Walrasian Economics - Donald A.walker

P1: JPJ
0521858550pre CB1007/Walker 0 521 85855 0 November 24, 2005 7:31

ii
This page intentionally left blank
P1: JPJ
Walrasian Economics
To understand the various strands of general equilibrium theory,
why it has taken the forms it has since the time of L eon Walras,
and appreciate fully a viewof the present state of general equilib-
riumtheorizing, it is essential to understand his work and examine
its inuence. The rst section of this book accordingly examines
the foundations of Walrass work, including his philosophical and
methodological approaches to economic modeling, his views on
human nature, and the basic components of his general equilib-
rium models. The second section examines how the inuence
of his ideas has been manifested in the work of his successors,
surveying the models of theorists such as H. L. Moore, Vilfredo
Pareto, Knut Wicksell, Gustav Cassel, Abraham Wald, John von
Neumann, J. R. Hicks, Kenneth Arrow, and G erard Debreu. The
book also provides a bibliography (relating to Chapter 11) of
recent models of many types regarding which their constructors
explicitly acknowledge Walrass inuence.
Donald A. Walker is University Professor and Professor of Eco-
nomics Emeritus at Indiana University of Pennsylvania. He is
the author of Walrass Market Models (Cambridge University
Press, 1996), Advances in General Equilibrium Theory (1997),
and many articles and reviews in professional journals, chap-
ters in collections, and entries in handbooks and encyclopedias.
Professor Walker is the editor of, among other works, William
Jaff es Essays on Walras (Cambridge University Press, 1983),
Money and Markets: Essays by Robert W. Clower (Cambridge
University Press, 1984), Equilibrium (2000), The Legacy of L eon
Walras (2001), and, with J.-P. Potier, La Correspondance entre
Aline Walras et William Jaff e (2004). He was president of the
History of Economics Society in 198788, editor of the Jour-
nal of the History of Economic Thought from 198999, and the
rst president of the International Walras Society, 19972000.
i
P1: JPJ
ii
P1: JPJ
Walrasian Economics
Donald A. Walker
Indiana University of Pennsylvania
iii
caxniioci uxiviisir\ iiiss
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, So Paulo
Cambridge University Press
The Edinburgh Building, Cambridge cn: :iu, UK
First published in print format
isnx-:, ,;-c-,::-,,,-;
isnx-:, ,;-c-,::-:o:::-
Donald A. Walker 2006
2006
Information on this title: www.cambridge.org/9780521858557
This publication is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
isnx-:c c-,::-:o:::-,
isnx-:c c-,::-,,,-c
Cambridge University Press has no responsibility for the persistence or accuracy of uiis
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
hardback
eBook (EBL)
eBook (EBL)
hardback
P1: JPJ
To Patricia, Valerie, and Anthony
v
P1: JPJ
vi
P1: JPJ
Contents
Preface page ix
Introduction 1
Part I. Walrass ideas
Section I. The foundations of Walrass mature
comprehensive model
1 General philosophy and methodology 25
2 Economic philosophy and methodology 58
3 Methods of evaluation of economic theory 85
4 Human nature 113
5 Basic sub-models 140
Section II. The written pledges sketch
6 Rationale for the written pledges sketch
and its characteristics 175
Section III. Walrass writings
7 Some bibliographical remarks 199
8 The denitive bibliography of the writings
of L eon Walras 207
Part II. Walrass inuence
9 Models constructed by Walrass contemporaries and
immediate successors 259
10 Models drawing upon the heritage of the written pledges
sketch, 1930 to 1971 288
vii
P1: JPJ
viii Contents
11 Concluding comments: Walrass ideas in modern
economics 313
References 317
References and Bibliography for Chapter 11 334
Index 345
P1: JPJ
Preface
To understand the various strands of general equilibriumtheory and why
it has taken the forms that it has since the time of L eon Walras, and to
achieve a view of the present state of general equilibrium theorizing and
the directions it appears to be taking, it is essential to understand Walrass
work and examine its inuence. In Walrass Market Models (1996),
I considered the technical details of his economic models in great detail,
but something more must be done to achieve a fully rounded view of
his economic contributions and a truly productive use of a knowledge
of them. The rst section of the present book accordingly examines
the foundations of his work, which were the essence of his heritage.
These were his philosophical and methodological approaches to eco-
nomic modeling, his views on human nature, and the basic components
of his general equilibium models, considered from the point of view of
what is interesting about them in light of the subsequent inuence of
his work. The second section examines howWalrass inuence has been
manifested in the theorizing of his successors, surveying the models
of economists such as H. L. Moore, Vilfredo Pareto, Knut Wicksell,
Gustav Cassel, Abraham Wald, John von Neumann, J. R. Hicks, Ken-
neth Arrow, and G erard Debreu. Thus this book and Walrass Market
Models are complementary. Taken together, they provide a well-rounded
statement of Walrasian economics (that is to say, the economics of Wal-
ras and the scholars who came after him who used his ideas), so it is my
hope they will be studied together.
I gratefully acknowledge beneting from the comments made on an
earlier draft of Chapters 1, 2, 3, and 4 by Jan van Daal, Jean-Pierre Potier,
and Deborah Redman, and the comments of two anonymous referees.
My thanks also go to Scott Parris and Peter Katsirubas for their tech-
nical expertise and willingness to accommodate my concerns in the
preparation of the book. Any deciencies of the nal version are my
responsibility.
ix
P1: JPJ
x Preface
I wish to thank the publishers for the following permissions:
Bibliography of the Writings of L eon Walras, 1st ed., History of
Political Economy, 19 (4), 1987, 667702. Permission granted by Duke
University Press to reprint this publication, with changes.
Some Comments on L eon Walrass Health and Productivity,
Journal of the History of Economic Thought, 21 (4), 1999, 43748.
Permission granted by the History of Economics Society and the editor
to reprint, with changes, material drawn from this publication.
D.A.W.
Indiana, January 2005
P1: PIG
0521858550int CB1007/Walker 0 521 85855 0 November 24, 2005 7:40
Introduction
An overview of the objectives of the book
This book is about Walrasian economics, meaning the economics of
L eon Walras and the economics in which other scholars have used his
ideas. It has two general objectives. One is to present an accurate account
of the foundations and basic components of Walrass models of general
economic equilibration and equilibrium.
1
The other is to trace the inu-
ence of that legacy on his contemporaries and successors. It will be
understood that when reference is made in this book to his legacy,
the word is used to mean his legacy of economic theorizing, not of his
normative ideas.
In treating the subject, I have made an effort to reach an audience
wider than general equilibrium specialists. I have therefore explained
some theoretical propositions that are elementary knowledge to gen-
eral equilibrium specialists but not to some students and not to some
economists who concentrate on other elds of research. The book is
as self-contained as is possible with subject matters that are innitely
extendable in various detailed respects. Virtually all the material in this
book is new research, undertaken specically for it, but treatments in
greater depth of a few of the topics, notably a few of the topics taken up
in Chapters 5 and 6, are given in Walker 1996.
1
Walras emphasized that his model dealt with the rst aspect as well as the second
(239, 1889, 2, p. 364; 1889, 2, p. 370; 1891, 2, p. 434; 1895, 2, p. 630), but normally,
except for purposes of emphasis, it will be referred to, as he did, simply as one of
general equilibrium. By contrast, some economists general equilibrium models deal
only with equilibrium and not with equilibration, and the context of the discussion
will make that clear.
The style of the citations of Walrass writings (as they appear, for example, in the
rst sentence of this note) is explained in the last part of this introduction.
1
P1: PIG
2 Introduction
L eon Walras and his economics
A biographical sketch
Walrass biography is an oft-told tale,
2
so here will be given only a brief
sketch of his life. He was, of course, the founder of the modern the-
ory of general economic equilibrium. He was born on December 16,
1834, in Evreux, France, and christened Marie Esprit L eon. He had an
excellent education during the period 1844 to 1854, covering such sub-
jects as logic, philosophy, ethics, history, physical sciences, chemistry,
mechanics, elementary mathematics, geometry, and calculus (Dock` es
and Potier 2001, pp. 1416). After his lyc ee education, however, he
failed the entrance examinations for the Ecole Polytechnique, and did
not have high enough grades to be retained in the program at the Ecole
imp eriale des Mines. Responding to his predominant interests at that
time, he abandoned the plan of becoming an engineer in 1856. He began
the pursuit of a literary career and, in fact, published a novel (2, 1858)
and a short story (7, 1859). In 1859, however, he began the fulllment of
a promise to his father, Auguste Walras, to give up his literary pursuits
in order to devote himself to the study of social science.
Of course, Walras had to earn a living for himself and his family,
so during the years 1860 to 1870 he worked rst as a journalist, then
worked in a bank and a railroad ofce, became managing director of a
cooperative association bank, and worked for another bank. During this
time he undertook economic research. That was a subject to which he
devoted himself for many years, so it is not surprising that the degree
of his technical sophistication and the character and quality of his the-
oretical work changed. His rst phase of intellectual activity in regard
to economic theory, beginning in 1859 and continuing until 1872, con-
sisted largely of journalistic applications of his knowledge of existing
theory and was experimental in regard to his original ideas.
3
Walrass
participation in a conference on taxation in Lausanne in 1860 (27, 1861)
and his publication of a book on property and justice in that year (15,
1860) one that mainly set forth the ideas of his father are examples
of his interest in economic topics. Among those ideas was the argument
that the state should own all land because land is not produced as the
result of the activity of any economic agent and because its value is given
2
Walrass autobiography is listed as entry 240 in Chapter 8. His antecedents and early
life are the subjects of Jaff e 1984. Other details about his life, both private and profes-
sional, are given in Jaff e 1935, Walker 1970, Walker 1987, Walker 1996, and Potier
and Walker 2004. Many of those details are mentioned and others added to them by
Pierre Dock` es and Jean-Pierre Potier in the most complete biography of Walras yet
written (249.V, pp. IXXCIV).
3
See the entries for the years 18611871 in the denitive bibliography (Chapter 8).
P1: PIG
Introduction 3
to it by the growth of population and of the economy. Walras also wrote
many analyses of then-current economic problems, notably contributing
articles to the journals La Presse, LInd ependant de la Moselle, Journal
des Economistes, and Le Travail, and giving and publishing public lec-
tures on his ideas about an ideal society. Those writings can be seen to be
his early efforts to adopt an analytical approach to the study of practical
economic issues, an approach that he rened in later years.
Moreover, at the start and again toward the end of that period, Walras
began to use mathematics in the construction of rudimentary models
dealing with exchange (16, 1860; 91, 1869; 95, 1871). He relates in his
autobibliography: The idea of creating mathematical economics, that I
had announced in my letter offering my services to the Council of State
of Vaud, never ceased to occupy my mind after 1860 (239, 1893, 1, p. 5).
He did not, however, have an adequate foundation in mathematics and
lackedsome of the concepts that he neededtoestablishthe foundations of
a theory of supply and demand and of the interrelationships of markets.
Nothing that he wrote during that phase of his career can be considered
a valuable contribution to economic theory.
Walrass 1860 paper on taxation drew him to the attention of Louis
Ruchonnet, a Swiss statesman. Ten years later, Ruchonnet recommended
successfully that Walras be offered an appointment, despite his lack
of formal credentials in economics, at the Acad emie (subsequently
Universit e) de Lausanne. Walras began his duties there as a profes-
sor without tenure in 1870, and the next year his appointment became
permanent. His subsequent attempts to obtain a position in a French uni-
versity were rebuffed for, after all, he still lacked the necessary educa-
tional credentials and, moreover, he was an exponent of mathematical
economics, which was regarded unfavorably by the academic establish-
ment. He therefore remained at Lausanne for his entire academic career.
He retired in 1892, and died on January 5, 1910, in Clarens, Switzerland.
Inuences on Walras
It should be recalled when reading the cursory statements here about
links between Walrass ideas and those of his predecessors that this book
is concerned with his theories and his inuence on other economists and
not withthe inuence of other scholars uponhim. Nevertheless, it is inter-
esting to note briey his intellectual predecessors and contributions of
theirs that helped himin conceiving and constructing his general equilib-
rium model. Although the ideas of a number of scholars were important
in preparing him for that achievement, Walras declared ungenerously
that the only two men who produced previous works which have aided
me are . . . A.-A. Walras, my father, and Mr. Cournot (239, 1874, 1,
P1: PIG
4 Introduction
p. 397). It is quite true that his father inuenced a number of L eons
philosophical and economic ideas. Auguste was a secondary school
administrator whose writings on economics, whose library of books on
the topic, and whose belief that economics could be transformed into a
mathematical discipline were all important in the formation of Walrass
thought. Auguste had tried to make some progress on the question of the
determination of prices, and bequeathed that problem to his son, along
with an interest in utility and scarcity as elements that should be used in
its solution. Auguste provided a number of concepts that were given an
important place in L eons economic models, providing himwith points
of departure or foundations for the establishment of the general theory
[sic] of economic equilibrium (239, 1903, 3, 217). These included a
concept of marginal utility, the distinction between durable goods and
those that are used only once, a classication of economic resources,
a treatment of productive factors as different kinds of capital that yield
services, and an aspect of the concept of a numeraire.
As for Antoine-Augustin Cournot (18101877), Walras expressed
very early in his career (30, 1863) his debt to that eminent scien-
tist. Admiring as Walras did Ren e Descartes analytical approach, and
regarding mathematical economics as being in the same class of scien-
tic achievements as Descartes geometry (118, 1876, p. 367), Walras
wrote that he was deeply impressed when, in 1853, he found a concrete
expression of that approach in Cournots work (1838). He learned from
Cournot that some aspects of economics could in fact be expressed in
mathematical terms, specically with the use of calculus (239, 1874, 1,
pp. 366, 398, 421), which pointed him down the route that he should
travel to achieve the objective of mathematizing economics. He wrote
that he saw, from that time on, that economics would not be a true
science until the day on which the theory of value in exchange would be
established . . . by means of the application of the calculus of functions
(213, 1905, p. 1, column 2). In 1853 he began to study geometry, cal-
culus, and mechanics as they were presented in the works of Descartes,
Isaac Newton, and J. L. Lagrange (240, 1893, in 239, 1, p. 2).
In specic regard to the mathematization of economics, Walras cred-
ited Cournot with introducing him to the idea that the demand of each
individual for any commodity can be mathematically expressed (240,
1893, in 239, 1, p. 5), and to the mathematical theory of monopoly
(123, 1877, p. 376; 376, p. 659; 239, 1877, 1, p. 535). He expressed
his indebtedness to Cournot again at the end of his career. Looking
back on it, he commented that he wanted to acknowledge that Cournots
application of mathematics to economics has had, on the direction of
my ideas and of my work, a particular inuence (213, 1905, p. 1,
columns 12).
P1: PIG
Introduction 5
Despite his acknowledging only the inuence of his father and
Cournot, and although he had many original thoughts on the functioning
and interdependence of all parts of the market system, Walrass ideas
beneted from a variety of contributions by other scholars. For example
in 1758, Fran cois Quesnay had constructed a rudimentary scheme of
the relationships that he thought existed between different parts of an
economic system. That was suggestive to writers that preceded Walras
and perhaps to him also, although he criticized the absence in it of any
notion of a theory of prices that is, of the functioning of markets (123,
1877, pp. 32829; 341, pp. 6056). Jean Baptiste Say, whose writings
were frequently mentioned by Walras, emphasized in the early 1800s
that the production and consumption sides of markets are interrelated, a
relationship that is fundamental in Walrass economic models. Say also
developed a theory of the entrepreneur that must have provided a stim-
ulus to Walrass thought, although Walras was critical of certain aspects
of Says theory (Walker 1996, pp. 28182). Achylle Nicolas Isnards
1781 treatise (see Jaff e 1969, p. 20, n. 8) probably provided Walras
with some ideas, inasmuch as Isnard devised a mathematical model of
exchange, used mathematics in his treatment of production and capital,
and dened a numeraire in precisely the way that Walras subsequently
did (Jaff e 1969; Klotz 1994).
The physicist Louis Poinsot, in his El ements de statique (1803), devel-
oped a model of general equilibrium of the phenomena in a physical
system that Walras declared was an important source of inspiration for
his theory of general economic equilibrium (239, 1901, 3, p. 148). I
opened the Statique of Poinsot one evening in 1853, he wrote, and
that theory of equilibrium achieved through the linking and unlinking
of forces and of connected elements seemed to me so luminous and so
straightforward that I read half of it in one sitting. The next day, I n-
ished off the second half (ibid.). Poinsots systems of equations treating
the physical universe, Walras wrote, showed him the methods by which
he could treat the economic universe. In actuality, he must have been
aware that a great many previous scholars had argued that the Newtonian
approach to physical science namely, analyzing the mutual determi-
nation of the values of the atomistic variables in a system should be
emulated in the social sciences.
It was certainly true, nevertheless, that Poinsots method was sug-
gestive to Walras in specic and detailed respects. His aim became the
construction of
. . . a new science: the science of economic forces analogous to the science of
astronomical forces. I cite astronomy because it is in fact the type of science
like which, sooner or later, the theory of social wealth ought to become. In both
there are natural facts, in the sense that they are and remain superior to social
P1: PIG
6 Introduction
conventions and that they impose themselves on the human will; laws equally
natural and consequently necessary, some of principal importance, fewin number,
the others secondary, quite numerous, varied and complex; facts and laws suitable
for an extensive and fruitful application of calculus and mathematical formulas.
The analogy is complete and striking (239, 1862, 1, pp. 11920).
Achieving a consensual understanding of Walrass legacy
Past interpretations of specic parts of Walrass legacy: Walrass fertile
mind delved into subjects that have remained pertinent, and it did so in
suggestive ways that led successive generations of economists to turn
to his writings. These later generations interpreted many aspects of his
work in the same way, but inasmuch as they had somewhat different
knowledge, experiences, interests, theoretical views, and methodologi-
cal predilections, they arrived at different interpretations and evaluations
of some particular aspects of his theories. In fact, some scholars intro-
duced their articles by stating that other scholars have erroneous views
on the part of his legacy they were about to examine (see, for exam-
ple, Burgenmeier 1994, Witteloostuijn and Maks 1988, Walker 1990,
Witteloostuijn and Maks 1990, Currie and Steedman 1990, Koppl 1992,
Burgenmeier 1994). Many of the ostensible disagreements arise because
the authors believe they are discussing the same subject when in real-
ity they actually are focusing upon different ideas presented by Walras
under the same name at different phases of his career. Some differences
of opinion reect different views about the degree of importance that
should be attached to one or another part of his writings. Other disagree-
ments concern the meaning or logical quality of one and the same group
of passages in Walrass writings viewed fromessentially the same frame
of reference, suggesting the possibility that some of the interpretations
are demonstrably false.
However, the majority of the disagreements, and the most signi-
cant ones, do not stem from those causes. Walrasian scholars, past and
present, have been highly capable economists, skilled in the exercise
of their particular professional specialties. Their disagreements have a
deeper cause than their abilities, educations, interests, and intellectual
experiences as academic economists which are, after all, very simi-
lar. It is that they have concentrated upon only some of his passages,
and have unknowingly neglected others that must also be considered in
order to assemble the full context necessary to understand correctly the
meaning of any of the relevant passages. This matter is illustrated in a
particularly striking way by the materials assembled in Chapters 1, 2,
and 3 that reveal that there has been a neglect of not merely an isolated
comment or two in which Walras argued in favor of empiricism, but
P1: PIG
Introduction 7
of passage after passage and of page after page on that topic, many of
which are in his major theoretical publications.
The approach taken in this book: The approach taken here therefore
follows the advice of Fran cois Bompaire. It is Bompaire to whom
credit must be given for having been the rst to declare that an under-
standing of Walrass theories (and Cournots and Vilfredo Paretos
as well) requires a study of all his scholarly writings, not just his
models and equations. Bompaire gave a detailed literary exposition of
Walrass general equilibrium economics as it relates to a laissez-faire
competitive economy, and examined Walrass equations. Regarding not
only Walras but Cournot and Pareto as well, however, he emphasized
that:
The reader should not expect to nd that our study has a purely economic and
mathematical character. We have studied our authors as they are, bringing together
all their writings, economic, mathematical, political, sociological, philosophic, in
order to attain the best possible knowledge of their true thought in regard to the
good and the bad effects of the principle of liberty. (Bompaire 1931, p. 7)
A dissenting view on the value of consulting Walrass writings other
than his principal theoretical models has been expressed by Michio
Morishima; that view must be considered because of its denial of the
value of the approach taken in this book. Regarding the conventional
concept of the history of economic thought, Morishima noted that the
aim. . . is to provide realistic portraits of great economists of the past;
it is then necessary for the historian to read collateral material such as
correspondence, recorded speeches and other published works in addi-
tion to the major work of the economist investigated, as Peach claims
(Morishima 1996, p. 92). Morishima explained, by contrast, that his own
. . . view of the subject of history of economic thought is very different. Its task
is not to present a precise and realistic portrait of a giant in history, but to reveal
hidden and concealed relationships in thought (vision and analysis) between great
economists, such as Ricardo, Marx and Walras in my own case.
4
Morishima suggested an approach that he believed is appropriate:
For this purpose, the traditional method which Peach supports is not very useful;
there was no possibility of correspondence betweenMarx andRicardo, or between
4
Heinz Kurz and Neri Salvadori argue that David Ricardos economics have nothing in
common with the ideas of Walras. According to them, Ricardos approach to the the-
ory of value and distribution is in fact extraneous to his [Walrass] way of thinking
(Kurz and Salvadori 2000, p. 982). Michio Morishima (1996) answered those writ-
ers, contending that there are important similarities of formulations and conclusions
between Ricardos and Walrass work.
P1: PIG
8 Introduction
Walras and Ricardo. It is almost useless to dig in the library in the hope of discov-
ering a newdocument. Instead, we have to read the major works of the economists
concerned as deeply as possible. . . . I believe that the history of economic thought
should not degenerate into a portrait museum; it should play the role of an infor-
mation centre which informs us, based on the examination of the works of the
economists of the past, how their various technical ideas and perspectives are
connected with each other, what their essential contributions to present-day eco-
nomics are, and how useful and suggestive they are for the future development
of economics (ibid.).
That is surely not a defensible position. To understand the major theo-
retical works of an economist as profoundly as possible, whether for their
own sake or to discern their relation to the work of other economists and
modern theorizing, it is useful to bring to bear all the relevant information
about those works. Indeed, Morishima has also expressed the opinion
that we can only judge the full value of [Walrass] pure economics by
considering it in the context of his original farsighted scheme, which,
Morishima noted, includedhis social andappliedeconomics (Morishima
1977, p. vii). Moreover, in the case of Walras, his correspondence with
many eminent economists is available and is very useful for understand-
ing the relation of his ideas to his predecessors and his contemporaries,
inasmuch as he discussed their work and his own in his letters and in
other writings also.
In short, therefore, in order to understand Walrass legacy, the part
of this book that is devoted directly to his ideas will draw not only
on his accounts of the behavior of the real economy of his day and
on his work on general equilibration and equilibrium, but also on his
philosophical and methodological ideas, his views about human nature,
his methodology, relevant passages in his books on social and applied
economics, his jottings, and his correspondence.
The foundations of a consensus: This book seeks to systematize the
aspects of that legacy and thus to construct the foundations for a wider
consensus on Walrass work than now exists; it capitalizes on the broad
agreement that exists about many important Walrasian matters. All
economists now recognize, of course, that his conception of general
equilibrium is his most important contribution, even though the agree-
ment is qualied because we do not all have the same understanding
of what that conception is. There is also general agreement about the
characteristics of specic aspects of his work: about his 1889 model
of exchange in a temporarily isolated freely competitive market, for
example, along with his theory of consumer demand and his treatment
of the economics of the rm. This book validates the realization by
modern Walrasian scholars that his contributions have a much wider
P1: PIG
Introduction 9
range, diversity, and richness than was thought to be the case in previ-
ous years, a realization that enables the introduction here of new and
wider perspectives on Walrass theoretical legacy. The book puts into
place the last major necessary rectication of the interpretation of his
scientic economic thought, namely an accurate and balanced account
of his philosophical and methodological views, one that recognizes and
meticulously documents the very extensive empiricist aspects of his phi-
losophy and methodology. With that rectication, and with the exposi-
tion of his views on human nature, the analysis of the major components
of his mature comprehensive model, the characterization of his written
pledges sketch, and the emphasis upon the structure and functioning of
his models rather than upon his equation systems, the major part of the
heuristic scheme falls into place and a new interpretative synthesis of
his economic theorizing is achieved. The synthesis provides a truly rich,
more complete, and coherent picture of Walrass contributions and thus
provides the foundation for a consensus on the character of his legacy.
Walrass legacy
The initial model: Walrass second phase of theoretical activity, his
period of high creativity and maximum theoretical prolicacy, spanned
the period 1872 to 1877.
5
It was during those years that he developed his
initial comprehensive model of general equilibrium. The starting point
of his investigations was his realization that the quantity that an indi-
vidual demands of a commodity does not depend just on its own price,
as Cournot had postulated, but also on the price of other commodities
and on the income of the consumer (239, 1, p. 5), and (Walras added)
5
For Walras, this phase came toward the end of the fourth decade of his life, after
the period that Joseph A. Schumpeter called the decade of sacred fertility of a
scientist (Schumpeter 1951, p. 87). Schumpeter apparently arrived at that thesis by his
studies of scientists professional lives. In presenting Eugen von B ohm-Bawerk as an
example that illustrates his thesis, however, Schumpeter credited the Nobel Laureate
chemist and philosopher Wilhelm Ostwald (18531932) with the development of the
theory of the fundamental importance, for the scientist, of the years following upon
the termination of physical growth (Schumpeter 1925 in 1964, p. 373; 1954, p. 388).
J. M. Keyness thesis on this matter poses the same threat to the self-esteem of a
high proportion of scientists at any given time. He also believed that the years before
thirty are the creative ones for a scientist, and thus would have interpreted Ostwalds
proposition as referring to the decade that comes right after the end of physical growth.
W. S. Jevons, Keynes wrote, did his creative work in his twenties, at an age when the
powers of pure originality are at their highest . . . (Keynes 1936 in 1956, p. 128; and
see Keynes 1925/1930, in ibid., pp. 11516). For Schumpeter and Keynes, therefore,
Walrass contributions were made long after what they considered to be normally the
creative period.
P1: PIG
10 Introduction
on his wealth or purchasing power (239, 1877, 1, p. 535). Another deci-
sive event in the development of Walrass initial model was his learning
from Antoine Paul Piccard, a colleague at the Acad emie de Lausanne,
a way of deriving an individuals demand function for a commodity
from his utility function (239, 1872, 1, pp. 30911). That construction
provided Walras with a way of introducing scientically the utility-
maximizing motive for economic behavior in markets and the essential
utility-maximizing condition of individual equilibrium.
By 1872, armed with those ideas, Walras was ready to begin his period
of great creativity and to esh out his conception of general equilibrium
in a functioning model. This he did in four memoirs (124, 1877), the
substance and much of the wording of which appear in the El ements
d economie politique pure (106, 1874; 123, 1877), a treatise that he
made the principal vehicle for the expression of his legacy in economic
theory. Thus Walras not only expressed the belief that all economic
phenomena are interrelated (which had been done by many economists
before him), he also specied their interrelations, offered an account of
their disequilibrium behavior, and described their conditions of equi-
librium. His initial model will not be discussed directly in this book
because it has not directly been part of his legacy. That distinction was
accorded to his later models. That is not to say that the initial model was
not important. It was his brilliant expression of pure originality, and he
was careful to preserve the parts of it that were valuable the majority
of it incorporating them into the model that will now be introduced.
The mature comprehensive model: In Walrass subsequent period of
maturity as a theorist, lasting from 1877 to about the middle of the
1890s, he prepared the greatest edition of the El ements, the second (176,
1889), in which he presented his mature comprehensive model of general
economic equilibration and equilibrium.
6
In that edition he improved the
realismand coherence of his work, rened and extended his concepts and
theoretical tools, and introduced calculus. The mature model is a model
of exchange, production and consumption of consumer commodities,
savings and production and pricing of durable capital goods, markets
for loans, and the role of money generally. In the chapters on Walrass
6
Schumpeter maintained that the work done by a scientist after conceiving the ideas
of his creative period was a development and renement of them (Schumpeter 1951,
p. 87). Keynes similarly contended, referring specically to Jevons, that The last
third of Jevonss life after he was thirty was mainly devoted to the elucidation and
amplication of what in essence he had already discovered (Keynes 1936 1956,
p. 128). Thus Walras exemplies that aspect of their theses by having spent his years
after 1877 developing, rening, and extending the ideas in his initial model.
P1: PIG
Introduction 11
ideas that pre-dated the fourth edition of the El ements (210, 1900), the
unqualiedwordmodel usedinreference tohis ideas means the mature
comprehensive model. When the plural word models is used in refer-
ence to his ideas, it is to draw attention to the sub-models that constitute
the mature comprehensive model, such as his models of the consumer,
of the entrepreneur, and of capital formation.
The purpose of the chapters in Part I is to provide the reader with
the guidelines and context that are necessary for an understanding of
the complex strands of Walrass legacy. Section I explains the building
blocks of his mature system of economic thought. These are his phi-
losophy and methodology, treated in Chapters 1, 2, and 3; his view of
human nature, treated in Chapter 4; and the ideas about the structure and
functioning of competitive markets that he incorporated into his mature
comprehensive model, treated in Chapter 5.
The written pledges sketch: The characteristics of Walrass theorizing
during his last intellectual phase are taken up in Section II of Part I.
That phase, which began around 1896, was one of decline. His output of
theoretical and other work had begun to diminish long before that year,
but soon afterwards his powers of concentration and analysis weakened
rapidly, and the lessening of his productivity and creativity became more
marked.
7
In 1899 he produced an incomplete model, a sketch of a vir-
tual process of equilibration, intended to eliminate the path dependency
of equilibrium that was generated by the disequilibrium transactions
and disequilibrium production in his mature comprehensive model. He
believed this could be accomplished by the device of written pledges
to supply goods and services, as described and analyzed in Chapter 6.
Walrass realization that a virtual adjustment process would make it pos-
sible to have a model that is not path dependent was the second major
part of his legacy.
Walrass nal phase could be said to have included the years from
1900 to his death in 1910, or they could be viewed as a period of repose,
for after 1899 he wrote not another word of theory and very little of
anything else. Thus as a contributor to economic theory, he bloomed
late at the age of thirty-seven or thirty-eight and faded early, writing
less each decade after 1877, and writing little and adding even less of
7
Schumpeter remarked that Walrass creative period roughly coincides with his tenure
of professional ofce (18701892) (Schumpeter 1954, p. 828). Schumpeter was in
this case using the word creative in a wider sense than when he was discussing a
scholars period of sacred creativity, namely using it in the wider sense of a period
of appreciable scholarly authorship. His comment makes it evident that he correctly
perceived the deterioration in Walrass abilities after 1892.
P1: PIG
12 Introduction
value after he was sixty, and nothing at all in the way of theoretical
contributions during the last eleven years of his life, as is demonstrated
in a methodical and quantitative way in the last part of Chapter 6.
Walrass inuence upon other economists
Opinions about the most important aspects of Walrass legacy
There have been signicant changes in what have been considered to
be the most important and inuential aspects of Walrass legacy.
8
For
many economists just sixty or seventy years ago, Walras was consid-
ered notable mainly for having developed marginal utility theory. J. R.
Hicks wrote that his position with Jevons and Menger as one of the
independent discoverers of the Marginal Utility principle is generally
regarded as L eon Walrass chief title to fame; and this no doubt justly
enough (Hicks 1934, p. 338). In the words of George Stigler, Walras is
best known as one of the discoverers of the theory of subjective value
(Stigler 1941, p. 228). Those statements do not mean that their authors
always agreed with the popular judgment. Stigler was quite clear that
Walrass fundamental contribution (ibid., p. 229), his greatest contri-
bution (ibid., p. 242), was the concept of general equilibrium (ibid.,
p. 229).
9
Hicks also in effect dissented from the popular notion and his
8
It certainly cannot be said that historians of economic thought, including some theo-
reticians acting in that capacity, have neglected Walrass writings. Nor can it be said
that the part of that literature that is in English has been neglected by the editors and
publishers of volumes of collected works. A collection of nineteen essays written
between 1935 and 1980 by William Jaff e and edited by the present author (Jaff e 1983)
treats the genesis and development of many of Walrass ideas, the meaning of his
work, and his place in the history of economic thought. A collection of twenty-ve
articles dating from the period 1931 to 1989 was prepared by Mark Blaug (1992).
John C. Wood (1993) edited a three-volume collection of sixty-eight articles, many
of them duplicating ones in the earlier collections. The present author edited a two-
volume collection (2001) of 65 articles mainly written after 1990, 54 of which were
newly collected, in accordance with the policy of the publisher not to include ones that
appear in previous collections unless it were absolutely essential to do so in order to
give coherence to a particular section of the volumes. A considerable number of the
articles, particularly in the rst three collections mentioned, are of mainly historical
interest. It is instructive, however, to read older opinions about Walrass contributions,
because some of themwere expressed by eminent economists, and some of themwere
sometimes read by economic theorists, who, indeed, on occasion thereby became
acquainted with misrepresentations of Walrass ideas and incorporated them in that
form into their own work.
9
It was not, however, as Stigler would have it, one of the few times in the history of
post-Smithian economics that a fundamentally new idea has emerged (Stigler 1941,
p. 229), inasmuch as the idea is to be found in the works of a number of Walrass
predecessors.
P1: PIG
Introduction 13
own statement in other passages, as will be seen in Chapter 10. J. A.
Schumpeter did not make the popular misjudgment, writing as early as
the time of Walrass death, that the theory of economic equilibrium is
Walrass claim to immortality . . . , although Schumpeter also believed
that the foundation stone of his structure was the marginal utility con-
cept (Schumpeter 1910 in 1951, pp. 7677). William Jaff e agreed that
the theory of general equilibriumwas Walrass greatest contribution, but
diverged from Schumpeters view in pointing out that Walras arrived at
his conception of general equilibrium before he developed his theory of
marginal utility and demand (Jaff e 1972, p. 396; 1976, p. 513).
Walras himself was quite clear in his own mind that the theory of
general equilibration and equilibriumwas his major legacy. For example,
when a correspondent praised his theory of money, Walras thanked him
but added:
I must state that that theory is not the one that I most cherish. The theory that
I would above all be happy and proud to see awarded the description of being
one of the greatest scientic efforts of the epoch, is the theory of the mutual
determination of the prices of products and of productive services (rent, wages,
and interest), based on the conception of the entrepreneur as an economic type
distinct from the landlord, the worker, and the capitalist, and on the conception
of the two markets for products and of services (239, 1887, 2, 213).
As recently as the 1960s, Walras was known by economists such as
Jaff e primarily as being concerned with static analysis and specically
for having been the rst theorist to develop a system of static equations
of general economic equilibrium. Such economists portrayed Walrass
general equilibrium models as being static, virtual, and timeless (Jaff e
1981, p. 328; in Jaff e 1983, p. 259). Indeed, some economists portray
his models in that way today (Rebeyrol 1999, 2002; Bridel 1997). In
this book it is emphasized that Walras did in fact develop, as is implicit
in his foregoing remark and as he often claimed explicitly, a dynamic
theory of the demonstration of the realization of that equilibrium by
the play of the rise and the fall of prices until the equality of effective
supply and demand is reached (239, 1895, 2, p. 630).
Walras also predicted that his general equilibrium theory would be
inuential. He was condent that he had laid the foundations for future
theoretical studies, and that there would not be fundamental changes in
economic theory that would displace his work. He declared repeatedly
that he had correctly described the economy in his model and that the
role of his successors would be to ll in the missing parts and to rene it
(239, 1891, 2, pp. 43435; 239, 2, 1892, p. 499; 189, 1892, in 191, 1896,
p. 477; 249.VIII, appendix I, 10, p. 705). Walras knew that he had pri-
ority regarding the development of general equilibrium theory but, he
P1: PIG
14 Introduction
explained, It seemed to me that in taking up for the rst time the general
problemof economic equilibrium, I should limit myself to trying to give
very broad outlines of its solution, and leave to another generation of
economist-mathematicians the task of correcting the details and lling
in all the empty parts (239, 1892, 2, p. 409). I have never claimed,
he wrote, to have nished the entire body of economic theory all by
myself, but only to have sketched it. Serious and capable people, rst,
will recognize that I have succeeded in making that sketch, and second,
will nish the picture (239, 1891, 2, pp. 43435). That, in a general
sense, is what happened, as will be seen; the two parts of Walrass theo-
retical legacy the mature comprehensive model and the written pledges
sketch strongly inuenced all later general equilibrium modeling.
Walras considered his theoretical legacy to be a positive body of work,
as he emphasized in mentioning Pareto and Enrico Barone as examples
of serious and capable people. He was aware that they disagreed strongly
with his normative views in general and his views about an ideal eco-
nomic system in particular and would never consider completing them,
but he noted that they had already begun the work of adding to his
positive economic model:
MM. Pareto and Barone . . . do not claim to refute me by proving that I have
not completed pure mathematical economics all by myself, and they do not set
themselves the task of redoing what I have done. They take the science at the
point to which I have brought it and left it, forming the goal of continuing it after
me; and they thereby reach important, new, complete, and denitive results (239,
1895, 2, p. 655).
Walrass contemporaries and immediate successors
The research into general equilibrium theory and applications of it
that have beneted from Walrass ideas are truly the measures of the
magnitude of his inuence. Those studies are examined in Part II. One of
the strands of competitive general equilibrium theorizing was the inves-
tigation of non-virtual economic systems that stemmed from Walrass
mature comprehensive model. Chapter 9 shows how the ideas he set
forth there were used by his contemporaries and immediate succes-
sors. The most important of them in regard to the Walrasian line of
theorization were Vilfredo Pareto (18481923), Knut Wicksell (1851
1926), Enrico Barone (18591924), Henry L. Moore (18691958), and
Joseph A. Schumpeter (18831950).
10
That strand went into a period of
10
Many of Walrass themes were developed by Irving Fisher (18671947), as for exam-
ple in his doctoral dissertation, in which he constructed a general economic equilib-
rium model replete with an equation system. Nevertheless, Fisher stated that the
original source of his inspiration in regard to marginal utility and mathematical
P1: PIG
Introduction 15
quiescence during the 1930s, was pursued with renewed vigor in the
form of general equilibrium treatments of Keynesian economics, culti-
vated in microeconomic research after 1960, and is ourishing at present
in the types of models listed in the rst part of the bibliography for
Chapter 11.
Virtual competitive models, 193071
The second strand of post-Walras investigation, namely general equilib-
rium models that are virtual in disequilibrium, or, as a special case, are
always in equilibrium, is examined in Chapter 10. They clearly bear the
marks of the type of general equilibrium model that Walras wanted to
construct in his last phase of theoretical activity. That strand dominated
the eld of theoretical inquiry from 1930 to 1971. The reasons for its
eclipse were difculties that were internal to the models and their lack of
relation to real economic processes. Those matters are discussed toward
the end of Chapter 10.
Summary and concluding remarks
Chapter 11 begins with a brief summary of Walrass legacy in the eld of
positive economic theorizing. The chapter then notes recent uses of his
ideas in theoretical models regarding which their constructors explicitly
treatment in general was W. S. Jevons. Fisher owed the notion of the symmetry of
supply and demand and of a rate of ow of a commodity to Rudolph Auspitz and
Richard Lieben. He also beneted from those three economists in numerous minor
respects (Fisher 1892, pp. 34). The true relation of Fishers general equilibrium
theorizing to Walrass is problematic. In the preface to his dissertation, Fisher wrote,
regarding his presentation of general equilibrium, that
These equations are essentially those of Walras in his

El ements d economie politique
pure. The only fundamental differences are that I use marginal utility throughout and
treat it as a function of the quantities of commodity, whereas Professor Walras makes
the quantity of each commodity a function of the prices. That similar results should be
obtained independently and by separate paths is certainly an argument to be weighed
by those skeptical of the mathematical method. (Fisher 1892, p. 4)
Does that word independently mean that Fisher, a doctoral student in economics
at Yale University interested in mathematical economics and therefore indubitably
familiar with a wide range of the relevant economic literature, was claiming to
have discovered the marginal utility principle and to have conceived of and con-
structed a general equilibrium model in about 1892 without having had knowledge of
Walrass treatise, published in 1874 and 1877 and followed by numerous expositions
and commentaries by Walras and other economists? And this despite Fisher having
noted: I have assumed that my readers are already familiar with (say) Jevons,
Walras, Menger or Weiser where illustrations and explanations regarding nal utility
abound (ibid., p. 5)!
P1: PIG
16 Introduction
acknowledge that they have used Walrass ideas, and recent empirical
applications of general equilibrium theory in which such acknowledge-
ments are made. It is evident from the profusion of these recent models
that Walrass legacy continues to contribute to advances in economic
understanding.
The place and use of the bibliography
of Walrass writings in this book
Walrass legacy is conveyed in the pages of his writings that were pub-
lished during his life and in his unpublished writings. His legacy likewise
has been preserved in subsequent republications and publications. The
annotated list of those writings is presented in Section III. They are the
record Bompaire instructs us to look at in order to form a sound exposi-
tion and interpretation of Walrass legacy. The denitive bibliography is
not an appendix to the book as a whole because the book deals not only
with his ideas but with the ideas of many other economists. Moreover,
it is not just a bibliography but also literally the basis of his legacy, and
in the annotations it presents a wealth of information about his writings
and thus about his intellectual biography. It is therefore properly placed
in a section under Part I. The general references, on the other hand, refer
to the works of authors other than Walras who are mentioned in both
the chapters on his ideas and in other chapters, so those references are
properly placed separately at the end of the book. Chapter 7 introduces
the denitive bibliography and Chapter 8 presents it.
It obviously would be redundant to list Walrass writings cited in this
booknot onlyinhis bibliographybut alsointhe list of references. Instead,
his works or passages from his work are cited by giving, in parentheses,
the entry number of the document in the denitive bibliography, the
date the document was published or written, and the pagination of the
cited material in the source in which it appears. It is unnecessary to note
Walrass name each time one of his writings is cited; it is evident that
any citation that begins with an entry number is an entry in the denitive
bibliography. With its simple entry number, this method refers the reader
directly to all the details of each of the cited publications, which are
frequently complex, with multiple dates of delivery of papers and of
publication.
With the exceptions noted below, Auguste and L eon Walrass CEuvres
economiques compl` etes
11
are not cited in this book, for three reasons.
