declared by exporters and importers. In most foreign countries, however, values are computed according to tables of prices officially drawn up. With a view to international comparisons, statisticians are exerting all the influence they can command in the direction of bringing about uniformity in methods of collecting sta- tistics in different places. Where complete accuracy is unattainable, it is impor- tant to be able to calculate the limits of the possible inaccu- racy. If we know the exact conditions under which the data were collected, and the precautions taken to ensure correctness, then to calculate such limits may be within our power, and due allowance for error can be made. We 336 THE INTERPRETATION OF STATISTICS. [CHAP. X. may add that if, when statistics are being collected, the use that is to be made of them is known and borne in mind, then although it may be necessary to make some allowance for the possible effects of bias they are more likely to be valuable for their purpose. 2. The interpretation of simple statistics. Apart from any inaccuracy in the actual figures, there is a con- stant danger of reading into statistics what when properly interpreted and analysed they cannot be shewn to imply. For in the phenomena to which they relate there may be differences, of which the mere figures taken by themselves yield no indication. A few simple illustrations may be given of the kind of errors likely to result, if heterogeneous and incommensurable quantities are treated as though they were homogeneous and commensurable. In the comparison of price-lists, and in many other enquiries, constant guard must be kept against overlooking differences in quality. This applies even to raw materials. The quality of corn, for example, as well as the yield of the harvest, varies with the season ; so that the Gazette average of wheat may itself be misleading. The fact of variation in the quality of raw materials is clearly recognised by Pro- fessor Rogers in his History of Agriculture and Prices. He purposely omits all notice of inferior grain, and in calcu- lating the average price of cattle neglects such quotations as evidently relate to animals much below the average quality. Similar omissions are made in the case of wool ; and it is pointed out that in this case, the difficulty is increased by the fact that even among the various kinds of best wool, there is so large a difference in value as to suggest a difference of breeds in different districts. It is NOTE.] THE INTERPRETATION OF STATISTICS. 337 clear that under such conditions as these, the most careful judgment in the selection and manipulation of figures is essential. The task of comparing prices, simple as it may at first sight appear, is seen to be one that needs for its adequate performance, not only freedom from bias, but also wide experience, and sagacity of a high order. When we pass from raw materials to manufactured goods, the difficulty is enormously increased. There is room for wide divergence in quality, when only one kind of material enters into the composition of the commodity in question ; and when materials are mixed, the proportion of the more valuable to the less valuable may in some cases vary almost indefinitely. The same names may even in course of time come to denote things that are practically different in kind. One or two instances may help to illustrate the point under discussion. It might have been anticipated that during the extraordinary inflation in the price of tin in the early part of 1888, the exports of tin-plates would decrease. As a matter of fact, however, they shewed a slight increase; and the explanation of the apparent anomaly seems to be that a large proportion of the goods classed as tin-plates in the Board of Trade returns have little or no tin in their composition. During the period when the price of tin was so high, the shipment of these thin iron plates largely predominated over those more thickly coated with tin 1 . Another example is afforded by an instance in whicli the value given for a consignment of shirtings to New Zealand was challenged as being so low as to be obviously 1 See the Economist, June 30, 1888, p. 823. K. 22 338 THE INTERPRETATION OF STATISTICS. [CHAP. X. incorrect. Invoices were accordingly produced, and they proved both that the figures were rightly quoted, and that the goods were really described as shirtings. Further enquiry, however, brought out the fact that their object was to serve as shrouds for the carcases of sheep