First, the original publication of any of Walrass writings given in this
11
This publication is described at the beginning of Chapter 7.
P1: PIG
Introduction 17
bibliography is the true primary source for it. The reprints of his writ-
ings differ from the original in format and pagination. For example, the
reprints that Walras put into collected volumes have pagination that dif-
fers from the original; then those collected volumes have been reprinted
in the OEC with formatting and pagination that diverges still further
from the original; and then in the case of the translations of two of those
collections, the formatting and the pagination diverge even further and,
of course, the language is different. Second, any of Walrass economic
writings can be found by consulting the index to the OEC. Third, the
dates of publication of the volumes in the OEC have nothing to do with
the dates when Walras wrote or published his texts. The exceptions are
that the location in the OEC is given for economic manuscripts that are
not published anywhere else, for some writings that Walras published
that would be extremely difcult to nd in the sources in which they orig-
inally appeared, and for cited material that appears in Walrass El ements
d economie politique pure, as distinct fromhis other writings. As is done
in this sentence in reference to the index (249.XIV), the volumes of the
OECare identied with their bibliographical entry number in Chapter 8,
in this case 249, followed by the volume number.
In references to the El ements, the edition that is cited rst is the one in
which the relevant material initially appeared, possibly with very minor
differences of wording between it and a subsequent edition. In very many
cases, the material that appears inall editions initiallyappearedinthe rst
edition, either in the rst part published in 1874 (106) or in the second
part published in 1877 (123). All such cited material became part of the
mature comprehensive model of general economic equilibrium that he
presented in the second (176, 1889) and third editions (191, 1896).
The second edition of the El ements is cited when reference is made
to the ideas in the mature comprehensive model that do not appear in
the rst edition. It is not necessary to cite the third edition because its
rst thirty-six lessons are the same as in the second edition and have
the same pagination.
12
When an edition of the El ements is cited, its
bibliographical entry number, its date of publication, and the pagination
of the cited passage are given, followed by a semicolon, and then the
relevant section number and the pagination in the OEC comparative
edition of the El ements (249.VIII).
That comparative edition is cited, with the pertinent section number,
for several reasons. First, it presents the textual variations of all editions.
For example, a passage in the rst edition may be documented with
the reference (106, 1874, p. 36; 32, p. 59); or, if the passage initially
12
The differences between the second and third editions are described under entry 192.
P1: PIG
18 Introduction
appeared in the second edition, with (123, 1889, pp. 106109; 8384,
pp. 12125). It will be noted that, to avoid useless repetition, the entry
number andvolume number of the comparative editionare not given. The
use of the section symbol indicates that reference is being made to it. The
only instances in which there is no section symbol for that edition is when
the passage appears in one of Walrass introductions to the El ements,
because his introductions do not have numbered sections. Second, the
section number in the comparative edition enables passages that are in
the fourth (210) and fth (224) editions of the El ements to be located
there, because they have the same section numbers as the comparative
edition. Third, the section numbers in the comparative edition make it
easy to nd an alternative to my English translation of a passage in the
El ements, or an English translation of a passage that I summarize, to be
located in entry 245, which is a translation of the fth edition. Fourth,
the comparative edition is cited in addition to the original sources of
the material because it is a readily available standard presentation of
Walrass treatise. The citations of it recognize the possibility that the
reader will wish to consult the French version or to read what Walras
wrote in the passages that surround the one that is cited.
More than one edition may be cited in front of the reference to the
comparative edition; for example, both 123, 1877 and 176, 1889. That
is done because the material rst appears in the 1877 edition, but Walras
changed it appreciably or added to it appreciably in the 1889 edition.
The body of the text of the OEC comparative edition is the fth edition
of the El ements. If the comparative edition is not cited after an earlier
edition, that is because Walras eliminated the relevant material from the
fourth and fth editions so that the pagination of the comparative edition
does not reect that material, or because Walras changed it so much that
it would not be proper to suggest that it is part of those editions. The
variations of all the editions of the El ements are reprinted in the pages of
the comparative edition, but in a variety of ways that, in many instances,
make it difcult to cite where they appear. Sometimes they are part of
the text but placed within crochets, sometimes reproduced as separate
pages, and sometimes as footnotes.
In a further effort to simplify the citations, the name of Walrass
correspondents are not given, nor is the number of the letter in entry
239. That information is irrelevant for an account of what Walras was
asserting and irrelevant for an understanding of the passages that are
quotedor mentioned. Moreover, that informationis unnecessarytolocate
the source of the citedmaterial, inasmuchas the volume andpage number
of entry 239 sufce for that. To further simplify the citations, the date of
the day on which Walras wrote a letter has been omitted, inasmuch as
P1: PIG
Introduction 19
that information is of no signicance for the purposes of this book, but
the year in which the letter was written is given in order to indicate its
general chronological position. The exceptions to this practice appear in
Chapter 6 in the treatment of Walrass health, where the precise dating
of his sequence of letters is of interest.
It is an accepted convention, adopted in this book, that quoted material
can be capitalized and punctuated in a way that improves readability and
helps the reader, without indicating such changes with brackets, as long
as nochange inmeaningoccurs. This is done toavoidthe unattractive and
cumbersome use of brackets and the insertion of a capital letter or lower-
case letter that the author did not write. An example is the translation of
a sentence that Walras wrote beginning with a capitalized initial word.
The sentence appears in the text as Walras wrote that it is like this that
things happen in the real world, not: Walras wrote that [i]t is like
this that things happen in the real world. Finally, regarding italicized
words in quotations, it will be understood that they always appear in
italics in the original unless it is explicitly stated that the emphasis is
added in this book.
P1: PIG
20
P1: PIG
0521858550c01 CB1007/Walker 0 521 85855 0 November 23, 2005 4:18
PART I
Walrass ideas
21
P1: PIG
22
P1: PIG
SECTI ON I
The foundations of Walrass
mature comprehensive model
23
P1: PIG
24
P1: PIG
CHAPTER 1
General philosophy and methodology
I. Introduction
There was a rationalist side to Walrass philosophical and methodolog-
ical thought and, in a certain sense and in certain limited respects, an
idealist side, but he was also a realist
1
and an empiricist. He had the lat-
ter attributes in the sense that he contended, rst, that there is an objec-
tive reality independent of ideas about it there is the personal reality
and there is la r ealit e impersonnelle (82, 1868, columns 210, 214, in
194, 1896, pp. 109, 119), the reality of physical phenomena (79, 1867,
column 185, in ibid., p. 89); second, that concepts should reect reality,
and that theories regarding all aspects of experience that are amenable to
scientic thought should be grounded rmly on empirical knowledge;
and third, that both the assumptions and the conclusions of theories
should be compared with reality and rejected or modied in the light of
newempirical evidence. In Chapters 2 and 3 it will be shown that Walras
made those contentions regarding the special case of economic science
and the even more special case of his mature comprehensive model
of general equilibration and equilibrium. It should be emphasized that
Chapters 1, 2, and 3 are expositions of Walrass views. They do not judge
them. They do not consider whether he was right or wrong in holding
his views, nor whether he or other nineteenth-century scholars opinions
on scientic method accurately reected how science is done or should
be done, nor whether his terminology regarding realism or empirical
knowledge or other philosophical and methodological matters was con-
sistent with modern usage or appropriate in the light of modern concep-
tions. His denitions of terms are explained and made evident by the
context in which they are discussed.
1
In the terminology of philosophers, realism was a Platonic doctrine of the reality of
ideas, of which things and beings are only the reection. Realism was also the name
of the medieval doctrine that universals are more real than perceptions of objects, and
of the doctrine that names denote the essences of things named. Those denitions
differ from the use of the word in reference to Walras.
25
P1: PIG
26 Walrass ideas
To understand the ideas of a scholar it is necessary to focus upon the
sense of the preponderance of his writings, and thus to discern his core
systemof thought. It is then possible to interpret his statements properly,
identify statements he made that are inconsistent with his core thought,
and relegate these to the category of irrelevant anomalies rather than
falling into the error of using them as the basis of an unrepresentative
exegetical edice. In the present study, Bompaires advice, explained in
the Introduction to this book, is followed. That course makes it evident
that the realist, empiricist, idealist, and rationalist aspects of Walrass
thought formed a coherent and unied philosophical and methodological
system, expressed many times, and that his exceedingly few statements
that seemingly contradict that system are isolated fragments that lie
outside of it. It will be shown that he developed his system of thought as
a synthesis in opposition to extreme or exclusive empiricismand extreme
or exclusive rationalism.
II. Some antecedents and contemporaries
Methodological tradition
In focusing on these matters, it will be seen that Walrass philosophical
and methodological views were much richer than has been supposed.
It will also be seen that although his views were rich, they were not
original. The reader who is familiar with the history of Continental and
British philosophy and methodology will be struck by the fact that vir-
tually any idea Walras had on those topics can be found in the writings
of scholars who preceded him. Whether he studied the writings of any
particular scientist or not, ideas that he shared in common with his prede-
cessors were widely disseminated in Western culture many years before
his time, and he drew liberally from a rich spectrum of Western thought.
It is therefore not surprising that some of his ideas can be traced back,
in a general sense, to the seventeenth (Dock` es 1996, p. 33) or eighteenth
centuries, as for example to the ideas of Ren e Descartes (15961650)
and Isaac Newton (16421727). The same can be said of the views of
any nineteenth-century scholar, such as the Newtonian aspects of the
outlook of John Stuart Mill (18061873). For some of his empiricist
ideas, Walras drew upon the philosophical school that owed much to
John Locke (16321704), whose writings were in his library. He also
owed empiricist ideas to Etienne Vacherot (18091897) (Debs 2004), a
philosopher who, contrary to the beliefs of some scholars, was not the
partisan of unalloyed idealism that he has been portrayed to be. Some
other contemporaries by whomWalras was inuenced or to whose ideas
P1: PIG
General philosophy and methodology 27
he reacted were Antoine-Augustin Cournot (18011877), Pierre-Joseph
Proudhon (18091865), J. S. Mill (18061873), and Victor Cousin
(17921867).
Walras and Descartes
Like many nineteenth-century French scholars, Walras subscribed to a
number of Ren e Descartes views (114, 1875, in 249.VII, pp. 30714;
Jolinck and Daal 1989
2
; Koppl 1992). There is also a striking analogy
between the history of interpretation of the philosophical thought of
Descartes and the interpretation of Walrass ideas put forth in this book.
Just as thorough research has exploded the myths about Descartes ideas,
so also does thorough research provide a balanced understanding of
Walrass epistemology and methodology. In characterizing Descartes
ideas, I cannot improve upon Deborah Redmans excellent exposition
and contributions to recent Cartesian scholarship (Redman 1997, pp. 21
35), so I will summarize the principal relevant points that she makes.
The truth of the assertions about Walrass ideas is demonstrated in this
and other chapters in Section I.
Descartes contended that there are aspects of science that are more
fundamental than mathematical propositions. He wanted to determine
the principle that provides a secure foundation for the use of math-
ematics in a science, and concluded that the principle is that mathe-
matics must be based on a clear understanding of the structure and
patterns of the relevant subject matter (ibid., p. 27). Walras similarly
believed that mathematics cannot be justied in economics unless it
is erected on a realistic foundation. Furthermore, we now realize that,
for Descartes, the demonstrations that make all elds of inquiry sci-
entic are not exclusively mathematical deductions, but any method
of showing and explaining the evidence, including a wide variety of
deductive and inductive procedures and analogy (ibid., p. 29); the
same was true of Walrass thought. In Descartes work and in all of
Walrass no inconsistency is alleged to exist between the rational and
the empirical aspects of theorization. Descartes understood the role
of experience that it is the source of principles and basic concepts
in physical science and that observation and experiment are necessary
(ibid., p. 31), and so did Walras. Descartes acknowledged that there are
2
Albert Jolink and Jan van Daal (1989) have pointed out that Walrass conception of
the use of mathematics in theoretical work was inuenced by Descartes Mathesis
Universalis. Other inuences in that regard are well known and have been noted in
this book, such as the work of Poinsot and Cournot.
P1: PIG
28 Walrass ideas
two ways of arriving at a knowledge of things through experience
and through deduction (Descartes, quoted in ibid., p. 31), and so did
Walras.
Like other philosophers and scientists in the seventeenth and eigh-
teenth centuries, by the geometric method Descartes meant not only
deduction as in geometry, but also the methods of pure mechanics and
the exact natural sciences; by the word deduction he meant not only
the deductive process in logic and mathematics, but also inferential
procedures such as induction and arguments by analogy (ibid., p. 29)
in short, any reasoning process by which we argue from whatever evi-
dence is available for the credibility of a given conclusion (Clarke 1991,
p. 241). Thus when Descartes wrote that the way of arriving at truth is
to use the process of deduction the rational method he was sim-
ply making an argument for providing as much evidence as possible to
support a conclusion (Redman 1997, p. 29). It will be seen that Walras
also maintained that deduction involves induction, and that conclusions
should be supported by evidence.
Descartes expression of the need for experimental procedures is
almost Baconian in its emphasis (Gewirtz 1941, p. 88). His recog-
nition of that need stemmed from his actual practice as a physicist
the parallel of the many studies of current economic problems made
by Walras during his journalistic phase and subsequently. As a result,
Descartes ends up qualifying his theory of natural philosophy to such
an extent that it brings him somewhat closer to the Aristotelian position
he was ghting (Redman 1997, pp. 3031). In Walrass case, he began
at the place that Descartes eventually reached, recognizing at the very
beginning of his career the importance of empiricism in his statements
of the fundamental principles of knowledge and scientic method, as
well, of course, as in his empirical investigations.
Descartes ambiguities, inconsistencies, and modications of his
views, like Walrass, will doubtless continue to fuel controversies, but,
Of this much, however, the philosophy and history of science is certain:
The characterization of Descartes as an archrationalist who did not
appreciate the experimental side of science is a false representation of
Descartes as scientist and philosopher (ibid., p. 33). Exactly the same
can be said of Walras.
The archrationalism thesis
Walras stated in one place that the deductive phase of scientic con-
struction in the physico-mathematical sciences does not have recourse
to infusions of new empirical considerations, and when it is completed,
those sciences re-enter the experiential realm after that, not to conrm,
P1: PIG
but to apply their conclusions (106, 1874, p. 32; 30, p. 53). He made
a similar statement about the specic eld of geometry, as will be seen.
Some scholars have relied on those statements and upon one that he
made about economic theory (see Chapter 2) to support their opinion
that Walras believed that science is a wholly deductive system whose
assumptions need not be true and whose propositions do not have to
be conrmed by data before being applied (Pokorny 1978, p. 391).
According to Pascal Bridel, There cannot therefore be any question, in
[Walrass] epistemology, of comparison of models to reality or verica-
tion of models by reference to reality, either on the level of assumptions
or of conclusions (Bridel 1990, p. 183, as paraphrased approvingly in
Lendjel 1997, p. 71; see Bridel 1997, p. 142).
Then howdo those scholars believe Walras establishes and judges the
truth of assumptions and conclusions? The answer according to Em eric
Lendjel is that Walrass writings bear witness . . . to an explicit choice
of a rationalist epistemology, consciously adopted along with its impli-
cations, according to which the truth of an assertion is not established by
experience but by the exercise of reason (Lendjel 1997, p. 49). Bridels
and Lendjels contentions on those matters will be called the archra-
tionalism thesis, and the word archrationalist will be used to mean
someone who espouses the epistemology they describe, or to indicate
a theory or model that exemplies it. In the accounts of the writers in
question, therefore, Walras was an archrationalist: Neither his assump-
tions, nor the facts entering into his models, nor his conclusions, nor
the verication of those elements refer to reality. His theoretical work is
detached from empirical considerations.
In this and in the following two chapters it will be shown that those
characterizations of Walrass philosophical orientation and his ideas
about scientic method are not tenable. The scholars who have made
them have neglected Bompaires advice, despite the fact that they have
repeated it by declaring that, in order to understand Walrass general
equilibrium model, everything he wrote must be taken into consider-
ation (for example, Bridel 1998, p. 232). The lesson of the history of
Cartesian heuristics is that scholars must examine received views crit-
ically in the light of the evidence. The application of that lesson to the
case of Walrass work results in shaking off the common but incorrect
notion that he did not appreciate the experimental side of theoretical
formulation and testing.
Walras and J. S. Mill
Mill was an enormously inuential philosopher and the preeminent
economist during the period of the 1840s until his death in 1873, and
P1: PIG
30 Walrass ideas
his ideas continued to be a part of Western intellectual tradition long
after that. In this and the following three chapters Walrass ideas will
sometimes be considered in connection with Mills, and the similarity
often striking of the language they used in discussing methodological
matters will sometimes be noted. This is not done to suggest that he
necessarily paraphrased Mill, or always agreed with him, or even that
he was familiar with all the passages in Mills work that are quoted. It is
done to show, by comparing the two scholars ideas, that Walras stud-
ied and subscribed to the major nineteenth-century strands of scientic
methodological opinion, which were well-represented in Mills writings.
It is done to showthat, in general respects, Walras thoroughly assimilated
and reected the mainstream philosophical and methodological thought
of his time.
3
It is also done to show that, even if Walras drew heavily
upon French traditions in those regards, he and Mill had many points
of contact with, and in common with, the scholarship of a wide interna-
tional intellectual community. Mill was certainly part of that community.
He had a residence in France, was uent in French, and was thoroughly
familiar with French philosophical traditions.
Walrass statements that the sciences use deduction and that they
re-enter the experiential realm. . . not to conrm but to apply their con-
clusions (106, 1874, p. 32; 30, p. 53; emphasis added) were not novel.
They were similar to views that had been expressed, in similar words, by
Mill in his inuential essay on economic methodology (Mill 1836/1844),
and it is likely that Walras was aware of that. Mill contended:
By the method ` a priori we mean (what has commonly been meant) reasoning
from an assumed hypothesis; which is not a practice conned to mathematics,
but is of the essence of all science which admits of general reasoning at all. To
verify the hypothesis itself ` a posteriori, that is, to examine whether the facts of
any actual case are in accordance with it, is no part of the business of science at
all, but of the application of science (ibid., p. 143).
In the same way that proponents of the archrationalism thesis character-
ize Walrass sentence as archrationalism, so also have those sentences of
Mills been alleged to enunciate the apparent position that the basic eco-
nomic theory is impervious to predictive failure (Hollander and Peart
1999, p. 369). In the light of the passage, Samuel Hollander and Sandra
Peart remark, it is scarcely surprising to nd it commonly alleged that
3
The truth of this assertion is not obviated by Walrass deprecating comments about
Mill as a philosopher and as an economist, reminiscent of W. S. Jevonss strictures.
Incidentally, those comments did not prevent Walras from invoking Mills authority
to support his own views when he found it advantageous to do so (for example, 239,
1891, 2, p. 434).
P1: PIG
Mill denied to verication the role of testing, and conceivably of modi-
fying theory . . . (ibid., p. 371).
4
Hollander and Peart disagree with that allegation and proceed to
demonstrate that Mill made clear allowances in both principle and
practice for the improvement of theory in the light of empirical evidence
(ibid., p. 395), as will be shown to be true also of Walras. Hollander and
Peart do not, however, do justice to their thesis, for their characteriza-
tion of Mills statement is an evident misreading of it. It is perfectly
understandable why he contended in that particular passage that con-
rmation of conclusions is not part of science. It was because there,
like Walras in his sentence, Mill was dening science as a process of
deduction and, at the time of writing that particular passage, he did not
regard testing as part of a deductive process. Nevertheless, he did not
say that the validity of theories is not tested by empirical evidence. On
the contrary, in that passage he said it is tested, and in that way, but he
called that process applications, and, in truth, the only way theories are
tested is by applying them. The applications reconnect the theory with
reality and often produce results that necessitate the modication of its
assumptions.
There are numerous writings in which Mill insisted on the necessity
of verication of theories, writings that are very similar to ones penned
by Walras. For example, Mill contended regarding social science that:
The conclusions of theory cannot be trusted, unless conrmed by observation; nor
those of observation, unless they can be afliated to theory, by deducing themfrom
the laws of human nature, and from a close analysis of the circumstances of the
particular situation. It is the accordance of these two kinds of evidence separately
taken the coincidence of ` a priori reasoning and specic experience which
forms the only sufcient ground for the principles of any science so immersed in
matter, dealing with such complex and concrete phenomena. (Mill 19811991,
8, p. 874)
In the same vein, he argued regarding all sciences that:
We cannot, therefore, too carefully endeavor to verify our theory, by comparing,
in the particular cases to which we have access, the results which it would have
led us to predict, with the most trustworthy accounts we can obtain of those which
have been actually realized. The discrepancy between our anticipations and the
actual fact is often the only circumstance which would have drawn our attention
to some important disturbing cause which we had overlooked. (ibid., 4, p. 332)
4
Similarly, Mark Blaug interprets Mill as believing that We never test the validity of
theories, because the conclusions are true as one aspect of human behavior, by virtue
of the assumptions, which in turn are true by virtue of being based on self-evident
facts of human experience (Blaug 1992, p. 68).
P1: PIG
32 Walrass ideas
Mill even dropped his distinction between deduction as science and
verication as part of applications, and dened deduction as including
not only induction but also verication. Asserting that deduction is a
mode of investigation, he argued that it consists of three operations:
the rst, one of direct induction; the second, of ratiocination; the third,
of verication (Mill 19811991, 7, p. 454; emphasis added).
Those phrases of Mills the combination of reason and experience,
the comparisonof theories withfacts are leitmotifs inWalrass writings.
Abundant quotations in this book will show that, just like Mill, Walras
contradicted his own statement about sciences not conrming theories
experientially, bringing his denition of science into accordance with
Mills mature views on the deductive mode of investigation. Indeed,
Walrass statement that, after the deductive phase of inquiry, the physi-
cal sciences properly speaking reconnect with experiential reality not
to conrm but to apply their conclusions (106, 1874, p. 32; 30, p. 53)
does not make sense. Walras was perfectly aware that to apply or attempt
to apply conclusions is to conrm or refute them, and he knew, as his
statement indicates, that they are applied or revealed to be inapplicable
by connecting them with reality. The statement was also totally incon-
sistent with his assertions that the physical sciences are experimental.
He knewand afrmed many times, as will be seen, that testing the valid-
ity of conclusions is standard scientic method, that in sciences such as
physics, biology, and chemistry, experiments are devised to determine
if assumptions, conclusions, and predictions are valid. Like Mill in his
mature work, Walras did not exclude empirical research observation,
experimental procedures in the case of some sciences, and induction
frompure science; on the contrary, he argued that it is an essential part
of it. Remarking on the obvious, he noted as an example that to design
machinery using the principles of pure mechanics is not pure science
(114, July 1879, pp. 1516), but he asserted that empirical research to
establish the principles of a pure science and to evaluate its conclusions
is scientic activity, as will be seen.
III. Concrete particular facts versus essences
Regarding all intellectual activity, Walras insisted on the necessity of
clearly distinguishing positive and normative studies. With time, and
thanks to sustained efforts, the human intellect determines at the same
time, with ever increasing certainty, in the order of natural facts, what
is; and, in the order of ethical facts, what ought to be (21, 1861, in
Walras 249.VII, p. 143). When he discussed a theory of science, he had
P1: PIG
reference to what is, to a positive body of knowledge. He contended,
furthermore, that a theory of science in general is indispensable for
constructing social science theory, economic theory, or the theory of any
other particular science (9, 1860, p. 197, n. 1). Accordingly, the right
and the duty of the economist is, before all and with care, to construct
the philosophy of science (106, 1874, p. 10; 8, p. 32). He maintained
that science is both experimental and rational, so a philosophy of science
must be both experimental and rational (9, 1860, p. 197, n. 1; 205,
1898, p. 453).
To give proper weight to the role of the experimental component
in Walrass view of science and its place in his modeling, his treat-
ment of facts must be examined. He has been portrayed by some writers
as dening facts in a fashion that is unlike the ordinary perception of
the meaning of that term. They contend that he adhered to a rational-
ist epistemology, which they contrast with an empiricist epistemology,
and which they allege indicates that he did not believe that hypotheses
should be based on facts. According to Pascal Bridel, Walras does not
accumulate details in order to build up realistic hypotheses: for him,
ideal types are purely abstract phenomena (Bridel 1997, p. 142). Thus
some writers contend that he did not intend that the components of his
model have referents in une r ealit e sensible and that it did not (Lendjel
1997, p. 81; and see Koppl 1992, Rebeyrol 1999, Tatti 2000, Huck 2001,
Bridel and Huck 2002). When L. Walras employs the word fact, it is
not in the ordinary sense. He draws a contrast between bodies, which
are material and changeable, and facts, which are the substance of
things in its ideal permanence (Dock` es 1996, 2526). For those writers
it therefore appears that when Walras referred to facts, such as com-
petition and exchange, he did not, to use Lendjels words, mean real
facts but genres, species, or essences (Lendjel 1997, p. 75; see
Chapter 2). The degree of representativeness of that view of Walrass
use of the word fact will now be examined.
It is true that Walras stated that in the world there are two orders of
real manifestations of matter: bodies and phenomena, or, if one wishes,
beings and facts (9, 1860, p. 196; and see 1874, p. 18; 16, p. 39), and
that bodies are temporary; facts endure (108, 1874, p. 18; 16, p. 39).
For bodies, the word beings can be substituted, and for phenom-
ena, the word facts can be substituted. In those passages, he declared
that bodies are the same thing as beings and are temporary, and that
phenomena are the same thing as facts and endure. Nevertheless, he
frequently did not employ those words in those senses. For example, in
extensive writings penned a few months before the 1860 passage just
P1: PIG
34 Walrass ideas
quoted, Walras presented very different denitions of them. He wrote:
I call bodies every instance that is limited in extent.
I call properties the characteristics of a body.
I call phenomena every instance that is limited in duration.
I call circumstances the characteristics of a phenomenon. (4, 1859, p. 30)
Therefore phenomena, that is, facts, are not enduring. Having thus
dened his terms, Walras explained that both bodies (beings) and phe-
nomena (facts) are real, concrete, and specic:
Reality is thus shared by two classes of real and concrete individual instances,
1st bodies possessing properties: a certain block of marble that is there on my
table, having certain dimensions, color . . . a certain oak tree in my courtyard,
of a certain height, having a certain number of branches, planted at such-and-
such a date. . . . A certain man that, at this moment, goes by in the street, with
such-and-such a name. . . . 2nd phenomena accompanied by circumstances: a
certain fall that happened to an object, lasting a certain number of seconds. . . . A
certain desire that I heard expressed the other day by such-and-such a person that
I know. . . . The slavery of blacks in certain provinces of the south of the United
States of America. (4, 1859, p. 31)
The distinction Walras made in those passages is therefore not between
bodies or beings that are temporary and phenomena or facts that
endure. Instead, quite the reverse: Bodies or beings are concrete par-
ticular things or persons; phenomena or facts are events, emotions, and
institutions, which are transitory and many of which are even ephemeral.
Walras went on repeatedly to make it clear that he perceived indi-
vidual facts to be exactly the same as most scholars and non-academics
consider them to be today, and used the unqualied word fact in the
ordinary modern sense; that is, to refer to particular concrete real things,
persons, processes, events, and situations. He did so when he criticized
those who do not know how to distinguish between an idea and a fact,
between lid eal et la r ealit e (51, no. 6, December 31, 1866, col. 183;
in 194, 1896, p. 15). He did so when he argued that if it is necessary and
sufcient, in order for a science to exist, that it rest upon a vast ensemble
of specic facts, then social science will exist (15, 1860, p. VIII; empha-
sis added). He did so when he described elementary phenomena. They
exist in reality and they are not ideas: The intellect that observes them
forms the idea of them and the ideas are expressed by words (4, 1859,
p. 8). He did so when he referred to all the individual and concrete facts
in society; that is to say, a social phenomenon occurring in reality that
could immediately be identied, attached to a cause equally individ-
ual and concrete . . . (15, 1860, p. IX). Those sentences provide another
example of Walrass practice of using the terms individual and concrete
facts and phenomena to mean the same thing, that is to say, material
P1: PIG
and changeable bodies, instead of saying, on the one hand, that bodies
are material and changeable, and, on the other, that facts or phenomena
are the substance of things in its ideal permanence.
Walrass habitual usages of terms also appear in his denitions of
natural facts and humanistic facts, the latter including industrial and
ethical facts. Natural facts have their origin in the blind and ineluctable
forces of nature. Humanistic facts have their source in the exercise of
the free will of humans. Humanistic industrial facts consist of relations
between persons and things so arranged as to subordinate the uses of
things to the objectives of persons (106, 1874, p. 18; p. 35, and 17, p.
39). Industrial techniques and economic production are complementary
examples of those types of facts (ibid., p. 38; 34, p. 61). Humanistic
ethical facts consist of relations of people with people (ibid., p. 13; p.
35). It will be noted that in that scheme of classication, Walras again
used the term facts to mean material and changeable bodies, such as
industrial techniques and economic production, as well as changeable
relations.
Walras further explained the nature of ordinary facts by noting the
difference between them and theories in this way: In good French, a
fact is one thing, and the theory of the fact, that is to say the study of
its nature, of its causes, of its consequences, and of its laws, is another
thing (108, 1874, p. 333). By a fact in that passage Walras did not
mean essences or some esoteric philosophical notion; he meant, in
good clear ordinary language, real things and events, as is indicated by
the example he used to illustrate the difference: In good French, to
identify and report facts, for example the facts of credit, to say what
credit operations are conducted today, to say how credit was organized
by the Jews, by the Phoenicians, the Carthaginians, in the Middle Ages,
is to undertake statistical studies; it is to establish the history of credit, it
is to produce a work of scholarship (ibid.). Those cannot be considered
examples of an idealist notion of the meaning of the word facts. There
are hundreds of other examples in his journalistic writings, in his studies
of real economic problems and policies, and in his theoretical writings
of his treating the things, persons, relationships, events, and processes of
the real world as facts in exactly the same way as academics and non-
academics alike do today.
IV. General facts
Walras also dened what he called general facts, and did so in just
the same way as his predecessors did and as is done today in scientic
contexts. John F. W. Herschel, for example, a scientist who was very
P1: PIG
36 Walrass ideas
inuential during the period 1830 to 1860, explained that laws apply
not to individual facts, but to general facts: When we have amassed
a great store of such general facts, they become the objects of another
and higher species of classication, and are themselves included in laws
which, as they dispose of groups, not individuals, have a far superior
degree of generality (Herschel 1831, 9495, p. 102). Science,
Walras similarly explained, constructs the theory of general facts
or groups of particular facts; that is to say, it indicates their nature,
their causes, their consequences, formulating their laws (239, 1901, 3,
p. 162). A general fact is therefore not a metaphysical concept; it is the
termHerschel and Walras used for a general subject matter. Examples
of general facts, Walras wrote, are vegetation, a natural fact studied
by a natural science; property, an ethical fact studied by moral science;
and civilization, a historical fact studied by historical science (9, 1860,
p. 196; 15, 1860, p. 4). There is nothing metaphysical or purely concep-
tual about vegetation, property, or civilization.
Like Herschel and many others, Walras indicated how general facts
are obtained, thus presenting the standard scientic view of one facet
of the connection between theories and reality. General facts are, as the
adjective indicates, generalizations, built upon observations of individ-
ual facts: When dealing with a series of individual facts between which
there are resemblances and differences, the scientic intellect eliminates
all of the qualities that are specic to each of the facts, gathers together
the qualities that are common to all or to several of the facts, and forms a
species (9, 1860, p. 196; 15, 1860, p. 4). What Herschel called higher
species of classication were then identied by Walras: In operat-
ing on a certain number of species in the same way that the scientic
mind has already operated on a certain number of individual facts, a
genus is attained. And so forth, attaining progressively higher levels
of generality (ibid.). In the botanical realm, for example, progressively
more general facts are pin oak trees, oak trees, hard-wood trees, decid-
uous trees, trees, and vegetation. Facts, at different levels of generality,
become the bases of hypotheses (ibid., p. 197; 15, 1860, p. 5).
Thus, Herschel and Walras were describing an aspect of the process
of induction, made possible by means of observation and classication
of individual real facts. Walras afrmed that general facts established by
scientic methods are universal, permanent, and irreducible abstractions
of whichthe concrete cases are particular manifestations (9, 1860, p. 196;
15, 1860, p. 4). For example, he would have said dendrological science
identies deciduous trees. The trees are a fact and not a theory. Any par-
ticular deciduous tree has a nite life: Ultimately it disappears, but decid-
uous trees endure as a class, each particular instance exemplifying the
characteristics identied in the scientic description of the general fact.
P1: PIG
V. Pure sciences
Subject matter
As part of his effort to construct a philosophy of science, Walras under-
took a classication of scientic disciplines according to their sub-
ject matter and methods. He altered it from time to time, dening and
classifying sciences in somewhat different ways. All the numerous vari-
ations he presented
5
are not the concern of this chapter, which is rather
to provide a picture of Walrass general outlook to aid in understanding
his views on the methods of the sciences and their relation to reality.
The purpose of that picture, in turn, is to aid in understanding his view
of the nature, methods, and uses of economic science (Chapter 2).
Reminiscent of J. S. Mills statement that science takes cognizance
of a phenomenon, and endeavors to discover its law (Mill 19811991,
4, p. 312), Walras asserted that science is conducted, not with bod-
ies but with phenomena, for which bodies are the theater. Facts, laws,
relationships, these are the object of science (9, 1860, p. 196; and
see 1874, p. 18; 16, p. 39). In that early passage, Walras used the
words phenomena and facts to mean general facts. Thus any pure
science formulates theories about the general facts of the real world,
for example, theories of the properties and behavior common to all
vegetation, theories that explain how deciduous trees function, theories
about the properties common to all oak trees, theories about how they
have become differentiated into many varieties. Walras illustrated the
approach taken by scientic inquiry with his example of the treatment
of the individual real and concrete facts of credit. To reason about
these facts how credit was organized by the Jews, the Carthagini-
ans, and so on is to identify and consider the general fact of credit,
the aspects of credit that are common to all of its particular manifesta-
tions in different cultures at different times. To reason about these facts
is to move to the theoretical plane of discourse: to explain the nature
of credit, to enumerate the kinds of credit, to demonstrate the rules of
credit, that would be to construct the theory of credit, that would be to
produce a work of science . . . (108, 1874, p. 333). With the passage
of time, Walras increasingly and ultimately almost exclusively used the
unqualied words the facts to mean real concrete individual facts,
and had little or no occasion to mention general facts directly or by
implication.
Walras explained that a distinguishing characteristic of a pure natural
science is that it studies the natural facts mentioned above, the facts and
relationships that have their origin in the play of the ineluctable forces of
5
These have been given a detailed exposition by Jean-Pierre Potier (1994).
P1: PIG
38 Walrass ideas
nature (141, August 1879, p. 246). He stated in 1879 that the pure nat-
ural sciences include mathematics, physics, chemistry, natural history,
anthropology (ibid.), and during the period 1879 to 1896, he classied
mathematics and economic theory also as pure natural sciences (see the
references given in Potier 1994, p. 235; and see 249.XIII, p. 563). All
those sciences use abstractions drawn from reality and have conclusions
that are connected to it. In 1898, Walras repositioned mathematics in
his scheme by identifying it as a tool, and made some other changes
in his classication, arguing that the pure natural sciences result from
the application of mathematics to physics, chemistry, and vegetal and
animal physiology (205, 1898, p. 452). In Walrass terminology, an
abstract science therefore turns out to be a pure science that is a method
or technique that is applied to various subject matters that are likewise
called pure sciences.
There are also pure moral sciences. They study, via the identication
of the characteristics of general facts, the individual humanistic facts
mentioned above, the facts and relationships that have their source in
the exercise of the free will of men (141, August 1879, p. 246). In 1879,
Walras identied those sciences as the history of human activity in all
its forms: languages and literatures, religions, art, science, industry, cus-
toms, law, war, politics (ibid.). In an elaboration that Walras introduced
in 1898, he stated that the pure moral sciences result fromthe application
of cnonics, by which Walras meant the theory of social behavior (198,
1897, p. 1019), and economic theory to psychology, history, sociology,
geography, statistics (205, 1898, p. 452).
Methods
Walras contended in his early work that there are two species of
sciences, those that are a priori and those that are experimental
[dexp erience]. The former apply the principle of necessity or contra-
diction; the latter apply the principle of order and of causality (14,
1860, in 249.VII, p. 58; and see 15, 1860, p. XIII; see Chapter 2). If by
the rst parts of those sentences he meant that there are a priori sciences
that do not have empirical components and connections, that was an idea
that he subsequently rejected, as will be seen. Indeed, in the light of his
other statements on abstract sciences, he probably meant, like Mill, that
the method ` a priori is a mixed method of induction and ratiocina-
tion (Mill 1836/1844, p. 143). In any case, Walras retained the notion
that experimental sciences compare theories to reality. Following the
ideas of the American sociologist Franklin H. Giddings (1896, pp. 47
50), Walras manifested that notion in his discussion of three types of
P1: PIG
science, considered from the point of view of their relation to reality
namely concrete, abstract, and pure sciences (205, 1898, pp. 44951).
According to Walras, they all have empirical components. Among the
orders of facts is sensory experience, that is to say the ideas by which
we attain cognizance of the subject by the intermediation of the senses
(82, 1868, column 208, in 194, 1896, p. 105).
Concrete sciences deal with empirical data; they are experimental
and identify particular facts. As for the abstract sciences, Walras made
it clear by the following line of reasoning that he believed they also
have empirical components. Their abstract quality does not mean that
they are not founded on reality. On the contrary, Walras contended, they
are so founded because, in Giddingss words, the abstract sciences
are not abstractions from nothing. They are abstractions from concrete
phenomena (Giddings 1896, p. 50). As Walras put it:
It is certainthat the physico-mathematical sciences, like the mathematical sciences
properlyspeaking, leave the empirical realmas soonas theyhave drawntheir types
from it. From these real types, they abstract ideal types that they dene; and, on
the basis of these denitions, they build a prioria priori all the scaffolding of their
theorems and proofs (106, 1874, p. 32; 30, p. 53).
Walras was maintaining, like all scholars before him, that those sciences
draw their basic components from reality. Like Thomas Hobbes, whose
writings were in his library, and a host of other scholars, Walras was
arguing that after the facts have been ascertained, those sciences follow
the deductive method.
Turning to the third type of science, Walras asserted that each pure
science results from the application of an abstract science to a concrete
one, and consequently has empirical foundations. All pure science is
constructed fromthe point of viewof pure truth, both rational and exper-
imental (ibid., p. 453; emphasis added to the last two words). All the
pure sciences therefore use both empirical and rational methods, induc-
tion and deduction. The pure natural sciences deal with an objective
external reality, and are positive, observational, and experimental. The
natural world is scrutinized by the senses in order to make it the object
of pure natural science (205, 1898, p. 492). He identied one of the
orders of facts as being sensory experience, that is to say the ideas by
which we comprehend the subject by the intermediation of the senses
(82, 1868, column 208, in 194, 1896, p. 105). Science is organized so
as to maintain the physical, chemical, botanical and physiological facts
distinct and separate in order to study them experimentally (203, 1898,
p. 1; in 205, p. 491; emphasis added). According to Walras, writing dur-
ing the period 1893 to 1895, The constancy of moral laws attests to the
P1: PIG
40 Walrass ideas
existence of the absolute personal reality, just as the constancy of natural
laws attests to the existence of the absolute external reality (205, 1898,
p. 493).
The pure natural sciences use the method of induction and deduc-
tion (151, July 1879, p. 15). They result from the association, on the
most extensive possible scale, of deduction with induction, of reasoning
with experience (205, 1898, p. 468). Reasoning was, for Walras,
somewhat broader than deduction, because reasoning includes not
only logical thinking to formconclusions and inferences but also the for-
mulation of concepts and assumptions. Induction is the use of real facts,
known by experiencing them, to form generalizations, and the method
of induction and deduction . . . is the true scientic method (151, July
1879, p. 26). Induction, the new method, has replaced speculation
(ibid., p. 28), and the deductive aspect of reasoning is based on facts.
Deduction therefore involves induction. Moreover, induction involves
deduction:
Perhaps it is well for it to be noted in passing that the method called induction
is both induction and deduction; there is no science that does not have recourse
to reasoning as well as to observation and experience to establish the relations
and the laws of the facts with which it is concerned. But, that having been said,
it is incontestable that there has been a marvelous development of the physical
and natural sciences and consequently of industry since the adoption of the new
method (141, July 1879, p. 28).
The process of empirical investigation and discovery
Walras expressed his rejection of archrationalism in his denition of
facts, in his treatment of the nature and methods of concrete, abstract,
pure, and applied science, and in the methods that he used in his own
work. He also manifested that rejection by drawing important distinc-
tions between, rst, what happens during the process of scientic discov-
ery, of nding real facts, forming abstractions, and identifying general
facts; second, the formulation of assumptions and hypotheses; third,
the subsequent deductive procedure; and fourth, the process of testing
assumptions and conclusions.
According to Walras, Facts occur before ideas and principles are
formed. Practice comes before theory, and above all, before there is a
nished, perfect, denitive theory (ibid.). Walras there expressed three
notions. First, individual facts are not concepts or ideas; they are not
essences; they are not scientic creations. They are generated and
experienced in the course of daily life. Second, facts precede theo-
ries. Third, theories are the results of a formative process; they do not
P1: PIG
spring full grown from the mind of their creator. As the author of sev-
eral sequentially improved theoretical tentatives (16, 1860; 91, 1869;
95, 1871), of many other pages of scribbled attempts that preceded the
El ements (106, 123), of emendations of the latter undertaken even when
it was in the stage of proofs, and of the extensive revisions in the sec-
ond edition, Walras was well aware of that. Thus, when he declared that
deduction the rational aspect of theorizing is used in working out the
structure and conclusions of a science (106, 1874, p, 32; 30, p. 53),
he was not describing the process of investigation and discovery, that is,
the process of identifying facts, striving to arrive at satisfactory abstrac-
tions, and formulating assumptions. When he referred to deduction, he
meant the interpretation of the inductively obtained hypothesis, the pro-
cedure of drawing its consequences and implications, thus obtaining the
temporarily nished structure of a theory or model. To Walras, deduc-
tive describes the form and logical characteristics of the interpretation.
In the journalistic, experimental, creative, and mature phases of his
career, Walras was opposed in principle to making assumptions that are
contrary to fact, which he regarded as a matter that is different frommak-
ing simplifying assumptions. Reinforcing the idea that the components
of theories must be drawn from reality, Walras explained that:
In effect, we assume:
1st that the world of ideas and of the ideal are the proper object and the true eld
of theory and of science. (The following rst being understood, namely that no
ideas and no ideal situation are admissible in theories and in science except on
the condition that they are drawn, by understanding and reasoning, from the facts
and the reality that experience furnishes) (51, December 31, 1866, column 180,
and in 194, 1896, p. 10).
Of course, by the world of ideas and of the ideal Walras did not mean
the best possible or the situation that he would like to see brought into
being.
6
He meant the theoretical plane of positive scientic discourse
about the reality that actually exists. In those contexts, his idealism
consisted of following the scientic method of the use of abstractions. It
can be confusing that, in contrast, he sometimes used the word ideal
to mean a state of affairs that he liked, as will be seen.
Walras divulged more about the stage of empirical investigation and
discovery in his consideration of how science studies general facts. In
this connection, he left no doubt as to his philosophical position: Real-
ity is the totality of everything that exists. Reality is a fact: it is the fact of
6
Nor, when Walras used the words ideal and idealism was he espousing the opinion
that the objects of external perception, in their intrinsic nature or as perceived, consist
of ideas, the doctrine called Platonic idealism.
P1: PIG
42 Walrass ideas
Existence. The fact of existence is a general fact which is manifested in
all places and at all times, which is universal and permanent (4, 1859,
p. 6). The science that studies it is ontology. In order to develop that
or any science, general facts must be studied, and, It seems to me,
he wrote, that to construct logically the theory of a general, universal,
permanent fact, it is simply necessary to pose and answer the following
ve questions regarding that fact (ibid.). In Walrass words, these
are:
1st What is the nature of the fact?
2nd What are its origins? In other words: what is its cause?
3rd Into how many principal species is it divided?
4th What are the laws that it follows, either in its most general form, or in its
principal species?
5th What consequences does it entail? In other words, what are its effects? (4,
1860, p. 6; and in 9, 1860, p. 196; and in 15, 1860, p. 5).
Walras went on to state that to each of these ve questions there are
corresponding methodical procedures that lead to their solutions. All
these questions being answered, the science is constructed; the theory
of the general fact is achieved . . . (9, 1860, p. 197; 15, 1860, p. 5).
Use of individual facts as the basis of assumptions
and to evaluate conclusions
What, Walras asked, are the principal methodical procedures that lead
to the solution of the questions that have been posed? He answered
that they are observation, experiments, inductions, hypotheses (ibid.).
For the second of these Walras wrote the French word exp erience. In
some contexts in his writings that word has the same meaning as the
word experience that appears in English sentences like, All knowl-
edge comes from experience, or Visiting the Grand Canyon is a won-
derful experience. Exp erience does not, however, have that meaning in
Walrass methodological writings. It is evident that, because he preceded
that word with observation, he did not redundantly mean experience
in the sense of obtaining knowledge by observation. He meant knowl-
edge obtained by the active involvement of the researcher in working
with the empirical subject matter, and that is best translated, in this case,
by experiments. Elsewhere Walras was more explicit and described
sciences as exp erimentales, using that word in the same sense as the
English experimental to refer to sciences like chemistry or physics that
conduct laboratory-style experiments. He was referring to such sciences
when he asserted that pure natural science studies physical, vegetal,
chemical, and physiological facts experimentally (203, 1898, p. 1; in
P1: PIG
205, p. 491). He was using the word experimental to mean founded
on and tested by empirical data.
According to Walrass statement, the procedures of observation,
experimentation, and induction are the methods by which the hypotheses
of a theory and, if it has been constructed logically, therefore its conclu-
sions, are made consonant with reality, and they are also the methods by
which the theory is tested. The methods involve the interaction of a the-
ory with the facts with which it deals, in such fashion that, in the course
of developing it, its assumptions and conclusions are repeatedly tested
by observations and experiments, and its assumptions altered if neces-
sary in the light of the empirical information. The closer the assumptions
and conclusions are to observations and experience, the more reliable
they are.
It is clear that Walras did not subscribe to what is called today the
hypothetico-deductive method.
7
A brief common denition that bears
on this subject distinguishes, Experimental science, which uses scien-
tic experiments (opposed to abstract science, hypothetico-deductive)
(Robert 2002, p. 1001). In its pure form, the latter method does not rely
upon induction to arrive at a hypothesis; instead, a hypothesis is made
creatively, perhaps as a hunch or an intuition. It is used to deduce var-
ious possible facts and is then tested by examining empirical evidence
to see whether those facts obtain in reality. If the predicted facts are
actually observed, then the hypothesis is at least tentatively conrmed.
That method is therefore one of conjectures that are conrmed or refuted
by empirical evidence. Incidentally, there is no discussion here of issues
such as observations being consistent with false hypotheses, the non-
conrmation of inductions by subsequently observed particular facts,
the methodological aspects of verifying hypotheses about unobservable
phenomena, and so forth, because Walras did not consider them.
The foregoing citations of Walrass writings have shown that, in
contrast to hypothetico-deductive methodology, he believed hypotheses
should be based on inductions, and thus on facts carefully accumulated
by observation and experimentation. He emphasized the experimental
foundations of the hypotheses made in abstract sciences: It is perhaps
the most delicate point, in the physico-mathematic sciences, to borrow
in this way from reality the experimental data on which the intellect
then establishes the series of rational deductions (103, 1874, p. 10).
7
In the Bacon-Mill tradition, induction was assumed to be both a method of discovery
and a method of proof of scientic laws. Today the doctrine that scientic hypotheses
are discovered by induction is known as inductivism. This view is opposed by the
hypothetico-deductive (HD) conception of science: scientic statements and theories
are free creations, hypotheses, that are tested by deducing empirical propositions from
them (Niiniluoto 1998, p. 248, emphasis added; and see Hands 1998, p. 376).
P1: PIG
44 Walrass ideas
The theory or model into which the hypotheses are incorporated is elab-
orated as a deductive structure, a process that is not experimental. The
nal outcomes of the structure are deductions that are predictions of pos-
sible facts. If further observation and experimentation determine that
they materialize in reality, Walras regarded them as conrming it. To
illustrate this, he gave the example of the planet Neptune. Observations
detected irregularities of the movement of Uranus. The hypothesis of the
existence of Neptune was based on those observations because Newtons
system required the existence of that planet to explain the irregularities.
That hypothesis could not be veried at the time it was made. What
completed the facts by nding the planet Neptune? Reasoning without
experience (239, 1893, 2, p. 573, n. 8). Walras did not really mean
nding, which would have contradicted reasoning without expe-
rience, because nding something is to experience it. He meant
hypothesizing the existence of the planet on the basis of the factual evi-
dence of the irregularities. Subsequently, the hypothesis was conrmed
experientially to be accurate by nding the planet. In fact, it is proved
by experience, and therefore, it is true for everyone, and for everyone to
the same degree, that each planet moves around the sun in a plane orbit
of which the vector rays make equal areas in equal lengths of time (85,
1868, column 274, in 194, 1896, pp. 14849). The reasoning to which
Walras referred was not the use of innate knowledge nor a hunch; it was
the deduction of the implications of the Newtonian model of the uni-
verse, which was ultimately based on observations and the other aspects
of the process of construction of theories that Walras described in the
ways indicated earlier in this chapter.
Reafrming that theories about general subject matters must be tested
by reference to individual facts, and thereby contradicting once more
his own statement about the non-conrmation of hypotheses and con-
clusions quoted at the beginning of this chapter, Walras noted that in
opposition to erroneous hypotheses in the physical and natural sci-
ences, simple everyday facts are provided by experience in the physi-
cal and natural realms and are used to refute them (jotting by Walras,
in Potier 1994, pp. 25354; emphasis added). Simple everyday facts,
he thought, are not concepts, genres or essences; they are individual
facts that exist in the r ealit e sensible. Theories are also refuted by
the facts identied by historical studies, although the evidence is dif-
cult to obtain: How much more difcult it is in regard to historical
studies to have recourse to reality against science! What are the facts in
that eld of study that are tortured by false hypotheses? (ibid., p. 254).
Thus hypotheses should be true, reective of the facts in the relevant
domain. Walras afrmed his belief that without a doubt, [reasoning]
P1: PIG
and science cannot be in contradiction with experience and facts (239,
1893, 2, p. 573, n. 8). In the same way, atly contradicting yet again the
archrationalism thesis, Walras declared that science uses facts to nd
laws, expressing themwith equations: It is in this way that the physico-
mathematical sciences formulate rationally laws that are conrmed by
experience (239, 1901, 3, p. 162; emphasis added).
VI. Applied sciences
If it were true that Walras did not believe that models can be compared
with empirical information, he ought to have thought that they cannot be
used for policy applications, because such uses would require the models
to be compared to reality. According to Walras, however, applications
of the theoretical conclusions of sciences are indeed made in order to
understand real particular situations, phenomena, and problems and to
guide the formulation of actions and policies. Changes suggested by a
policy must be applied to real conditions, and determining whether the
policy is efcacious requires comparing theoretical conclusions with
its real outcomes. Walras had the same opinion as Mill, who pointed
out that if a science is not to be useless, practical rules must be capa-
ble of being founded on it, and an art would not be an art, unless
it were founded upon a scientic knowledge of the properties of the
subject-matter . . . (Mill 1836/1844, p. 124). Walras likewise argued
that a theory can be used in the formulation of policies designed to alter
real situations only if it has correctly identied their fundamental fea-
tures and structural characteristics (239, 1874, 1, p. 443; 141, July 1879,
pp. 1522). Pure science, he wrote, is thereby enabled to be the basis of
applied science and practice. He remarked that it is only foolish people
who need results in order to understand the signicance of pure science.
Any scientist worthy of that name saw the beauty of Newtons system
without thinking that it would be of use in helping to navigate a boat in
the sea (239, 2, 1893, p. 573, n. 8). Nevertheless, The physical and
natural sciences have industry as their ultimate goal (141, July 1879,
p. 16).
Accordingly, Walras identied applied science as another aspect of
all disciplines. Agreeing with Mills afrmation that science is a col-
lection of truths; art, a body of rules, or directions for conduct (Mill
1836/1844, p. 124), Walras used the same language to describe applied
science, writing that one of its branches is applied natural science, which
furnishes rules for the conduct of persons with respect to impersonal
things (141, August 1879, p. 246). It is both rational and experimental
(205, 1898, p. 453). It considers what ought to be from the viewpoint
P1: PIG
46 Walrass ideas
of expediency, usefulness, and material well-being (ibid.). Examples
are agricultural technology, industrial technology, and medicine (141,
August 1879, p. 246).
The other branch is applied ethical science, which furnishes rules
for the conduct of persons with respect to each other (ibid.). It con-
siders what ought to be from the viewpoint of what is equitable and
just in relation to the phenomena that result from the exercise of human
will (106, 1874, pp. 1617; 15, pp. 3839). Examples are individ-
ual and social ethics (141, August 1879, p. 246). Following the teach-
ings of many scholars, including Victor Cousin, whose ideas in turn
reected the inuence of Immanuel Kant (see Rebeyrol 1999, pp. 14
17), Walras believed that, like the rules of applied science, propositions
regarding ethics, or what ought to be, should be founded on a scien-
tic basis; that is, on the basis of a theory or model that accurately
reects the relevant aspects of reality. He probably also read that notion
in the work of a philosopher in many respects different from Kant,
namely John Locke (Redman 1997, pp. 6869; see Chapter 4 of the
present book). Walras contended that applied ethical science is exclu-
sively rational (205, 1898, p. 453), but he also declared that, as is
true of other sciences, individual facts are essential for it: It is easy
to recognize that the method of deduction, when exercised on a con-
siderable accumulation of facts gathered beforehand by observation of
humanity and society, gives birth to the nest and most fruitful social
ethics, but, when it operates in a vacuum, produces only completely
vain and ridiculous utopias (141, August 1879, p. 238). That would
appear tobe contradictorywere it not for Walrass convictionthat rational
thought contains empirical components. In short, the operations of erect-
ing the nished structure of applied ethical science are deductive, but its
assumptions are based on empirical observations of society and human
nature.
Finally, Walras identied industry the actual practice in everyday
life of the rules of usefulness, practicability, and convenience furnished
by applied natural science as distinct from the study of them; and
mores the actual practice in everyday life of the principles and rules
of applied moral science as distinct from the study of them (205,
pp. 453, 458). In Walrass lexicon, mores are the manner of functioning
of the institutions of property, the family, and government.
Walras deduced far-reaching consequences from his classication of
studies, as, for example, in his portrayal of the situation of German
scholars in the nineteenth century. They distinguished between only the
theoretical side of political science and the practical one. That was a
consequence, he believed, of their lack of an adequate philosophy of
P1: PIG
science. They would otherwise have been led to the distinction between
pure science and applied science, and the distinction between natural
science and moral science (141, August 1879, pp. 22930). That would
have greatly strengthened their theoretical work. They would also have
devised reasonable practical arrangements, such as appropriate admin-
istrative placement of the programs in German universities, avoiding
incongruities like the teaching of both constitutional policy and the pro-
cesses of animal births in the division of administrative and nancial
sciences (ibid., p. 245).
VII. The real and the ideal
Real and ideal types
The notion of real and ideal types is a recurrent theme in Walrass
methodological writings and it has an important place in his economic
modeling. Real types, according to Walras, are phenomena identied by
scientists who study concrete reality. The real types must be properly
identied and accurately described so as to give a true representation of
it. Ideal types are abstracted from real types and are dened neglecting
perturbations, transitory conditions, and unimportant variations found
in particular cases (123, 1877, p. 267; 222, pp. 33435). They are
therefore free of the idiosyncrasies of specic empirical instances of a
phenomenon (106, 1874, p. 32; 30, p. 53). Walrass use of the term
ideal types is an instance of his use of philosophical terminology in
a way that differed from traditional meanings. His ideal types are,
of course, actually nothing other than scientic abstractions used, he
believed, as the basis of chains of deductive reasoning.
Similarly, Walras chose to use the words perfect, permanent,
and universal in reference to scientic abstractions. That was in the
terminological tradition of scholars who, like Descartes and Locke, used
those adjectives in reference to geometry. What Walras meant, however,
is the same as what a modern scientist means when he says simply
that he has formulated an abstraction. Geometrical forms, he stated,
are ideal types; and those ideal types are the proper object of pure
geometry. (I understand well that there are no ideal types that are
admissible, in social science as in pure geometry, except those drawn by
the understanding from the real types that experience furnishes.) (51,
December 31, 1866, columns 18182, and in 194, 1896, p. 12). Thus
when Walras stated that an ideal type is perfect, he was not really saying
anything more than that it has the properties that the scientist denes it
to have, whereas the more or less similar phenomena found in reality
P1: PIG
48 Walrass ideas
do not have exactly those properties, or have those properties but also
have unimportant additional features. He gave the following examples to
illustrate this obvious matter (106, 1874, p. 32; 30, p. 53). The concept
of a circle as dened in plane geometry is perfectly round, will exist
as long as humanity has concepts, and is the same in all cultures and
epochs in which circles are contemplated. Empirical phenomena that are
circular, however, are approximately round, each particular case being
awed or deviating in some respect from the form of a perfect circle and
being more or less transitory. In any empirical circumference, Walras
continued, the radii are not all exactly equal to each other. Likewise,
in any actual triangle drawn by a human, the sum of the angles is not
exactly equal to the sum of two right angles. The exact properties of the
geometric forms therefore cannot be proven by empirical measurements
of their similar real counterparts. Walras thus emphasized in that passage
as in others that there are propositions that are developed on the basis
of pure reasoning that cannot be veried by experience in the sense of
empirical measurements and experiments, but that are logical according
to the rules of the system of logic. In those reections, Walras was
arguing along the same lines as J. S. Mill, who had observed that the
conclusions of geometry are not strictly true of such lines, angles, and
gures, as human hands can construct. But no one, therefore, contends
that the conclusions of geometry are of no utility. . . . No mathematician
ever thought that his denition of a line corresponded to an actual line
(Mill 1836/1844, p. 145).
The history of civilizations shows that Walras was not accurate in
stating that real geometrical types are easily obtained from observation
of reality furnished immediately by experience (51, December 31,
1866, column 182; and in 194, 1896, p. 13). Nor was he accurate in
contending that ideal geometrical types are relatively easily obtained, as
compared with ideal types in the social sciences, by an immediate a
posteriori synthesis of experience (ibid.). It took many centuries for the
development of constructions such as cycloids, epicycloids, roulettes,
deltoids, astroids, involutes, logarithmic and parabolic spirals, not to
mention three-dimensional forms such as hyperboloids of one sheet,
hyperbolic paraboloids and quadratic cylinders; most societies did not
develop them at all. Furthermore, those forms did not, for the most part,
have real counterparts in nature but were rather internal developments
in the elds of geometry and related mathematics. In Walrass defense,
however, it should be noted that he realized, as indicated by his listing of
great scientic thinkers of the past, that scientic bodies of knowledge
like mechanics and astronomy have required many years of the labor of
many scientists (210, 1900, p. XX; p. 22).
P1: PIG
Walrass view of geometrical forms also illustrates his thesis that
abstract and pure sciences are connected to reality not only by their
basing ideal types on real types, but also in regard to the empirical con-
rmation of their conclusions. In 1874, he wrote regarding those forms
that reality does not conrm these denitions and demonstrations; it
permits only a rich application of them (106, 1874, p. 32). That sen-
tence would appear to provide some evidence that the archrationalism
thesis may have some validity in reference to Walrass view of mathe-
matics, but in fact it did not represent his considered conclusion about
even that specic aspect of science. Regarding the geometrical de-
nitions, he emphasized that they must be based on real types drawn
from reality (51, December 31, 1866, column 182; and in 194, 1896,
p. 13), as has just been seen. Regarding both denitions and demonstra-
tions, he rewrote his 1874 sentence sometime during the period 1896
to 1899 to reect his rejection of the opinion that reality does not con-
rm them: It is only approximately, he asserted, that reality conrms
these denitions and demonstrations; but it permits a very rich applica-
tion of them (210, 1900, p. 30; p. 53; emphasis added). Thus reality
does conrm them if they are true, although it does so approximately;
the exact properties of geometric forms, for example, are conrmed
approximately by empirical forms. If that were not the case, the geo-
metrical or other concept or theory would not be applicable. It would
have to remain part of a contrived logical structure, a sort of intellec-
tual game, or it would have to be reformulated. Indeed, Walras allowed,
approximate conrmation is all that empirical tests can ever provide
for theories (see 239, 3, p. 183, n. 2; Chapter 2). Part of his wording,
however, was not well chosen, inasmuch as he referred to approximate
conrmation and application in the foregoing quotation as though the
degree of fruitfulness of the denitions and demonstrations is somewhat
independent of the conrmations. In fact, he frequently expressed the
opinion that numerous, diverse, and successful what he called rich
applications of the denitions and demonstrations are conrmations of
them.
Walras added that some physical functions, unlike geometric forms,
can be deduced from scientic knowledge without immediate recourse
to empirical information, but experience conrms them. Regarding a
function f(t) expressing velocity, Thanks to the mathematical method,
we can arrive at it by pure reasoning, without the help of experience,
which subsequently intervenes to conrm, if one wishes, but not to verify
the result (249.VII, p. 251) a distinction without a difference. All of
the dozens of other quotations in this book that bear on the issue make
it clear that in the contexts in which he employed them, Walras used
P1: PIG
50 Walrass ideas
the verbs to conrm and to verify as exact synonyms, as indeed
they are in similar contexts in ordinary and scholarly usage in English
and in French.
8
For example, he explained that since we are unable to
reviewall the social facts to nd in themalways the persistent agreement
of material interests and Justice, we are simply going to put before us
the fact of society and the principal social facts and, with respect to
each of them, verify the exactitude of our principle in several rapid lines
that will be easy to interpret (249.XIII, p. 175). Again, he declared
that consideration of a curve had the advantage of indicating a proper
method of experimentation to establish by direct experience the reality
of the law[of gravity], if one did not knowit, or of verifying it if one had
discovered it by another means (176, 1889, p. 15; in 249.VII, p. 251).
So, the law is veried by experimentation.
Similarly, Walras evaluated Malthuss theory by verifying it: an
increase in the quantity of capital goods properly speaking, one which
precedes and exceeds the increase in the number of persons, [occurs] in
order to compensate for the constancy of the amount of land. Here the
theory of Malthus on population and subsistence levels is veried (249.
XI, p. 500). Regarding his analysis of the different effects on the level of
rents of industries producing necessities and of luxury goods industries,
he remarked: Let an effort be made to verify seriously this assertion,
and one will be convinced of the essential difference in the economic
conditions in which the two varieties of industries in question operate
(12, 1860, in 249.VII, pp. 99100), and then went on to evaluate the
statistics that bear on the issue.
8
The dictionary translation of the rst meaning of v erier is to check, and the
second, which is the relevant one for the contexts of this book, is: (=conrmer,
prouver) [+afrmation, fait], to establish the truth of, to conrm(the truth of), to prove
to be true; [axiome] to establish ou conrm the truth of; [+t emoignage] to establish
the truth ou veracity (frm) of, to conrm, [+soupcons, conjecture] to bear out, to
conrm; [+hypoth` ese, th eorie] to bear out, to conrm, to prove (Collins Robert 2002,
p. 1070). In English, the rst or principal meaning of To conrm is 1. To support
or establish the certainty or validity of; verify (American Heritage Dictionary 1991,
p. 308). To verify is 1. To prove the truth of by the presentation of evidence
or testimony; substantiate. 2. To determine or test the truth or accuracy of, as by
comparison, investigation, or reference: conduct experiments to verify a hypothesis
(ibid., p. 1343; and see Robert 2002, pp. 509, 2756). Those are exactly the senses in
which Walras used to conrm, conrmation, and to verify. In fact, there are
few instances in which he used the word verify and still fewer in which he used
verication, and in some of those, the context was such that he meant to check,
as when he referred to dates at which he and another writer had published books,
and being uncertain, wrote (Verify these dates) (249.XIII, p. 536), and as when he
wrote V erier

El ements, 2
e
edition (ibid., p. 550), and wrote, regarding inventories
of goods, la v erication, le conditionnement, larrimage, la bonne conservation et
la r eexp edition de toutes sortes de marchandises (249.XII p. 621).
P1: PIG
Imagination
Walrass remarks about the imagination are not illuminating. He used
the words representations of the imagination to describe concepts on
several occasions, as for example in an undated jotting
9
regarding the
notion of perfect circles (249.XIII, p. 557), despite his identifying
perfect circles elsewhere as abstractions from real types, as has just
been seen. In 1868 (82, 1868, cols. 21112; and in 1886, pp. 11014),
he contended that perceptions of the imagination are not abstractions.
Being nothing more than collections of sensations or of feelings, any
attempt at abstraction made in regard to them would only end up in
annihilating them, without producing any idea (ibid., col. 212; 1886,
p. 113). In an article written ve years before his death, however, he
took the opposite position:
I will say only that [Cournot] greatly contributed to convincing me that all the
sciences (mathematical, physical, or other) are based on concepts (names, dia-
grams, matter, forces, etc.) without metaphysical signicance, which are nothing
other than representations of the imagination or mental syntheses, but which are
nonetheless the indispensable base on which these sciences erect their edice of
facts and of relationships that are very truly objective. (213, 1905, p. 1, column
1; and in 237, 1962, p. 61; emphasis added)
Compounding the confusion, in making those retrospective comments
Walras forgot that he had asserted that the bases of scientic construc-
tions are ideal types based upon real types that are identied as existing
in reality, not representations of the imagination. Perhaps, however, in
the foregoing quotation he meant no more than that the representations
are categories that are used to identify and classify the objective facts;
that is, are used to erect a scientic edice with those facts. He used
the word imagination in the ordinary sense when he attributed great
scientic discoveries and erroneous notions to its action (jotting, 1909,
249.XIII, p. 619), and when he declared that it is as necessary for the
scholar, the engineer, the economist, and the politician as for the artist
in order to have aspirations that transcend what exists at present (198,
1897 in 205, 1898, p. 271). He also used the word imagination in the
sense of Etienne Vacherot (1858), without dening it clearly, to indicate
a subjective sort of cognition that adds unity to what is perceived or that
aids in forming perceptions (90, 1868 in 249.IX, p. 98). Exterior and
inward personal perceptions constitute the domain of a rst intellectual
faculty, the imagination; let us investigate how these perceptions of the
9
This matter is further discussed in Chapter 4, where it is again noted that the jottings
are not adequate nor reliable support for the view that Walras believed concepts are
created by the imagination.
P1: PIG
52 Walrass ideas
imagination are formed in the mind by means of external and internal
experience (253, in 249.XII, p. 129). Information is transmitted through
the senses and hence processed by the brain, so what is experienced is
in part, a product of the imagination (205, 1898, p. 493). It has been
seen that a fundamental tenet of Walrass philosophical and method-
ological system was that ideal types are rmly based on real types that
are drawn precisely and scrupulously from real facts, so it was an aber-
ration for him to declare on two occasions that ideal types are products
of the imagination without precise correlation to real phenomena, albeit
related to them (see Chapter 3).
Idealism and realism
Walras described himself at different times as having different opinions
about idealism and realism. In this regard, William Jaff e was far from
making sweeping and shallowgeneralizations. He gave a carefully qual-
ied accurate description of Walrass relation to idealism. Jaff e would
not say that Walras was an idealist, allowing only that he leaned toward
it, and not for long, but only in his early youth (Jaff e 1980, p. 349, n.).
Nevertheless, the passage that Jaff e quoted to support that description
does not do so, and his misuse of it must be added to the list of mis-
interpretations of Walrass thought. In the passage in question, which
evidently dates from early in his career, Walras declared: I am an ide-
alist, but he went on to explain in his next sentence that by the word
idealist in that particular passage he meant that he was idealistic,
which is a very different matter. He was not using the word idealist in
its epistemological sense but in the sense that, like Jesus, he had noble
aspirations for the future of humanity despite their impracticality and the
consequent difculties in the way of their realization. He railed against
accepting facts as masters and as laws, and complained that empiricism
triumphant reigns supreme, by which he meant that the harsh realities
of the facts of the existing social and economic situation should not stie
hopes and dreams of changing it:
I am an idealist. I believe that ideas transform the world into their image and
that the ideals conceived by a man, by a school of thought, impose themselves
on human affairs. I believe that the world has taken eighteen centuries to try
without success to realize the ideals of Jesus and the rst Apostles. I believe that
the world will take another eighteen or perhaps twenty centuries to try, without
better success, to realize the ideals conceived by the men of 89 ideals seen
more clearly by us and claried by our successors. . . . It is possible that my ideal
is narrowly conceived. It is less so, however, than one would be led to believe
P1: PIG
by the imperfect rendition of it given by my words. However that may be, I take
refuge in my ideal. It is my sanctuary against the invasion of brute facts. . . .
10
Walras was there using the words idealism and ideals in the norma-
tive sense to mean a liking for a state of affairs that he wanted to see
come into existence.
Turning to the question of epistemological idealismat about the same
time as writing that passage, Walras stated, in explanation of his onto-
logical views, that he was neither an epistemological idealist nor a real-
ist. Epistemological idealism is not normative; it is a positive doctrine
regarding the nature of knowledge whether pleasant or distasteful,
desired or disliked and how it is obtained. He contrasted the posi-
tions, rst, of an idealist, second, of a realist, and, he wrote, thirdly (to
emphasize the distinction) of himself. He was, he explained, nothing,
for the moment, but a sincere man (4, 1859, p. 3). Walras argued, with
literary ourishes, that common sense should rule on ontological ques-
tions. You, Realists, you do not want to hear any discussion of Beauty,
nor to know what it is. But will you therefore deny the light of day?
The idea of Beauty is universally understood. You, on the other hand,
you Idealists, you are scarcely more sensible. Why be contemptuous of
bourgeois people and of Reality, and want to speak only of princesses
and viscounts? (ibid., p. 5).
Syntheses of doctrines and methods
Walras asserted that he wanted to create a reconciliation and synthesis of
rationalismand empiricismand believed that he had done so in both prin-
ciple and in practice. Scholars who believe that is not so have neglected
and denied the existence of his writings in favor of the synthesis. They
have argued that the idea that he wanted such a fusion, and consequently
that he believed that theories should be realistic and judged by their con-
sonance with reality, is a confusion of methodological and ontological
aspects of his thought (Lendjel 1997). They describe his epistemology
and his work as idealist (Koppl 1995; Dock` es 1999, p. 20; Tatti 2000,
p. 17), not recognizing the limited sense in which he stated that was
true nor noting the realist aspect of his thought. One such writer alleged
that if Walras entertained the idea of the synthesis at all, it was not to
effectuate it, but merely to glimpse that it may be possible. Moreover,
10
The jotting in Walrass handwriting is in my possession. It is published in 249.XIII,
p. 551. Another English translation is given in Jaff e 1980, p. 533, n. 14; reprinted in
Jaff e 1983, p. 349, n. 14.
P1: PIG
54 Walrass ideas
that scholar has argued, if Walras had done so, it would have been
from the perspective of a metaphysical synthesis of concepts and real-
ity rather than from that of a methodological synthesis concerned with
constructing and improving theories: If Walras glimpsed the possibility
of a fusion of concepts and reality, it is not so much from the perspec-
tive of verication but from a metaphysical perspective (Lendjel 1997,
p. 74).
In contrast to those notions, by following Bompaires advice it will
become evident that there is no confusion of Walrass methodology and
ontology in the contention that he wanted to achieve a synthesis or in
the other aspects of the exposition presented in this chapter. Walras
described himself in his creative phase as both an idealist and a realist,
asserting that he had achieved a synthesis of those two doctrines. That,
with a recognition of the domains to which he applied those words and
with proper interpretations of them, is a correct characterization of his
orientation during his creative and mature theoretical phases. In this
connection, it is important for an understanding of his philosophical and
methodological positions to note, as has been abundantly documented
above, that by his idealism he did not mean that he believed there is
knowledge that consists of innate ideas, conceived independently of
experience.
It is alsocrucial torecognize that his reference tohis idealist andrealist
views and to their synthesis was made with respect to the domains of
applied economics and policy. He declared: I have been very occupied
by the study of these questions and I have reached a resolution of the
question of idealismand of realismby a synthesis that seems to me more
and more felicitous and fertile as time passes. That synthesis rests on the
distinction between science and policy applications. I am an idealist in
regard to science; I am a realist in regard to applications (239, 1874, 1,
p. 444). He thenexplainedina similar passage that byscience he meant
applied science, and by policy applications, he meant practice. He
made this clear by stating that, in his system of thought, the distinction
between the Ideal and the Real did not relate to pure science contrasted
with the Real, but to the distinction between applied economic science
(the theory of the art of producing wealth) and policy applications (the
practice of that art) (239, 1877, 1, p. 542):
I call applied science the theory of an art, and consequently, applied economic
theory the theory of the art of producing wealth. With respect to the practice of an
art, it deals only with what I call policy; it is not therefore between pure science
and applied science but between applied economic theory and policy that the
distinction between the Ideal and the Real is to be made (ibid.).
P1: PIG
Reinforcing the point that pure science is not involved in that distinction,
Walras went on to say that with respect to my pure economic theory,
it studies purely and simply the fact of the determination of prices . . .
(ibid.). Thus, in the context of the foregoing remarks, when he referred
to the Ideal he certainly did not mean that pure science is or could be
a representation of what is thought to be good or desirable conditions
of the physical and social worlds. Applied science, however, does deal
with the Ideal in the sense of that which is desirable; applied science
is prescriptive; it formulates rules of conduct to achieve the goal of
efciency.
Walras also claimed to have made another type of synthesis, one
relating exclusively to the content of pure science. That was the syn-
thesis of rationalism and empiricism (239, 1874, 1, p. 374). His idea of
that synthesis was a central thrust and unifying principle of his orga-
nized system of philosophical and methodological thought, clearly and
steadily seen, articulated, and carried out by him. He afrmed that his
epistemological views and the pure scientic work that they shaped had
not only rational but also empirical components. As has been seen, his
fusion of rationalism and empiricism called for theories that are based
upon realistic foundations and that use deduction to reach realistic con-
clusions. Believing as he did in the value of reconciling differences of
approach and method and the possibility of doing so, Walras wanted his
epistemological position regarding the activity of engaging in pure sci-
entic research and its content to be viewed as effectuating that fusion.
He rejected again the idealism that maintains there are innate ideas:
When the idealism of Hegel came on the scene with the aim of creat-
ing the physical and moral world a priori, the moment was ripe for a
reaction in favor of experience (ibid., pp. 37374). The true place of
the intellect is the rational method, that is, deduction, which must be
synthesized with empiricism:
Now that empiricism runs unchecked, it would be an excellent idea to proclaim
the true place of the intellect and for there to be a reconciliation of these great
methods: the experimental method and the rational method. And what eld would
be better chosen for that synthesis than that of the social sciences. . . . As for me,
my keenest desire is to see that beautiful philosophic and scientic synthesis in
action and my dearest ambition is to take part in it (ibid., p. 374).
The synthesis provides a balance of methods: It is as narrowly exclusive
to want to do away with the rational method and thereby privilege the
experimental method as to do away with the experimental method and
thereby privilege the rational method . . . (253, in 249.XII, p. 206).
P1: PIG
56 Walrass ideas
VIII. Conclusion
Walras did not merely glimpse the possibility of the syntheses of reality
and concepts, of realism and idealism in regard to applied science and
policy, of the experimental and the rational methods in pure science, of
induction and deduction. He stated that his point of view and his syn-
thesis of rational and experimental also were, to use the word he used to
describe the synthesis in the foregoing quotation, scientic (ibid.). Far
from being an archrationalist, Walras espoused a systematic and coher-
ent system of philosophical and methodological thought of which those
syntheses were integral and important parts. Whether he called a science
abstract or pure, or both, as in the case of economic theory and geometry,
his view of the matter was that in either case it has not only rational but
also empirical components. Experience nourishes and supports reason,
and reason sheds light on and guides experience . . . (79, 1867, column
185, in 194, 1896, p. 94). He did not say that the construction of abstract
and pure sciences is a rational process and that their applications are
experimental. He asserted that they use both rational and experimental
methods in the construction of their theories. In order for a fact to be
given scientic status, it must be both real and rational; it is necessary
that observation and experience establish that the fact exists and that it
exists in such and such a manner, and that reasoning shows that it cannot
fail to exist and that it cannot exist in any other manner under normal
conditions (154, 1880/1881, in 194, 1896, p. 338). Prior to the last
phase of his career, he consistently argued that general facts, real types,
ideal types, assumptions, and hypotheses should be drawn fromthe facts
furnished by experience, that deduction is used to interpret hypotheses,
and that experiments and observation are used to evaluate conclusions.
He recognized that deduction is not discovery of new knowledge but is
instead a process of interpreting empirical knowledge and of arriving at
an understanding of its implications. Walras did not, therefore, subscribe
to the view that science does not verify its hypotheses by reference to
facts. Nor was there a place in his organized system of philosophic and
methodological thought for his own isolated and internally contradictory
statement that the physico-mathematical sciences do not consult reality
for conrmation (106, 1874, p. 32; 30, p. 53; see above).
Walras thought that valid theories and facts cannot be contradictory.
That means that assumptions must be compared with the facts on which
they are based, and must be modied if they are not consistent with
them. Once deduced, he believed, the conclusions of theories and models
must be tested by reference to the facts that they are intended to explain
or predict, and modied if they are not consistent with them. In most
P1: PIG
disciplines, he noted, and particularly in the social sciences, those activ-
ities are not simple acts, but rather a difcult and often lengthy process.
They require repeated comparisons of the formative assumptions with
reality, and progressive modication and reformulation of them and of
their conclusions in accordance with additional and improved empirical
knowledge arising fromresearch and fromattempts at their scientic and
practical applications. Theories, Walras asserted, are therefore refuted or
achieve a degree of probability of conrmation by reference to empirical
evidence. The process of theoretical formulation and testing continues
until the evidence leads the theorist to believe that he has arrived at sound
representations and explanations of reality.
Thus Walrass statements to the effect that he was a Platonist, that
he was an idealist, and that, in his normative economics, he conceived
idealistically of an ideal society do not mean that he was not also a
realist or that he did not try to understand and model human behavior
in the real world. Notwithstanding some of his purely terminological
idiosyncrasies, Walras was squarely in the methodological tradition of
nineteenth-century scientists. There were, of course, variations of detail
in that tradition, from which he chose some aspects and rejected others,
but in general respects, he was a scholar of his own time, which indeed
no philosopher or scientist can escape being.
P1: PIG/... P2: IWV
CHAPTER 2
Economic philosophy and methodology
In this chapter and the following one Walrass economic philosophy
and methodology are discussed. They are special aspects of the general
philosophical and methodological views that he developed during the
creative and mature phases of his theoretical activity. He drew most
of his ideas about economic philosophy and methodology from the
work of his predecessors, including his father, J.-B. Say, J. S. Mill,
A.-A. Cournot, and others mentioned in Chapter 1, and from Pelegrino
Rossi (Diemer 2002) and Jules Dupuit (ibid). In regard to the specic
features of his economic theorizing, Walrass beliefs about economic
science were the same as those of many other important economists,
such as those mentioned above, in the respects that will be indicated
in this chapter. He tried to implement the methodological ideas of
nineteenth-century physical scientists which were not the same thing
as their mechanics or their physics.
1
They, of course, were indebted for
many of their methodological ideas to the scientists of previous peri-
ods. Whatever their origins, Walrass methodological practices, his use
of mathematics, and his modeling techniques were far in advance of
anything that economists in centuries previous to his own could have
imagined.
1
Philip Mirowski (1989) believes that Walras not only wanted to see the methods of the
physical science used in economics, which is true, but also believes that Walras drew
upon specic concepts of nineteenth-century physics and particularly its mechanics
subdivision in order to construct his economic models. Other scholars offer a more con-
vincing analysis. Albert Jolink (1993) indicates that Mirowskis account of Walrass
ideas about physics and their relation to his economics is distorted in such a way as to
seem to support Mirowskis thesis but that Walrass work does not in fact do so. Jolink
and Jan van Daal (1989) show persuasively that Walrass use of mechanical analogies
did not result froma knowledge of or misunderstanding of nineteenth-century physics.
He rst constructed his models following the general methodological precepts of the
physical scientists of his time induction, mathematization, deduction and then
subsequently tried to nd analogies between his constructions and specic physics
concepts (ibid.).
58
P1: PIG/... P2: IWV
Economic philosophy and methodology 59
Walrass ideas about how to undertake economic theorizing, imple-
mented in his mature comprehensive model of general equilibration and
equilibrium, constitute a system of economic methodological thought
designed to achieve his goal of understanding the economy of his time.
The system can be seen to be constituted of a carefully reasoned log-
ical sequence of steps. For purposes of clarity, his six principal steps
are outlined rst, with subsequent appropriate discussion and documen-
tation. First, Walras believed that the economy is an objective reality.
Second, he advocated and followed procedures by which it should be
studied, namely observation, experimentation, induction, abstraction of
ideal types, hypothesizing, deduction, and verication. Third, his study
of the real economy convinced him that the vast majority of markets
were freely competitive. Fourth, he wanted his mature comprehensive
model to reect that reality, so he had to abstract the features of freely
competitive markets from the real economy and assume them to prevail
throughout his model economy. Fifth, he contended that his model could
be made to reect reality in progressively greater detail by incorporat-
ing additional components drawn from the real economy, and that doing
so would enhance the models applicability to the solution of real eco-
nomic problems. Sixth, he indicated the criteria and the methods by
which theories and models should be judged, and evaluated the results
of the foregoing steps.
I. The rst step: Assertion of objective economic reality and
the corresponding positive nature of economic theory
Objective economic reality
Walras believed that every aspect of the world, and hence the econ-
omy, is an objective reality (as documented in Chapter 1). He wrote
that he wanted to distinguish . . . humanity, society, the individual, the
real state, in a word, social reality, and he identied each of the ideal
types based on those real phenomena (51, 1866, columns 18182, in
193, pp. 1213). For example, economics, both positive and normative,
exists; it exists not only in the ideal realm of the mind, but also in
the reality of things (76, 1867, column 145, in 194, 1886, p. 59). The
subjects of economic studies are in the real world: Let us now cast our
eyes around us, he exhorted his readers. We will see that there are
in reality two distinct markets: a market for productive services and a
market for products (152, 1880, in 249.VII, p. 437). Of course, when
he referred to reality, he meant his perception of it. He recognized that
P1: PIG/... P2: IWV
60 Walrass ideas
different theoreticians may experience the real economy from different
perspectives and may therefore have divergences in their conceptions
of economic phenomena (239, 2, 1901, p. 176).
The corresponding positive character of economic theory
Walras was concernedwiththe denitionandclassicationof the various
aspects of economic studies (106, 1874, pp. 344; 139, pp. 2566).
Applying his general methodological distinction of positive and nor-
mative studies to the specic eld of economic studies, a distinction
that was a foundation of his work, he identied three economic subject
matters (15, 1860, p. XVI; 89, 1868, in 194, 1896, p. 31; 106, 1874,
p. 22; 20, pp. 4243). These deal respectively with the true, the use-
ful, and the just; that is to say, they are respectively positive, utilitarian,
and normative (239, 1862, 1, p. 120). He undertook his examination of
economics from that triple point of view (ibid.). Moreover, declaring
(like his predecessors) that, Economics is a science (9, 1860, p. 196),
Walras contended that, like every particular scientic discipline, it has
the various aspects discussed in Chapter 1: that is, briey, pure science,
applied science, and practice. The branch of economics that establishes
what is true, Walras asserted, is the rst of those subject matters, namely
pure economic theory. He announced early in his career his intention to
study it and to develop a rigorous analysis and synthesis of real economic
and social phenomena. This he would do by examining the nature, laws,
causes, and results of social wealth; industrial activities; and the distri-
bution of income (239, 1862, 1, p. 120). The economic theorizing should
be conducted along the lines of the physico-mathematical sciences. The
Physiocrats, and J.-B. Say, David Ricardo, and J. S. Mill, he pointed out,
considered that economics was that kind of a science. The deciency
found in their work is that they tried to use verbal reasoning in connec-
tions where only mathematics could lead to a successful analysis (114,
1875, in 249.VII, pp. 3045).