sent to Europe in refrigerating chambers. For this purpose an article much inferior in quality and price to those ordi- narily classed as shirtings was required 1 . In some cases the differences in quality are now in one direction, now in the other, so that over any fairly large area they practically cancel one another. Error may then be avoided by simply taking averages. But no such re- source is available where the changes tend to be all in one direction, as in the improvement in the quality of some manufactured goods, and in house accommodation. In certain cases there is a progressive change of quality even in raw materials. "An ox or a sheep," Professor Marshall remarks, " weighs now more than twice as much as it used to ; of that weight a larger percentage is meat, of the meat a larger percentage is prime meat, and of all the meat a larger percentage is solid food, and a smaller percentage is water". 2 An instance of a somewhat different kind may be added, in which mere price-lists are likely to mislead those who have not special knowledge of the trades in which the lists are used. Where the price of the raw material is subject to considerable fluctuations, it is not unusual for the nominal wholesale price of finished goods to remain un- changed that is to say, there is no alteration in the pub- 1 British Association Report for 1885, p. 870. 2 Contemporary Review, March, 1887, p. 375. NOTE.] THE INTERPRETATION OF STATISTICS. 339 lished price-lists while there is, nevertheless, an actual alteration in prices through a modification of discounts and in other ways. For reasons of this kind, price-lists may sometimes be even worse than useless, unless supplemented by additional information. Another obvious case in which care is necessary in the interpretation of statistics need be touched upon only briefly. Whenever there are changes in the prices of commodities, it is clear that values, say of exports or im- ports, afford no adequate measure of amounts. For in- stance, in 1872 we exported iron and steel to the value of 35,996,167, and cotton yarn to the value of 16,697,426; while in 1882 the figures had fallen to 31,598,306, and 12,864,711. The amounts were, however, only 3,382,762 tons and 212,327,972 Ibs. in the former year; while they were 4,353,552 tons and 238,254,700 Ibs. in the latter. Here then is a further reason why all statistics involving prices need to be interpreted with caution. It is now generally recognised that between distant periods any bare comparison of prices is worthless ; we are practically more in danger of being led astray when we are comparing statistics over a long series of consecutive years. Passing to statistics of wages, it is to be observed that the figures themselves usually give nothing more than nominal time wages, and hence afford a very uncertain criterion both of real wages and of task wages. Other variables are involved in the determination of both of these, and supplementary statistics are therefore required if a comparison is to be made either of the well-being of the working classes, or of the cost of labour, at different periods or in different localities. Account must, for in- 222 340 THE INTERPRETATION OF STATISTICS. [CHAP. X. stance, be taken, not only of variations in the prices of those commodities upon which wages are habitually spent, but also of variations in their quality, thus again introduc- ing the difficulties to which reference has already been made. We need not dwell upon the importance of also taking into consideration variations in hours of labour, intensity of labour, and so on. The necessity of having regard to such points as these shews, however, the difficulty attend- ing arguments from wage-statistics to the condition and progress of the working classes. Any argument bearing upon this problem should if possible be cumulative, the same result being obtained from different points of view and from quite independent figures. In other questions relating to wages it may be neces- sary to take account of the quality of the work done ; and the neglect of this consideration may equally vitiate an argument from statistics. Professor Cliffe Leslie, in seek- ing to controvert the doctrine that wages tend to an equality, remarks that Dorsetshire, Somersetshire, and Devonshire labourers have for the last fifty years been earning less than half what the same men might have earned in Northumberland. Granting, however, that agri- cultural wages in Northumberland have been twice those of the south-western counties, it does not follow that the same men, who earned ten shillings a week in Devonshire, would have been able to earn twenty shillings if they had migrated to the north ; for the very fact that the northern labourers had for two or three generations been earning higher wages would be likely to make them more efficient workers and more valuable to their employers. A further difficulty arises from the fact that in the NOTE.] THE INTERPRETATION OF STATISTICS. 