Walras put the central core of his creative and mature work on eco-
nomic theory into his treatise, the El ements d economie politique pure
(106, 1874; 123, 1877; 176, 1889). Some writers, like William Jaff e
(1977), have contended that the El ements describes a utopia, constructed
in part with features that Walras took from his knowledge of the real
economy, but a normative scheme nonetheless. Most economists, how-
ever, have noted his desire, expressed in the El ements and elsewhere
(as documented in Walker 1984), to devise a positive model, or with-
out discussion have regarded that objective as self-evident. Referring
P1: PIG/... P2: IWV
to passages that belong to the mature comprehensive model, they have
analyzed Walrass general equilibriumtheory fromthe point of viewthat
it is an attempt to understand the real economy of his time.
2
Moreover,
readers of Walrass many descriptive and analytical studies of empirical
subjects (see Chapter 8) have perforce recognized that he had an evident
concern with economic reality.
Given his belief in the objective reality of the economy and his desire
to analyze it as it is, Walras naturally asserted that his model of it is
positive. Economic theory, he maintained, should be constructed by
abstracting from all considerations of material advantage or of justice
(239, 1873, 1, p. 345). When he stated, for example, that the price of a
commodity rises or falls in accordance with the conditions of supply and
demand in its market, he was taking note of an observed truth entirely
and absolutely independent of any considerations of either usefulness or
equity (90, 1868, in 194, 1896, p. 30). His system of pure economic
theory consists of an exposition of the mechanism of free compe-
tition from a purely objective point of view (239, 1894, 2, p. 624,
n. 5). My model, he declared, concludes nothing in favor of nor
against that regime, and I think it is necessary to abstract theory com-
pletely from the ethical point of view (239, 1877, 1, p. 542). In this
way, when the time comes to introduce it, Walras asserted, the eld
of study will be found to be free of and unburdened by any preconcep-
tion (ibid.). Thus he intended that his economic theory, as expressed in
the El ements, be free of normative considerations, and he believed that
it is.
Pure economic theory could not, in Walrass view, be otherwise than
free of normative considerations, because when he made those state-
ments and when he constructed his initial and mature models of general
equilibrium, he considered economic theory to be a pure natural science,
like physics or chemistry, dealing with objective reality. Before 1896,
Walras didnot place economic theoryamongthe moral sciences, because
he believed it was concerned with phenomena that are not inuenced
by free will. He called economic theory an abstract science, and a pure
science (205, 1898, p. 454). It will be recalled that Walras also specied
that economic theory and cnonics are applied to disciplines such as his-
tory and sociology to generate the pure moral sciences (Chapter 1). The
2
For example, Kuenne 1961, Howitt 1973, Morishima 1977, Walker 1984, Brems 1986,
Witteloostuijn and Maks 1988, Gijsel 1989, Negishi 1989, Currie and Steedman 1990,
Witteloostuijn and Maks 1990, Ingrao and Israel 1990, Syll 1993, Hilton 1995, Herland
1996, Morishima 1996, Walker 1996, Potier 1998, Rebeyrol 1998.
P1: PIG/... P2: IWV
62 Walrass ideas
status of economics in that presentation results in a convoluted account
of its relation to other disciplines. Inasmuch as he had argued that his
pure economic science results from the application of mathematics to
economic studies, it is unconvincing that he then treated pure economics,
like mathematics, as one of the methods-and-techniques sciences that
are applied to various elds, thus generating psychology, history, soci-
ology, and so on.
Support for the idea that economics is a natural science can be found in
the writings of previous scholars such as John Locke, who believed that
the laws of human nature are like the laws that govern the movements of
planets, and that economic relations are similar to natural phenomena
(Redman 1997, p. 69), just as Walras declared in his comparison of
economic theory to astronomy (see Introduction and following text).
Walrass basis for his viewwas that he reasoned that in freely competitive
markets there is a given supply and a competitive pricing process of such
character that the price of each commodity is determined independently
of the wishes of any of the participants. Wheat, for example, has a price
because it is useful and limited in amount, two natural circumstances
(106, 1874, p. 29; 28, p. 50). Its price does not result from the will
of the seller, nor the will of the buyer, nor from an agreement between
the two. They would each like a more favorable price than the one
determined by the market, but, under the given circumstances, wheat is
worth neither more nor less (ibid.).
Walrass claim that his economic theory is positive is supported by
his treatment of economic facts. According to the idealismthesis, when
Walras speaks of the fact of free competition, of the fact of exchange,
or of the fact of the transformation of services into products, he was
not referring, as Walker would have it, to real facts, but to genres, to
species, or to essences (Lendjel 1997, p. 75). How Walras treated
facts with respect to science in general has been described in Chap-
ter 1. His treatment of the special case of economic facts will now
be considered, and it will become evident that the idealism thesis is
in contradiction to what Walras meant by the word fact in the con-
texts in which he was referring to the real economy, and specically
in reference to free competition, exchange, and the transformation of
services.
To stress the character of some economic facts, Walras used the word
natural. The relation of utility to something fromwhich it is derived is
a natural fact in the sense that it depends on the nature of our needs being
such and such and on the nature of things being able or not to satisfy
them more or less (15, 1860, p. XV). Value in exchange therefore has
the character of a natural fact (106, 1874, p. 29; 28, p. 50) and has its
P1: PIG/... P2: IWV
place, like all subject matters in which free will does not play a part, in
the category of the natural sciences:
The fact of value in exchange is a natural and ineluctable fact; since, although
it is generated partially as a result of the presence of mankind on the earth, it is
above all generated as a result of the limitation of the quantity of useful things,
and should be considered just as independent of our psychological liberty as are
the facts of gravity, of vegetation, etc. (9, 1860, p. 201).
Among the economic facts are found the fact of value in exchange and the fact of
exchange which are essentially natural facts just like the facts of heat, of illnesses.
They are the primitive and direct object of political economy, a natural science as
completely independent of justice as is physics or pathology (15, 1860, p. XV).
Using the methods of the physical sciences, the economic theorist studies
the natural facts, including quantitative relationships, that are manifested
in exchange in freely competitive markets. It is obvious that by eco-
nomic facts such as competitive markets and exchange, Walras meant
ordinary primitive that is, basic, fundamental, and intellectually
unprocessed matters of everyday life, like plants, heat, and illnesses,
not genres, essences, or transcendental entities.
Believing that his economic theory establishes what is true, and in
accordance with his general methodological tenets, Walras did not, in his
mature phase of theoretical activity, make assumptions that he thought
were contrary to fact. That is not to say that he did not make assumptions
he believed could simplify without distorting the results. In his mature
comprehensive economic model, he assumed, for example, that there are
no disequilibrium transactions during a market day, although he knew
there were. He described that, however, as a hypothesis no scientic
mind will hesitate to concede to the theoretician (239, 1895, 2, p. 630),
thereby indicating his (erroneous) belief that the assumption in question
was not really contrary to the facts but was rather a simplication that
claried the essential features of the process of price formation in a
freely competitive market. His use of abstraction and simplication in
relation to free competition will be discussed shortly, and his use of them
in regard to utility will be discussed in Chapter 4.
The aspects of objective reality treated by economic theory
Walrass denition of his pure economic theory as being the study
of exchange and value in exchange (91, 186970, in 249.XI, p. 343)
appears to be narrow. It appears to be a statement that it is the theory
of pure exchange. In 1877, for example, he described the aspect of eco-
nomic theory that he had chosen to emphasize in his own work in this
P1: PIG/... P2: IWV
64 Walrass ideas
way: With respect to my economic theory, it studies purely and simply
the fact of the determination of prices or the ratios of exchange under a
hypothetical regime of absolute free competition (239, 1877, 1, p. 542).
On various occasions he repeated that denition (for example, 176, 1889,
p. XII; 249.VIII, p. 11). How should it be interpreted? There are two
questions to consider. One is the implications of the last six words of
the quotation. The other question is what the statement indicates about
Walrass belief about the scope of economic theory.
Some scholars have interpreted those six words to mean that his model
was not concerned with the real economy of his day. They argue that the
model is a ction because the nineteenth-century economy was not char-
acterized throughout by absolute free competition; and hypothetical,
they believe, means that the model is a sort of imaginative creation
without realistic referents:
Put in another way, Walras, contradicting the evidence provided by observation,
imagines a ctional world constituted by a vast market composed of special mar-
kets governed by free competition; and the preface of the second edition of the
El ements (1889) species completely explicitly that pure economic theory is
essentially the theory of the determination of prices under a hypothetical regime
of free competition. (1874, p. 11).
3
The consequences of that unrealistic (non
r ealiste) hypothesis competition are that the El ements constitutes a pure
theory, like pure mechanics, that cannot claim any descriptive realism (Diemer
and Lallement 2004, p. 6).
4
The meaning of hypothetical is that which is of the nature of a
hypothesis, and a hypothesis is a proposition intended to character-
ize or explain phenomena. Was Walras using the word hypothetical to
refer to an imaginary or ctional notion or to a hypothesis based on facts.
In the latter case, to what degree did Walras believe his hypothesis about
free competition was so based and what did he claim was the quality of
its empirical foundations?
As a brief preview of this matter, Walrass explanation of the specic
wording of the statement will be examined. It will be seen that the
3
That sentence is not found on page 11 or anywhere else in the rst edition. The authors
meant page 11 of bibliographical entry 249.VIII. The sentence appears on page XII
of the second edition (176, 1889), with the difference that the last three words are
actually absolute free competition.
4
The authors go on to assert that, In order to prevent any ambiguity and all possi-
ble errors of interpretation, L eon Walras, in his article Une branche nouvelle de la
math ematique (1876), recalled the necessity of clearly distinguishing three different
things regarding competition: the observable facts, the ideal type, and the normative
doctrine (Diemer and Lallement 2004, p. 6). It is true that Walras recalled that neces-
sity in his 1876 article, but his intention in doing so was to guard against just such an
interpretation of his writings as is made by Diemer and Lallement. The conclusion of
Chapter 3 shows that he intended his 1876 statements to make clear that he believed
his model was about economic reality.
P1: PIG/... P2: IWV
meaning of his statement is quite otherwise than the foregoing long
quotation asserts. This was made clear by his contentions about the
economy, about methodology and scientic hypotheses, and about the
characteristics of his model and its relation to the real economy. By a
hypothetical regime he did not mean a non-existent regime but rather
an assumed one assumed because of the realism of that assumption.
He meant an abstraction and generalization based upon observations of
the real economy and used to make deductions about it that are veried
by reference to it. He had converted the economic theory of a hypothet-
ical regime of absolute free competition [into] an exact science and one
of the most beautiful of all the exact sciences (239, 1888, 2, p. 251.).
There was not, so far as Walras was concerned, any such thing as an
exact natural science that is a ction. He explained that hypotheses could
be more or less conjectural and realistic, and claimed that his hypothesis
was on the more realistic end of the scale. Indeed, he declared, although
he did not presume to claim that his theory of general economic equilib-
rium was as great as Newtons, he was as glad to have produced it as he
would have been to have produced Darwins. The hypothesis of a freely
competitive equilibrium system, Walras continued, is less conjectural
and more rmly based on facts than Darwins and it is at least as impor-
tant as Darwins in respect of being practical knowledge about the real
world:
It is less hypothetical and more denitive, and inspires at least as lively an interest;
for it is of at least as much importance to us to know how the prices and the
quantities of commodities to be produced and to be exchanged tend to determine
themselves naturally, as to know if we descend from monkeys or if we are a
distinct species (ibid.).
Walras stated that his own hypothesis was drawn scrupulously
from real markets:
Denition of the market and of competition; effective
demand curves
Before all, it is necessary to dene with precision the mechanism of the free
competition that we assume governs our market. For that, let us go to a free
market and see how it functions. Let us, for example, go into a wheat market,
and let us take note scrupulously of the operations that are conducted there. It is
perhaps the most delicate point, in the physico-mathematic sciences, to borrowin
this way fromreality the experimental data on which the intellect then establishes
the series of rational deductions (103, 1874, p. 10).
5
5
See Chapter 1, where the last sentence of the quotation is also used. Walras added the
title in the republications of the text, rst in italics, as shown here, and subsequently
in roman capitals. See the annotation of entry 103.
P1: PIG/... P2: IWV
66 Walrass ideas
Let us recall, moreover, that it is always a matter here, not of posing
and solving in reality in any given case the problem in question, but
exclusivelya matter of conceivingscienticallythe nature of the problem
that generates itself and that solves itself empirically on the market
(176, 1889, p. 134; 110, p. 161). In the light of statements like that,
of which there are many, it cannot reasonably be supposed that Walrass
reference to a hypothetical regime means that he was asserting that
his model did not capture the essentials of real competitive markets. The
words unrealistic and ctional are used by Diemer and Lallement,
but Walras did not use those words or share their view. He wrote: I
continue to believe that my conception of the equilibrium of production
is not a ction but an abstraction . . . (239, 1893, 2, p. 598; emphasis in
the original), and declared that his model was very much in conformity
with reality (239, 1879, 1, p. 628, n. 3).
As for the scope that Walras believed his theory to have, a rst matter
to consider, in relation to his other writings, is his assertion in the forego-
ing statement that economic theory studies the determination of price or
the ratios of exchange. In partial explanation of his model, Walras char-
acterized pure economic theory, in a passage that he wrote in 1872 and
republished in subsequent years, as, The Theory of Value in Exchange
or Theory of Social Wealth. In the rest of that particular passage, he
relegated The Theory of Industry or Theory of the Production of Social
Wealth to the domain of Applied economics (98, 1872, in 249.XI,
p. 419). Pure economic theorystudies, The totalityof all things, material
or immaterial, which are susceptible of having a price. . . . That is why
economic theory is also the theory of social wealth (176, 1889, pp. XII
XIII; p. 11). The totality of such things includes durable goods, primary
materials, non-durable consumer goods and services, and the services of
labor, natural resources, and capital goods. Thus Walrass denition of
value in exchange as the same subject matter as social wealth indicates
the broad signicance that he attached to economic value, giving a much
wider perspective to the meaning of studying ratios of exchange.
Walrass statement classifying the theory of production as applied
economics did not clarify his meaning, but his subsequent denitions
are clear and broader in scope. He stated that pure economic theory
does not deal exclusively with the exchange of commodities that are
already produced and on the market. It is rather the pure and simple
study of the natural and necessary effects of free competition in regard
to production and exchange (239, 1873, 1, p. 345; emphasis added).
Again, he specied that his system of pure economic theory reveals,
from a purely objective point of view, the mechanism of free competi-
tion in regard to exchange and production (239, 1894, 2, p. 624, n. 5),
P1: PIG/... P2: IWV
that pure economic theory rst establishes the theory of exchange and
then the theory of production, setting them forth in systems of equations
(114, 1875, in 249.VII, pp. 298304). Thus Walras referred to the whole
class of effects of free competition and included production as well as
exchange in pure theory, and, of course, a large part of the El ements deals
precisely with the theory of production. Indeed, in one place he referred
only to production, writing that the goal that he set himself in his pure
economic theory in mathematical form was to present and explain, for
the rst time, the mechanismof production by reducing it to its essential
elements (180, 1892, in191, 1896, p. 477; 210, 1900, p. 479; appendixI,
10, p. 705). Walras therefore included in his pure theory such matters
as an analysis of what happens to production functions and to the average
cost of production during the process of adjustment of prices, services
hired, rates of production and consumption, and quantities exchanged.
He called what happens during that process a tatonnement, a word
meaning reiterated hesitant efforts to nd something in the case of
the economy, to nd equilibrium. Clearly, in considering these matters,
Walras was not dealing with what he considered to be ctional and unre-
alistic elements. He also indicated the breadth of his denition, insisting
as it does on the effects of free competition, by classifying a central part
of his welfare economics as pure theory, namely his attempts to demon-
strate that freely competitive markets generate a relative maximum of
utility for society (123, 1877, p. 305; 26364, pp. 42325).
II. The second step: The methods by which knowledge
of the real economy is obtained
Observation, experience, and experiments
in economic theorization
Toward the end of his life Walras claimed that he had drawn his ideas not
from other books but from my brain, thus failing on that occasion as on
numerous others to acknowledge his many debts to his predecessors. He
then went on to state that he followed the Meth[od] of reasoning and not
of experimentation (249.XIII, p. 560). The vast preponderance of the
evidence will now be considered. In contradiction to that jotting, Walras
developed and subscribed to a system of thought that fully recognized
the place not only of reasoning but also of obtaining and using empirical
knowledge in constructing economic theory, and he stressed that he drew
upon such knowledge. In fact, in the rest of this jotting he contended
that empirical knowledge is half of scientic method. He wrote that
he was not contemptuous of facts and experience, but that is only the
P1: PIG/... P2: IWV
68 Walrass ideas
half of it (Germans, French sociologists). Indeed, in this jotting he was
primarily interested in emphasizing the use of reason as the method
that is indispensable when considering justice and interests concern
for material welfare (ibid.). Even when describing social ethics as an a
priori discipline, he asserted that its assumptions are based on experience
(249.XIII, p. 569).
According to Walras, there are economic hypotheses similar to some
of the functions of physical science (Chapter 1) that are formulated on
the basis of the previous state of the science, and therefore without being
directly based upon a new collection of facts. Proceeding in economic
theory by the same means as in physics, we will arrive at the same results,
of which the most important will be to link together economic relations
and laws, in such a fashion that, certain of these relations and these
laws being given by experience, the others are deduced by reasoning
(249.VI, p. 257). Experience shows that an economic relation exists;
reasoning explains it. For example, The fact of surplus value of land
rent in a progressive society is well established by experience and well
explained by reasoning . . . (163, 1885, p. 80). The rst of these condi-
tions of scientic truth is that the matter in question rest on a demonstra-
tion either rational or experimental . . . (85, 1868, column 274, in 194,
1896, pp. 14849). The law of the variation of equilibrium prices is
similarly expressed purely and simply by noting that, in accordance with
observation and experience, when the utility of the commodity increases
or its quantity diminishes, the price rises; that when the utility decreases
or the quantity increases, the price falls. This law is demonstrated by
making clear why and how this happens (249.VI, p. 436). Whether the
foundations are the existing science or new empirical knowledge, the
hypotheses and the conclusions deduced from them are subjected to
the test of conrmation or refutation by empirical evidence.
J. S. Mill contended that, The backward state of the Moral Sciences
can only be remedied by applying to them the methods of Physical Sci-
ence, duly extended and generalised (Mill 19811991, 8, p. 833), and
specically that the methods of physical science are the proper mod-
els for political economy (Mill 19811991, 1, p. 173). Agreeing with
that view(14, July 1879, p. 15), Walras declared that the methods of pure
economic theory include observation and exp erience (239, 1885, 2,
p. 64; and see, for example, 9, 1860, p. 197; and see below). His argument
that scientic methods should be used is an argument that economic the-
ory should deal with reality. By the rst of those French words, used
in application specically to economic science, Walras meant exactly
what the same word in English means in scientic contexts, namely
scrutiny of real phenomena. There is no doubt that he always thought
P1: PIG/... P2: IWV
that economic science is observational in that sense. He believed that
economics, unlike some sciences, deals only with observable external
facts and observable personal subjective facts (106, 1874, p. 103; 102,
p. 146; 198, 1897, in 205, p. 270), which he called natural facts. Thus
he did not consider the methodological problems that arise in connection
with the testing of theories about unobservables. Illustrating the meaning
of observational, he asserted in an early essay that, for example, the
detailed description of the techniques of modern industry empirical
knowledge, obtained by observation lays foundations of unshakable
solidity for economics (21, 1861, in Walras 249.VII, p. 141). The eco-
nomic researcher follows scientic methods to establish and document
the accuracy of observations, but he has a neutral role in that process.
Walrass use of the word exp erience in economic research contexts
is the same as in his discussions of science in general (Chapter 1). If
he had used only that word to describe economic method, there may be
some reason to regard it as meaning empirical studies or something
else less categorical than experimentation. His precise wording, how-
ever, was that the methods of economic theory include observation ou
exp erience la plus rigoureuse (239, 1885, 2, p. 64), thereby differenti-
ating the two methods as he did in the quotation in the next paragraph
of this chapter, and indicating that the latter term alludes to a second
aspect of economic research. That aspect, in that context, does not have
the meaning of the English word experience, which cannot be rigor-
ous. The French exp erience la plus rigoureuse instead indicates that
the researcher enters actively into the situation of interest, handling and
interpreting economic facts with careful scientic methods, identifying
real types that are important, using them to form abstractions, isolating
the true characteristics of the variables, tracing their effects in order to
uncover causal relationships, and testing the verisimilitude of assump-
tions and conclusions. In short, the word exp erience in Walrass eco-
nomic writings in the context under discussion, just as in his discussions
of his general philosophical and methodological views, translates as
empirical research techniques. The use of those techniques is also the
meaning of Walrass term experimental.
Walras argued in one place that economics is observational but not
experimental in the sense of the laboratory sciences when he criticized
economists who contended that economics is experimental but who,
in actuality, did not have a clear idea of what a science is or what an
experiment is (198, 1897 in 205, 1898, p. 270). He explained that
. . . economics can be an observational science but not an experimental
one. Mechanics, physics, chemistry, physiology experiment. Because of the
P1: PIG/... P2: IWV
70 Walrass ideas
impersonal nature of the facts that they study, they can isolate specic facts
and make them occur, a hundred percent of the time, in a relation to other facts.
By induction, they arrive at the judgment that the relation is a physical or natural
law. Economics does not experiment; it observes personal facts (ibid.).
It is clear that in that particular statement, Walras regarded observation
and experimentation as different forms of research, and that by the latter
he was referring to controlled experiments, usages that should be borne
in mind in the following discussion.
In 1861, however, Walras asserted, contradicting the passage just
quoted, that, Economics is an experimental science . . . (27, 1861,
p. 92). It has been suggested that he expressed that view when under
the inuence of his father (see Potier 1994, p. 256), as though he held
it temporarily and not as a basic conviction. In fact, Walras repeated
it during his entire career. By experimental science he did not mean
controlled laboratory experiments. He meant, as he stated in the rest of
the sentence, that it observes natural facts with the greatest rigor, dis-
covering their order and relationships (27, 1861, p. 92). Determining
causal relationships order and relationships is a principal objective
of experiments, so he used that phrase to evoke that activity. Detecting
those relationships is a principal objective of observation. Walras was
preserving the difference between experimental and observational
and was stating, over the course of the two clauses of his sentence, that
economics employs both methods.
There are many other passage in which Walras asserted unambigu-
ously that economic theory is experimental in the active participatory
sense of that adjective. It has been seen in Chapter 1 that, for most of
his career, he classied economic theory along with physics, chemistry,
and biology as a pure natural science. He used the words experimen-
tal and observational in reference to natural sciences, as when he
wrote about the experimental sciences which, through the observation
of facts attain, by induction or by hypothesizing, progressively more
profound knowledge of laws and relationships (15, 1860, p. XIII). It is
probable that in that passage he meant, not that observational sciences
by convention can be called experimental, but that scientists conduct
experiments and observe the results. After 1896 he described economic
theory as a pure moral science, thus asserting before and after that date
that it is a pure science, and contended that, All pure science is done
from the point of view of pure truth, rational and experimental (205,
1898, p. 453; emphasis added to the last two words).
Walras remarked that a certain scholar made a faulty distinction
when he described physical and natural sciences as observational, and
P1: PIG/... P2: IWV
physiology and social science as experimental, but Walras did not mean
that social science is not experimental. He contended rather that the nat-
ural sciences are not only observational but also experimental and that
social science is not only experimental but also observational (249.XIII,
p. 558). He argued that, The physical and natural sc.[iences] properly
speaking are just as experimental as physiology. Physiology will also
be both rational and experimental. Social science, id (ibid.; empha-
sis added.). The Latin idem, meaning the same, indicates his belief
that economics, as a social science, is also not only rational but also
experimental in the sense that the physical and natural sciences are. For
example, it determines that the intrinsic value of land is a scientic fact
from the points of view of reasoning and of experience (215, 1907, in
249.VII, p. 471). Then in 1900, Walras summed up his views on this
matter by declaring that, It is now very certain that political economy
is, like astronomy, like mechanics, a science that is both experimental
and rational (210, 1900, p. XX; p. 22). Walrass declaration implies that
he thought that economics has empirical bases and is tested empirically.
Walrass assertions about the methods of economics make it evident
in yet another way that he rejected the archrationalism thesis. He main-
tained that, as he practiced it, economic science is like all the other pure
sciences in using observations and experiments. He afrmed that it does
so in order to attain knowledge of reality and that those methods are
undertaken precisely in order to compare assumptions and conclusions
with reality. The theories and models developed in the experimental sci-
ences, of which he declared economics is one, are compared to reality.
Those sciences apply the criterion of the conformity of theory with
perceived reality [de lordre pens e avec lordre per cu], or the principle
of order and of causality (14, 1860, in 249.VII, p. 58; and see 15, 1860,
p. XIII), and they apply that criterion by using observation and exper-
imentation. Thus he believed that in regard to their relation to reality;
that is, in regard to the ontological status that he conceived their subjects
matters to have, all the pure sciences are the same. According to Walras,
they all, including economics, deal in a scientic way with une r ealit e
sensible, not in a metaphysical way with a metaphysical realm, which is
indeed explicit in his viewthat they are all experimental or observational
or both. Whatever nuances he placed on the words experience, exper-
imental, and observational, it is clear that by using them to describe
the methods of economic science, he was doing so to indicate that it is
empirically based and empirically veried. Walras therefore contended
that he arrived at his model of a hypothetical regime of absolute free
competition by using the correct scientic methods of observation of
the real economy and by experimentation to arrive at his abstractions.
P1: PIG/... P2: IWV
72 Walrass ideas
To summarize: First, Walras believed that knowledge of reality is
obtained by sensory experience and thus he dissented from the view
that knowledge that is certain comes from innate ideas, referring to the
natural world that is scrutinized by the senses in order to make it the
object of pure natural science (203, 1898, in 205, p. 492; see Potier
1994, p. 268; Chapter 1). Second, he believed that all the pure sciences
not only have a rational aspect but are also experimental or observational
or both. An attribution to him of the opinion that pure economic science
does not compare its assumptions or conclusions with reality is therefore
contradictory to his thought. That attribution would imply that Walras
had a separate epistemological and methodological outlook with respect
to just one of the pure sciences, namely economics; that he believed that
economic activities in the real world, alone of all the subjects treated
by science, are ontologically different and are not scrutinized by the
senses to make them the object of pure economic theory; and that pure
economic theory, alone of all the sciences, is neither observational nor
experimental. The archrationalism thesis cannot be true unless Walras
thought that all pure sciences are empirically based, deal with reality,
and are compared to it, except the pure science of economics. That, it
has just been shown, is contradictory to what he believed.
III. The third step: Recognition of the near-ubiquity
of freely competitive markets in the real economy
Walrass main objective as an economic theorist was to understand the
behavior of the markets that functioned in the economy of his time,
and he therefore studied what he considered to be the two major types,
namely monopoly and freely competitive markets. He did not integrate
monopoly and freely competitive markets into a single model, nor did
he want to do so. To be logical, it is necessary to proceed from the
general case to the particular case, and not from the particular case
to the general case, as would be done by a physicist who, in order to
observe the sun, carefully chooses an overcast period instead of taking
advantage of a cloudless sky (106, 1874, p. 51; with changes in 176,
1889, p. 70; 43, p. 73). The secret of science is to bring the general
case to the foreground and to relegate particular cases and exceptions
to the background . . . (239, 1891, 2, p. 434). Applying that general
scientic method to economics, he contended that the methodologically
correct procedure is to put aside the special cases that he considered
imperfectly competitive markets to be, and to analyze the characteristics
of the majority of markets (1874, p. 49; 41, p. 71; 1877, p. 267; 222,
pp. 33435).
P1: PIG/... P2: IWV
What were the majority? What was his idea of the character of the
real economy of his day? Walras surveyed it and concluded that it was
principally, rst, one in which there was private ownership of economic
resources and other commodities; second, a market economy; and third,
one in which free competition in regard to exchange is the almost uni-
versal regime (239, 1891, 2, pp. 43435). It is true, his observations
led him to recognize, that production and exchange in the real economy
are subjected to legal and administrative regulations, and are inuenced
by custom. The regulations are becoming less important, but the force
of custom remains (114, 1875, in 249.VII, p. 298). Nevertheless, free
competition is the principal mode, practiced on all markets with more
or less precision and therefore with less or more frictions (239, 1895,
2, p. 630). He reinforced the clarity of the meaning of those statements
by drawing an analogy between the subjects studied by physical scien-
tists and economists. The mathematics, he observed, may not be under-
stood by everyone, but we would all agree that Newton and Laplace had
described the world of astronomic facts conforming to the principle of
gravity. Walras was writing about the real cosmos, about what those
scientists and he believed was a valid description of it. Then, writing
about the ordinary facts of the real economy, he went on to ask: Why
not accept in the same way the description of the world of economic facts
conforming to the principle of free competition? (123, 1877, p. 365;
370, p. 651).
If Walras had tried to deal with the frictions and particular modes of
behavior that are found in imperfectly competitive markets as well as the
behavior of highlycompetitive markets inthe same model, he wouldhave
found it impossible to achieve the goal of constructing a comprehensive
model of an entire economic system. His subject matter would have been
too complicated for him to analyze with his mathematical skills or even
with the most advanced mathematical techniques that were available
(123, 1877, p. 267; 176, 1889, p. 252; 222, pp. 33435).
IV. The fourth step: The construction of a freely
competitive model
The assumption that the markets in the model
are freely competitive
Walras wanted his mature economic model to reect the fundamental
characteristics of the real economy. That is demonstrated in all aspects
of his theorizing, because the way he treated each topic is an expres-
sion in one particular respect or another of the parts and implications
P1: PIG/... P2: IWV
74 Walrass ideas
of his system of scientic thought, and that system rests on connections
between theory and reality. His views about the real economy and scien-
tic methodology led Walras to the fourth step in his construction of his
system of economic theorizing, namely to making the assumption that
all the markets in his model are freely competitive. He recognized the
complex facts of reality, he explained, but in a scientic investigation
of the determination of the relation of the market for products to the
market of services and of the equilibrium prices of those commodities,
the soundest procedure is to assume their markets are regulated by free
competition alone (114, 1875, in 249.VII, p. 298). I prefer, Walras
wrote, for my part, to start with unlimited competition, which is the
general case, to arrive nally at monopoly, which is a special case . . .
(176, 1889, p. 499; 381, p. 665). Explaining this matter further, Walras
wrote: I have reduced the economic mechanismto its most essential ele-
ments by abstracting from all the accessory complications (239, 1892,
2, pp. 50910). Regarding competition, Walras repeated he had as a
rst step, reduced that mechanism to its essential elements (239, 1894,
2, p. 624, n. 5).
Bythe economic mechanism, he was referringtothe real economythat
his observation and empirical research presented to him. He was identi-
fying the fundamental aspects of the economy, not taking the rst step
toward constructing a ction or a utopia. He contemplated that empiri-
cal totality and then arrived at his model by a subtractive and extractive
process; that is, by abstracting elements of that totality, discarding those
that were not essential parts of its structure and functioning. In contrast,
when a writer makes up a ctional or utopian economy, he undertakes
an additive process, conceiving of one component after another, many
of them ctional, and putting them into a model. Of course, a positive
model does not spring forth fully formed from the mind of the theoreti-
cian, and thus in discussing the history of the process of its formation,
the assembling and articulation of its parts is referred to as construct-
ing the model. By contemplating the nished system and comparing
it to the reality to which it is related, however, it becomes clear that its
parts are simplied extractions from a complex reality.
The functioning of freely competitive markets, Walras argued, is then
seen to determine simultaneously the rate of production and prices of
products, and the rate of production and prices of productive services.
Explaining and demonstrating this is the object of pure economic theory
(114, 1875, in 249.VII, p. 298). In the question of the tatonnement,
for example, I take the almost universal mode of free competition in
exchange and the associated phenomenon of Walrasian pricing, which
is the changing of the price in the same direction as the sign of the
market excess demand (239, 1891, 2, p. 434). It is for the same reasons
P1: PIG/... P2: IWV
that Walrass assumption of free competition was adopted by all general
equilibrium theorists for decades, and is still used in many models.
Real referents of the components of the model
It has been alleged that to suppose that Walras wanted to develop a model
that would explain the real economy of his day is an error occasioned
by an inability to distinguish between his real types and his ideal types.
Walras, it has been contended, did not intend the ideal types in his model
to have referents in the real economy of his day and they did not (Koppl
1995, Lendjel 1997, Huck 2001, Bridel and Huck 2002). Those char-
acterizations of Walrass philosophical and methodological outlook in
general and of his economic model in particular are not defensible, as
has been indicated in the foregoing exposition of his approval of empir-
ical methods and components of economic science and his conclusions
regarding the character of the economy of his day, and as will now be
made clear in an examination of his real and ideal economic types.
The manner by which Walras arrived at his economic abstractions
demonstrates that he habitually referred to economic facts as being indi-
vidual concrete facts of the real economy, that he explained how they
are apprehended and organized so as to identify real types, that the ideal
types that he used as components of his model are based on real types,
and that he therefore established the connections between the ideal types
and the real economy. As Walras phrased matters, the components of
his model are ideal types based on real types and are therefore rmly
connected to reality. No one, he wrote,
. . . can prevent us from dening, that is, drawing by abstraction from experience,
the ideas of social wealth, capital and income, productive services and products,
proprietors, workers, and capitalists, entrepreneurs, the market and of prices; the
idea of humans in society, of their division of labor, of their quality of being
reasonable and free. Now, on the foundation of those ideas, we can, we should,
establish the theory of production and that of the distribution of the social wealth
among the people in the society (89, 1868, in 194, 1896, p. 187).
As in geometry, he said, the only ideal types admissible in social
science are those that are extracted by the understanding from the real
types that experience furnishes (251 in 249.XII, p. 166).
Just as, in taking up the problemof the mathematical determination of the prices of
products, we have had to dene with precision the mechanismof free competition
in regard to exchange, similarly in taking up the problem of the mathematical
determination of the prices of productive services, we have to examine carefully
the facts and experience to obtain from them the exact notion of the mechanism
of free competition in regard to production (123, 1877, p. 222; 178, p. 276;
emphasis added).
P1: PIG/... P2: IWV
76 Walrass ideas
Thus Walras drewthe mature models basic components fromthe reality
that he perceived, which, as has been seen, was the near-ubiquity of free
competition. His conviction about economic reality explains why he
constructed a model of a freely competitive economy, and makes clear
that he did not do so to construct a hypothetical in the sense of a ctional
model (see Diemer and Lallement 2004), or a utopia whose conditions
necessitated the operation of that type of system (see Jaff e 1977).
Induction and deduction
It has been shown that Walras believed that induction using empirical
information, gained by experimentation and observation, as the basis
of assumptions, hypotheses, and scientic laws is an important part
of investigation in all science. Therefore he believed that the process of
induction is a crucially important aspect of economic science. It is true
that he wrote in 1896 that social science is an abstract and deductive
science as much as and more so than the mechanics of material forces
(194, 1896, p. VI). Economics, in particular, is incontestably, in any
case, a science that is not only abstract and deductive, but also mathemat-
ical (205, 1898, p. 450).
6
In the foregoing sentences, however, Walras
was, of course, referring to the deductive phase of theorizing; he was
not asserting that social science does not rest on empirical knowledge
and is not tested by it.
On the contrary, like the other social sciences, Walras contended, eco-
nomics must be cultivated by scholars of general culture who are accus-
tomed to handling, at the same time, induction and deduction, reasoning
and experience (210, 1900, p. XX; p. 22). Studying the physical and
natural sciences gives the social scientist the habit of using the method
of induction and deduction that is the true scientic method (141, July
1879, p. 26). Walras commented regarding that habit: Is it not permit-
ted to believe that the same method which has so well succeeded in the
physical and natural sciences would have equally good results in the
moral and political sciences . . .? (ibid., p. 15). Adopting the method of
using both induction and deduction,
. . . pure economic theory should drawfromexperience the real types of exchange,
supply, demand, market, capital goods, incomes, productive services, products.
Fromthese real types, theory should abstract ideal types by a process of denition,
and reason on the latter . . . (106, 1874, p. 32; 30, p. 53).
6
Mill was another example of the many economists who contended that social science
is deductive and abstract and that political economy, too, was Newtonian because of
its deductive, abstract nature (Redman 1997, p. 325).
P1: PIG/... P2: IWV
The word real in that quotation perfectly describes the types in ques-
tion. It indicates that Walras believed they exist in the real economy and
are apprehended through experience of it; they are facts in the ordinary
sense of the word, not essences. Most of Walrass references to facts
are references to real concrete individual facts; only the intellectually
processed facts he called ideal types are abstractions that describe what
is common to groups of similar real facts and essential to them only in
that sense. Another example of his meaning of the word ideal is his
statement that this state of equilibrium of production is, like the state
of equilibrium of exchange, an ideal and not a real state (123, 1877,
p. 231; 188, p. 283). Walras did not mean ideal in the sense of a
normative goal; he meant a state envisaged through scientic abstrac-
tion that would be reached if the parameters identied in the model were
actually to remain constant for a sufcient period of real time. Because
the ideal types in his mature comprehensive general equilibrium model
are based upon real types, the ideal types rest on an empirical founda-
tion. They have referents in une r ealit e sensible, and that, for Walras,
is their justication.
Thus Walras was not following arcane philosophical directives in his
choice of methods to construct his economic models. Scientic thinking,
he observed, requires that subjective elements be discarded in order to
leave in our perceptions only what is objective or absolute (205, 1898,
p. 492). Neither metaphysical nor fanciful, his procedure and his descrip-
tions of it embody the manifestly accurate perception that the concepts
used in scientic theories, and thus the theories themselves, are ideal-
izations of real situations, events, and things. As Giddings remarked
(see Chapter 1), abstractions are not abstractions from nothing; they
are taken from reality. The ideal economic types are therefore idealiza-
tions in the sense that they are, like perfect circles, free of the irrelevant
idiosyncrasies and perturbing events and forces that affect corresponding
particular real instances of the phenomena. That is what Walras meant
when he wrote that he had reduced the economic mechanism to its most
essential elements. The world is not perfect. Perfection is a pure idea.
The relation of perfection to imperfection is the relation of the ideal to
reality (249.XIII, p. 560). The ideal types found in Walrass mature
comprehensive model of economic equilibrium do not differ in charac-
ter and purpose from those used by economists and scientists in other
elds. Like scientists in all disciplines, in order to obtain his economic
abstractions (his ideal types), Walras stripped away from his portrayals
of empirical economic phenomena (his real types) the aspects that were
not relevant for his model. He then expressed the results in scientic
form.
P1: PIG/... P2: IWV
78 Walrass ideas
Walras thought that the task of obtaining economic ideal types is
arduous. In the social sciences, he asserted, there is a lengthy pro-
cess of interplay of scientic reasoning and the reality with which it
deals: The real type in the theory of society is not immediately fur-
nished by experience, and the ideal type can be drawn from it only by
means of an a posteriori synthesis that is very long and very laborious
(51, December 31, 1866, column 182; and in 194, 1896, p. 13; and
see 15, 1860, p. XIII). The process involves repeated comparisons of
the formative scientic abstraction with the real type found in reality,
improvement of perceptions of real types by continued observation of
facts and experimentation, and consequent repeated modications of
the abstraction. In regard to the economic and social sciences, the ideal
types must be sought with difculty in a complex amalgamated mass
of data; there is, we would say, a problem of complexity. It is therefore
necessary to make an empirical effort of decomposition or analysis and
of reconstruction and synthesis in order to obtain the real types and attain
the ideal types (Dock` es 1996, pp. 5960). Walras would have thought
that this is a good description of the processes by which all scientists
arrive at their abstractions.
Real referents of the components of the model in practice
Comparing Walrass knowledge of the facts of the economy of his day
with the ideal types in his models renders it objectively ascertainable that
his ideal types have real referents, that is, the components of his mature
comprehensive model correspond to elements in the real markets he
described in many writings. That could not be otherwise than the case
because he abstracted the former from the latter. To suggest that Walras
did not think that the masses of empirical detail about actual markets
he provided in his writings are ordinary facts would be ludicrous. He
was not writing about some metaphysical intellectual construction, but
about the real world, the r ealit e sensible.
Some examples, selected from literally hundreds that appear in his
writings, will illustrate this matter. He provided abundant information
about his knowledge of the institutions and behavior of real competitive
markets. Among many others, he discussed commodity exchanges, grain
markets, sh markets, (106, 1874, pp. 4849; 41, pp. 7071; 249.XII,
passim in pp. 445714), and the institutions, participants, rules, types
of securities, and different sorts of operations on the Paris Bourse (13,
1860; 65, 1867; 147, 1880; 249.XII, pp. 67180; 249.XII, p. 890, n. 21).
In those real freely competitive organized markets, Walras noted, prices
are determined by the forces of supply and demand without collusion
P1: PIG/... P2: IWV
and are changed by buyers and sellers in the same direction as the sign of
the market excess demand, which is what Walras meant by the termfree
competition (106, 1874, p. 50; 42, pp. 7172). He likewise asserted
that the workings of free competitionalsooccur inmanyreal unorganized
competitive markets (106, 1874, pp. 4849; 41, pp. 7071).
The ideal types in his model bonds, brokers, criers, buyers, sellers,
prices, supply, demand, the physical setting of the markets are there-
fore, via real types, abstractions from his knowledge of real markets. He
used those ideal types, together with a few simplifying conditions, to
obtain his model of an organized market, le march e type (147, March
1880, p. 460; 147, April 1880, p. 78; Walker 2000b) that he assumed in
all his theoretical formulations of freely competitive markets. He was
not like those economists, he declared, who
. . . have not furnished the denition of either the mechanism of free competition,
or of the diverse elements of that mechanism. They have not extracted from expe-
rience, by observation and abstraction, the types of these elements: the market,
demand, supply, the raising of prices, the lowering of prices, the equilibrium
price, products, the productive services: labor, the services of land, the services
of capital goods, the services of the entrepreneur, etc. (114, 1875, in 249.VII,
p. 301).
In that passage, as in many others, it is patently obvious that Walras
was saying that the real economy is characterized by a system of freely
competitive markets; that he abstracted its features to obtain the ideal
types that are in his model; and that those ideal types therefore have a
one-to-one correspondence with elements in the real economy. Thus he
not only said that economic science should base its ideal types on real
phenomena, he did precisely that in his theoretical work. It is clear that
on the occasions when he described his work as dealing with a hypo-
thetical regime of competition, he did not mean that its hypotheses were
not drawn from reality.
Again, examining in his model the impact of the substitution of a
commodity in place of one whose price has risen, Walras remarked that
the behavior is what is seeneveryday (106, 1874, p. 154; 152, p. 238).
In order to construct his purely competitive model of the capital market,
he declared that it is necessary to borrow some decisive circumstances
from reality and from experience (123, 1877, p. 302; 269, p. 434).
In one analysis he temporarily assumed that capital is lent in kind, but
then, indicating from whence he obtained his ideas about the savings of
capitalists, noted that in the real world, capital is lent in money, for the
reason that it is in this form that the capitalist accumulates his capital
by saving (123, 1877, p. 252; 208, p. 313). As for the problem of
the determination of the price of a commodity money, his geometrical
P1: PIG/... P2: IWV
80 Walrass ideas
method solves it as it is done, in the real world, by the tatonnements
that are very well represented by his diagram (176, 1889, pp. 38586;
283, pp. 48384).
Walrass procedure is also illustrated by his treatment of market
imperfections. He preferred free competition for most economic activi-
ties in his ideal society and in the real economy, but he recognized that
it was not the only real market structure. Indicating his interest in the
empirical situation, Walras pointed out that whatever is said, or seems
to be said, quite frequently by economists, free competition is not the
only possible form of organization of economic activity; there are oth-
ers, such as regulation, tariffs, economic privileges, monopolies, etc.
(123, 1877, p. 369; 371, p. 655). He identied the characteristics of
industries that he considered to be natural monopolies in particular, the
economics of railroads and set forth a model of their behavior in order
to analyze their pricing and output policies (115, 1875). His model of
monopolyis anideal economic type; it uses concepts abstractedfromreal
monopolies (123, 1877, pp. 37785; 37084, pp. 66068). Although
Walras knew that the quantity demanded of any monopolys product is
inuenced, however slightly, by the actions of other rms, to obtain the
ideal demand function for his model monopoly, he assumed that the dis-
turbing inuences of their pricing and output policies do not exist. Real
monopolies, Walras would have said, therefore imperfectly resemble the
ideal pure monopoly found in economic theory.
Fictional versus real types
Some of the components of Walrass constructions are not ideal types
because they are not abstractions, not idealizations, not based on real
types found in the real economy. They are ctions. This is clear in
general but obscured in some instances by Walrass terminological
inconsistency. Ina fewcases, he usedthe wordimaginary torefer toc-
tions. In others, he used it merely to indicate that creative mental activity
enters into the scientic formulation of ideal types that, as abstractions,
do not have precisely the features of particular real instances (Chap-
ter 1). In a few cases, he described ideal economic types as imaginary.
In others, he described imaginary components as ideal. Regarding the
net income commodity, for example, he wrote that he had imagined it
that to produce its effect, it will be sufcient to imagine it; but he
also wrote that it is an ideal commodity (210, 1900, p. 250; 242,
p. 359).
Despite those terminological variations, the ctional components and
the ideal economic types in Walrass writings are easily distinguished.
P1: PIG/... P2: IWV
He had no choice but to use the ideal economic types that appear in his
mature model, whether he approved of them or not, because they were
based on important elements of the real economy. The ctional com-
ponents were not. Like the ctional written pledges made by suppliers
and the ctional perpetual net income commodity, they appear in the
theorizing that Walras introduced in the fourth edition of the El ements,
abandoning the aim of realism. Fictional components also appear in
abundance in Walrass writings about his ideal society.
Walras used the ctional components because he liked them, and he
was aware that in matters of policy it is necessary not to lose ones head,
to keep oneself free of illusions, and not to believe that ones desires are
realities (141, July 1879, p. 5, in 249.VII, p. 377). His ideal society did
not actually exist, but he believed it would be achieved by implementing
the reforms that he recommended. In referring to it, he was dealing with
a normative conception, and therefore when he used the word ideal in
connection with it, he did so in the sense of desirable or the best. Its
institutions, economic agents, and rules of economic life are the ones that
he desired, and the justice and social utility to which he referred were
his notions of the meaning of those terms. Obviously the social ideal,
which he regarded as perfection in regard to justice and social utility
(Dock` es 1994, p. 280), is in reality his social ideal, an elevation of his
personal preferences to the level of what he considered to be universally
valid metaphysical principles.
There is an evident contrast between the ctional situation in Walrass
ideal society in which all natural monopolies are owned by the state
and the real situation in the nineteenth-century European and American
economies. Privately owned monopolies existed in the real economy
and were important, so Walras had to recognize them in his economic
theorizing, but he chose not to put them into his ideal society. Natural
monopolies are there owned by the state. Similarly, the institution of the
possession of land by private individuals existed in the real economy
and it was an important institution, so Walras had to abstract it and
feature it in his economic model. In contrast, he did not put it into
the rational society that is in conformity with the ideal society of
his youth (Dock` es 1994, p. 322; Dock` es 1996, p. 221). Into the ideal
society, he put the ctional condition that all land is owned by the state;
the private ownership of land and therefore landlords do not exist there.
In the same way, he provided theoretical analyses of taxation policies in
the real economy, but for his ideal society he chose the ctional condition
that there are no taxes, along with the ctional condition that the state
supports itself exclusively with the rents that it charges for the use of
land.
P1: PIG/... P2: IWV
82 Walrass ideas
Realism and applications
Walrass mature comprehensive model diverges from reality in the
respect that it includes only freely competitive markets (and in some
other ways), but he nevertheless asserted or implied many times that it
was realistic. He meant that in the sense that it provided an understand-
ing of the real economy of his day and was applicable to the solution
of real economic problems. Walras was greatly concerned with those
matters. During his journalistic phase and his entire subsequent career,
he analyzed a wide variety of empirical economic situations. When he
started work on the El ements, he declared that one of his major objectives
was to construct a model of general equilibrium in order to understand
and evaluate the consequences of free competition, to devise sound poli-
cies to encourage it, and to enable the evaluation of alternative forms
of economic organization in activities in which it does not work (123,
1877, pp. 26768; 176, 1889, p. 254; 223, pp. 33536). After he
nished his treatise, he demonstrated that he had succeeded in devel-
oping empirically useful theoretical constructions by applying them to
the analysis of such matters as the Paris Bourse, banking, agriculture,
monetary policy in Europe and India, and railroads. I have shown,
he wrote, by its application to a single question (that of money) how
the new method permits an improvement over the old solutions. I will
show the same thing in regard to credit, speculation, [property, income
tax] etc., in the El ements d economie politique appliqu ee and in the
El ements d economie sociale (239, 1889, 2, 36465).
7
At the end of
his life, Walras declared that he had worked during his entire career to
develop theories that were applicable to real problems (Walras 1909,
pp. 58183), to put the social question onto its true basis and, if pos-
sible, to solve it scientically (ibid., pp. 58081). In other words, he
believed his work was realistic; that was why his economic theory, he
contended, provides us the means of achieving in actual practice both
justice and economic objectives (ibid., p. 583).
V. The fth step: Introduction of more empirically
derived detail
The fth step Walras took in the development of his model was to assert
that once he had presented the essential elements of the economy in it,
7
It will be observed that Walras used the word El ements in that passage. Subsequently,
of course, he chose the word Etudes.
P1: PIG/... P2: IWV
it could be made progressively less abstract, and thus to correspond in
progressively greater detail to reality, by incorporating additional con-
ditions drawn from the real economy and from other aspects of society.
Walras explained that it was possible to do that because the fundamental
structure of his model was sufciently in conformity with reality (239,
1895, 2, p. 629). The idea he expressed there is that a theory will not
be valid or useful unless it has that quality; and, of course, determining
whether that is the case requires comparing its assumptions and con-
clusions with reality. Its verisimilitude made it possible for additional
empirically derived material secondary elements, as he called them
to be introduced into it. That, he wrote, allows economic questions to
be studied in mathematical form in any degree of detail that is desired
(ibid.).
The secondary elements, according to Walras, should be added: It is
appropriate to introduce into [my] conception one by one all the compli-
cations that reality presents (239, 1894, 2, p. 624, n. 5). He was referring
to economic details and psychological and sociological elements that he
had excluded from his abstract economic model (239, 1875, 1, p. 435;
Chapter 3). In the same sense, he wrote that we should pass from the
assumption of an annual periodic market to that of a permanent market;
that is to say from the static to the dynamic state. Why should that be
done? Walrass answer was, in order to bring us closer and closer to
the reality of things (123, 1877, p. 310; 322, p. 579). He was again
advocating that a progressively more detailed correspondence of a model
to the real economy should be achieved and thus again manifesting his
belief that theories should be judged by reference to the degree of their
consonance with reality. Moreover, Walras indicated that in order to
undertake applications of economic theory, it is essential to take into
account the ndings of other disciplines (141, July 1879, pp. 1617).
Economic theory is only the theory of social wealth. To pass from it
to applied economics and social economics, it is necessary to introduce
into the picture the physiological, psychological, individual, and moral
aspects of human nature. You must believe that is something I know
well how to take into account (239, 1874, 1, p. 366). He vehemently
rejected the charge that he tried to apply pure theory directly without
considering the aspects of society and individual behavior studied by
other disciplines: On the contrary. To give directions about the distri-
bution and production of social wealth by men in society, I have taken
care to develop the general theory of society and to combine the princi-
ples with those of the theory of social wealth (239, 1901, 3, 173). Thus
although he was on occasion (ibid.; 249.XIII, pp. 567, 568) critical of
P1: PIG/... P2: IWV
84 Walrass ideas
Paretos method of successive approximations as it referred to achiev-
ing progressively more realistic theoretical formulations (Chapter 9), he
agreed with the aspect of it that called for the involvement of disciplines
in addition to pure economic theory in the analysis and solution of real
problems.
P1: PIG
CHAPTER 3
Methods of evaluation of economic theory
I. The sixth step: Criteria for evaluations
In Chapter 2 it was shown that Walras was guided by his philosophical
and methodological views about economic theorizing in the steps that
shaped the development of his mature comprehensive economic model.
His sixth step was to develop the criteria he believed were valid for
judging the results of the preceding steps in his system of thought and
for judging the work of other economists, and to make those judgments.
Walras contended that all economic models (once it is established
that they are logical) should be judged by comparing their assumptions,
structure, contents, and conclusions with the real economy. He declared,
therefore, that his own model should be evaluated by comparing it with
the reality from which he abstracted it. His statements cited in Chapter 2
to the effect that his mature model of general equilibriumis in conformity
with reality are such comparisons. His belief in the necessity of evalu-
ating theoretical work and the criteria that he believed should be used to
make the judgments were consequences of his intellectual orientation,
part of his organized philosophical and methodological system.
Theorizing and reality
There is an interesting passage in the El ements (quoted in Chapter 1) in
which Walras declares that theory should nd real types in the real world,
base ideal types upon them, and then build the structure of scientic
reasoning on the latter, to which he adds the remark, not returning to
reality until the science has been constructed and doing so in order to
make applications. We will thus have, on an ideal market, ideal prices
that will be in a rigorous relation with ideal demand and ideal supply
(106, 1874, p. 32; 30, p. 53). Is that, as has been claimed (Bridel
1990, p. 183; Lendjel 1997, p. 70), evidence supporting the opinion
that Walrass rationalistic approach . . . excludes any confrontation of
the results of pure economics with reality for validation (Bridel 1997,
p. 142)?
85
P1: PIG
86 Walrass ideas
In fact, there are four ideas in that passage. The rst is an insistence
upon the necessity of basing the building blocks of a theory on empirical
knowledge, which is an explicit assertion of the necessity of constructing
a realistic theory. The only way to determine if a real economic type
and hence an ideal economic type is a proper representation of reality
is to compare it to the real economy. The second is the implication that
economic science is elaborated as a deductive process. In that regard
Walras was naturally aware that a given deductive chain is a closed
system. New empirical data differing from the original set would lead
to new assumptions, which naturally must be introduced at the level
of assumptions. Obviously, the new components contained in the new
assumptions cannot properly be introduced into the middle of a chain
of deductive reasoning based on the old set of components. Thus in
making his statement about economic theory not returning to reality
until the science has been constructed, Walras was referring to the
working out of the deductive structure once a given set of ideal types
and assumptions has been formulated. It was elsewhere that he described
the phase of laborious abstraction, formulation, comparison of the basic
components and assumptions with reality, and reformulation of them. It
was elsewhere that he referred to the phase of comparing the conclusions
of the theory to the facts, and to the phase of introducing more detail
into it. The third idea in the passage is that theories should be applied,
and that means that they must return to reality, and that implies that
they must be realistic. The fourth idea, expressed in the terminology
of ideal types, unfashionable today, is that the elements, relationships,
and functioning of economic models are clearly dened and precise,
an afrmation that, using modern terminology, any modern theoretician
would apply to his own models. Thus, far from supporting the notion
that Walras was an archrationalist, the passage contradicts it.
There is nevertheless a sentence in which Walras ostensibly contra-
dicted his view of the necessity of empirically testing economic theory.
He was reacting to a passage in which Cournot wrote that economic the-
ory, like mathematics, allows and stimulates the control of reasoning
by experience in all the parts that are susceptible of scientic treatment
(Cournot 1863, p. 16). Walras jotted at the bottom of the page on which
that statement appears that economic theory nattend pas des conr-
mations de lexp erience (Walras undated, in Potier 1994, p. 242, where
Cournot is also quoted). He meant that economic theory does not need
to be conrmed by experience, that the use of it does not need to await
conrmation by experience.
Walrass jotting is not representative of his thought. First, it makes
no sense internally. He did not think that economic theory would be
P1: PIG
Methods of evaluation of economic theory 87
developed and then neglected, that it would have no uses in reference
to the real economic past, present, or future. Indeed, his sentence, like
all his writings, indicates his view that economic theory is used. To use
it, however, is to apply it, and to apply it is to conrm it or contradict it
by experience. Second, the jotting is incompatible with Walrass philo-
sophical and methodological system of thought. He afrmed on many
occasions his belief that economic theories are conrmed (or refuted)
by empirical information and that if the theories are to be used for
applications, conrmation is necessary. He granted, however, that some
economic hypotheses cannot be conrmed because the phenomena to
which they refer are not measurable. His expression of those opinions
was part of his organized system of economic philosophy and method-
ology, whereas his jotting in Cournots book is a poorly conceived frag-
ment that lies outside of that system and is incompatible with it. This
will now be demonstrated in yet another way, namely by showing how
Walras believed that assumptions and conclusions should be evaluated
and how he actually judged examples of them found in the works of
various economists, including his own.
Logical experiments and historical evidence
Walras was fully aware that it may be difcult to determine whether
a particular real phenomenon exists under the conditions assumed in a
model. He knew that if a particular fact of the real economy contradicts
a conclusion of a model, it may not be because the model is wrong
but because the conditions assumed to be constant or not to exist in
the model are violated in reality by exogenous events and unknown
endogenous inuences. If reasoning on the one hand, and observation
and experience on the other hand, are contradictory, it is necessary to
try to bring them into agreement if it is desired to elaborate the science
with the use of the fact in question; but if the reasoning is correct, it is
not destroyed by that contradiction, because possibly it is the fact that
should be either more accurately identied or explained by exceptional
circumstances (154, 1880/1881, in 194, 1896, p. 338). He pointed out
that, In practice, that determination, namely of the causes of changes
of variables in the real economy, could be more or less difcult (106,
1874, p. 103; 102, p. 146), even with respect to impersonal facts
such as the quantities of commodities held by economic agents (ibid.).
Economic theory not only deals with such external experiences (203,
1898, p. 1), but, unlike the physical sciences, it also observes personal
facts (198, 1897 in 205, 1898, p. 270) intimate reality (203, 1898,
p. 1; in 205, 1898, p. 492). Personal facts are experienced by economic
P1: PIG
88 Walrass ideas
agents, and pose an even greater problembecause a given set of personal
facts is always associated with other personal facts so that economic
theory can never say with certainty if the set occurs because of the
other facts or although the other facts are present (198, 1897 in 205,
1898, p. 270).
Walras therefore conducted logical experiments: It follows that, in
this eld, deduction must always come to the aid of induction (104,
1874, in 205, 1898, p. 251). Regarding the determination of cause and
effect in empirical situations, that determination could be more or less
difcult; but nothing forces us to declare that it is impossible in theory
(106, 1874, p. 103; 102, pp. 14647). The outcomes of the deductive
phase of his modeling are the conclusions of his analysis. In his com-
parative static logical experiments, for example, he varied the quantities
of commodities held by the participants, or their utility functions, and
deduced the consequences for the equilibrium prices (for example, 106,
1874, pp. 13437; 176, 1889, pp. 15963; 13738, pp. 20914. 176,
1889, pp. 12728; 103, pp. 149, 151. 106, 1874, p. 104; 103, p. 151.
123, 1877, p. 294; 176, 1889, p. 309; 265, p. 430).
Walras then compared the conclusions about both impersonal and
personal economic facts with reality. His means of doing so was obser-
vation of them. Even at the beginning of his career, when he rst identi-
ed natural facts as the subject matter of economics, he declared that
it observes them with the greatest rigor (27, 1861, p. 92). It is true,
he recognized, that in some cases, Science pronounces truths that expe-
rience is unable to conrm, an example of which, according to Walras,
is the proportionality of prices to marginal utilities (239, 1893, 2, p. 573,
n. 8), but in that instance by experience he meant objective quantitative
methods. He thought that the proportionality, like all the facts studied by
economics, is nevertheless observable. Like satisfaction, he believed, it
is observable by introspection.
Employing the laboratory of historical experience, Walras explained
that economists nd periods in which, rst, the constants in the model
are constants or nearly so in reality and in which, second, the interesting
variable changed. The consequences of the change are then traced in the
historical record. The case can even be recognized in which the prime
cause of a variation of a price draws itself, so to speak, to the attention
of the observers (106, 1874, p. 103; 102, p. 147). An example, Walras
pointed out, would be the rise of the price of a commodity just when
a remarkable feature of the commodity is discovered or an accident
destroys part of the supply of it. One cannot do otherwise than connect
one or the other of those events withthe rise that occurred (ibid.). Insofar
as it is a question of examining statistical data, however, observation is
P1: PIG
not an easy process because the data may be erroneous or not complete
and therefore only approximately representative of the real situation
(104, 1874, in 205, 1898, p. 251). Additional examples of Walrass use
of historical evidence are given in the following section.
Walrass evaluations of his own model
Limitations of space dictate that only a few of Walrass evaluations of
the assumptions and conclusions of his model can be given here, but
they nevertheless constitute convincing evidence of his methodological
beliefs and, like his applied economics, demonstrate that he regarded his
model as dealing with the economy.
In discussing the behavior that establishes the existence of individ-
ual supply and demand functions, Walras wrote that it is like this that
things happen in the real world (106, 1874, p. 152; 151, p. 235);
and after further explanations, he repeated that assertion (ibid.). Like-
wise, the settlement of nancial claims between, for example, Brussels,
Amsterdam, and Frankfurt, is done in a market that acts as a vast clear-
inghouse, thus manifesting in the real economy the law of supply and
demand which regulates all these exchanges (106, 1874, p. 208; 316,
p. 538). Why did Walras choose to put in his model the condition that
entrepreneurs purchase productive services in the country in which they
sell their products? He did so, he wrote, because it conforms to real-
ity, and that was his evaluation of the scientic worth of the condition
1
(123, 1877, p. 258; 176, 1889, p. 243). He wanted to judge his analysis of
speculative nance, and he did so in this way: In conclusion, it remains
to take a look at the facts surrounding [that activity] in order to nd in
them, if possible, conrmation of our ideas (147, April 1880, p. 85;
emphasis added). He proceeded to examine the empirical evidence in
great detail. Walras stated as an abstract theoretical proposition, as a rst
approximation, that the offer of services by each worker is a function of
the prices of all services and products. Thus he used an ideal type, but
cautioned that the economist should not be dupe of his abstractions
(198, 1897, in 205, 1898, p. 275), by which he meant that the theorist
must evaluate his theories by examining reality. He proceeded to com-
pare the proposition with the conditions found in the real economy, and
showed that it is true in some occupations but not realizable in others
(ibid.).
1
Walras had written assumption in 1877 but replaced it with condition in 1889,
implying a less abstract relation to reality. He also changed the last part of the sentence.
He eliminated the passage in the fourth edition (see 249.VIII, p. 322).
P1: PIG
90 Walrass ideas
Walras compared his theory of prices and incomes to the empirical
evidence regarding the historical course of those variables, and judged
his theory favorably because it arrived at the reality of things (104,
1874, in 205, p. 257). Similarly, he evaluated the validity of two con-
clusions of his theory by comparing them with reality: This second
conclusion seems to me, like the rst, perfectly in conformity with the
surrounding facts (ibid., p. 251). He called upon empirical evidence,
upon a real fact, to explain apparent exceptions, namely the fact which
has already been sufciently described by me . . . (ibid.). He evaluated
his theoretical classication of the elements of social wealth landed,
personal, and articial capital and of the payments for their services,
by comparing his theory with the facts: It is, he wrote, the only clas-
sication that conforms to the nature of things and that is justied by
the comparative history of political economy and of modern industry
(21, 1861, in 249.VII, p. 142). He decided to study a particular aspect of
monetary behavior because, In the real world [Dans la r ealit e], that use
of a commodity as money has a certain inuence on its value . . . (106,
1874, p. 150; 150, p. 228). He judged by an empirical consideration
the justication of his assuming that money is used, as has been seen
(Chapter 2), asserting that he was thereby adopting the hypothesis con-
forming to the reality of things (106, 1874, p. 149; 149, p. 227). He
evaluated his notion of something that is both a numeraire and money
in this way: By using that notion, we bring ourselves more and more
closely to the reality of things because money is also the numeraire in
the real economy (106, 1874, p. 151; and, referring only to the numeraire,
in 210, 1900, p. 157; 151, p. 233). His concept represents the matter
as it takes place in reality (249.XI, p. 493).
As for the solutions of his systemof equations of general equilibrium,
Walras compared them with the real economy. That is how he judged
the scientic worth of the system: The sequence of events in the real
economy constitutes the empirical solution of this system of equations
(123, 1877, p. 365; 370, p. 651). He believed (erroneously, as will
be seen in Chapter 5), that the result of the ow of the interrelated
phenomena of reality is truly the empirical solution of his system of
general equilibriumequations (123, 1877, p. 365; p. 651), and asserted
many times that his model of tatonnement was realistic (Chapter 5, and
see Walker 1996, Chapters 12, 13, 14). He developed theories of pricing
of products and services, of the rate of net income and the pricing of
landed, personal, and capital factors of production, and of money prices.
Separately considered, these are all abstract theories, but which, he
wrote in judging their value by the criterion of their realism, upon inter-
penetrating each other, put us, by means of a methodical synthesis, into
P1: PIG
full reality (210, 1900, p. 172; 164, p. 254). His compared his mature
comprehensive model of general equilibrium with the real economy and
concluded, as noted in Chapter 2 in another context, that it was very
much in conformity with reality (239, 1879, 1, p. 628, n. 3), the image
of the economy of his time (179, 1891, p. 47; 191, 1896, pp. 46970;
appendix I, 4, p. 698). He insisted on la vraisemblance et la plausi-
bilit e the verisimilitude and the plausibility of his economic theory
(239, 1888, 2, p. 251). Verisimilitude is the quality of appearing to be
true or real. That was the case, he explained, because the model contains
components such as a market, supply, demand, a pricing process, and
equilibrium prices of services and products; and, he continued, these
concepts, it should be emphasized, are rigorously in conformity with
the facts, with observation, with experience. Indeed, thanks to the inter-
vention of money, the two markets, for services and for products, are
perfectly distinct in the reality of things, as they are from the viewpoint
of science (ibid., p. 230; p. 282; emphasis added). Walras was at pains
in that passage as elsewhere to indicate by the juxtaposition of the two
italicized words that they have distinct meanings. Concepts in that
context are ideal types; the facts are features of the real economy.
There are, for example, the facts of the real markets for services and for
products, and there is Walrass model of those markets, which contains
idealizations of them. Likewise, he judged his model of price formation
by comparing it with reality: It was a scientic way of taking note of
an observed truth (90, 1868, in 194, 1896, p. 30; emphasis added).
He supported his analyses of product differentiation, price discrimina-
tion, and monopolistic practices by references to empirical conditions
(123, 1877, pp. 38485; 384, p. 667). He compared to real markets his
assumption in a theoretical situation that all of the supply of a good is
offered at whatever price it may bring: This case, so simple, is in reality
extremely frequent . . . (106, 1874, p. 155; 153, p. 239).
Yet another example is Walrass discussion of the consequences of
certain rms increasing their production above the average quantity for
the industry. Those rms will make prots, whereas rms that produce
less than the average will make losses and go bankrupt. Thus the con-
sequence of xed costs has the enormous importance that it condemns
small entrepreneurs to disappear beneath the big entrepreneurs and that
it leads free competition to end up in monopoly. Walras then asserted:
Very well, if my theory concludes that such a process occurs, it is a
true theory because that process certainly happens (239, 1891, let-
ter 1027, n. 7). Thus Walras described an empirical process, compared
his theory with it, stated that his theory identied that process, and
then asserted that the reason his theory is true is that its conclusion is
P1: PIG
92 Walrass ideas
veried by the empirical evidence. If it had not been, he would have had
to change the theory to bring it into consonance with reality. He contin-
ued by asserting that free competition is not self-sustaining; there must
be an exterior power that maintains it. I will undertake to establish in
applied economics the consequences of that truth of economic theory.
In the meantime, in the domain of economic theory, I will not go beyond
my data (ibid.). Walras was writing about the real economy of his day,
about a theory based on the fact, as he believed it to be, that free com-
petition in the real economy is not self-sustaining, about a process that
certainly happens, not some metaphysical supposition. He was dealing
in his theory with a r ealit e sensible, the tendency in the real world for
freely competitive industries to become increasingly concentrated, not
with a hypothetical or ideal situation, and asserted that he intended, for
the bases of his theorizing, to stick to the factual information of which
he had knowledge.
Walras evaluated his theory of capital in the same way. It has been
seen that he claimed that he drew its assumptions from reality and
from experience, thus comparing them with reality (123, 1877, p. 302;
269, p. 434). He then did the same for the nished structure: It is
precisely what a theory of that nature should be: the faithful expression
and exact explanation of the phenomena of reality (176, 1889, p. XXII;
p. 19), a statement that he modied to read an abstract expression and
rational explanation of the phenomena of reality (191, 1896, p. XXII;
p. 19). Chapter 1 has shown that by phenomena in such a context he
meant concrete individual facts. The machinery, buildings, and produc-
ers goods he discussed in his theory of capital cannot be supposed to
be anything else. The fact of the division of labor is not a theory or an
abstract concept. It is a fact the evident existence of which is immedi-
ately apparent from a glance at society (106, 1874, p. 36; 32, p. 59).
Likewise, Walras compared his modeling of why money is held with
the real monetary economy and judged that the model reected it: This
conception conforms to reality, but is rendered rigorous by means of
scientic reasoning (210, 1900, p. 300; 273, p. 443). That conformity
was his justication of his modeling. As compared to barter, Walras
chose to adopt the assumption of the intervention of money, which is,
when all is said and done, closer to the reality of things. . . . In the reality
of things, the producer of wheat sells his wheat for money, the producer
of coffee does the same thing; and with the money obtained thereby, the
former buys coffee, the other buys wheat. That is what we are going to
assume here (106, 1874, p. 149; 149, p. 227). Judging his monetary
theory by comparing with the facts its conclusion about the relation of
the quantity of precious metal and its utility, he wrote: And, in fact,
P1: PIG
that is truly what happens, as demonstrated by experience (123, 1877,
p. 186; 176, 1889, p. 463). In the same vein, he declared that his model-
ing of the holding of assets reects real economic behavior because, In
reality, in a functioning society, a consumer, landlord, worker or capital-
ist knows very approximately what he needs to hold (ibid.). Economic
theory must explain the real economy, as, for example, the case of differ-
entials in the prices of foodstuffs in industrial and in agricultural centers.
Comparing his theory to the facts, We thus see that we have arrived,
quite easily, by reasoning and theorizing at the reality of things (104,
1874, in 205, 1898, p. 257). He compared his treatment of accounting
and inventory control with the facts and concluded that our theory of
production is well-founded on the nature of things (123, 1877, p. 235;
191, p. 289). His denitions dealing with the accounting practices of
the rm are established theoretically and practically (1877, p. 244;
198, p. 297). Walras presented a model with xed coefcients of pro-
duction, but he judged it by reference to the real situation, and declared
that in reality they are not and that he would take that into considera-
tion (123, 1877, p. 249; 2204, p. 305). Supporting the verisimilitude of
his theory, Walras noted that different parcels of land can have, as we
will see that they have in reality, greatly differing marginal utilities and
value (123, 1877, p. 219; 174, p. 270). Walras did not want to assume
that very large emissions of currency be made because he judged that
that supposition is not in conformity with the reality of things, and that
is why we assume an emission on a much smaller scale (148, 1879,
in 205, 1898, p. 355).
In summary, Walras believed that his mature comprehensive model
was a realistic abstraction, not a ction. Friedrich von Wieser agreed
with him:
It does not copy nature, but gives us a simplied representation of it, which is no
misrepresentation, but such as to sharpen our vision in view of the complexities
of reality. [It is] like the ideal picture which the geographer draws on his map, as a
means, not to deception, but to more effective guidance, he meanwhile assuming
that they who are to prot by the map will know how to read it, i.e. to interpret it
in accordance with nature (Wieser 1891, p. 108).
Walrass evaluations of the theories of other economists
It will now be shown that Walras demonstrated in yet another way his
belief that the assumptions and conclusions of theories should be evalu-
ated by reference to their consonance or dissonance with reality, namely
by confronting the theories of other economists with empirical evidence.
One example is his examination of David Ricardos theory of rent.
P1: PIG
94 Walrass ideas
Walras considered the proposition that innitely small increments of
capital result in an innitely small reduction in the rate of production.
Of course this might not be true, he remarked, but until it is shown that
this case is the general case or the exception, not by arbitrary assump-
tions, but by observations drawn from experience, we have the right to
reason theoretically on the basis of the former supposition (123, 1877,
pp. 34243). He was contending that the empirical evidence should
determine whether Ricardos assumption can be judged to be true or
untrue, and hence be retained or rejected. He kept that wording for
twenty-three years, and then, in 1900, modied it to read: . . . the gen-
eral case, by reasoning or by experience . . . (210, 1900, p. 404; 354,
pp. 62425). Thus he allowed that reasoning logical thinking to form
conclusions or inferences could be used, which would have to be done
on the basis of the evidence and to establish the logical soundness of the
argument. He continued nevertheless to indicate that experience that
is, new empirical research should be undertaken to determine if the
proposition is consonant with reality. In another instance, he evaluated in
this way the hypothesis of the English school that value depends upon
labor: The observation of facts formally contradicts that hypothesis
(15, 1860, p. L). What was Bastiats cardinal fault regarding the English
school? It was not illogical qualities of his reasoning. It was that he
wanted to make others accept those consequences that were the most
contrary to the reality of the facts (106, 1874, p. 160; 160, p. 249).
Walras judged T. R. Malthuss theory about the potential rate of increase
of population by noting that it is an estimate that is less than, rather
than greater than, is true in reality (123, 1877, p. 317; 328, p. 593).
He was comparing the theory to the facts, and evaluating it by reference
to them. He followed the same method regarding Malthuss contentions
about the rate of increase of the means of subsistence: They are not
founded in either reason or experience (ibid., p. 317; 328, p. 594).
The error of Condillac and J.-B. Say in regard to utility was that their
denition of utility is too broad. This can be seen by examining the facts:
It attributes value to things that, in reality, do not have it (106, 1874,
p. 157; 157, p. 245).
Walras noted that it sufces to compare J. S. Mills second and third
categories of commodities with the real economy to see that they can-
not be innitely multiplied (ibid., p. 337; 349, p. 616). The English
economists have a formula for the determination of the average rate
of wages, but is it so determined in reality? (ibid., p. 357; 365,
p. 643). An examination of the facts of the economy reveals that it is
not (ibid., pp. 35758; 365, pp. 64344). Walrass criticism of the
English economists analysis of special productive services is based on
P1: PIG
a comparison of the real economy with their classications. Examples
of real occupations, he asserted, such as the services of artists, singers,
doctors, and surgeons, refute their analysis. Walras also confronted their
analysis with the facts that different products come from different types
of land, that grapes are grown on some types and wheat on others, but
that the land could nevertheless be used to produce alternative crops
(ibid., pp. 33435; 347, pp. 61314). A comparison of the Physio-
cratic theory with the real facts shows that it is erroneous because it
failed to recognize that the income from the land in towns is a true net
product of industry and commerce (ibid., pp. 32728; p. 605). The
Physiocrats conclusion about the economic behavior of the sterile class
is wrong. How did Walras arrive at that judgment? Was it because he
discovered logical faults in the deductive structure of the theory? No,
it is wrong because it violate[s] the reality of the facts (123, 1877,
p. 326; 338, p. 604). Walras argued that it is therefore observation
and experience that confounds and ruins the hypothesis of Rousseau and
of all the XVIIIth century of the existence of the state of nature anterior
to and superior to the social state, at least from the economic point of
view, the only one that we are concerned with here (253, in 249.XII,
p. 121). With the assumption of trading bodies, Jevons abandons the
realm of reality and places himself on the terrain of ctitious means.
Walras preferred to stay with what is realistic: Wanting to remain, so far
as we are concerned, on the former grounds, we cannot accept Jevonss
formula except for a special case (176, 1889, p. 190; 163, pp. 252
53). Likewise, Walras believed that he had refuted the English theory of
interest by following the same procedure that is, not by proving that
it is illogical but by nding that it conicted with the situation in the
real economy. It is very certain that, in the reality of things, a given
individual can have two or three roles (123, 1877, p. 228; 184, p. 280),
and a realistic approach must recognize that the roles are distinct. The
English theory, he argued, fails to distinguish between the entrepreneur
and the capitalist, economic agents who are clearly different in reality.
Some entrepreneurs borrow the funds that they need for their rms and
therefore are obviously not capitalists, and there are many capitalists
who are not entrepreneurs (123, 1877, p. 359; 366, p. 645). Walras
clearly thought that he had not failed to make the distinction, that the
entrepreneur in his model was correctly based on the entrepreneur found
in the real economy.
2
J. S. Mills theory that concluded that the system
2
It has been contended by one analyst that the entrepreneur is not a real personage in
[Walrass] epistemology. He is an ideal type and an analytical principle . . . (Lendjel
1997, pp. 7879). That provides another example of the consequences of neglecting
P1: PIG
96 Walrass ideas
of tenant farmership was benecial appears wrong in the light of casual
empirical data, and, Walras wrote, some facts, more precise and more
general, will make it apparent how the results of experience agree with
the conclusions of [my] theory (254, in 249.XII, p. 557). Walras judged
the conclusions of applied economic science by the same criterion. For
example, contemporary applied science reduces to the formula Do
nothing! Unfortunately that conception receives fromreality every day,
under our very eyes, overwhelming refutations (169, 1886, in 205,
1898, p. 124).
What more clear and eloquent statements of the interplay of theoret-
ical economics and empirical evidence through time could Walras have
made than in the following passages?
Industry must furnish facts to economics; economics must accept these facts,
generalize them, draw laws from them, and from these laws, deduce rules
for the use of industry. In a time when occupations become more and more
specialized, when the division of labor is unlimited, it is more essential than
ever to maintain the communication of practice to theory, of theory to prac-
tice, of facts to principles, of principles to facts; otherwise intelligence and
consciousness go to one side, material activity to another, and the harmo-
nious unity of human destiny which alone creates its energy nds itself, for
the time being, disrupted. Industrialists are assuredly wrong to be disdainful of
economic teachings: by being so, they condemn themselves to routine activities
and light-heartedly place themselves in unfavorable conditions of production and
of sales; but the economists are even more ill-advised to not x their eyes atten-
tively on industrial developments: it is by so doing that they narrow the scope
of science instead of enlarging it, going backwards when it is necessary to go
forwards, closing themselves up in antiquated systems and, neglecting the great
problems, linger on questions of detail which they often do not even succeed in
understanding (21, 1861, in 249.VII, pp. 15152).
Similarly, We have seen how modern industry had served the theory of
value in exchange by introducing into reality [dans la r ealit e des choses]
all the kinds of social wealth, and in constraining science, in a way,
to recognize, to enumerate, to describe those kinds successively; how
modern industry had served the theory of property in the same way . . .
(21, 1861, in 249.VII, p. 149).
When Walras wrote about the facts that industry must furnish to
economics and the inuence of industry on economic theory (ibid.,
p. 137), it is evident he was referring to ordinary industrial facts, not to
abstractions, concepts, or essences. The same is true of his remarks
the link between ideal types in Walrass economic model and the economy of his day.
It is true that the entrepreneur in the model is an ideal type, but it has the quality of
ideality abstractness and generality precisely because Walras abstracted it from a
real type, namely real entrepreneurs.
P1: PIG
about industrial practice as the basis of theory, and the application of
accounting procedures to agricultural production:
We have here a striking example of the manner in which theory and practice
should aid one another; for it is certain that industrial practices, expressed by
accounting methods, can successfully serve to establish the theory of production;
and it is equally certain that that theory, once it is established, can serve no less
felicitously to express agricultural practices by means of accounting methods
(123, 1877, p. 238; 194, p. 293).
Walras was certainly not glimpsing a metaphysical relation between
economic theory and industrial reality.
II. Applications of economic theory
Rationale of applications
Walras realized that applications of pure economic theory to normative
or applied matters may involve considerations drawn from psychology,
sociology, political science, and other studies (Chapters 2, 4). It might
be thought that applications therefore do not test his basic economic
model but rather a wider, policy-oriented construction; or it might be
thought that there is a difference between testing hypotheses and testing
the basic model. Walras did not agree with those views. He thought that
if applications showed that a wider economic model is true, then they
showed that its pure theoretical core the basic model is also true, and
he thought that tests of its hypotheses are tests of it.
Walras believed that economic theory can be applied to practical
problems that, indeed, it is a prerequisite for any economic and social
applications (239, 1874, 1, p. 443; and see 205, 1898, p. 454). Reject-
ing archrationalism, therefore, Walras recognized that theory has to be
realistic in order to reveal the aspects of the economy that are sus-
ceptible to manipulation by policies, and to explain the functioning
of the economy so that policies could be designed that could modify
it. Theorizing about land use, for example, must satisfy both reason
and experience. To achieve that, land use must be studied closely
to ascertain its real characteristics; if these are supposed arbitrarily,
without relation to reality, the resulting science is pointless, without
any possible application (249.XI, p. 362). One can construct a sci-
ence for its own sake, Walras noted, and make no practical use of it,
But one will see that these truths of pure economic theory provide
the solutions to the most important, strongly debated and least clear
problems of applied and social economics (106, 1874, p. 32; 30,
pp. 5354).
P1: PIG
98 Walrass ideas
Social economics
Acting upon his methodological conviction, Walras used his positive
economic theory as a basis for policies to achieve his notion of social
and economic justice. He introduced a normative scheme in the eld
of argumentation, utopian speculation, and prescription that he called
social economics (194, 1896). Policies should be based on sound theory.
For example, so long as economists are content purely and simply to pro-
claim laissez-faire, laissez-passer as a sacrosanct dogma, their attacks
on the socialists are no better founded than the opinionated proclama-
tions of the socialists themselves (114, 1875, in 249.VII, p. 303). Walras
observed that in order to have a basis for preferring one type of market
structure rather than another and to devise sound economic policies, or
simply out of scientic curiosity, it would be necessary to study the
natural and necessary effects of the various different possible forms of
organization of society (123, 1877, p. 369; 371, p. 656).
Walras believed that the policies he recommended would establish
the proper functions of the individual and of the state, and solve the
social problem, namely the problem of the poverty of the working
class (see Jaff e 1980, p. 532, n. 10; and Jaff e 1983, p. 347, n. 10). The
policies included the purchase of all land by the state, state ownership
of natural monopolies, laws to ensure the continued existence of freely
competitive markets, a respect for individual ownership of non-land
property and private income, and the abolition of income taxes (194,
1896). Thus his recommendations constitute an extensive program of
reformulation of economic institutions.
3
Applied economics
Walras also wanted to improve the functioning of the real economy,
which led him to applied economics, the third subject matter in his
aforementioned well-known triad of economic studies. Applied eco-
nomics is, he explained, an applied natural science dealing with the
relationships between people, not as moral beings, but in their role as
workers performing tasks and thus having relationships to things. It is
3
Walrass opinions about a preferred society were strongly inuenced by the philo-
sophical outlook of the scholars of his century, which was naturally in turn inuenced
by the ideas of previous scholars. His fathers notions about natural law, derived from
Hugo Grotius and Samuel von Pufendorf (Jaff e 1977, p. 29, and Jaff e 1983, p. 103),
led L eon to believe that there are inherently true principles of economic justice that
spring necessarily from the nature of things and of economic interrelationships. He
adopted Etienne Vacherots opinion that taxation is unjust and Victor Cousins belief
that differing socioeconomic interests can be reconciled.
P1: PIG
concerned with material well-being and expediency. In the majority of
his economic writings approximately 60 percent (Chapters 6, 8)
Walras used economic theory to explain the workings of the economy,
to discover features of the economy upon which policies can operate to
solve empirical problems, and to predict the results of parametric and
structural changes. Applied economics seeks the most favorable con-
ditions for agriculture, industry, commerce, and credit, or the theory of
the production of wealth (74, 1867, column 115, in 194, 1896, p. 30). It
enunciates the rules of usefulness (205, 1898, p. 453) and prescribes
the rules of conduct (141, August 1879, p. 246) to achieve economic
efciency. For example, maximizing the output of the economic system
increases material well-being; applied economics is concerned with how
to achieve that outcome.
Economic theory can and should be the scientic basis of policies.
Walras deplored, in the work of contemporary economists, the divorce,
more and more marked, between economic theory and concrete reality
(220, 1910, p. 1, column 4, in 239.VII, p. 516). Developing and apply-
ing pure economic theory is the way to arrive at practical answers to
economic problems (141, July 1879, pp. 1522).
4
For example, because
it is not self-evident, according to Walras, that the consequences of free
competition are the most desirable for society, one of the purposes of
pure theory is to furnish the analytical means for applied economics
to inquire into whether free competition promotes the general welfare.