341 course of time the character of the work itself may change; so that men who are called by the same name are not necessarily doing the same work. For instance, the rates of pay of postmen in India have increased considerably in recent years. The work required of them is, however, found to be of a different character, involving greater re- sponsibility, and demanding a better education, and a higher degree of intelligence. In 1855 they had only to deliver letters, and many of them could not even read their own vernacular. Now they have to pay money orders, and some of them are expected to read English as well as the vernacular language of the district '. From what has been said in this and the preceding section it follows that if a collection of statistics is to be of scientific value, it should not be a mere list of figures. The method of collection and the principle of compilation should be carefully explained ; and, if possible, notes should be added touching on any peculiar influences that have affected the phenomena themselves, or the accuracy of the returns, during the period over which the statistics extend. There should in particular be evidence that any apparent anomalies have been made the subject of special investi- gation, so as to exclude the possibility of their being simply due to error of some kind. The more experienced the statistician, the more scepticism he shews about all statis- tics, in regard to which the above conditions have not been complied with. 3. The range of statistics. Another danger to be guarded against in the use of statistics is that of basing con- clusions upon an incomplete survey. It need hardly be said 1 See Barbour, Theory of Bimetallism, p. 125. 342 THE RANGE OF STATISTICS. [CHAP. X. that if we are seeking empirically to determine the effects of any cause, either our facts and figures should be gathered from the whole area or period over which the operation of that cause is felt, or we should at least have adequate grounds for believing that those statistics to which we confine our attention are typical and representative. A similar precaution is necessary when by the aid of statistics we investigate the course or progress of economic pheno- mena. From this point of view it is necessary again to refer to the importance of considering the manner in which statistics are collected. The individual statistics may be perfectly correct, but they may be unrepresentative. It is, for in- stance, unsatisfactory to obtain statistics of wages by means of circular letters forwarded to workmen with the request that they will fill in various details. Only a few send answers, and these are likely to be the better off and the more intelligent. There is, therefore, no guarantee that the statistics so obtained are really typical. Another point to notice in connexion with wage- statistics is that in comparing wages in different occu- pations and at different times, average yearly earnings constitute the only satisfactory unit. The regularity of employment varies enormously in different trades and at different periods ; and a mere record of the daily or weekly wages earned by those who succeed in getting work may, therefore, be misleading, in so far as it gives no indication of the extent to which workmen are liable to be temporarily unemployed. Professor Thorold Rogers's view that the fifteenth century and the first quarter of the sixteenth may be regarded as " the golden age of the English labourer " NOTE.] THE RANGE OF STATISTICS. 