Applied economics is thereby enabled to study the effects of free com-
petition in agriculture, industry, trade and credit (114, 1875, in 249.VII,
p. 301). It is because of the lack of a clear distinction between pure
theory and applied theory and the lack of a rigorous analysis of the
4
Peter de Gijsel (1989) notes Walrass assertions that his general equilibrium theory
provided the positive analytical basis necessary for his policy formulations, which jus-
tied in Walrass mind the term scientic socialism as a description of his approach
to remedying the ills of society. Roger Koppl (1995) believes that Walras paradox-
ically thought that his general equilibrium model was both positive and normative,
although Koppl was unable to point to any statement of Walrass to that effect. Koppl
tries to support his thesis by taking the indirect route of maintaining that a resolu-
tion of the paradox that he alleges exists is made possible by an understanding of
the ideas of Etienne Vacherot, who argued in favor of the reconciliation or synthesis
of seemingly contradictory doctrines. Koppl contends that Walras, inuenced by that
notion, effectuated a synthesis of positive and normative ideas in his El ements. Using a
special denition that makes normative ideas hardly distinguishable from positive
ideas, Koppl maintains that Walras considered all sciences to be such a synthesis,
which directly contradicts what Walras wrote many times (Chapter 1). Nevertheless,
as is true of other writers and ultimately setting aside his own thesis, Koppl regards
Walrass general equilibrium model as a positive foundation, dealing with what is, for
his explicitly normative conceptions of states of affairs that did not exist but that he
thought should exist.
P1: PIG
100 Walrass ideas
principle of free competition that political economy was mired, accord-
ing to Walras, in so unsatisfactory a state. It is obvious, Walras remarked,
that applied mechanics, for example, cannot be undertaken without a
prior development of pure mechanics, but it has not always been as
clearly seen that the practical elds of knowledge such as natural law,
agricultural, industrial, commercial, and nancial economics, are pre-
ceded and governed by an entire group of pure sciences, namely meta-
physical and humanistic philosophy, human physiology and psychol-
ogy, history, criticism, pure economic theory (ibid., pp. 1516). Thus
applied economics and social economics, which dominate politics and
business, are themselves dominated by pure economic theory, philos-
ophy, and historical studies (ibid., p. 17). It is necessary to know
the natural properties of social wealth. We cannot command value in
exchange, as is true of gravity, except by obeying them. From this fol-
lows the necessity of constructing pure economic theory just as applied
mechanics necessitates pure mechanics (239, 1874, 1, p. 443; and see
198, 1897 in 205, 1898, p. 271). In regard to the social and economic
order the organization of industry and property Walras noted that
it could be altered. To do so, however, a theoretical analysis of the real
economic situation must be undertaken. Only in that way can theory
be of use in the development of applied science, which is necessary for
the wise guidance of actual practice (ibid.; 141, July 1879, pp. 1517).
It follows that economic theory must itself be sound, and to achieve
that condition, Walras followed other nineteenth-century scholars like
J. S. Mill, John Herschel, and Auguste Comte in believing that it was
necessary to apply scientic methods to it.
As other examples of real problems that need to be studied with the
aid of realistic theory, Walras considered whether mines, railroads, and
the emission of bank notes should be in private hands (ibid., pp. 18
20). Similarly, he examined the use of a commodity money in the real
economy, asking whether it is a good idea to have a unique or a double
monetary standard. That cannot be known without answering a prior
question of pure science, thereby learning the effects of the two types
of standard (ibid., p. 18). In that connection, Walras illustrated the dif-
ferences between economic theory, applied economics, and practice in
this way:
Howdoes the double use of a commodity as a commodity properly speaking and
as a mediumof exchange as money contribute to the total value of that commodity?
That is a question of economic theory. Is it appropriate to have two monetary
metals or only one? And, in the latter case, should it be gold or silver? That is a
question of applied economics. Finally, in a given country, at a given time, where
there is the double monetary standard of gold and silver, if silver is demonetized,
P1: PIG
what will be the probable inuence of that policy on prices? That is a question of
practical economic policy (114 1875; 249.VII, p. 305).
It must be admitted that in this particular example, what Walras called
applied economics and what he called practical economic policy appear
to be virtually the same type of approach, one of predicting the results
of the adoption of a policy.
The usefulness or expediency of economic policies does not establish
their justice or injustice (106, 1874, p. 16; 15, p. 38; 1898, p. 458), as
Walras showed with an example: If commodities are sold at high prices
to the rich and cheaply to the poor . . . there will be a great increase in
effective utility, which would be expedient. Normative considerations,
however, reveal that the goal to strive for is not simply maximum eco-
nomic welfare, but the maximum material well-being compatible with
justice. The fact that you may be hungrier than me does not confer
upon you, in and of itself, the right to eat my dinner (239, 1885, 2,
p. 50).
Nevertheless, the objective of both expedient and just policies is to
change the real situation in accordance with goals, and that takes time
and sustained effort. Walras expressed that obvious matter in the philo-
sophical terminology of which he was fond. He also, confusingly, used
the word theory in a few passages to mean a normative scheme. For
example, he wrote that in the realmof applied social science and practice,
It is an essential point of my doctrine that no economic and social theory
can be put immediately and completely into practice, and that true, cer-
tain and denitive progress always consists in a slow and patient routing
of reality towards the ideal (239, 1902, 3, p. 201). This is the case
because economic life takes place and has certain well-entrenched fea-
tures that develop prior to and hence independently of policies to reform
the economy. Money is minted; railroads are constructed; mines are
exploited; banknotes are issued; wealthis distributedamongindividuals;
the state is supported by the income tax. In such conditions, the facts
arise outside of principles and are contrary to them; practice is unlinked
with and even contradictory to theory. If policymakers come to believe
that those features should be modied to achieve various social goals,
Then, later, it is necessary to align the facts with the principles, to make
practice conform to theory (141, July 1879, p. 28); that is, conform to
the normative scheme. As is true of many others, these passages provide
more examples of instances in which what Walras meant by the facts,
in this case minting money, constructing railroads, exploiting mines,
and so forth, are not intellectual constructions indeed, are not anything
other than features of the real economy.
P1: PIG
102 Walrass ideas
Walras would have been happy, he said, to see questions of normative
and applied economics examined in a journal consecrated to practical
questions (239, 1902, 3, p. 202). Institutions, such as the property sys-
tem, should be established not only rationally but experimentally
(ibid.). He wanted idealistic tendencies to be tempered by a recognition
of what is pragmatically feasible, and practical policies, which introduce
a normative goal, to be based on sound positive theory:
We people of the Latin race are reproached for being to an excessive degree
theoreticians and idealists; and that reproach is well-founded: our revolution, our
socialism proves that. Let us therefore become practical and realistic. I would
very much like that to happen, basing our practical policies on a serious theory
and leading our reality toward a high ideal of justice and material well-being
(ibid.).
III. Applications and evaluation of mathematics
in economics
Justication
Whatever reservations Auguste Walras may have had about the use of
mathematics in economics, his declaration that economic theory is a
mathematical science (Auguste Walras 1831, p. 278), and that knowl-
edgeable persons rightly expect to see it one day ranked among the
mathematical sciences (ibid., p. 284), was the aspect of his ideas on
the subject that caught L eons attention and that became rmly rooted
in his mind. He therefore developed the method of using mathematics
to describe and analyze some of the features of his general equilibrium
model and to formulate and examine many real economic problems and
did so without formal training in economic modeling, without having
taken more than a fewmathematics courses, and lacking the company of
other similarly inclined economists. That method was one of his major
legacies. He was its principal proponent in the nineteenth century and
was directly responsible for it being used by his immediate successors
and indirectly responsible for its use by mainstream economic modelers
down to the present day. He expressed his belief that he was bequeath-
ing the mathematical method and a positive body of scientic work to
future generations of economists.
5
I have not, he wrote, the least
doubt about the future of my method and even of my doctrine, but I
know that success of this sort does not become clearly apparent until
after the death of the author (239, 1904, 3, pp. 25051).
5
He was not suggesting, as he had done very early in his career (239, 1859, 1, 3539),
that subsequent generations should subscribe to and rene his scheme for a just society.
P1: PIG
Walras thought that the opponents of mathematical economics do not
have a defensible case. To support their belief that the application of
mathematics to scientic subject matter is in general impossible, they
would have to show that Descartes, Newton, Fourrier, Amp` ere, Horn,
Clausius, Zeuner, and Tyndall were wrong in applying mathematics to
mechanics, astronomy, physics, and chemistry. That, Walras commented
ironically, would be difcult to prove (114, 1875, in 249.VII, p. 325).
Alternatively, if they grant that those applications were possible and
were made, then to sustain their position against the use of mathematics
in economics they would have to showthat economics is not a science of
quantitative relations between variables. He contended they cannot do
that because economics unequivocally is, to a large degree, such a sci-
ence (ibid.), a newbranch of mathematics (ibid., p. 291). It is, he stated
or implied, a physico-mathematical discipline (106, 1874, p. 31; 30,
p. 53; 114, 1875, in 249.VII, p. 304). He later came to believe that eco-
nomics is a psychico-mathematical discipline (205, 1898, p. 464), then
reverted to contending that it is entirely like the physico-mathematic
sciences (210, 1900, p. 29; 30, p. 52), and then still later stated once
more that it is a psychico-mathematical discipline (219, 1909, p. 2).
He was careful to qualify that by saying, Of course, it is understood
that we are not speaking of doing violence to the nature of things by
treating qualitative facts at all costs as quantitative facts (114, 1875, in
249.VII, p. 329). Moreover, Without a doubt, and even in that essen-
tially abstract, general and theoretical activity of analysis, there is a point
at which mathematical operations should be halted in recognition of the
multiplicity and diversity of individual preferences and idiosyncrasies,
and of the fact of free will (103, 1874, p. 102).
In Walrass opinion, inasmuch as many aspects of reality can be sym-
bolized mathematically and inasmuch as mathematics enables more pre-
cise deductions than does the literary method, the rational or deductive
phase of theorizing attains its clearest and most efcient form in math-
ematical formulations (ibid., pp. 32829). The employment of the lan-
guage, of the method, of mathematical truths has no goal but that of
pursuing a more rigorous, more penetrating analysis than that which is
permitted by ordinary language and thus obtains more exact conclu-
sions (239, 1874, 1, p. 371). He argued that the legitimacy and the utility
of the use of mathematics in economics has been principally demon-
strated to me by its results, that is to say by the complete clarication
of theory [formerly] wholly obscure: for example, the theory of value
(239, 1897, 1, p. 742). Walras predicted that the algebraic and geometric
representation of fundamental economic variables was destined to trans-
formpolitical economy just as it had transformed mechanics (114, 1875,
P1: PIG
104 Walrass ideas
in 249.VII, pp. 31011), and he continued until the end of his life to insist
upon the parallels of economic theory with the mathematical formula-
tions of pure mechanics (219, 1909; 249.XI, annex III, pp. 62550).
One objection to mathematical economics, Walras noted, is that the
general public cannot understand it. Nevertheless, the use of mathemat-
ics in economics should not be abandoned on that account, because it is
essential for the theoretical study of many questions and because of the
important practical respects in which it is useful (114, 1875, in 249.VII,
p. 328). Walras exhorted his readers to rejoice in the mathematization of
the parts of economics that can appropriately be treated in that way (114,
1875, in 249.VII, p. 329). Even though the mathematical theorems and
proofs of pure economic theory may be incomprehensible to the gen-
eral public, the conclusions of economic research, he contended, can be
expressed in a literary fashion. Non-economists then have no difculty
in understanding them and hence in understanding the bases of policy
recommendations (ibid).
Theoretical uses of mathematics
Walras examined the ways in which mathematics can be used in eco-
nomic theory. In that case, the coefcients are not numeric and the
mathematical functions are general (ibid., pp. 31112). In economic
theory, the formulas not only can but must remain indeterminate, in
order to remain at the same time general and permanent (ibid., p. 306).
The mathematics has no goal other than the scientic determination
of the relations between magnitudes: quantity, utility and marginal util-
ity, demand, supply, the price of commodities (239, 1875, 1, p. 435),
and the initial endowments of commodities and resources (114, 1875,
in 249.VII, pp. 304, 31213). Inasmuch as cardinally non-measurable
variables can be treated mathematically, Walrass list, it will be noted,
included utility and marginal utility. His model deals with those vari-
ables and with supply and demand, establishing, he claimed, both the
existence of equilibrium and its stability:
The formulation of pure economic theory in the mathematical formis an operation
which essentially ought to consist in furnishing a system of equations based upon
the double condition of the maximum satisfaction of wants and of the equality of
supply and demand for services and products, and to show that the mechanism
of free competition leads precisely to the solution by tatonnement of that system
of equations . . . (239, 1889, 2, p. 365).
The use of the mathematical language and method, he claimed,
enabled me thus not only to demonstrate the laws of the establishment
P1: PIG
of equilibrium prices, but also to undertake logical experiments,
namely, to demonstrate the laws of the variation of those prices, and
to analyze the fact of free competition and in so doing to establish it
as a principle (123, 1877, pp. 36566; 370, p. 651). Thanks to the
concurrent use of the analytical formulation and of the geometric rep-
resentation, he wrote, we obtain both the idea and the image of the
phenomenon of the determination of prices on the market in the case
of the exchange of several commodities for each other; and, in achiev-
ing that, in my view, we at last possess the theory (179, 1891, p. 8;
appendix I, 4, pp. 69899).
Equations based on economic reality
Walras asserted that not only the verbal components of his model had
real referents, but the symbols and relations in his equations did also. He
was fully aware that mathematics can be used improperly, deluding the
economic theoretician into thinking that he is achieving a more precise
description of reality than is achieved with words, when in fact he is dis-
torting it. That defect of economic modeling, Walras explained, results
from using mathematical symbols and equations that do not have refer-
ents in the economic phenomenon (239, 1897, 2, p. 727) the func-
tioning of the real economy a procedure that he vigorously opposed.
He stated: I believe it is essential to focus on the real economy and by
proceeding in that way to avoid that great stumbling block to the appli-
cation of mathematics to economics, namely of putting something other
than economic reality into the equations (ibid., p. 728). His model, he
declared, has
. . . the advantage of noting mathematically all the phases of the economic
phenomenon: prices cried at random demands at these prices supplies corre-
sponding to demands raising or lowering of prices in accordance with the excess
of quantity demanded over quantity supplied or of quantity supplied over quantity
demanded, and, nally, exchange at equilibrium prices (ibid., pp. 72728).
Continuing with that theme, he indicated that the equations and their
associated theories or models are not the same thing:
I nd one thing very striking. It is that the economists who are mediocre mathe-
maticians, like Jevons, have produced very worthwhile theories, while the mathe-
maticians who are not very good economists, like Edgeworth, Auspitz and Lieben,
say many foolish things. From that, I conclude that it is essential to establish the
foundations of economic theory very solidly before building mathematical con-
structions on the foundations that are called upon to support them (239, 1890,
2, p. 404).
P1: PIG
106 Walrass ideas
What makes a theory foolish, Walras was explaining, is its failure to be
based on realistic foundations. It is ironic that, despite his methodologi-
cal convictions and his efforts, successful in many respects, to develop a
realistic, mature economic model, he did not build his equation system
upon that foundation insofar as that system did not take account of the
dynamics of the tatonnements in his model, as will be seen in Chapters 5
and 6.
Walras described himself as one of the economists who did not have
a high degree of mathematical skill. Advanced mathematical analysis is
necessary but, he confessed, I am, for my part, not much at ease with it.
I have studied treatises on the innitesimal calculus and I can read works
in which calculus is used, but, not having had much occasion to practice
differentiation and integration, I am embarrassed at trying to use them
myself (239, 1892, 2, p. 510). That did not disturb him. He believed
that, for an economist, the acquisition of mathematical prociency is
a matter of secondary importance, the knowledge of economic reality
being primary. He complained that criticisms of his use of mathematics
had been made by persons who had not read his work, and asserted that if
they were to study it, they would see that his economic equations differed
from his knowledge of the entities and functioning of the real economy,
and that he wanted the equations to symbolize the latter. Anyone who
reads his articles on money would recognize immediately, he wrote to
Jean Cheysson, an engineer and economist, that, I am as far as possible
from engaging in what you call purely mathematical abstractions and
speculations; that I am much more an economist than a mathematician;
and that I have never had recourse to mathematics except to give true
expression to facts and to economic relationships that are essentially
quantitative, but are always taken from observation or the most rigor-
ously conducted experimentation (239, 1885, 2, p. 64; emphasis added
to the last twelve words).
Walras emphasized again that the mathematical symbols and rela-
tions used in economic theories should correspond to real entities and
relations. He called upon experience for the notions of products and of
productive services, and so forth, and then expressed mathematically
the quantities possessed (114, 1875, in 249.VII, p. 314).
All these operations: choice of variables and of functions, adoption of notations
that are both simple and expressive; the use, as Descartes said, of geometrical
methods to better consider particular relationships and to represent them more
distinctly to the imagination and to the senses; the use of algebraic methods
to retain them in the mind or to understand several of them at the same time;
employment of the method of deduction or of synthesis; these are difcult opera-
tions for the success of which the essential thing is not so much having some fairly
P1: PIG
elementary mathematical knowledge but rather being in possession of economic
data that are always carefully drawn from experience (ibid., emphasis added).
The equations are not the model; they are the mathematical expres-
sion of some of its features; and the essential thing is to get the model
right because mathematical expressions of an erroneous model are of no
value. I have, he wrote, been able to develop all this theory, although
very unaccomplished as a mathematician, in my capacity as a good
economist, because the mathematical difculties are insignicant and
only the economic difculties are serious (239, 1897, 2, p. 728).
Mathematics and applications
Exploring the limitations of mathematics in connection with establishing
truth, Walras observed that it is possible to use mathematics in applied
economics only because equations can have real referents. The equa-
tions could not otherwise have empirically derived numerical coef-
cients. Applications, however, are restricted by the nature of the subject
matter (114, 1875, in 249.VII, p. 315). Regarding marginal utility func-
tions, for example, it is impossible to replace theoretical ones with ones
containing empirically derived numerical coefcients because utility is
not quantitatively measurable (ibid.). The use of mathematics is also
restricted with respect to measurable magnitudes by the limitations of
the available statistical techniques and data. During Walrass lifetime,
those limitations made it difcult to establish statistical demand and
supply functions, but he speculated that improvements of methods and
data would one day enable that to be done reasonably accurately (ibid.).
Indeed, he remarked, there were many economic phenomena for which
numerically specied functions could be established immediately for
the purpose of practical applications. That was true regarding actuarial
matters (ibid., p. 327). Likewise, such questions as the imposition of a
tax or an import duty could be studied using empirical equations, thereby
enabling the impact of those changes on various variables to be deter-
mined quantitatively (ibid., p. 317). To answer the third question posed
above regarding the demonetization of silver, numerical calculations
could be used in that case (ibid., pp. 306, 328). Such calculations would
provide information that would enable the formulation of economic poli-
cies that have sound theoretical foundations (ibid., pp. 31617). It may
even transpire, he argued, that the application of mathematics to eco-
nomics will one day include the substitution, in certain given cases, of
computations in place of the mechanism of the raising and lowering of
prices on the market (239, 1874, 1, p. 374). In all those considerations,
P1: PIG
108 Walrass ideas
it is patently evident that Walras was asserting that his model had ref-
erents in the real economy. It could not otherwise have been applied to
the functioning of real markets and especially not to the determination
of prices by mathematical operations on empirical supply and demand
functions.
Walras believed that, like geometrical gures, mathematical reason-
ing can be only approximately conrmed empirically. Emile Bouvier
expressed Walrass view in this way: To the objection that mathemati-
cal economists disagree among themselves regarding their own mathe-
matical results, L. W. had conceded the point and rested his case on the
observation that the question was never one of nding absolute truth,
but of whether mathematics could help the economist arrive at a better
approximation to it (Jaff e summarizing Bouvier 1901, in 239, 3, p. 183,
n. 2).
Walras was also aware that the use of mathematics can be unproduc-
tive, adding nothing to economic knowledge. Employing mathematics
for its own sake, he observed, results in sterile exercises. He remarked
that our pure mathematicians, who are nothing more than integrating
machines, give me the impression of being constricted intellects (239,
1901, 3, p. 149). He believed that he had avoided that trap by linking
mathematics and literary analysis, thereby creating a new and worth-
while mode of discourse. He wanted words to be used to describe the
institutional, technological, procedural, and behavioral foundations of
economic models, and mathematics to be used, where possible, to ana-
lyze economic interactions and deduce their outcomes. Economic theory
and applied economics should be developed by men with minds and a
culture that are at the same time scientic and literary; for only then will
they be subject to the true scientic method (141, July 1879, p. 27). He
praised that type of synthesis in the work of one of his correspondents:
By causing a mathematical theory of economics to be taught and thus
restoring the intimate alliance of science with letters, you will per-
haps save French intellectual creativity from decadence (239, 1901,
3, p. 149).
IV. Conrmation of hypotheses and conclusions
by experience
In accordance with those considerations, Walras insisted that economic
theories and models had to meet two tests. One is the test of reason
that is, to use sound thinking in forming inductions and assumptions and
sound logic in deducing conclusions. The other is the test of realism.
At every juncture, he turned to reality as the basis of his hypotheses
P1: PIG
and as the criterion of the verisimilitude of his conclusions, and he crit-
icized other economists theories that he considered to be decient in
those regards. Thus to Walras, reasoning had an empirical connection
not only at the stage of empirical verication but also at the level of
its foundations, inasmuch as it should be properly exercised on facts
to form inductions that correctly represent and generalize them. His
synthesis of the experiential and rational methods in economic theoriz-
ing, his applications of theory to real problems, his comparisons of the
assumptions and conclusions of his theories to reality, and his evaluation
of the theories of other economists by comparing them to the facts, all
reect the realist side of his intellectual orientation. To emphasize the
truth of those remarks, it is worth recalling his words: Theories formu-
late the laws of general facts or groups of particular facts and use a
deductive structure of equations; and it is in this way that the physico-
mathematical sciences which economic theory is entirely like (210,
1900, p. 29; 30, p. 52) rationally formulate laws that are conrmed
by experience (239, 1901, 3, p. 162; emphasis added). Regarding a
conclusion of his theory about the results of free trade, Walras asserted:
That rational indication is conrmed by experience (197, 1895/1897,
in 205, 1898, p. 293), and declared that observation and experience
conrm it (205, 1898, 261). Similarly, economic theory demonstrates
` a priori and economic history conrms ` a posteriori the effects of the
evolution of economies (ibid., p. 474). In a progressive economy land
rent rises, a circumstance that is demonstrated by reasoning and con-
rmed by experience (218, 1909, in 249.VII, p. 513). Walrass theory
indicates that the price of land should rise, and the fact that it has done so
furnishes us a striking experimental conrmation of our theory (154,
1880/1881, in 194, 1896, p. 304; and in 249.XI, p. 233). According to his
theory, the true character of the land registry is exclusively scal and the
record shows that history is a striking conrmation by experience of the
predictions of the theory (102, 1873, in 194, 1896, p. 400). Regarding
the fact that the land tax is not a tax on income but a co-ownership
of land by the state . . . we have constructed, in economic theory, an a
priori and rational proof of this fact; we add here that the a posteriori or
experimental proof is found in all the pages of history (253, in 249.XII,
p. 371; and see 102, 1873, in 194, 1896, p. 407).
We do not have to fear, Walras reassured his readers, that the
conrmations of experience are not joined to the indications of reason
in the matter of the French monetary system (205, 1898, p. 110). With
respect to the volume of legal tender bank notes, Experience agrees also
with reasoning regarding the necessity of having a mandatory circula-
tion to avoid catastrophes (254, in 249.XII, p. 647). Science properly
P1: PIG
110 Walrass ideas
speaking has fullled its proper function, in a country in which it is
believed that good principles are the basis of fruitful applications and that
there are no contradictions between theory and practice, when science
has determined those principles. . . . If the principles are not respected,
there is only error, confusion, disorder; and that is what the history of the
railroads would prove, in case of need, by way of conrmation . . . (205,
1898, p. 232). The liberty of the trade in cereals provides an admirable
experimental conrmation of the truths that economics establishes a pri-
ori (254, in 249.XII, p. 598). Regarding the system of corporations and
regulation of them, The examination that we will undertake shortly of
that system will add the conrmation of the experimental proof to the
demonstration by reasoning that we are making here of the superiority of
the system of liberty of work and trade (ibid., p. 475). Economic his-
tory is innitely precious; it furnishes economic theorists either with
points of departure for theoretical formulations the empirical bases of
hypotheses or with empirical conrmations of theories empirical
tests of their validity once they have been developed (198, 1897, in
205, 1898, pp. 27071). To test his ideas about nancial speculation,
Walras noted that to nish [the study of it], it remains for us to take a
look at the facts surrounding it in order to nd, if possible, the conrma-
tion of our ideas (147, April 1880, p. 85; emphasis added). Regarding
Jevonss theory of business cycles, Posterior observations have not only
conrmed the reality and the regularity of the phenomenon in question,
but also established that they are not national but rather international or
universal (205, 1898, p. 135). The mechanism of commercial credit
institutions is of a degree of perfection so well demonstrated by science
and so well conrmed by experience that there is no reason why it would
not serve as an exact and durable model for industrial credit institu-
tions (70, 1867, pp. 174445). Emphasizing once more the importance
of empirical conrmations of conclusions, Walras afrmed that justice
is conrmed by its concordance with material interests; rational mate-
rial interests are conrmed by experience; and until that conrmation is
obtained, science is not denitively achieved (198, 1897 in 205, 1898,
p. 270; emphasis added). Of course, it must always be remembered
that, according to Walras, theories, which arrive by rational deduction
at conclusions, have their bases in factual considerations.
V. Conclusion
Walras wanted to understand the behavior of the market system in
the economy of his time. Consequently, his principal objective as
an economist, during his mature phase of theoretical activity, was to
P1: PIG
construct a model of general equilibration and equilibrium whose ele-
ments were drawn, as he said, scrupulously fromreality. He contended
that using them enabled him, through a process of deduction, analysis,
and empirical verication, to develop an abstract but accurate represen-
tation of the real economy of his day, and therefore one that reected
and predicted its outcomes. In Chapter 2 and in this chapter, an account
has been given of the methods by which he thought he had achieved that
objective.
Walras neatly summarized three aspects of a competitive economy
and its study in the following way (114, 1875, in 249.VII, p. 298). There
is the fact of free competition as it exists in more or less imperfect
conditions. That was his characterization of the ordinary facts of the
real economy of his day, identied in Chapter 2 as the third step in his
system of economic thought. Next, there is the idea or the conception
of absolute free competition as it could possibly exist, excluding any
intervention of notions of utility or equity. That is his positive model,
identied in Chapter 2 as the fourth step of his system of thought. It
incorporates ideal economic types, free of the frictions and imperfectly
competitive elements that characterize some real markets. It is also the
modern conception of perfect competition. Lastly, there is the prin-
ciple of free competition as it should be constituted so as to satisfy
those notions. That is his applied and normative economics, concerned
respectively with utility and equity.
6
It is also exemplied in the work
of Vilfredo Pareto and A. C. Pigou, and in the new welfare economics
and the work of many modern economists.
The adherents of the archrationalism and idealism theses have con-
centrated on Walrass fourth step, his conception of absolute free com-
petition, developing a special interpretation of it that neglects the third
step, and thus neglects the connection that Walras insisted it had with
competition in the real economy. Jaff e, in his later years, concentrated
on the normative aspect and did not believe that Walras had worked on
the third and fourth steps. The truth is that Walras paid attention to all
the aspects of the study of competition and made clear the ways in which
he thought they were different and were related, declaring that there
is too much simplication. All these operations are necessary. Consider
free competition as a fact, as a concept, and as an economic and ethical
principle (249.XIII, p. 560).
This chapter and the two preceding chapters have shown, among
other things, rst, that when Walras referred to economic facts he meant
6
An excellent treatment of howWalras passed fromone to another of these four concepts
is given in Arena and Ragni 1994.
P1: PIG
112 Walrass ideas
real facts; second, that the components of his model were intended
by him to have referents in the real economy; and third, that according
to Walras, again with respect specically to economic propositions, the
truth of an assertion is established by logically correct reasoning and by
experience; in other words, that his philosophical and methodological
system requires comparison of models to reality in order to verify or
falsify them, on both the levels of assumption and conclusion. In the
light of the unambiguous and abundant evidence presented in this book,
contentions to the contrary seem very strange. If Walrass name were
not invoked in connection with them, it would be impossible to believe
they were written about him, so strongly do they contrast with the letter
and the spirit of his work. Walras emphasized in his creative and mature
intellectual phases that he was not an archrationalist in either principle
or practice with respect to his own theorizing or his evaluations of it, or
with respect to the criteria that he used to evaluate the work of others.
P1: KsF
CHAPTER 4
Human nature
I. Introduction
Walrass treatment of the general characteristics of human nature was an
application of his general philosophy and methodology to that aspect of
reality. His treatment of human nature in relation to economic situations
was an application of his economic philosophy and methodology to
that specic aspect of reality. His ideas on both of these matters had
many similarities with the ideas of scholars who preceded him, but he is
more responsible than anyone else for a particular conception of human
nature being imprinted indelibly on neoclassical economics and on the
vast majority of all modern economic models. It is interesting, therefore,
to study his ideas about human nature in general, and to examine how
he arrived at his conception of an economic agent and his justication
for it.
Walrass ideas about human nature have been studied by a number
of writers,
1
but there are some respects in which our knowledge of and
understanding of his ideas can be deepened, and this chapter undertakes
to do that. The numerous aspects of Walrass ideas considered here
include his recognition that human nature has many different facets, his
difculties in dealing with the notion of free will, his identication of
real and ideal types regarding human nature, the contrast between his
knowledge of human nature and his concept of economic man, and his
use of the latter in a series of steps of reasonings about the functioning
of his economic model.
Among other benets, the results of the inquiry contribute to a fuller
comprehension of Walrass approach to modeling economic behavior
and aid in an assessment of the scientic value of his models. He has been
thought to have had virtually the same ideas about consumer demand
and exchange as did W. S. Jevons and Carl Menger.
2
The study presented
1
See especially Dock` es 1996 and the references given in that book.
2
J. A. Schumpeter accepted that judgment in a general way but observed that the
agreement of his results withthose of Menger andJevons is as strikingas the differences
in their starting points and methods (Schumpeter 1910 in 1951, p. 77). Friedrich
113
P1: KsF
114 Walrass ideas
here of Walrass ideas on human nature provides the basis for judging
his similarity or otherwise to those economists on those subjects. The
study also enables a comparison of his concept of human nature with
the ideas on that topic of other neoclassical economists, such as H.-H.
Gossen and Alfred Marshall, and with the ideas of modern economists.
Fortunately, it is not necessary to infer Walrass conception of human
nature, because he discussed the subject explicitly and at length. He did
so at the beginning of his career as an economist in his work Recherche
de lid eal social (18671868, in 90, 1868), an important source for this
chapter. He evidently retained the ideas expressed there, inasmuch as he
republished it (194, 1896, pp. 25147) without changes when he was
62 years old. Another important source is his Esquisse dune doctrine
economique et sociale (205, 1898, pp. 44995), where the ideas he
expressed on the subject were unquestionably his mature ones, because
he wrote that text during the period 18931895. Three other important
sources for this chapter are his El ements d economie politique pure, his
correspondence (239), and some of his jottings.
3
II. Aspects of human nature
Distinction between animal and human nature
Like the other parts of his economics, Walrass treatment of human
nature was the outcome of his observations and vicarious knowledge of
both external and internal aspects of human behavior (203, 1898, p. 1).
Accordingtohim, tounderstandhumannature it is necessaryrst of all to
recognize the distinction between the characteristics shared by humans
and animals, and those reserved exclusively for the former. That enables
the identication of specically human behavior. Walras noted that a line
Hayek repeated that observation, writing that their work is so clearly distinct in
general character and background that the most interesting problem is really how so
different routes should have led to such similar results (Hayek 1934, p. 394). William
Jaff e (1976), however, developed the argument that the ideas of the three economists
on demand were signicantly different, thus initiating a durable controversy. The
controversy is reviewed and reconsidered in a recent article by Philippe Fontaine
(1998). From a different point of view, Roger Koppl (1995) arrived at the conclusion
that Walras should not be considered a typical neoclassical economist.
3
Most of Walrass jottings cited in this chapter are undated. I originally used typescripts
and photocopies of Aline Walrass transcriptions of them, identifying them by the
collection in which they are kept. Since I wrote this chapter, however, Walrass jottings
were published in 249.XIII, pp. 509622. Inasmuch as the originals are not readily
accessible, I have located most of the cited jottings in that source and have referred
to it. If I was unable to nd the jotting there, I have cited the collection in which it is
found; for example, the Fonds Lyon. The collections are described in the introduction
to the published version (ibid., pp. 5038).
P1: KsF
Human nature 115
of demarcation exists between the physical characteristics of humans on
the one hand and their intellectual and ethical characteristics on the other.
On one side of that line, a persons physiological and psychological life is
on a purely animal plane; on the other, it is distinctively human (90, 1868,
in 194, 1896, p. 101). It was not with the physical but with the moral
aspects of human nature that Walras was concerned.
Multiple facets
In any specic respect, Walras of course recognized that there are dif-
ferences in behavior in different societies and times, that humans in all
societies reveal themselves as such in a variety of ways. As everyone
does, he knew that there are differences in the personality of differ-
ent individuals, that different persons experience different sentiments,
passions, outlooks, and attitudes and are motivated in contrasting ways
(249.XIII, pp. 51966, passim). He noted the sadness of humans and
their melancholy (ibid., 1862, p. 589). Some people are perfectly snob-
bish (ibid., p. 546; and see p. 515). Some try to increase their wealth
and improve their social position at the expense of others, and some are
obsequious if they nd that being so is useful for obtaining favors (ibid.,
p. 524). Some persons use force and treachery to appropriate wealth
(106, 1874, p. 40; 35, p. 62). There is attery, intrigue (249.XIII,
p. 515; and see p. 521, no. 19), trickery, honesty, vanity, pride (ibid.,
p. 616), and avarice (ibid. p. 529). There is unintelligence and ill-will.
Stupidity and jealousy (ibid., p. 614). Some individuals are incapable,
lazy, spendthrift; others are skilled, hardworking, and economical
(205, 1898, p. 473). There is intolerance, mediocrity, and foolishness
(249.XIII, pp. 614, 620). Some individuals undertake the necessary tasks
of life, while others depend upon the generosity of others (205, 1898,
p. 473). The world is full of people who do their duty too much or
too little (jotting, Fonds Lyon). Walras also observed that humans are
artistic, scholarly, moral, and industrious, and therefore sociable and
progressive (194, 1896, p. 190).
Malleability
Did Walras believe that human nature is eternal and everywhere identi-
cal (Dock` es 1996, p. 31)? He did not, if by that is meant that a persons
nature is xed, innate, and independent of his personal history, society,
and culture. Human nature, he believed, is malleable. Behavior changes
in a given society over time and differs from one society to another
according to the conditions that form the personalities of its members
P1: KsF
116 Walrass ideas
(194, 1896, pp. 189). There are differences relative to place and time
(ibid.). Sadness and melancholy, for example, are not intrinsic features
of human nature but rather an effect of the iniquity of social laws, that,
plunging some into insurmountable poverty and others into excessive
wealth, do not furnish anyone the conditions of true happiness, which
actually consist of material and moral well-being achieved by work
(249.XIII, 1862, p. 589).
According to Walras (writing in an era when instinct psychology was
widely accepted), a part of human behavior is inuenced by instincts.
That, he believed, is the part dominated by the sensations and sentiments
that have their source in self-love (194, 1896, p. 107)). He also empha-
sized, however, that humans are not ruled exclusively by instincts. They
experience a class of disinterested emotions unknown to animals and
can therefore behave in ways that are not selsh (ibid.). In explaining
some of the respects in which human behavior transcends instinctive
behavior, Walras evidenced again his belief that human nature is mal-
leable. His basic position was that human nature can be modied because
it has the non-instinctive components and because humans can control
their instincts. For example, religion can change human nature. No
one, he declared, should believe that I misjudge the role of Christian
spiritualism in human behavior. I understand that role, and I admire it.
The spiritual aspect of behavior has raised persons and citizens above
the animal level by the struggle of the free and responsible personal-
ity against instinct; the moral human being has created himself through
Christianity (249.XIII, p. 588). Through the renovation of society
we will arrive at the renovation of individuals (ibid., p. 581). Scien-
tic knowledge and reasoning can also change human characteristics:
astronomy . . . has done the most to draw up mankind from its cow-
ering state aficted by fear, and to raise it to the rank of a being that
understands and dominates the physical universe (ibid., p. 577).
Walras noted that the abilities and accomplishments of human beings
are not the automatic results of innate characteristics. The fact that a per-
son may be born with a high degree of intelligence, for example, does
not guarantee that he or she will create important scientic or artistic
works, which rather are the products of effort and study (jotting, Fonds
Lyon). The full development of human faculties comes about through
the effort that is required by the difculty of making a living, and in con-
sequence, through hardship (249.XIII, 1862, p. 589). In this context,
Walras added the curious remark that nature gives us intelligence, but
we give ourselves talent and genius by an effort of free will (ibid.,
p. 583), an idea that is contrary to the truth. Talents and genius are inher-
ent characteristics of people who are so endowed, and, in recognition
P1: KsF
Human nature 117
of that fact, Walras elsewhere mentioned talent . . . received . . . as an
inborn characteristic and not acquired through energy, persistence and
work (249.XIII, 1862, p. 589).
Essential unity in behavior
Despite his remarks about the diversity of the aspects of human nature,
Walras thought that underlying the differences relative to places and
times, there is an essential unity in human nature (194, 1896, p. 189). He
must have encountered that idea in many of his readings. David Hume,
for example, asserted:
It is universally acknowledged that there is a great uniformity among the actions
of men, in all nations and ages, and that human nature remains still the same,
in its principles and operations. The same motives always produce the same
actions: The same events follow from the same causes (Hume 1777/1975,
p. 83).
4
Walras thus shared to a signicant degree the belief of the writers of the
Scottish Enlightenment that certain human traits were timeless and uni-
formand thus could be anticipated or predicted (Redman 1997, p. 186).
Walras had reference to general propensities, drawing his knowledge of
them from his experiences and his readings. He concentrated his atten-
tion on the behavior of people in Western societies but also mentioned
others. He declared that in all times and in all places among which
he referred to Africa, India, and Tahiti as well as European countries
you will nd humans associated with humans and will nd them more
or less likely to be concerned with that which is beautiful, that which is
true, that which is useful, and that which is good (194, 1896, p. 121).
They have economic and ethical relationships with each other (ibid.,
p. 126). All these statements are truths of observation and of experi-
ence. It is observation that teaches us that humans are never found in a
solitary state, without relationships with other humans (253, in 249.XII,
p. 120).
According to Walras, other universal aspects of human nature are
an esthetic sense, revealed in artistic efforts and poetry, and a capacity
for reasoning and understanding, expressed in language and science. Yet
another is the exercise of free will in the development of the institutional
arrangements of private property, the family, andgovernment (205, 1898,
pp. 45758; 194, 1896, pp. 1012, 104). Walras discussed at length
4
If Walras studied the books he owned, he must have become familiar with Humes
philosophical notions. They are discussed in the works of many philosophers and
numerous histories of philosophical thought that were in his library.
P1: KsF
118 Walrass ideas
the facets of civilization science, art, industry, and so on that are
expressions of human nature. All aspects of human physiology and of
human psychology, he declared, are manifested in an aptitude for the
division of labor and in the personality of humans (194, 1896, p. 128).
The essential unity is therefore manifested in the exercise of human
faculties in art, science, industry, and ethics (ibid., p. 189).
These considerations reveal the dual character of Walrass account
of human nature. He remarked upon the existence of differences in
personalities, genetically transmitted characteristics, and motivations.
Those are differences in the nature of humans. It turns out, therefore,
that he was making a distinction between human nature as manifested
in a certain class of traits he believed to be shared by all humans, and
other traits of human nature that vary from one person to another. To say
that some of those behavioral traits are part of human nature and others
are not would not be tenable, so Walras wisely chose to say that there
is an essential unity in human nature, rather than to make the stronger
categorical statement that it is everywhere and in all epochs identical.
That allowed him to maintain that there are similarities on the general
level of motivations, interests, and preoccupations, and also differences
in the nature of individuals in specic respects.
III. Free will
Walras took several positions on the question of whether or not free
will is an aspect of human nature. One such position was that humans
have freedom of choice. Pure materialism, he asserted, is absurd.
Not to see in the life of humanity, in the work of society, in history, and
in progress anything more than an exercise of physical functions and a
series of physiological facts is anerror (249.XIII, p. 591). Determinism
is metaphysics, not science (ibid., p. 599). He contended that there is
in the universe a force which knows itself and possesses itself: that is
human willpower (106, 1874, p. 18; 16, p. 39). In their economic
activities, humans do not obey the blind inevitability of natural forces,
but rather the motivations of human will (194, 1896, p. 124). It is
free to be exercised in the realm of production as in the realm of the
distribution of social wealth (106, 1874, p. 38; 34, p. 61) and is
aware of being exercised and can act in several ways (106, 1874, p. 19;
17, p. 40). It is a force that is clairvoyant and free (106, 1874, p. 18;
17, p. 39; and see 194, 1896, p. 141).
5
5
See Potier 1994, pp. 227, 229, 235, 249, 26768, where that idea of Walrass is
emphasized.
P1: KsF
Human nature 119
Nevertheless, Walras took another position on the issue of free will.
He thought that the universe obeys the laws of cause and effect, and
accordingly he subscribed to the opinion that free will is the inevitable
illusion of a being who is at the same time a product and an agent [an
effect and a cause], and who is aware of himself as a cause, but not
as an effect (Edmond Sch erer quoted in 205, 1898, p. 451; the words
in brackets were inserted by Walras; emphasis added). Accepting that
notion with some reservations (249.XIII, pp. 562, 569), Walras described
the experimental fact of the illusion of human liberty as a fact of
unique importance (205, 1898, p. 452). In one instance, seemingly
not wishing to commit himself denitively on the matter, he identied
the fact of freedom or of its illusion as one that is indispensable for
understanding the sciences (ibid., p. 453).
That did not solve the problemfor Walras. He continued to be troubled
by what he believed to be the seeming incompatibility of the ideas of
free will versus determinism, andreturnedtothe subject numerous times,
arguing in favor of one or the other of the views or trying to reconcile
them.
First, he proclaimed that free will is not an empirical fact but a notion,
a creation of the mind intended to describe and explain the apparent
liberty of choice. In this respect, Walras was inuenced by Etienne
Vacherots view that experience is a synthesis of objective and sub-
jective elements, and accepted Vacherots opinion that a subjective ele-
ment is contributed not only by the mind but also by the imagination
(205, 1898, p. 460). Accordingly, Walras contended in a youthful jot-
ting, persons withfree will, moral person[s], workers dividingtasks are
representations of the imagination rather than unintellectualized phe-
nomena that are objectively present in the universe ((249.XIII, p. 557)).
Repeating his phrase about the imagination, he wrote that the concept of
humanity possessing free will is one of the ideal types, perfect, absolute,
precisely because they are creations of the mind, not without a relation
to reality, but without a correlative object in reality. Perfect circle, free
will, two representations of the imagination (ibid.).
6
Second, in his phase of intellectual maturity as a theoretician, Walras
adopted the contrary point of view that the existence of free will is an
empirical fact: I have always taken the fact of the free will of mankind
as fact of experience not a concept or an essence without giving it
either more metaphysical value or less scientic value than the facts of
6
Walras used the words representations of the imagination again in connection with
science toward the end of his life, but in that instance made no mention of free will
(213, 1905, p. 1, column 1; and in 237, 1962, p. 61; see Chapter 1).
P1: KsF
120 Walrass ideas
vegetation and of life (194, 1896, p. VI; and see 249.XIII, p. 562). He
praised the ancient Greeks for recognizing the facts of experience in all
their diversity, and among them, the fact of the free will of humanity . . .
(ibid.).
Those two approaches of Walrass to the problem were uncharac-
teristically ill-assorted. Not much weight can be given to the youthful
jotting just quoted, however, inasmuch as Walras maintained categori-
cally in writings published in his subsequent mature intellectual period,
as has just been seen, that free will is a fact of experience. He used the
word fact in the modern sense as something found in objective reality.
Furthermore, the jotting is unconvincing as an argument that free will
is a creation of the imagination, because it also concedes that free will
does have a relation to reality, which implies that it has a referent in
reality. Moreover, Walras afrmed even in his earliest writings that the
bases of scientic constructions are ideal types drawn from real types
identied as existing in reality, and are not creations of the imagination
(see Chapter 1). Indeed, Walras argued in his mature intellectual period
that perfect circles have referents in reality, namely the imperfect real
circles fromwhich they are abstracted as idealizations (106, 1874, p. 32;
30, p. 53). Perhaps in his youthful note he simply wanted to say that
the representations of the imagination are categories that are used to
identify and classify objective facts, that is, used to erect a scientic
edice.
Third, Walras claimed that the problem of free will versus determin-
ism is resolved by identifying a crucial difference between the physical
world and the moral activities of humanity. The former, he maintained,
is subject to the law of causality. Things and natural forces do not have
the awareness of acting, and even less than that are they able to act oth-
erwise than they do (106, 1874, p. 19; 17, p. 40). They are blind and
inevitable (106, 1874, p. 19; 17, p. 39). Humans, on the other hand,
are not subject to that law. A human is a free and reasonable being, that
is to say a moral person, as distinct from all the other beings which, not
having reason nor being free, are only things (194, 1896, p. 33). What
is a moral person? Walras dened him and her by their characteristics:
a superior awareness that links a human to his destiny, a superior intel-
ligence which enables him to understand it, free will which permits him
to involve himself with it; these are the three faculties which cause a
human to be a moral person (194, 1896, p. 141). Walras added that
. . . in the presence of the action of causes which impinge upon him and the
reactions which his faculties call into being, a human raises himself progressively
above that action and that reaction by the awareness of his unity, of his identity.
He dominates his emotions and his ideas and he dominates his decisions. He is
P1: KsF
Human nature 121
aware of himself, he knows himself and he possesses himself. He has an objective
and he pursues it, but he does so knowing that he has an objective and that his
obligation is to pursue it (ibid., p. 115).
Walras repeated that it is for those reasons that we say that a human
being achieves a clairvoyant and free destiny, and that he is not a thing
but a person (ibid.). A problematic aspect of Walrass third line of
reasoning is that it reduces to an assertion that free will exists without
disposing of the notion that the phenomena that make up the universe,
among which are human beings, obey causal laws.
Fourth, Walras decided to preserve the notions of both free will and
determinism with respect to human behavior. His effort to do this took
the form of maintaining that although free will exists, it is not operative
with respect to all human behavior, that perhaps no more than 30 or
40 percent of human decisions are free. Humans are able to tread only a
narrow path of volitional liberty, a path bounded on each side by strictly
causally determined phenomena. Perhaps that force, the free will of a
human being, does not know itself and does not possess itself as much
as it supposes (106, 1874, pp. 1819; 17, p. 39). The freedom of
choice of humans is subject to a superior determining principle. Their
liberty consists only of making their conduct conform to the demands
of the principle that reective thought enables them to identify (194,
1896, p. 124; 1990, p. 110). That is the principle of utility or of concern
for material well being (ibid.).
Walras then repeated, with some variations of language (ibid.), that
whereas things and natural forces are subject to the law of causality,
the decisions of humans are not completely subject to it. The problem
that he struggled with was not vanquished by that argument, however. It
is a supposition, not a scientic demonstration. If a persons decisions
are responses to events and situations, either exterior to or within the
person, and are inuenced by personal, social, and biological history as
recorded by the individuals physiological structure, and if the decisions
are the manifestations on the conscious plane of a series of chemical
and electrical cellular processes that produce neurological changes and
cerebral effects, why would the decisions not all be strictly determined
consequences; that is, the consequences of causes?
Economists used to wrestle with the question of the relation of
free will specically to economic science, and they produced differ-
ent answers. In Karl Marxs opinion, for example, free will makes it
impossible to achieve a scientic study of much of social and economic
behavior. Sound methodology, he believed, restricts the domain of sci-
ence to the study of mechanical causation and hence makes it impossible
P1: KsF
122 Walrass ideas
to explain free teleological behavior scientically (Marx 1858, p. 46;
1867, 1, p. 10). J. S. Mill, on the other hand, thought that human behav-
ior is causally determined, so he had no difculty in coming to the
conclusion that social science, including economic science, is possible
(Mill 19811991, 8, pp. 835, 898).
In the light of those considerations, Walras nally adopted a position
about reality that, he believed, resolved apparent contradictions between
the ideas of the autonomy of human will and determinism so far as con-
cerns economic science. Like Marx, he regarded pure economic theory
as applying to the domain of causal relationships, and the domain of
free will as lying outside of economic science, but he also believed that
a great deal of social and economic behavior is amenable to scientic
analysis. Walras therefore disagreed with critics who declared that his
economic models were invalidated by the operation of free will. They
argue that one of the factors in the determination of the price in free
competition is human free will, whose decisions cannot be predicted
mathematically (176, 1889, p. 252; 222, p. 334). The critics were
incompetent, he contended, to pronounce on that matter: What a type!
The cretin of the Institute [of France], the gentleman with the rosette
of the Legion of Honor, accumulating fourteen academic appointments,
knowing beyond that neither a word of psychology nor of mathematics,
who says to you that human free will does not allow itself to be put into
an equation (249.XIII, p. 533).
Walras responded that the criticism is based on an incomprehension
of what he did with mathematics in his model. It may be, he conceded,
that the force of free will, which is exercised regarding various matters
prior to the formation of prices, cannot be expressed mathematically.
He granted that mathematical operations in economic theorization
can be pursued only up to the point at which they should be halted in
recognition of . . . the fact of free will (103, 1874, p. 10). Mathematical
economists are not upset by that circumstance, however, because it does
not invalidate what they are doing (114, 1875, in 249.VII, p. 326). Walras
pointed out that we have never tried to predict mathematically the
decisions made by humans with free will; we have tried only to express
their effects mathematically (103, 1874, p. 10).
Free will, Walras contended, may act in the economic decisions made
during pre-market behavior, but once those decisions have been made,
market processes unfold independently of themand hence independently
of the agents will. In an economic model, he argued, it is irrelevant
whether preferences are expressions of free will or deterministic forces.
He explained that economists put the question of free will aside, in
order to investigate the economic effects of preferences described by
utility curves. People may have no inuence over the utility derived from
P1: KsF
Human nature 123
commodities, or limitless inuence, or a limited degree of inuence,
but, he contended, those possibilities have no relevance for his scientic
methodbecause that methodonlycomes intoplayexactlyat the moment
when the inuence of individual preferences, idiosyncrasies, and free
will ceases (114, 1875, in 249.VII, p. 326). No matter how they are
determined, he pointed out, the utility curves and hence the demand
curves are what they are. When they change, the effect is that prices
change. That is discovered and explained by economic theory, which is
therefore
. . . focused completely, not on human free will, but on its effects. Does the degree
of freedomof will prevent its effects fromobeying natural and mathematical laws?
Not at all. Whether or not you are free to throw a stone, something that you are
certainly not free to do is to cause that stone, once you have thrown it, to fall
otherwise than in obedience to the law of gravity. It is just the same in that regard
for value as for gravity (ibid.).
Furthermore, Walras pointed out, the existence of free will would not
render predictions of behavior impossible. Aggregates of people follow
some uniform and hence statistically predictable patterns of behavior,
and individual behavior is predictable with respect to such uniformities
of it as exist. Those who oppose mathematical economics in the belief
that it is incompatible with free will are inconsistent because they use
birth-rate and suicide-rate curves and make predictions based upon them,
even though free will appears to operate in individuals decisions about
having children and committing suicide:
It is argued that the transformation of social science into a mathematical science
transforms human beings into machines, etc., etc. . . . But truly, it is we ourselves
who thus learn, not without astonishment, that birth and death rates completely
escape the inuence of human will. We believe we are at least as free to marry
or not, to prolong our lives by a healthy life style or abbreviate them by bad
habits, as to exist without houses, without furniture, without clothing and without
food. We believe we are even free to commit suicide, if we so choose, which
does not prevent statistics from indicating that there is, on average, only a certain
percentage of us who will use the liberty to do so. This reveals clearly enough
that even the effects of the free will of humanity are not entirely outside of the
purview of prediction and calculation (114, 1875, in 249.VII, p. 327; and see
249.XIII, pp. 53435).
Walras nally abandoned further consideration of the problem and
thereby arrived at a position that is scientically superior to that of either
Marx or Mill. He did not want to base his economics on the notion of the
existence or the absence of free will. He declared that whether free will
is accepted as a fact or considered to be an illusion is of little importance.
Scientically, he maintained, to say that free will is the illusion of a
being who is aware of himself as a cause but not as an effect, is to
P1: KsF
124 Walrass ideas
say nothing (205, 1898, p. 451). Natural scientists are not constrained
in their research by the question of free will and engineers build dams
without thinking about it. Economists and politicians are not intellectu-
ally paralyzed by the problem of its existence or nonexistence and can
similarly design models and apply economic policies without worrying
about whether or not their activities are the uncontrollable unfoldings
of a determinate universe (ibid.). In accordance with the principle that
economic science does not need to consider the questions raised by the
issue of free will or the illusion of it, and should not do so, Walras
declared that for my part, I used the needs [utility] curves as facts [on
which I base my reasoning] without investigating the inuences that the
nature of things and human will could respectively have (239, 1874, 1,
p. 435). All social science can be based on the fact of moral personality
without knowing what free will is (249.XIII, p. 557). Economic sci-
ence, Walras correctly concluded, can exist regarding much of economic
behavior whether there is free will or not.
IV. Society and the individual
Critique of individualism
Walrass idea of a human being shows that he did not belong without
reservation to the school of individualist thought, if by that school is
meant the belief that the characteristics of society are without exception
the products of the motives and ideas of individuals. He asked: Is it thus
also in the individual, or in the group, that the force is found which at
times throws nations off the path of progress and into political reactions,
and at times precipitates them into war and revolutions? He answered:
We believe it is only by making abstractions that it is possible to con-
ceive of either the individual without the state or of the state without
the individual, and that humans in society, individuals in the state, is
the only reality (194, 1896, p. 91). Concrete unintellectualized reality
in those regards is constituted of individuals who are integral parts of
society. Humans function in a societal context, so the inuence of the
latter upon individuals must be taken into account.
The lives of all persons in a society are made interdependent by the
activities they undertake to satisfy their needs (106, 1874, p. 36; 32,
p. 59). The division of labor implies the interdependence of workers, and
indeed the aptitude of humans to undertake work in that way enables
them to earn their livelihood and is therefore the condition that makes
possible their existence (ibid.). Likewise, withrespect toprivate property,
a subject to which he devoted much thought, Walras maintained that the
P1: KsF
Human nature 125
appropriation of things depends, it is well understood, not upon each of
us in particular, but on all of us in general. It is a fact of human life that has
its origin not in the individual will of each person, but in the collective
activity of the entire society (106, 1874, p. 39; 35, p. 62). Walrass
account of the history of appropriation does not, however, represent it as
originally being a consequence of collective activity and certainly not
of cooperation. The division of wealth at the beginning of civilization,
he contended, was determined by force, trickery and chance, although
nevertheless not totally divorced from rational conditions. The bravest,
the most vigorous, the most able, the most fortunate have obtained the
largest part, and the others have had the rest, that is to say nothing or
very little (106, 1874, p. 40; 35, p. 62). Even so, Walras believed
that the manner in which objects are appropriated depends on the stage
of development of the society and that societies have evolved in the
direction of a distribution of wealth based upon principle (ibid.).
Human nature as the basis of social phenomena
Despite those considerations, Walras subscribed to methodological indi-
vidualism insofar as he thought that some aspects of human nature are
not formed by social interactions but are instead fundamental in the
sense that they give rise to various forms of social behavior. That viewis
a foundation for Walrass opinion that humans manifest the same gen-
eral features in all societies at all times. There are, he contended, three
inherent characteristics of human nature that inuence the character of
society: human will, need, and wants (205, 1898, p. 451). These are
the causes of the moral and economic activity of humanity (ibid.), and
as such must be studied in order to understand it.
Contending again that aspects of human nature inuence society
rather than the other way around, Walras sought in the former what
he called the principle of the distinction between art and science, and
the principle of the distinction between industry and custom (194, 1896,
p. 128). Likewise, he believed it is in the study of human nature that
we must look for the principle of the concordance between art and sci-
ence, and the concordance between industry and custom (ibid.). Addi-
tionally, human nature determines certain important features of history:
There will be antagonismor harmony in human destiny depending upon
whether there is antagonism or harmony within human nature (ibid.).
To decide which of those alternatives exist,
Let us therefore return to the question of human nature: to physiological nature,
that is to say to the aptitude for the division of labor, and to psychological nature,
P1: KsF
126 Walrass ideas
that is to say to amicable and beautiful love, to understanding and to reason,
to free will; and let us see whether all these faculties exist like metal springs
functioning in opposing ways, or like the well-organized parts of a smoothly
operating mechanism (ibid.).
The bases of human nature and therefore of human behavior are to be
found in the nature of the human brain and body.
V. Deductions about society from an understanding
of the general features of human nature
A theme found in the writings of many scholars is that the characteris-
tics of society and hence social science can be deduced from an under-
standing of human nature. For example, the eighteenth-century Scottish
scholars, including Adam Smith, proposed to arrive at a knowledge of
human nature by using empirical methods in the spirit of Bacon and
Newton. . . . With certain principles of human nature established, pre-
dictable outcomes, conduct, and behavior could be deduced as typical
or natural (Redman 1997, p. 114). Through a study of human nature,
philosophers could identify key customs and institutions on which to
build a science of man (ibid., p. 113). It has been seen in Chapter 1 that
J. S. Mill also believed that there are invariable laws of human nature,
and that discovering them made possible a science of society. He had
the Newtonian goal of formulating secondary laws of a science of soci-
ety derivable from the simplest primary laws of human nature (ibid.,
p. 325): The conclusions of theory cannot be trusted, he asserted,
unless conrmed by observation; nor those of observation, unless they
can be afliated to theory, by deducing them from the laws of human
nature, and from a close analysis of the circumstances of the particular
situation (Mill 19811991, 8, p. 874; emphasis added).
Walras was squarely in that tradition. He believed that inasmuch as
there are aspects of human nature that shape the development of society,
it is possible to deduce the characteristics of society from the char-
acteristics of human nature. Therefore, to establish the principal fea-
tures of the structure and functioning of society and the economy, he
argued, the rst thing that has to be done is to construct an accurate
denition of a human being (194, 1896, p. 122); in other words, to
achieve an understanding of human nature. The way to do that is to use
empirical research methods to examine real human behavior and then to
abstract the denition from reality: Thus, to develop the economic or
moral theory of society in general, and to develop it in an a priori and
rational fashion, it is rst necessary to obtain the denition of a human
by abstraction from experience (194, 1896, p. 189). The human, not
P1: KsF
Human nature 127
the human-animal, but the human properly speaking who is the theatre
of humanistic facts, has been revealed to us both by external or phys-
iological experience and by internal or psychological experience (82,
1868, column 214, in 194, 1896, p. 119). Walras accordingly observed
human behavior in the real world, identied the important features that
were common to different individuals, and discarded their idiosyncratic
features. We are therefore, he declared, in possession of a denition
of the ideal human extracted with care from a correct analysis of the
real human (194, 1896, p. 122). Comparing his denition with the real-
ity from which he drew it, he asked regarding it: Do we depart from
observation and experience? and he answered: No (ibid., p. 189). He
asserted again that he had arrived at the denition without leaving the
realmof either observation or experience, raising ourselves above reality
only by scientic abstraction (ibid., p. 190). Walras declared that from
that experimental denition, it remains only to deduce, by a series of
analytic judgments, all the theorems of social science (ibid., p. 122),
experimental meaning founded on empirical information.
Walras also joined the long list of his predecessors who appealed to
natural law and the moral character of humans in order to reach sound
scientic knowledge. As John Locke viewed this matter:
Natural law, the will of God, is a body of rules conforming to rational nature. In
order to show that rules of moral obligation could be derived from nature (i.e.,
human nature), ethics had to be demonstrable. And the demonstrability of moral
principles, thought Locke, is analogous to the demonstrability of mathematical
principles: while in mathematics reasoning starts with properties of gures and
numbers, in ethics it begins with the idea of man as a rational creature. (Lockes
view explained in Redman 1997, pp. 6869; see Locke 1954, p. 54.)
In the same manner, Walras argued that scholars could examine care-
fully the scientic results that can be reached by starting with the moral
personality of humans, on the grounds of the good old natural law, if
only they employ a little thought and reasoning (194, 1896, p. VII).
It is not surprising that many of the results Walras deduced from his
understanding of human nature were the ones he preferred. For example,
he concluded that private property, an institution of which he approved,
is one of its consequences. The appropriation of scarce things or of
social wealth is a humanistic and not a natural fact: it has its origin in the
exercise of human will and activity, and not in the forces of nature (106,
1874, p. 39; 35, p. 61). The next step in Walrass reasoning is normative.
On the basis of his principles of moral philosophy, he introduced the
idea that human nature is a legitimizing condition. Its consequences are
morally and thus legally right: The right of private property is founded
on the fact of the personality of humans (194, 1896, p. 33; emphasis
added to right). In other words, humans naturally want to own things,
P1: KsF
128 Walrass ideas
and a rule is meet and proper if it expresses a natural proclivity, at least
in the case of property rights. It is a natural law, Walras contended,
that humans own themselves and therefore their faculties; thus a human
is also, by natural law, owner of the things that are part of social wealth
for which he has exchanged his labor (ibid.).
VI. Homo conomicus
Human nature in Walrass economic models
Walras brought together the results of his inquiry thus far in this way:
Fromthe analysis of human nature and of the classication of the social
world that we have established . . . , it follows that there are two groups
of facts and social relations: economic facts and economic relations, and
moral facts and moral relations (ibid., p. 123). The economic phenom-
ena obey and express what is useful or what is of material interest and
are studied by economics, a social science. The moral phenomena obey
and express the idea of justice and are studied by a moral science, which
is social science properly speaking (ibid.). Walrass treatment of human
nature in relation to economic phenomena will now be examined.
What is the facet of human nature that Walras extracted from a
correct analysis of the real human (194, 1896, p. 122) and featured in
his economic model in a setting drawn from a correct analysis of the
real economy? He identied it by distinguishing three aspects of human
nature in this way:
In fact, the person who has needs, who takes part in the division of labor and who,
in order to maximize satisfaction of his needs sells services and buys products
in quantities such that their marginal utilities are the reciprocals of the poten-
tially exchangeable quantities of services and of products, lhomo conomicus,
is also the person who is endowed with sympathy and an esthetic sense, with
understanding and with reason, with a will that is conscious and free, lhomo
ethicus; and they are both the person living in society, cultivating art, engaging in
science, following customs and undertaking industrial activities, in short, lhomo
cnonicus (205, 1898, p. 450).
The abstraction that Walras featured in his economic model is homo
conomicus, just as has been done by most economists before and after
him. His use of it does not mean that he abandoned consideration of
economic reality. He argued that deductions from human nature cannot
be made in isolation from a situation in which humans act, and he was
not interested in speculating about the consequences of human nature in
articial situations. He therefore criticized the situations and the behav-
ior of the economic actors in the models constructed by H.-H. Gossen
P1: KsF
Human nature 129
and W. S. Jevons. They deduced the consequences of humans who desire
a maximumof pleasure with a minimumof trouble (163, 1885, p. 77),
just as Walras did, but their deductions were irrelevant because they did
not place their economic actors in a setting drawn fromthe real economy.
Perhaps, Walras commented, matters occur as those scholars describe
on Robinson Crusoes island or in an isolated or primitive society. Nev-
ertheless, they do not happen like that, I will not say in our actual social
and economic state, but in the abstract and ideal social and economic
state which is the one about which pure economics constructs theories
(163, 1885, p. 77). That statement provides a perfect example of the
meaning in Walrass writings of the word ideal in connection with the
nature of economic models. He was not referring in that quotation to the
ideal society that he preferred and that he wanted to see created. He was
alluding to the idealized situation that his economic models constitute as
a result of their being constructed with humans and markets abstracted
from the real economy via the identication of real types in it. In his
criticism of Gossen and Jevons, he was asserting that, unlike them, he
was not dealing with articial situations in which the interactions of
persons in a market economy are absent, but with the real economy on
the abstract level of economic theory.
Walrass homo conomicus mirrored the classical economists con-
cept. J. S. Mill, for example, asserted that political economy does not
treat of the whole of mans nature as modied by the social state, nor of
the whole conduct of man in society (Mill 1836/1844, p. 137) what
Walras called the conduct of homo cnonicus. Mill went on to say that
economic theory is concerned with him solely as a being who desires
to possess wealth, and who is capable of judging of the comparative
efcacy of means for obtaining that end the motivation and sagac-
ity of Walrass homo conomicus. Therefore economic theory predicts
only such of the phenomena of the social state as take place in conse-
quence of the pursuit of wealth (ibid., pp. 13738). Mill also remarked:
Not that any political economist was ever so absurd as to suppose that
mankind are really thus constituted . . . (Mill 1836/1844, p. 139). Sim-
ilarly, as has been shown above, Walras knew, just as everyone knows,
that humans have more goals than the simple acquisition of material
wealth, that human nature has many other facets. Mankind seeks not
only what is useful and materially satisfying, he seeks also truth and
beauty and justice (194, 1896, pp. 100101). Thus, as several writers
have pointed out (see, for example, Dock` es 1996, p. 80), it would be
incorrect and indeed, to use Mills word, absurd to suppose that Walras
believed that humans act solely as calculating mechanisms whose only
objective is the maximization of utility.
P1: KsF
130 Walrass ideas
Economists before and after Walras have given the same justication
for the abstraction homo conomicus. Mill, for example, explained that
it is used in economics because this is the mode in which science must
necessarily proceed (Mill 1836/1844, p. 139). Otherwise no general
propositions could be formulated and economic science could not exist
(ibid., p. 146). Economic theory deals only with the aspects of behavior
in which the acquisition of wealth is the main motive, so assuming that
to be the only motive in those connections, of all hypotheses equally
simple, is the nearest to the truth (ibid.). Walras justied the abstraction
homo conomicus in the same ways. First, he argued that it is a neces-
sary simplication, one without which there could not be a science of
economics, and he noted that the abstraction is of great importance not
only for economic theory but for applied economics also: In the latter,
we assume theoretically that humans always know their interests and
seek them; and practically it is necessary to act as though that were true:
it is the best way of arriving at the truth (jotting, Fonds Lyon). Second,
he justied it on empirical grounds, namely that it captures an impor-
tant part of reality: Personal material interest remains an individuals
great stimulus (205, 1898, p. 485). Personal interest should not be
completely eliminated as a human motive under the pretext that per-
sonal interest is not the only motive of human beings (249.XIII, 1891,
p. 611). On the contrary, like Mill, he contended that it is the motive of
behavior in economic contexts.
Walras rejected the charge that his use of mathematics in connection
with his model made it impossible for the model to take account of cer-
tain relevant aspects of human behavior. When Charles Renouvier, for
example, wrote to him that psychological, social, and other conditions
are of such a character as to introduce a difference between the pre-
dictions of mathematical economics and the establishment of economic
facts, and maintained that the difference is considerable (239, 1874,
1, p. 396), Walras replied that sound theoretical method necessitates
holding those conditions in temporary abeyance. He explained that in
his El ements d economie politique pure, exclusively a work of theoreti-
cal economics, I believed that I could abstract fromthe psychological,
social, and other conditions of which you speak, on the grounds that
they are disturbances of secondary importance (239, 1875, 1, p. 435).
Moreover, The economists who take issue with the right to abstract
from perturbing circumstances do themselves abstract from them every
day: it is a scientic necessity (249.XIII, p. 555). Walras declared that
he wanted to recognize such conditions but that he would do so only
in the proper context: It is only, it seems to me, when it comes to
the point of applying numerical calculations to accurate data that the
P1: KsF
Human nature 131
place is reached at which account of those conditions should be taken
(239, 1875, 1, p. 435). By psychological conditions in those passages
he obviously meant aspects of human nature other than those of eco-
nomic man, inasmuch as the motives and sensations of economic man
are psychological characteristics.
VII. Steps in Walrass economic reasoning about
preferences and market behavior
Walras followed a chain of reasoning that connects the characteristics
of economic man with the functioning of markets, explaining the links
between the phenomena of utility, of effective demand and supply, and of
the equilibrium price (123, 1877, p. 388; 387, p. 671). It is very clear
that in his discussion of those links he was referring to real people and the
real economy, and to abstractions that he drew from them. He used his
perceptions of human nature and his knowledge of the structure and
procedures of real markets to establish the links between the former and
the functioning of the latter, and he made his account of those matters
directly a central part of his model. In regard to those links, there is
therefore no distinction between his account of reality, carried out on an
abstract level, and his model.
The rst step
On the basis of his observation of personal facts, Walras rst asserted
that real persons have desires for commodities, that the satisfaction of
desires generates utility, and that there is a functional relation between
each commodity and utility. He featured that condition in his model:
Each transactor, in our theory, can be assumed to establish his curves
of utility or need in accordance with his preferences (176, 1889, p. 252;
222, p. 334). Different needs are felt to have greater or lesser intensity
in comparison to others, which inuences the real costs that economic
agents are willingto incur to procure commodities that satisfytheir wants
(106, 1874, p. 78; 72, p. 105). Following the principle that science
should not consider the questions raised by free will or its illusion,
Walras explained that for my part, I took the utility curves as facts
[used as my starting point]
7
without investigating the inuence that the
nature of things and free will may respectively have (239, 1874, 1,
p. 435).
7
The words in brackets are William Jaff es effort to transcribe nearly illegible writing.
P1: KsF
132 Walrass ideas
The second step
Walras then identied four properties that he believed utility to have in
reality, each of which are aspects of human nature:
A. He granted that utility is not a directly cardinally measurable mag-
nitude (114, 1875, in 249.VII, p. 315), but he nevertheless believed that
it is measurable and therefore that to treat it as such was not contrary
to fact. He justied that treatment in three ways. First, he explained
that for a mathematical expression used for the purpose of undertaking
a theoretical study it seems to me sufcient to suppose that utility is
measurable in economics by dening it as the intensity of the last want
satised, just as it is assumed that mass is measurable in mechanics by
dening it as the number of molecules or the quantity of matter in a
physical body (239, 1901, 3, pp. 17172). Second, he argued that utility
can be rendered measurable, as is done for temperature, by express-
ing it by an arbitrary function that always increases with it (Pareto-
Poincar e). That is advantageous for a numerical evaluation aimed at
practical applications (ibid., p. 171). Therefore, I assume, he wrote,
that there exists a scale for the measurement of the intensity of needs
or of intensive utility, true not only for similar units of a given type of
wealth but also for different units of the various types of wealth (106,
1874, p. 79; 74, p. 107). Third, he believed that once having used that
assumption, he was able to dispense with it. He wrote to Henri Poincar e
that:
You have made me very happy by explaining to me with the authoritativeness that
you command that I was justied in representing the satisfactions of individuals
by functions, even arbitrary ones, but always increasing with the satisfactions
represented, on the condition of eliminating those functions, as I effectively do,
as soon as I deduced from them the curves of demand and of supply together with
their essential properties (239, 1901, 3, p. 167).
B. Walras then adopted the hypothesis that the utility that a consumer
derives fromeach commodity is independent of the amount he consumes
of others (123, 1877, p. 246; 201, p. 302; 176, 1889, p. 99; 75, p. 111).
Walras recognized, however, that utility is not always a function of
a single variable (239, 1901, 3, p. 172), namely of each individual
commodity separately. I have studied and made notes regarding the
utility of interdependent commodities (and also of commodities that
are substitutes for each other) (239, 1892, 2, p. 498). Nevertheless,
he defended his abstraction. But I put all that aside with the goal of
simplication (ibid., p. 499). He regarded it as a simplication that
captures an aspect of consumer preferences, and pointed out that the
features of the total real situation that it left out could be taken into
P1: KsF
Human nature 133
account in a more detailed representation of reality: It is very simple to
assume that [utility] is a function of the quantities consumed of several
commodities. The utilities are then proportional to the partial derivatives
of the total utility function with respect to the quantities consumed of
the various commodities (ibid.).
When Walras assumed independent utilities, he did not mean that, in
reality and in his model, the maximumamount of money that a consumer
will pay for, say, a bottle of wine, depends only on the utility of the wine:
It depends also in part on the utility to the consumer of bread, of meat,
of clothes, of furniture; for according to whether that utility increases
or decreases, the maximum sacrice that will be made in exchange
for wine will diminish or increase (123, 1877, p. 38788; 387,
p. 670).
In technical terms, the maximum pecuniary sacrice that a consumer is disposed
to make to obtain a unit of a product is a function not only of the utility of the
product, but also of the utility of all the other products that are on the market, and
is also a function of the consumers nancial resources (123, 1877, p. 388; 387,
p. 671).
C. Walras initially focused on the case of continuous utility func-
tions, putting aside, also on the grounds of a desirable simplication,
the fact that some preference functions are discontinuous (239, 1901,
3, p. 172). He then explained what happened in real pricing situations
if demand functions are discontinuous, and made his model reect that
condition (106, 1874, pp. 6162, 13334; 52, pp. 8587, 133, p. 203;
1889, pp. 1069; 8384, pp. 12125; 1889, p. 156 and Fig. 5; 133,
pp. 2034).
D. Walrass fourth characterization of preferences was that humans
experience decreasing marginal utility, which he believed was estab-
lished, like his other ideas about preferences, by observation of real
interior and exterior facts: All these successive units have, for posses-
sor (1), an intensity of utility that decreases from the rst unit, which
lls the most urgent need, until the last, after the consumption of which
satiation occurs (176, 1889, p. 98; 74, p. 107).
The third step
Walras then contended that important aspects of real human behavior are
that individuals seek the maximum satisfaction of needs, or, the maxi-
mum effective utility (106, 1874, pp. 8687; 80, p. 116) from all the
commodities they consume, and that in many cases real economic agents
calculate rationally and accurately how to do so. He therefore featured
P1: KsF
134 Walrass ideas
that behavior in his model and assumed that other aspects of human
nature are not operative. Referring to his analysis of pure exchange in
real competitive markets, for example, he observed that taking as a point
of departure the utility of each of the commodities to each of the trans-
actors, expressed by curves that are decreasing functions of the quantity
consumed, I demonstrated . . . the condition of the maximum satisfaction
of needs (163, 1885, p. 68). Walras also believed that consumers in the
real economy think about their future needs and keep stocks of com-
modities and money awaiting the due dates of their rental incomes,
of their wages, of their interest receipts, or in order to buy new capital
goods (210, 1900, p. 300; 273, p. 443). Producers similarly await
the payment for the products they have sold (210, 1900, pp. 300301;
273, p. 445). He therefore assumed in his model that economic man
also maximizes utility by being economical and prudent.
Nevertheless, Walrass letters and jottings make it clear that, like
everyone else, he was perfectly aware that humans may be shortsighted
in regard to providing for their future and may make errors, be inde-
cisive, and be unskilled at formulating efcient plans for maximizing
prot and utility. He knew, like everyone else, that humans can be irra-
tional in the sense of behaving in a way that frustrates the achieve-
ment of their own consciously held goals, and that they may act with-
out understanding or being aware of their true motives. Walras turned
to reality to evaluate his theoretical considerations. He noted that, in
the real economy, complications such as imperfect abilities, imperfect
knowledge, and various frictions and obstacles often prevent the com-
plete attainment of economic plans and therefore maximum utility. Real
individuals are not absolutely selsh and thus do not strive to max-
imize utility to the fullest possible extent, and they are not perfectly
discerning, perspicacious, and wise. The preceding reasoning, about
monopolies, is based on the hypothesis, generally accepted regarding
economic questions, that private interests are both selsh and clairvoy-
ant. But that hypothesis, we have recognized, does not conform strictly
to reality. For egoists, private interests are always both; but for clair-
voyant persons, matters are different (196, 1875/1897, in 205, 1898,
p. 226). With respect to assumptions, it is very certain that it is neces-
sary to be careful when passing fromabstract theory to reality. In reality,
there are frictions in the economic mechanism; and, additionally, men
are neither perfectly egoistic nor perfectly clairvoyant (239, 1901, 3,
p. 167). Walras indicated that he introduced those considerations in his
applied and social economics, thereby treating a fully realistic situation
(ibid.).
P1: KsF
Human nature 135
VIII. The consequences of the characteristics
of economic man
The fourth step
Walrass identication of those consequences constituted the next link
in his account. Continuing to present it as a description of what happens
in the real economy and as a component of his model, he explained how
the utility curves and economic conditions give rise to the specic forms
of an economic agents demand and supply functions: Each owner of a
commodity who enters the market in order to exchange a certain quan-
tity of that commodity for a certain quantity of some other commodity
has his trading desires, either potential or effective, susceptible of rig-
orous determination (106, 1874, p. 59; 50, p. 82). Real consumers,
professional traders, and wholesale and retail merchants, he explained,
add to their holdings of each commodity, or sell out of their stocks of
commodities, until they hold the batch of commodities that maximizes
their utility. The buying and selling may be done as a reaction of the
individual to quoted prices, but these dispositions are no less real in
the potential and non-effective state with respect to prices that are pos-
sible but not quoted (106, 1874, p. 60; 50, p. 83). A person in the real
economy can predict all the possible values of p
a
, from zero to innity,
and consequently determine all the corresponding values of d
a
, express-
ing them in some manner, and therefore has a demand function (ibid.).
Walrass additional characterization of demand functions, also based on
observation of real behavior, was that any individuals demand for a
commodity is, in principle, a function of the prices of all commodities
(106, 1874, p. 111; 110, p. 161; 123, 1877, p. 247; 201, pp. 3034).
8
The fth step
Walrass objective was to demonstrate scientically how prices and all
other economic variables are mutually determined in freely competitive
markets (Chapters 2 and 5). Once having derived individual demand
8
John Creedy (1999) notes that Walrass individual demand curves are not partial
equilibrium constructions. In consideration of the interrelationships of commodities
and their prices, Walras identied general equilibrium demand and supply curves.
Creedy notes that those curves are treated in the work of C. F. W. Launhardt, and
that they have been independently rediscovered several times. Creedy examines their
derivation and suggests, in the light of their potential central role in the analysis of
exchange, that they could be very useful, particularly when employed with nonlinear
mathematical methods.
P1: KsF
136 Walrass ideas
functions, his next step was therefore to show the part he believed they
play in the functioning of real competitive markets and hence in the func-
tioning of the markets in his model: Having thus successively shown: 1
o
how current or equilibrium prices are determined by the demand curves
and 2
o
how the demand curves themselves result from the utility and
the quantity of the commodities, I revealed the relationship which links
utility and the quantity of commodities to their price on the market
(163, 1885, p. 69; 194, 1896, pp. 35152). Walras was not asserting
that he created the relationship in a model by making postulates; he was
stating that it exists in reality and that he revealed it.
Putting together his observations of the motive of utility maximiza-
tion, the demand functions, and the existence of markets and their pricing
procedures, Walras explained that the condition of maximum satisfac-
tion [is] always the determining condition of the supply of services and
the demand for products and for net income (123, 1877, p. 285; 244,
1988, p. 369). The services and products acquired by consumers
. . . are those which interest people in their roles as individuals freely occupying
themselves with achieving their preferred personal conditions; that is to say, with
the satisfaction of needs that are diverse and unequal for each of them. Each
consumer calculates the number of units of services or of products such as foods,
clothing, furniture, etc., that he wishes to consume. He compares the intensity of
the utility, not only of similar units of a given service or product, but of different
units of diverse kinds of services or products. Once the prices are cried or posted,
he sees how he should allocate his income among the diverse commodities in
order to obtain the greatest possible total utility. And, nally, he demands those
quantities of the various products or services (115, 1875, in 196, 1897; and in
205, 1898, p. 197).
That allocation of income among the various available commodities is
done in such a manner that no redistribution of it would produce a greater
total amount of satisfaction (106, 1874, pp. 8687; 80, p. 116; 176,
p. 297; 261, pp. 407, 409, 411). The consumer does that by buying
the quantity of each commodity that makes the utility received from the
expenditure of a unit (a marginal amount) of money on it equal for every
commodity purchased.
Walras explained that similar considerations are true of the behavior
related to saving and investing. For example, with respect to the produc-
tion of newcapital goods, economic agents seek the maximumeffective
utility of new capital goods (175, 1889; 25758). Walras thus found
in the desire to obtain the maximum satisfaction of needs, the deter-
mining cause of the offer of productive services and of the demand for
products by landowners, workers and capitalists (106, 1877, p. 231;
176, 1889, p. 214; 188, p. 284; and see 106, 1877, p. 284; 244,
P1: KsF
Human nature 137
p. 369). The supply and demand functions, Walras maintained, are part
of the real economic mechanisms of exchange, production, consump-
tion, saving, and investment. He integrated the functions into the totality
of the interdependent activities that constitute his mature comprehensive
economic model of general equilibrium.
IX. Conclusion
Many scholars before and after Walras have argued that scientic meth-
ods should be applied to the study of human nature, and believed that
when that was accomplished, social science could be used for the scien-
tic solution of policy questions. Walras subscribed fully to those ideas,
but early in his career he did not think that the study of social science had
produced benecial effects similar to those that astronomy had brought
about in humanitys ideas about its place in the universe. He did not,
however, exclude the possibility of that happening: Perhaps construct-
ing social science as a positive science will not produce less progress in
the ethical ideas of humanity (249.XIII, p. 557). Scientic training, he
argued, is essential for that purpose. He asked rhetorically:
Would we see so many socialists or economists limiting themselves to inadequate
and vague denitions of mankind and of society if they had learned to borrowfrom
observation and experience the rigorous and precise denitions of the different
minerals, types of vegetation, or animals? And would we see them so powerless
to draw the series of social relations from a denition of mankind and of society,
if they knew how to extract the series of geometric properties from the denitions
of forms? (141, July 1879, p. 26).
By 1891, Walras had become optimistic about those matters:
It is by proving rigorously the elementary theorems of geometry and algebra, and
then the resulting theorems of the calculus and mechanics in order to apply them
to experimental data, that the marvels of modern industry are achieved. Let us
proceed in the same way in economics, and we shall doubtless succeed in acting
upon the nature of things in the economic and social realm just as we have in
the physical and industrial realm (179, 1891, p. 48; and in 192, 1896, appendix I,
p. 471; 249.VIII, 1988, appendix I, 4, p. 699).
In fact, Walras thought he had achieved that goal. The method of
natural scientists, he asserted, is to discover the fundamental causes of
a phenomenon and by analyzing them arrive at an understanding of it
(194, 1896, p. VII). By applying that approach to the study of human
nature, Walras believed he had achieved an understanding of it and had
used that knowledge to construct the study of economic behavior as
a positive science. Experience, by the intermediation of the concrete
and inductive sciences such as physiology, psychology, history, rst
P1: KsF
138 Walrass ideas
provides it with the denition from which positive science extracts its
fundamental formula by the exercise of reasoning, and then provides the
data regarding the great problems to which positive science applies that
formula by using the rational method (194, 1896, p. VII). He thought
that doing so made it possible to nd scientic solutions of social and
economic problems, andhe was convincedthat he hadprovidedprecisely
those sorts of solutions.
Walras described human nature as having many facets some
admirable, some deplorable as being malleable, as differing in detailed
respects from one person to another, and as having the same general
characteristics in all societies in all epochs. He believed on the basis
of his observations that the feature of human nature that operates in
economic situations is the objective of maximizing satisfaction. He
therefore abstracted the economic motive from the complex reality of
human behavior and showed how it actuates the behavior of economic
agents. The desire to maximize utility, he asserted, has many conse-
quences, such as efforts to obtain income and wealth and the expenditure
of income on certain kinds and amounts of commodities. Walras visual-
ized real economic agents as acting in a setting of real market institutions
and procedures, and then deduced the consequences of their supply and
demand desires in that setting. In his model, he presented consumers,
service suppliers, and entrepreneurs as being clairvoyant and as acting in
a rational fashion to consider all available information and their prefer-
ences in order to maximize their utility, as knowing how to do so, and as
predicting correctly the utility that they will derive from commodities.
He summarized his modeling of the action of human nature in economic
situations in this way: From the condition of maximum effective utility
is deduced rationally the demand function or the effective supply func-
tion depending upon the price . . . ; and from the condition of equality of
supply and demand is deduced rationally the ruling equilibrium price.
Thus all of economic theory can be constructed as a mathematical sci-
ence (205, 1898, p. 467). He was nevertheless cognizant of the fact
that real persons are not totally egoistic and that they make mistakes
of economic judgment, and he contended that he took account of those
facts in his writings on social and applied economics.