343 has from this standpoint been criticized by Dr Cunning- ham and others, on the ground that it rests not only upon the interpretation of prices, but also upon the assumption that the labourer's income is fairly represented by three hundred times his daily wages. Turning to a different department of economic enquiry, it is unsatisfactory, in seeking to investigate the effects of gold discoveries on prices, to attend only to statistics of prices in a few principal markets. Cliffe Leslie adduces evidence to shew that the gold discoveries of 1850 coincided with the opening up of backward places through improve- ments in means of communication. This tended to bring about what there had not previously been a level of prices between the hitherto backward places and the great centres of commerce and industry ; and the new gold enabled the process to be carried out by a levelling-up instead of a levelling-down, i.e., it was made possible for prices to be raised in the former localities without their being at the same time lowered in the latter. It was, accordingly, in backward places for example, on the new lines of railway in the inland parts of Ireland and Scotland, and similarly in many countries of Europe that the effect of the gold discoveries in raising prices was most apparent; it could not be properly estimated by merely considering prices in great towns such as London and Paris 1 . Considerations of a somewhat similar kind are pertinent in relation to the effects of diminished gold supplies. Pro- fessor Nicholson holds that a falling off in the supplies of gold is likely to exert its primary influence in countries 1 Essays in Political and Moral Philosophy, 1888, pp. 282, ff. 344 THE GROUPING OF STATISTICS. [CHAP. X. on the margin of the commercial world, in which credit is comparatively little developed ; and that we cannot, there- fore, properly investigate the operation of the diminished production of bullion by simply attending to statistics of gold reserves, and movements of the precious metals, in the great centres of commerce. 4. The grouping of statistics. While our statistics must not be partial, we must also seek to fulfil the more difficult requirement of so grouping or " weighting " them, as to bring out correctly their relative importance. In studying the course of prices, for instance, while careful not to overlook any markets occupying a unique position or affected by peculiar influences, we must also be careful not to assign the same relative weight to small as to large markets. A simple example will serve to indicate what is here meant. Suppose that on two successive days the price of corn in any market is thirty-two shillings and thirty-six shillings a quarter ; then a bare consideration of these figures gives thirty-four shillings as the average price for the two days. But suppose further that three times as much corn is bought and sold on the first day as on the second, then for most purposes it would be more correct to say that the average is thirty-three shillings. Speaking more generally, if amounts a, b, c are on different occasions sold at prices x, y, z respectively, we weight our price re- /y _L jy _L "gj turns accordingly, and take as our average not - , 9 but -P- Where data for the determination of a + b + c a, b, c are not available, it is usually impossible to ensure NOTE.] THE GROUPING OF STATISTICS. 345 that our averages are not sometimes unduly enhanced, and sometimes unduly depressed, by reason of an exaggerated importance being assigned to small transactions of an exceptional character. In such a problem as that of measuring changes in the general purchasing power of money, the task of assigning varying weights to our primary figures becomes at the same time of increasing importance and increasing difficulty 1 . In taking the average of a series of averages there is a special risk of error, somewhat analogous to that just discussed. Thus, if there is a relative increase in the number of workers in more highly paid occupations, the average wages of all labour may rise much faster than the average of representative wages in each trade, or may even rise while the latter is stationary or falling. To quote an illustration given by Professor Marshall "If there are 500 men in grade A earning 12s. a week, 400 in grade B earning 25s., and 100 in grade C earning 40*., the average wages of the 1000 men are 20s. If after a time 300 from grade A have passed on to grade B, and 300 from grade B to grade C, the wages in each grade remaining station- ary, then the average wages of the whole thousand men will be '28s. Gd. And even if the rate of wages in each grade had meanwhile fallen 10 per cent., the average wages of all would still be about 25s. 6t/., that is, would have risen more than 25 per cent". 2 The taking of averages is thus frequently complicated 1 The need for a " weighted index-number " is, however, dimin- ished, if the number of commodities taken as the basis of the calcu- lation is made very large indeed. - Principles of Economics, vol. i. p. 725 note. 346 THE GROUPING OF STATISTICS. [CHAP. X. by the fact that an equal significance ought not to be attached to all the figures ; and the difficulty of the problem is still further increased when some kind of average less easy of calculation than the arithmetical is appropriate. In the great majority of economic investigations the arith- metical average is, indeed, the most suitable as well as the simplest ; but there are some exceptions. For instance, if population doubles itself in twenty-five years, it is obviously incorrect to say that the average annual increase is four per cent. 1 There are some cases in which to take an average at all may be misleading. If, for instance, we take an average of men's wages and children's wages, or of the earnings of professional men and manual labourers, or of house-rents in Whitechapel and in the neighbourhood of Hyde Park, we are averaging things that belong to different categories, neither of which is really represented in the result. Even if such an average can be of service for any special purpose, it will become delusive when used in ignorance of the diverse nature of the data on which it is based 2 . If there is continuity in our data, so that although the extremes are far removed from one another we pass be- 1 On the different kinds of averages, and on the subject of averages in general, see Venn, Logic of CJiance, 1888, Chapters 18, 19. 2 It has been said, e.g., by Mr Longe, that there is no such thing as an average or general rate of wages. In one sense this admits of easy disproof, since we can take the arithmetical mean of any series of quantities whatever. What is meant, however, is that wages in different occupations are unrelated to one another and disparate, so that their average is a mere number and has no practical signification or importance. This view, although it may be rejected, will serve to illustrate the meaning of what is said above in the text. NOTE.] THE GROUPING OF STATISTICS. 347 tween them gradually and through all intermediate grades, then it is a different matter. But in any case the value of an average is enormously increased if at the same time the range of the variations on both sides is clearly stated. The truth is that an average from its very nature lets drop a considerable amount of information. In itself it tells nothing as to the manner in which the data from which it is obtained are grouped. It may, therefore, advantageously be accompanied by a supplementary statement on this point, giving not only the extreme deviations from the average as above suggested, but also the average deviation 1 . This leads to the remark that in treating of fluctuations from an average, considerable importance attaches to the particular range over which the average has been calcu- lated. In dealing, for instance, with the statistics of some phenomenon over a term of years, we may seek to establish a periodicity in the movements towards and away from the average ; but the average, if taken for successive periods of years, may itself be subject to progressive variations, and unless these are correctly calculated and due allowance made for them, our conclusions may be seriously vitiated. Thus what has been called the " par " of trade, that is, the level which indicates neither prosperity nor depression, is itself ever gradually shifting its position. The exports and imports of any given year, the railway traffic receipts, the production of iron and the like, must not be compared simply with the corresponding statistics of some former period ; for they may all shew an increase, and yet because the normal level has itself risen in the interval the later 1 Compare Venn, Logic of Chance, pp. 444, ff. 348 THE GROUPING OF STATISTICS. [CHAP. X. date may coincide with the lowest tide of depression, whilst the earlier coincided with the highest tide of prosperity 1 . On the other hand, if we are studying the secular move- ments, it is equally necessary to have analysed the periodic variations. Averages of terms of years have to be com- pared, and the periods over which these averages are taken should be such as to eliminate as far as possible inter- ferences caused by the periodic movements 2 . It is not intended to give here a systematic discussion of the technique of the statistical method ; and enough has now been said to indicate the nature of the difficulties to which the treatment of statistics may give rise. The theory of statistics, which investigates in detail both the principles in accordance with which statistics should in the first instance be collected and arranged, and also the right methods of taking averages and dealing with fluctuations, is, as we have already pointed out, a department of applied logic or methodology. As such, it demands a distinctive treatment, though it seems hardly appropriate to speak of it as an independent science. 1 If we merely compare a single year at one period with a single year at another, we are practically committing the fallacy referred to in the preceding section, namely, of arguing from partial data. 2 Compare Jevous, Investigations in Currency and Finance, pp. 34, ff. INDEX. Absolutism of theory, 282. Abstract economics, 137 140 ; 284; 2936. Abstraction, method of, 115 123; 206; 212. Achenwall, G., 311. Adams, H. C. , 45 n. Agreement, method of, 190, 1. Algebraic methods, 238 n. Altruistic motives, 43; 123. Ambiguity of terms, 153 6. Analysis of conceptions, 149. A posteriori arguments for free trade or protection, 187 9. Applied economics, 57 ; 90; 138 n. Arithmetical average, 346. Arithmetical examples, 239, 240. Art, 35; 39 n. Art of political economy, 36; 538; 7280. Ashley, W. J., on the price of iron in the fourteenth century, 287; on economic legislation, 288 n. ; on the historical con- ception of political economy, 304. Assize of Bread and Ale, 289. ' Austrian School ', the, 21 ; 250. Averages, 330; 3448. Axiomata media, 168. Bacon, Francis, on science and its practical applications, 48 n.; on the explication of concep- tions, 146 ; on experiments 173 n. Baden-Powell, Sir G., 182 n. Bagehot, W., on the scope and method of political economy, 12 sqq. ; on the abstract cha- racter of political economy, 113 .; on the a posteriori method, 197 ; on the functions of ob- servation, 217; on the assump- tions of economic science, 228 ; on the method of science, 230 ; on the Bank of England, 254 ; on the relativity of economic doctrines, 282 ; his limitation of political economy to the theory of modern commerce, 289293 ; his use of statistics, 328. Bank of England, 253, 4. Banking, art of, 73. Barbour, D., 341 H. Black Death, 178, 9; 257260. Blanqui, J. A., 275 n. Bonar, J., on Malthus, 181 H.; on Eicardo, 226 n. Bramwell, Lord, 32 H. 350 INDEX. Cairnes, J. E., on the scope and method of political economy, 12 sqq. ; on applied political economy, 58 n.; on laisgerfaire, 70 n.; on rent, 86; on second- ary influences in economics, 114 n.; on 'mental experiment', 17271.; on a posteriori evidence, 183 ; on slave labour, 194 ; on the inductive method, 197 ; on the hypothetical character of political economy, 213 n.; on the ultimate premisses of econ- omic science, 216 n., 229 n.; on mathematical methods, 240, 1 ; on the effects of the Australian gold discoveries, 255 n. ; on po- litical economy and statistics, 324. Capital, 149; 150 n'; 160; 162; functions of, 67; 84; increase of, 128,9; 193, 4. Carlyle, Thomas, 69 n. Cartograms, 322 n. Causation, laws of, 205 sqq. Causes in history, 271. Caveat emptor, rule of, 127. Ceylon, 268 n.; 291. Chamberlain, J., 125 n. Chances, doctrine of, 321. Cherbuliez, A. E., on the con- sumption of wealth, 106 n.; on scientific division of labour, 131. Clark, J. B., on the definition of wealth, 95; on distribution and exchange, 100 n. Classification, 151 ; 157; 1668. Cohn, G., on induction and de- duction, 28 71., 164, 5. Colloquial use of terms, 153 5. Commercial motives, 116. Common sense, 142 5. Communism, experiments in, 1757. Comparative method, 268 n. Competition, 31; 38; 42, 3; 83; 288, 9; in the Middle Ages, 257; 260; 286,7. Competition, hypothesis of, 65; 227; 230 n.; 278,9. Competitive prices and customary prices, 284 n. Comte, A., on political economy and sociology, 108 sqq.; his use of the terms 'statics' and 'dynamics,' 141; on definitions in political economy, 147. Concomitant variations, method of, 191; 321. Concrete economics, 138 140; 284. Conring, H., 311 n. Consols, price of, 200. Consumption of wealth, 101 7. Continuity of economic phe- nomena, 246. Conventional morality, rules of, 126, 7. Co-operative production, 195. Corn Laws, effects of their repeal, 188, 9; 222. Cosmopolitanism of theory, 282 71. Cosmopolitical economy, 74, 5. Cossa, L., on applied political economy, 57 n.; on the premisses of economic science, 229 n. ; on mathematical political eco- nomy, 238 7i., 241. INDEX. 351 Cost of production, 61; 67; 100, 1; 138,9; 157 n.; 214 n.; 247. Counteracting causes, 205 8. Cournot, A., on unproductive consumption, 103 n. ; his use of the hypothesis of pure mo- nopoly, 227 n.; his use of algebraic methods, 238 n. ; on the application of mathematics to political economy, 242 4, 249. Cunningham, W., on economic science, 52 n. ; on changes in the use of terms, 161 n. ; on political economy as an em- pirical science, 167 ; on the importance of economic litera- ture to the historian of eco- nomic facts, 274 n. ; on mediae- val prices and rents, 286; on Thorold Rogers, 343. Curves, method of, 322. Custom, 256, 7; 260; 284. Customary prices, 284 n.; 286 8. Deduction and induction, 27, 8; 164, 5; 183; 193; 196, 7. Deductive method, 204 235. Definition, 146163. Dependent and independent changes, distinction between, 206. De Quincey, T., on Bicardo, 281. Descriptive economics, 166 8. Descriptive School of statisticians, 311, 2 n. Desire of wealth, 115 122. Dewey, D. B., 322 n. Diagrammatic methods, 238 n.; 2449. Diagrams, statistical, 321 3. Difference, method of, 170, 1; 175; 176; 178190. Diminishing Beturn, law of, 82 ; 172; 227. Diminishing Utility, law of, 227 ; 294. Distribution of wealth, 42 n.; 78, 9; 98101; 197,8; 253. Disturbing causes, 210 ; 217. Dogmatism in definition, 156. Dunbar, C. F., on the opposition between different schools of economists, 29 n. ; on observa- tion and the deductive method in economics, 216 n. Dynamics of political economy, 1402. Economic activities, subject to moral laws, 41. Economic activity, definition of, 96. Economic development, 44, 5; 135, 6; 1402; 253, 4; 267, 8; 3035. Economic history, 252 sqq.; philo- sophy of, 267; 3035. Economic ideals, 32, 3; 36; 44, 5; 53; 60, 1; 7680. Economic laws, 3 ; 36; 83; 86,7. Economic legislation, 32 n. 'Economic man,' the, 16; 19; 24; 113122; 138; 206. Economic motives, 115; 124 9. Economic precepts, 32, 3 ; 36. Economic theories, history of, 273280. Economics, 2; 52 .; 89 n. Economy, 1; 96. 352 INDEX. Edgeworth, F. Y., on mathe- matical reasoning, 243 n. ; on the treatment of variables, 246 ; on the application of mathe- matics to political economy, 248 n., 249 n., 250 n. Egoism and altruism, 115 123. Ely, E. T., on the ideal of political economy, 79. Empirical generalizations, 5 ; 190203 ; 326, 7. 'English School,' the, 20, 1; 28, 9. Ethical conception of political economy, 23; 25, 6; 37,8; 46, 7; 51, 2; 59. Ethics, 35 n. Ethics of political economy, 36; 5861. Exchange of wealth, 67, 8; 93 n.; 99101; 197200. Exclusiveness, fallacy of, 8, 9. Experiment, 169178. Experimenta lucifera et fructi- fera, 173 n. Experimental legislation in the Middle Ages, 175 n. Export and import statistics, 333; 335; 339. Export duties, 75. Extension of demand, 249. Fallacies in political economy, 7,8. Farrer, Sir T., 182 n.; 329 n. Financial crises, 129, 130. Fluctuations from an average, 347, 8. Foxwell, H. S., on the use of mathematical analysis, 245. Free trade and protection, 74 n.\ 132; 181, 2; 184, 5; 1879; 328, 9. 'German Historical School,' the, 20 sqq. ; 79; 120 n.; 298 sqq. Giffen, K., on the method of difference in economic reason- ings, 185, 6 ; on the use of the deductive method, 202 n. ; on Malthus, 327; on the use of statistics, 329 n., 332 . Gold discoveries, 189; 255; 343. Gold supply and prices, 202; 208; 343, 4. Graphic method of statistics, 3213. Gresham's law, 36 n. ; 290. Guy, W. A., 318 n. Harrison, F. , on poli tical economy and sociology, 109 n. ; on the operation of religious motives, 128 71.; on the postulates of political economy, 228 n. Hedonics, 88 n. Hildebraud, B., 20. Historical conception of political economy, 296 309. Historical illustrations, 255 261. Historical method, 24, 5; 142; 254; 2639; 296309. Historical method of definition, 161. 'Historical School,' the, 20 sqq.; 142; 262; 297 sqq. 'Historismus,' 298. History and statistics, 191. Hooper, W., on statistics, 314 H., 319. INDEX. 353 'Hypothetical', application of the term to deductive political economy, 205213. Hypothetical precepts, 55; 76, 7. Illustrative hypotheses, 215. Incidence of taxes on commo- dities, 221. Inconvertible currency, 129. India, 75; 268 n.; 284 n.; 2902; 341. Indirect taxation, 162. Induction and deduction, 27, 8; 164, 5; 183; 193; 196, 7. ' Inductive ', use of the term, 192 n. Inductive method, its varieties, 169. Industrial legislation, art of, 58 ; 73 n. Ingram, J. K., on political econ- omy and sociology, 26, 7, 109 n. ; on the use of mathematics in economic reasoning, 241; on the historical method, 297, 303 n. Intensification of demand, 249. Interdependence of economic phe- nomena, 98 sq