Walrass treatment of homo conomicus and thus of preference func-
tions and of the principle of utility maximization was enormously inu-
ential. This was because, instead of conning his analysis to the inves-
tigation of consumption and simple exchange, he went far beyond the
work of other developers of utility principles by using the principles in
connection with the behavior in multiple markets of a variety of partic-
ipants undertaking different economic functions. Walrass conception
P1: KsF
Human nature 139
of economic man was therefore a foundation of his model of economic
behavior, and it is for the same reasons as those advanced by himthat the
abstraction homo conomicus has been and continues to be employed
by all general equilibrium theoreticians, albeit divested of some of his
contentions about preference functions.
P1: KsF
CHAPTER 5
Basic sub-models
I. Walrass approach
The purpose of this chapter is to identify and describe the fundamental
components of Walrass mature comprehensive model of general equi-
libration and equilibrium.
1
To convey fully their character, it will, on
occasion, be necessary to analyze these components critically. The way
he developed the model is in sharp contrast to his approach in his writ-
ten pledges sketch (Chapter 6), which is the same as the approach of the
theorists who created the modern virtual general equilibrium models.
2
The latter is typied by the work of G erard Debreu. In his Theory of
Value (1959), for example, Debreu constructed an equation system on
the basis of a number of axioms. He did not mean for them to describe
the actual behavior of real economic agents acting in a real economy.
His equations are not descriptive of economic behavior, as indicated by
the fact that he made no mention of the institutions, technology, spa-
tial features, regulations, pricing procedures and conventions, and other
structural characteristics of markets. Such matters, he explained, are
irrelevant for the logical development of the theory (ibid., p. 35).
Walras, on the other hand, devoted page after page to the description
of what he perceived to be the behavior of real economic agents in the set-
ting of real markets with specic characteristics, using the techniques of
observation, abstraction, and generalization discussed in previous chap-
ters of this book. He presented that description, and his explanations of
howthe phenomena are generated and their consequences, as his mature
comprehensive model. Thus, as was true of his treatment of economic
1
These topics are treated in much greater detail in Walker 1996, but there is novel
content and manner of exposition in each section of this chapter.
2
Avirtual model is one in which no economic activities occur in disequilibrium, except
the quotation of prices and the manifestation of the associated desired (ex ante) supply
and demand quantities. Trade, hiring of resources, production, consumption, savings,
and all other such economic activities occur only when the general equilibrium set of
prices is found, and that is when the functioning of the model becomes non-virtual.
Special cases of a virtual model are ones in which use is made of oral or written
pledges to trade if the price is the equilibrium price.
140
P1: KsF
Basic sub-models 141
man, it cannot be said that Walras rst described the real economy in the
El ements and then drew upon that description to construct his model.
It is evident from his explicit assertions and from the way he presented
the material in the book that he did not think there was any difference
between his account of what he perceived to be real economic phe-
nomena and the content of his model. There were some exceptions to
that approach. One of them one that was not only unrealistic but that
also distorted his account was his modeling of behavior in regard to
new capital goods, as will be seen. Other exceptions were, he believed,
simplications that identied phenomena found in reality but dispensed
with certain related aspects that he deemed unnecessary for a model that
concentrated on the broad outlines of economic behavior, and that he
asserted could be introduced as needed (Chapters 2 and 4).
Following the procedure adopted by Alfred Marshall, many theorists
concentrated upon the analysis of entities such as an industry, a market,
a consumer, and a rm on the assumption that they are analytically tem-
porarily independent of the rest of the economy, thus analyzing particular
equilibrium situations. Walras asserted, in contrast, that it is necessary
to consider simultaneously the interaction of all parts of the economy
in order to understand the behavior that transpires in any part of it, and
for that reason he developed general equilibrium analysis. He was the
rst economist to construct a complete general equilibrium model that
included the major aspects of an economic system. Doing so required
him to examine exchange, production, consumption, investment, sav-
ings, and money, and to explain many different economic processes that
take place within each of those aspects, and he did that in ways that
were replete with useful tools, concepts, and analyses. Thus he not only
expressed the belief that all economic phenomena are interrelated, which
had been done by many economists before him, he also constructed a
model that specied the interrelations of its parts and demonstrated the
implications of the interrelations in regard to the equilibrating process
and the equilibrium values of the variables.
Walras assumed that the type of structural features and important
aspects of the pricing process found in the nineteenth-century Paris
Bourse (Walker 2000b, 2001b) characterized all the markets in his
model. On any market day, brokers pledged orally, for their own account
or for clients, to buy or sell certain quantities of a security at whatever
price was currently quoted. They called out their pledges loudly, creat-
ing what he called the tumult of the Bourse (106, 1874, p. 51; 42,
p. 72). ibid.). In Walrass words, the brokers bid the price up or down
(106, 1874, p. 50; 42, p. 72) and they give orders. Would-be buy-
ers demand at that price; would-be sellers offer to sell at that price
P1: KsF
142 Walrass ideas
(ibid). Although in the real Bourse, irrevocable disequilibrium transac-
tions occurred within a market day, Walras assumed they did not. He
then expressed the interrelatedness of real markets in a presentation in
his model of the dependence of supply and demand quantities for each
commodity upon the prices of all commodities, and analyzed the forma-
tion of prices and the associated volume of sales (176, 1889, 13941,
147; 116, pp. 171, 173; 123, p. 187).
Turning to the question of ows of production and consumption,
Walras described the connections that he identied in the real economy
among the behavior of consumers, entrepreneurs, workers, capitalists,
and landlords, explained the relationships between input and output mar-
kets, described how the ows of goods and services are adjusted in the
real economy, and made all those features a part of his model. He asserted
that there are, for each industry, market supply functions for productive
services (123, 1877, p. 255; 213, p. 319; 123, 1877, p. 287; 245,
p. 371), market demand functions for outputs (123, 1877, pp. 24647;
2012, pp. 3024; 123, 1877, p. 287; 245, p. 371), and market
supply functions for output (123, 1877, p. 254; 21011, pp. 315
17). He made those descriptions part of his model, and from them he
deduced the resulting connections among the prices and rates of output
of different commodities, analyzing the pricing, production, and acqui-
sition of productive services, primary materials, consumer commodities,
and capital goods. He gave an explanation of the role of money in the
economy, analyzing loan markets, cash balances, metallic currency, and
at money (176, 1889, pp. 375424), and made his treatment of all
those phenomena a part of his model (see Marget 1931, 1935; Kuenne
1961; Walker 1970; Howitt 1973; Morishima 1977; Hall 1983; Hilton
1995).
II. The entrepreneur
Economics of the rm
Walras appears to have developed independently the modern idea of
a rms production function. He dened the coefcients of production
as the quantities of each of the kinds of productive services that are
used in the production of one unit of the product of a rm (123, 1877,
p. 248; 203, pp. 3045). In much of his modeling, he assumed that
the coefcients are xed (123, 1877, p. 249; 204, p. 305), but he
also considered the case of variable coefcients in relation to general
equilibrium (123, 1877, p. 249; 204, p. 305; 123, 1877, pp. 313
15; 176, 1889, pp. 31921, 49092; 32425, pp. 58486, 72122).
Similarly, in connection with the production function, he analyzed the
P1: KsF
entrepreneurs disequilibrium experiments with different combinations
of inputs to maximize prot (191, 1896, pp. 49092), and made that
analysis part of his mature model. Making additionally clear that he was
referring to irrevocable disequilibrium behavior, Walras explained that
the experimenting entrepreneur (lentrepreneur qui t atonne) adds or
decreases each productive service according to whether the value of the
increment of the service is less or greater than the value of the [addi-
tional] output that the increment generates (191, 1896, pp. 49091). He
thereby nds the minimumaverage cost of production by tatonnement
(239, 1895, 2, p. 651).
3
Walras derived the equation for a rms average
cost from the production function, the prices of the inputs, and, implic-
itly, the rms budget (123, 1877, pp. 24850, 253; 2035, pp. 3056;
209, p. 315). He expressed the rms supply function for output ver-
bally and mathematically.
The marginal productivity theory of distribution sprang fromWalrass
work on the economics of the rm (see foregoing citations and 123,
1877, pp. 31415; 32526, pp. 58687), as was evident in Vilfredo
Paretos (1896/1897, 2, pp. 8490) and Enrico Barones (1895, 1896)
sophisticated treatments of that theory.
The entrepreneurs role
Walras contended that understanding the role of the entrepreneur in the
real economy is the key to all economic theory (239, 1887, 2, p. 212)
because of the essential functions the entrepreneur undertakes. Explain-
ing his or her behavior and its consequences, Walras contended that
the entrepreneur is instrumental in the processes by which the interrela-
tionships of economic variables are established and by which the econ-
omy moves toward equilibrium. Walras maintained that one connection
between real input and output markets is established by entrepreneurs
through the circumstance that the average cost of production is altered on
the input side of the market by their actions and the price charged on the
output side is in part a consequence of the supply they bring to market.
Another connection established by entrepreneurs between those markets
is that they make payments to the owners of the economic resources that
they purchase and hire, and those sums are spent by their recipients on
consumer commodities and capital goods properly speaking produced
by the entrepreneurs (123, 1877, pp. 22830; 18586, pp. 28182).
3
The use of different amounts of services in disequilibriumcannot take place in a virtual
model. This is remarked upon here, as in other connections in this book, because the
contrary belief that the mature comprehensive model is virtual has been expressed
(Rebeyrol 1999, pp. 13435; 2002, pp. 54149).
P1: KsF
144 Walrass ideas
Walras incorporated all the foregoing perceptions of the functions of real
entrepreneurs and the consequences of their activities into his mature
comprehensive model.
One problem in Walrass formulation is that he gave different
denitions of the entrepreneur. More than once, he portrayed the
entrepreneurs function as exclusively that of buying labor, land services,
capital-goods services, and rawand semi-nished goods, and selling the
rms product (239, 1887, 2, p. 212; 239, 1895, 2, p. 629). When Walras
took that point of view, he asserted that controlling the productive pro-
cess is a managerial function and that the Walrasian entrepreneur is not
a manager. Taking part in the capacity of a director or otherwise in
the operation of the transformation of services into products is not an
entrepreneurial activity (239, 1887, 2, p. 212). In Francis A. Walkers
work, Walras charged, the entrepreneur is totally confused with the
director of the rm (249.XIII, p. 565).
On the other hand, in various passages Walras described the
entrepreneur as having precisely the role of directing the transforma-
tion of services into products.
4
Controlling the productive process, he
combines the factors to produce consumer commodities or capital goods
(123, 1877, p. 233; 189, pp. 28788). Using productive services, he
acts in his role as entrepreneur to transform them into products (123,
1877, p. 246; 176, 1889, p. 229; 201, p. 302). The entrepreneur, having
applied the productive services to the primary materials, sells the prod-
ucts (123, 1877, p. 233; 189, p. 287). That is not all. As has been seen in
the previous section, Walras also specied that one of the entrepreneurial
functions is choosing a technology for the productive process for exam-
ple, xed or variable coefcients of production. In the latter case, another
entrepreneurial activity is experimenting with the production function by
changing the proportions of the inputs to maximize output for any given
cost (191, 1896, pp. 49092). Still other entrepreneurial functions are
ordering that output be diminished or increased and directing the owof
resources toward protable lines of production and out of unprotable
ones (123, 1877, p. 231; 188, pp. 28384).
Walras maintained that the entrepreneur receives payment qua
entrepreneur only for the activities of purchasing and combining the
factors of production and selling the output, and only in disequilibrium.
He asserted that the remuneration of the entrepreneur is prot. Prots are
made when average cost is less than price. The entrepreneur makes a loss
in the reverse case (123, 1877, pp. 23132; 188, pp. 28385). Walras
4
Claude Mouchot (2000) has pointed out that Walras stated that combining inputs to
produce an output is also an entrepreneurial function.
P1: KsF
explained the accounting practices that entrepreneurs use in real rms
to gauge various aspects of their condition, and in particular, their prof-
itability (123, 1877, pp. 23344; 18999, pp. 28798). It is obvious
that he had reference to the economy in which there are disequilibrium
transactions and disequilibrium production because, apart from the fact
that there are no prots or losses in a virtual purely competitive economy,
Walras indicated that he was dealing with the activities of entrepreneurs
and the operations of rms in the real economy. He then asserted that his
model used that treatment as one of its foundations (123, 1877, p. 235;
191, p. 289).
Walras identied the conditions that would obtain if equilibriumwere
actually reached, and, mixing in his wider conception of entrepreneurial
functions, explained that entrepreneurs adjust production from one dis-
equilibrium rate to another, thereby altering average cost and price until
they become equal and prot per unit of output therefore falls to zero.
When that occurs in all enterprises and supply equals demand in all
markets, a state of equilibrium obtains (123, 1877, pp. 22829, 230
32; 189, pp. 19497). It follows, Walras asserted, that in equilibrium
the entrepreneur has no income qua entrepreneur (123, 1877, p. 232;
188, p. 284), and receives remuneration only for whatever functions
he performs as a worker, including in the latter category performers of
managerial functions for which they receive a salary, or as a capitalist,
or as a landlord. Thus in equilibrium he subsists on wages or interest or
rent, or more than one of those types of income (ibid.).
That gives rise to a second problem in Walrass treatment of the
entrepreneur. The activities of buying and selling that Walras called
entrepreneurial are not conned to disequilibrium. The Walrasian
entrepreneur buys productive services and sells output in equilibrium
also, thus contributing to the value of the output by performing those
functions. His remuneration as entrepreneur should therefore continue,
but prots from which that remuneration would have to be paid
are zero. The way out of this seeming theoretical predicament, which
Walras did not explore because he did not realize that it existed, is in
two parts (Walker 1996, p. 299). First, it should be recognized that there
are routine activities of buying and selling in both disequilibrium and
equilibrium, and that for performing them the entrepreneur, in his role
as a managerial worker, receives a salary. Second, some of the buy-
ing and selling decisions in disequilibrium require that the successful
entrepreneur have exceptional abilities of judgment and perspicacity,
because in that situation there are risks, unforeseen events in the ser-
vice markets, and problems in securing supplies. The payment to the
entrepreneur for making those decisions wisely is a part of the prots
P1: KsF
146 Walrass ideas
of the rm, and the penalty for making bad decisions is a reduction or
elimination of his bonuses and stock options and, in the worst case, dis-
missal. In equilibrium, the demanding and risky aspects of buying and
selling disappear, so the part of the entrepreneurs prot that is paid for
undertaking those aspects becomes zero.
Yet another problem in Walrass exposition is that he did not spec-
ify that the entrepreneur receives payment as entrepreneur for his other
entrepreneurial functions choosing the technologies used in the rm,
experimenting with the proportions of the inputs, changing the ows of
resources out of unprotable rms and into protable ones. That over-
sight is easilyrectied. InWalrass model, those functions are undertaken
only in disequilibrium. Walras properly identied their equilibrating
character. His exposition can therefore reasonably be supplemented by
recognizing that the Walrasian entrepreneur receives a part of the prots
of the rm for undertaking them. In equilibrium, he does not perform
those functions, so his remuneration for themin that state becomes zero.
Walras was right to have said that the word entrepreneur should be
reserved for an economic agent who is, in his entrepreneurial capacity,
not a worker, a capitalist, or a landowner. He should therefore have found
some word other than entrepreneur for the person who undertakes the
routine activities of buying, producing, and selling in both disequilib-
rium and equilibrium, and some word other than prot (b en ece) for
the payment for them. Manager, managerial labor, and wages of
management would have been appropriate because those payments are
under the average cost curve, a necessary payment for the services of
a worker in both disequilibrium and equilibrium.
The entrepreneur and losses
Walras should also have consistently recognized that losses are borne
not by the entrepreneur as such but by the owners of the rm, and that
the entrepreneur is ordinarily one of them. He should have incorporated
that notion into his theory of the entrepreneur. Instead he either identied
it as an empirical matter of which his theory did not take account, or
insinuated it, all the while stating that the entrepreneur as such bears
losses: And, in effect, is it not evident that if he [the entrepreneur]
obtains for his own productive service, in his own business, a price that
is greater or lesser than the one he would obtain elsewhere, he gains
a prot or suffers a loss from the difference between those prices?
(123, 1877, p. 232; 188, p. 285; emphasis added).
Walras evinced the recognition in question in discussing coopera-
tive associations, whose members, he pointed out, should be envisaged
P1: KsF
separately in their double role of shareholders who are, in his words,
entrepreneur-capitalists and also either employees in production coop-
eratives or clients in consumer, credit, and real estate cooperatives. As
shareholders capitalists they either bear the losses, or preferably,
obtain the prots of the rmin proportion to their ownership of the capi-
tal (205, 1898, pp. 28485). Thus in that passage, Walras specied that
in the real economy, prots are the income of capitalists and losses are
borne by them.
He then made it clear that an entrepreneur must, in practice, also have
the role of being an owner of something of value, and that it is only in
such a capacity that he can have a negative remuneration. It is necessary,
Walras wrote,
. . . that the entrepreneur own something that he can give up in order to cover
the losses of the rm. Theoretically, it can be assumed that the entrepreneurs
are nothing other than entrepreneurs; practically, matters are rarely like that. It
happens as an exception that an entrepreneur whose ability and honesty are per-
fectly well-known obtains credit without guarantees, or, as is said, with personal
guarantees; but, in general, the entrepreneur must be at the same time landlord or
capitalist, and moreover his land or his capital goods must be employed in his rm
or in another. Then, it is understood that if his rm, instead of generating prots,
generates losses, the losses will be covered by the liquidation of his possessions
before the capitalists equity is reduced (205, 1898, pp. 31819).
Evidently, if the entrepreneur in Walrass model is to bear losses, then
he must also, in the model, be a capitalist.
III. Tatonnement
Equilibrium and stability of the real economy
The basis of the dynamics of Walrass mature comprehensive model
was his conception of the real economy as being a system in which the
equilibrating processes are incremental, irrevocable, time-consuming
adjustments of prices, of the volume of irrevocable disequilibriumtrans-
actions, and of irrevocable disequilibriumproduction, consumption, and
savings. He understood that in the real economy, prices and quantities of
goods and services exchanged, produced, and consumed are always dis-
equilibrium magnitudes from the general equilibrium perspective (123,
1877, p. 310; 322, pp. 57980). The real economy is always tending
toward equilibrium without ever arriving there because it cannot move
along a path toward it except by tatonnements, and before the taton-
nements are nished they have to begin again, following new routes, all
the parameters of the problem. . . having changed (123, 1877, p. 310;
322, p. 580).
P1: KsF
148 Walrass ideas
The parameters of the real economy, according to Walras, are util-
ity functions and the quantities of goods and services possessed by the
participants (106, p. 102; 102, p. 146), and conditions such as tech-
nology, the amount of land, and the size of the work force (123, 1877,
pp. 146; 237, p. 353). Inasmuch as many of the parameters change
before equilibrium is reached, the state of equilibrium of production
is, like the state of equilibrium of exchange, an ideal state and not a real
one (123, 1877, p. 231; 188, p. 283). In the real economy, it never
transpires that the price of products is absolutely equal to their aver-
age cost of production in productive services, nor does it ever happen
that the quantities supplied and demanded of productive services or of
products are absolutely equal (ibid.). It can happen, and in reality it
happens frequently, [that] the price stays always above the average cost
of production (123, 1877, pp. 31011; 322, p. 580).
Walras explained that in the real economy, the variables undergo
mutually interdependent adjustments in the situation that follows a
change in the basic conditions, as determined by the interrelationships
between excess demands for different commodities, the behavior of mar-
ket participants, and the process of pricing. Prices are changed repeat-
edly, inducing changes in the amounts of commodities produced, sup-
plied, and demanded at any given time. Buyers and sellers in real markets
do not knowthe equilibriummagnitudes of prices and quantities of com-
modities produced and traded, and are concerned only with their own
equilibrium and not the particular equilibrium of their market or of the
entire economy; their activities nevertheless cause the values of the vari-
ables to move in the direction of the equilibriummagnitudes. They do so
by quoting one set of prices after another in such a manner as to diminish
excess demands, and through reacting to each set by producing, buying
and selling, and consuming different amounts of the commodities.
In other words, the real economy cannot move in a single adjustment
from an initial disequilibrium position to one of equilibrium. Neverthe-
less, if the adjustment processes in the real economy were not disturbed
by parametric changes, it would converge, Walras maintained, to a stable
equilibrium of all variables: The operations of the raising and lowering
of prices, of the increases and decreases of the quantities of commodi-
ties produced, etc., on the markets are nothing other than the solution
by tatonnement of the equations of exchange, of production, and of
capital formation (239, 1889, 2, p. 364; emphasis added), the solution
being the equilibrium set of variables. He did not arrive at the idea that
the real economy is an equilibrating system by studying mathematics
but by observing the economy and reecting on it and on the ideas of
other economists, and through the verbal reasoning expressed on many
P1: KsF
pages of his writings (for example, 123, 1877, pp. 22244; 18999,
pp. 28798).
General equilibration and equilibrium of the model
Various controversies surround the nature of Walrass treatment of taton-
nement. As is frequently the case in Walrasian exegesis, the major differ-
ences of opinion on the character of tatonnement have been generated by
the fact that he introduced the written pledges sketch of a virtual process
in the fourth edition of the El ements, but nevertheless retained in contra-
diction to that process much of the non-virtual mature comprehensive
model. For example, it has been argued (see articles cited in Walker 1996,
Chapter 12) that he failed to depict disequilibrium behavior, whether in
connection with stability or otherwise, that he did not construct a model
of economic tatonnement, and that he was uninterested in the economic
adjustment processes that characterize real competitive markets. It has
been contended that he was instead concerned with a technique of math-
ematical iteration for the purpose of nding the solutions to the equations
of a model of general equilibrium. WilliamJaff e, for example, expressed
the view that the tatonnement in all Walrass models is a static, timeless
mechanical process (Jaff e 1981). Roger Koppl believes that Walras did
not model disequilibrium behavior, that he thought that any discus-
sion of disequilibrium falls, by necessity, outside the province of pure
economics (Koppl 1992).
There is a very different view. Drawing upon the features of Walrass
mature comprehensive model, economists such as Martin Currie and Ian
Steedman (1990, ch. 3) have expressed the opinion that he wanted to
develop a realistic model of time-consuming economic tatonnement in
a freely competitive economy. They correctly argue that he modeled a
dynamic equilibrating process in real time of price formation, exchange,
production, consumption, capital accumulation, and monetary adjust-
ment. It has likewise been shown that, in his model of an oral pledges
market (Walker 1996, Chapters 3, 4, 5), he devoted a great deal of atten-
tion to the equilibrating behavior of markets. He incorporated into that
model a number of real trading practices, explaining who changes the
price, how offers to buy and sell are made, how agents act on behalf of
their clients, how the quoted price converges to equilibrium, and how it
is known when the market excess demand is zero. A. van Witteloostujin
and J. A. H. Maks (1988, 1990) and Michio Morishima (1996) similarly
explain that the textbook presentation of Walrass theorizing in regard
to statics and dynamics does him an injustice. Walras based his model,
they demonstrate, on his theories of who changes prices, of iterative
P1: KsF
150 Walrass ideas
pricing, of entrepreneurial behavior, and of a presumed convergence of
the variables to equilibrium. Rather than offering only a static theory,
they point out, Walrass model moves through a dynamic sequence of
adjustments.
Many economists have stated that Walras assumed that prices in his
model are cried by an auctioneer (for example, Hildenbrand and Kirman
1988, pp. 1023). Some writers believe that he assumed there is a single
central market in which all commodities are traded and that it is directed
by a single central auctioneer. Michio Morishima had no difculty in
discerning that Walras assumed that each commodity is sold in its own
market, but Morishima developed his own original notion to the effect
that Walras assumed there are many auctioneers, one in each separate
market (Morishima 1977, pp. 19, 31). In fact, Walras did not mention
or imply a central price quoter or decentralized price quoters in any
connection.
Many economists have contended that in Walrass model the buyers
and sellers are all always price takers. Claude M enard (1990, p. 114), for
example, asserts that one of the major assumptions of Walrass model
is that entrepreneurs are always price-takers. Economists who believed
that is true and who have also recognized that Walras did not mention
a central price quoter have contended that models like his therefore do
not contain any mechanism for the changing of prices (Arrow and Hahn
1971, pp. 266, 325). They have thereby gratuitously created for them-
selves a problem of how prices are changed. Arrows solution was that
pricing must either be exogenously determined in a purely competitive
market, that is, conducted by a central price quoter, or that it must be done
endogenously by imperfectly competitive rms (Arrow 1959, p. 43).
In fact, Walras drew from the real economy the pricing behavior he
used in his mature model, and accordingly, in all his many treatments
of price formation he explicitly portrayed buyers and sellers, including
entrepreneurs, as quoting and changing prices in decentralized mar-
kets (see, for example, 1877, p. 22931; 18587, pp. 28183; com-
plete documentation is given in Walker 1996, pp. 5556, 8589, 1012,
26466).
Walrass verbal reasoning about the equilibrium
and stability of the model
Having achieved, as he thought, an understanding of the behavior of
buyers, sellers, consumers, savers, and producers of goods and services
in the real economy, and having incorporated that behavior into his
mature comprehensive model, Walras then examined the question of the
P1: KsF
existence, uniqueness, and stability of the models general equilibrium.
He thereby initiated the major agenda followed by general equilibrium
theorists for sixty years following his death.
Regarding the stability of the mature comprehensive model, Walras
declared that the object and goal of economic theory consist above all
andbefore all inthe demonstrationtowhichI amreferring (239, 1889, 2,
p. 364), namely the explanation of how the disequilibrium behavior of a
freely competitive economy would, ceteris paribus, lead to equilibrium.
Bypublishinghis initial results inthe years 1874to1877, Walras became,
he pointed out, the rst economist to study the stability of a general
equilibrium model (239, 1895, 2, p. 630). He presented his reasoning
about the real economy as the equilibrating characteristics of his model
(for example, 123, 1877, pp. 22244; 176, 1889, p. 280; 18999,
pp. 28798). He did not, he remarked, represent the tatonnements in that
model in the virtual way that he did in the later written pledges sketch
(210, 1900, p. VIII; pp. 5, 8; see Chapter 6). Instead, he represented
them as they are made . . . effectively (ibid.; emphasis added); that is,
as they occur in reality, and therefore as being irrevocable adjustments
of the disequilibrium variables.
By means of verbal reasoning, Walras then traced out the process
of adjustment toward equilibrium in his model. On one occasion he
preferred to say, responding to criticism, that stability was probable
(176, 1889, p. 246) rather than certain (123, 1877, p. 261). He had
reference there to a market for services in a special model in which there
is no production of new capital goods nor markets for capital goods
(see Walker 1996, pp. 113154). Regarding his mature comprehensive
model, however, he was convinced it is stable, on the grounds of the
following reasoning. The participants do not actually trade unless the
price in their market is found to be one at which the market supply and
demand quantities are equal (106, 1874, p. 50; 42, p. 71; 239, 1895, 2,
p. 630; Chapter 3). If they are unequal, the price is changed in accordance
with the Walrasian pricing rule (see Chapter 4). When the market-day
equilibrium price is found, the traders fulll the verbal pledges they
have made at that price (ibid.). It may be, however, that the equilibrium
is imperfect (176, 1889, p. 135; 111, p. 161). Walras explained
that the price at which the equality of supply and demand quantities
occurs in any market and the transactions that take place at that price are
actually disequilibrium magnitudes if either of two conditions obtains:
If other markets are not in equilibrium, or if the price and the average
cost of production of the commodity are not equal (123, 1877, p. 264;
220, p. 329). If the rst of those conditions obtains, he indicated,
the participants in any given market, whose supplies and demands are
P1: KsF
152 Walrass ideas
functions of all prices, will be affected by the changes of prices in the
other markets that are undergoing disequilibrium adjustments. That will
have the effect of shifting the individual demand curves and hence the
market demand curve in the given market either before or after the
equality of supply and demand as a function of its price is reached in it. In
either case, Walrasian pricing in the various markets will continue. Thus
although exchange takes place only at market-day equilibrium prices,
irrevocable non-virtual behavior in disequilibriumnevertheless occurs in
the mature model of general equilibration and equilibrium of exchange
because trade takes place in any particular market as soon as the supply
and demand quantities are equal in it, although other markets are not yet
in equilibrium. When their prices subsequently change, the particular
market becomes disequilibrated as the participants react to those new
prices. It is revealed that the trade that has occurred in the particular
market was actually a disequilibrium amount from the viewpoint of the
equilibrium of the system as a whole. In studying these matters, Walras
became the rst economist to present a model of the determination of
market-dayprices andof the quantities exchangedina systemof multiple
interrelated markets.
The other part of the adjustment process is that the inequality of
average cost and price leads to changes of supply functions. According
to Walras, the rate of production of the commodity in each market and
therefore its market-day (temporary) equilibriumprice are progressively
altered on each of a series of subsequent days until general equilibrium
is reached. The aspect of the tatonnement in the real economy and in
the model that takes place with respect to new quantities produced is the
progressive diminution, as a result of the changes of the price in the
output market and of the prices of inputs, of the difference between
price and average cost of the product. The rate of output in each rm
and hence of the output of the industry is changed in the same direction
as the sign of that difference. The aspect of the tatonnement in the real
economy and in the model that takes place with respect to the price and
sales of each commodity is the increase of the quantity if the price is
greater than average cost and the reverse if the price is less, and hence
is the movement of the market-day equilibrium price of the commodity
toward its average cost on each successive day. As those processes occur,
the system of new quantities manufactured and new prices . . . is closer
to equilibriumthan the previous one, and it is necessary only to continue
the tatonnement in order to approach it more and more closely (176,
1889, p. 241; 212, p. 318).
Walras then considered the equilibriumconditions for the entire set of
markets. It was in this context that he set forth Walrass law or the basis
P1: KsF
for that law, rst given that name by Oskar Lange. It is curious that Lange
(1942, p. 51, n. 2) cited 116 of the 1900 edition of Walrass El ements
(210), and that Don Patinkin repeated, albeit rather questioningly, that
citation and also offered 122 of the rst edition (106, 1874), because
in neither of those places, nor in the other places mentioned in a general
way by Patinkin, did Walras state the lawin anything like a direct way. In
fact, it was on page 256 of the rst edition (123, 1877; 213, p. 323) that
he presented a version of the law, limited by the consideration that he
treated only markets for productive services. Letting O be the quantity
supplied, D the quantity demanded, p the price, and the various sub-
scripts respectively indicating the services of land, personal faculties,
and capital goods, the relation is:
(O
t
D
t
) p
t
+(O
p
D
p
) p
p
+(O
k
D
k
) p
k
+ = 0
In that place, Walras presented the law as an equilibrium condition: In
the general equilibrium of a multi-market freely competitive economy
in which there is production, the sum of the values of all excess demand
quantities for commodities is zero (assuming, he should have added,
that all prices are positive), because all the excess demand quantities
are zero. Unless that equation is satised, it is not possible to have
equilibrium (ibid., p. 257), which indicates that the equation is not
satised in disequilibrium. In the El ements, he used primes on the terms;
the equation without primes just given is exactly how Walras presented
the lawin 1897 (198, in 205, 1898, p. 277). There, he also asserts that the
mechanism of free competition solves the equation, implying that it is
true in equilibrium and that in disequilibrium, before the functioning of
that mechanism, the sum of the values of the excess demand quantities
is not zero.
Walrass additional remarks are interesting not only for the con-
tent that they add to the law, but also because they reafrm, just two
years before his adoption of the virtual written pledges sketch, his
view of the real economy and of his model as being systems in which
there are irrevocable disequilibrium transactions and production. He
explained:
That equation permits, in itself, the afrmation that, excluding the case of a general
crisis, if at a given moment the supply of certain productive services exceeds the
demand for them, the demand for certain other productive services must exceed
their supply (ibid.).
It appears that his implication is that the algebraic sum of the values of
the excess demand quantities for all commodities is also zero in a state
P1: KsF
154 Walrass ideas
of disequilibrium. He went on to examine the adjustments of prices in
disequilibrium and the concomitant changes in production and of the
supply and demand quantities that lead those quantities to equality in
the case of each service. His discussion included noting that services
that are not transformable into others, as is the case with land services
in relation to labor, undergo changes in prices, and that services that are
usable in different ways, such as a type of labor that can be altered to
performa different task, mayundergoprimarilychanges intheir amounts
in each line of production. He concluded with respect to wages that to
lead to or to maintain the equilibrium of production in regard to wages,
it is necessary in the last analysis to redirect labor from rms in which
wages are tending to fall towards rms in which wages are tending to
rise (ibid., p. 278).
The state of equilibrium, Walras declared, is the normal state in
the sense that it is the one to which things tend automatically under
the conditions of free competition in production and exchange (123,
1877, p. 231; 188, p. 283); that is, under the conditions that he thought
obtained in the vast majority of real markets and that obtained in his
model. Then the equilibrium prices are those at which the quantities
demanded and supplied of each service or product are equal, and at
which, moreover, the price of each product is equal to its average cost
of production (176, 1889, p. XV; p. 13). The equilibrium prices lead
each economic agent to initiate behavior that they then want to con-
tinue buying, selling, investing, and consuming certain amounts (see
for example, 123, 1877, pp. 25051; 206, pp. 3067). The mutually
determined sets of equilibrium prices, average costs, and quantities sup-
plied and demanded are harmonious; they fulll the plans of all the
participants in the model. No one who could change the prevailing eco-
nomic conditions wants to do so, and no one who wants to change them
is able to do so.
Entrepreneurs, capitalists, and capital formation
Walras analyzed saving and investment behavior in the real economy
(123, 1877, p. 284; 176, 1889, pp. 27071, 307; 24142, pp. 35763). In
the course of his study of those matters, he formulated a macroeconomic
savings function, being the rst economist to do so. In his examina-
tion of capital accumulation, Walras discussed additional ways in which
entrepreneurs connect different economic sectors and markets. Capital-
ists in the real economy transfer their money savings to entrepreneurs
through purchasing stocks and bonds (123, 1877, pp. 3023; 176, 1889,
pp. 30910; 269, pp. 43436), a process that Walras examined in
P1: KsF
studies of credit markets (205, 1898, pp. 30736). The incomes paid to
owners of economic resources by entrepreneurs are partially saved by
the recipients in their role as members of households, and the savings are
used by entrepreneurs in capital-goods industries to acquire raw mate-
rials and productive services in order to construct new capital goods.
These are bought by capitalists and used by entrepreneurs in rms to
make other capital goods properly speaking and consumer commodities
(123, 1877, pp. 29495; 176, 1889, pp. 28687; 25455, pp. 38891).
Walras analyzed how those goods are priced and are employed in the
most protable uses, thereby resulting in an increase in levels of pro-
duction of different types of commodity (123, 1877, pp. 273311; 176,
1889, pp. 261317; 23170, pp. 345436). He likewise developed
an explanation of the determination of the rate of net income gener-
ated by the use of capital goods, and of the determination of the market
and equilibrium rates of interest (123, 1877, pp. 278311; 176, 1889,
pp. 261312; 23171, pp. 345436). He made his account of all those
matters a part of his mature comprehensive model.
Conditions of a reprise
Walras identied a tatonnement in his model as being a complete adjust-
ment process, beginning with a new initial disequilibrium and ending
with equilibrium. Within a tatonnement, he distinguished each reprise
of tatonnement (176, 1889, p. 280) the several stages that occur dur-
ing the course of the overall process. He had in mind an initial period of
price adjustments and associated disequilibrium behavior, followed by
the participants pausing and assessing their individual situations and
market conditions, a recognition by some participants that they can
improve their economic situation, followed by a second period and a
second assessment, followed by additional periods until the equilibrium
is reached. Thus each of the reprises eventuates in a temporary dise-
quilibrium of the entire system.
5
Each such successive state is closer to
the eventual general equilibrium than the previous state, and the entire
tatonnement ends when equilibrium is reached.
5
Witteloostujin and Maks (1988, 1990) believe that a Walrasian tatonnement in the
mature model consists of a series of temporary market equilibria between which there
are intertemporal disequilibria. If they had reference to the equilibrium in a particular
market achieved by the equality of the quantities supplied and demanded on a market
day, and the subsequent series of particular equilibriummarket-day prices, that is true.
It is not, however, true of the nal state of each disequilibrium reprise of the general
equilibrium tatonnement. Except for the last one, each reprise ends in disequilibrium,
which is why it is suceeded by a new reprise.
P1: KsF
156 Walrass ideas
A reprise takes place under these conditions:
It is always a question of arriving at the equilibrium of capital formation in the
same manner that we arrived at the equilibrium of exchange and at the equilib-
rium of production, that is to say by assuming that the given conditions of the
problem are invariable during all the time that the tatonnements take, except that
the assumption will subsequently be made that those given conditions are vari-
ables in order to study the consequences of their variations (ibid., p. 279). . . . We
must assume that, at each reprise of tatonnement, our entrepreneurs that produce
products [non-durable goods] and new capital goods nd, in the country, land-
lords, workers, and capitalists possessing the same quantities of services, having
the same needs for services and products, and the same dispositions to save
(ibid., p. 280).
It wouldhave beensimpler if Walras haddeclaredthat the supplyfunc-
tions for capital-goods services are unchanged, rather than the amounts
of those services. When he wrote that the capitalists have the same
quantities of services, he must have meant that the maximum potential
amounts of capital-goods services are the same at the beginning of each
reprise of tatonnement. He must have been referring to the capacity of the
capitalists to offer services, and similarly regarding the other resource
owners, because he stated that the services that are tendered on the mar-
ket are functions of prices and not a xed amount. The amounts provided
change during the tatonnement as prices change (123, 1877, p. 247; 176,
1889, p. 230; 201, p. 303). They each vary with all the prices of all
the services and products and are determined, after the determination
of those prices, by the maximization of effective utility (205, 1898,
p. 275). If it were the case that the maximum potential amounts of capi-
tal goods services are the same, that condition, together with unchanging
preferences of the capitalists, would ensure that the supply functions for
capital-goods services are the same at the beginning of each reprise.
Walras asserted that disequilibrium production of non-durable goods
and of net new durable capital goods occurs during each reprise of the
tatonnement:
A certain rate of net income and certain prices of services being cried, and cer-
tain quantities of products and of new capital goods being produced, if this rate,
these prices and these quantities are not an equilibrium rate, and are not equilib-
rium prices and quantities, it is necessary not only to cry another rate and other
prices, but also to produce other quantities of products and of capital goods. . . .
At a rate of net income cried rst at random, and then raised or lowered accord-
ing to the circumstances, the entrepreneurs that produce new capital goods will
fabricate certain quantities of new capital goods determined rst at random
and then increased or diminished quantities according to the circumstances
(176, 1889, p. 280).
P1: KsF
The net new investment that occurs during the equilibrating process
changes the magnitude and composition of the capital stock. The
new durable capital goods are sold during the disequilibrium phase
at disequilibrium prices: Those goods, being produced in the general
tatonnement then being brought onto the capital goods market and being
sold there in accordance with the mechanismof free competition, would
be sold at prices that differ fromtheir average costs of production, but are
closer to being equal than the previous prices were to being equal to their
average costs of production (ibid., p. 292; emphasis added). The aver-
age costs and prices of output being generally unequal, the entrepreneurs
who produce new capital goods will make prots or losses . . . (123,
1877, p. 295; 176, 1889, p. 289; 256, p. 393), phenomena that do not
exist in a virtual model. Those are additional robust statements of irre-
vocable disequilibrium production and resources allocation, and of the
identity of the model with the real economy.
Another fundamental change in the situation is the redirection of
resources by the entrepreneurs during the phases of the tatonnement,
which alters the composition of output:
Now this tatonnement is precisely that which generates itself on the market,
under the regime of free competition, when the entrepreneurs who produce new
capital goods, just like those that produce products, direct resources toward the
enterprises or away from them according to whether prots or losses are made in
them (123, 1877, p. 298; 176, 1889, p. 298; 258, p. 401).
Walras specied that the capital-goods services are provided exclu-
sively by the capital goods that were in existence at the very begin-
ning of the entire tatonnement process. In some discussions of taton-
nement, he stated that the new capital goods do not function during
the period (249.XIII, p. 265). Moreover, he placed his statement that
production and capitalization will be able to continue, but, of course,
with the changes resulting from the existence of the new capital goods
(176, 1889, p. 280) after his assertion that equilibriumhas been reached,
implying that the new capital goods produced in disequilibrium are not
used until equilibrium is found. He postulated arbitrarily that the new
capital goods lie idle in disequilibriumto try to insure that there would be
no changes in the maximumpotential amounts of capital-goods services
during the course of the tatonnement, and hence in the supply functions
for them. Otherwise his equations, in which those amounts are param-
eters, and in which the functions are assumed to be unchanging, would
have solutions that are not the equilibrium values of the variables of the
model.
P1: KsF
158 Walrass ideas
Contradictions in the model
Despite accepting [the] necessity (176, 1889, p. 280) of irrevocable
disequilibrium phenomena and specically of the fact that new capital
goods are produced during the phases of the tatonnement, Walras did
not accept the necessity that the new capital goods are used during
the tatonnement. The structure and functioning of his mature model
is thereby rendered less than coherent in regard to the equilibrating
process and equilibrium by the logical problems created by his arbitrary
postulate.
It cannot be that the new capital goods are held in inventories by
their producers, because Walras stated that they are sold to capitalists
during each reprise. For the supply functions of capital goods services
at the beginning of each reprise to be unchanged, it would have to be
the case, therefore, that the buyers of the new capital goods hold them
in inventories during the tatonnement and do not try to make money
from offering their potential services for sale. That condition is not an
outcome of the structure and functioning of the model and cannot be
made consistent with it. Walras gave no reason founded in economic
behavior, rational or irrational, for supposing that the new capital goods
are not used during the tatonnement. His arbitrary postulate contradicts
his theory of saving and investment, which is based on the supposition
that the reasoncapital goods are acquiredis tomaximize utilitybyrenting
themout or using themin their own rms. Their owners therefore would
not let the newcapital goods lie idle. The supplyfunctions for the services
of the new capital goods of each type would be added into the total of
capital goods services of each type offered for sale. That total would
change during the tatonnement and would, specically, be different at
the beginning of each reprise. The given conditions of the problem
cannot be invariable during all the time that the tatonnements take.
Moreover, the supply functions for old capital-goods services cannot
be the same at the beginning of or during each reprise of tatonnement.
Walras assumed, to deal with the most interesting case, that aggregate
investment is positive (123, 1877, p. 284; 176, 1889, p. 271; 242,
p. 361), although if it were negative there would still be the following sort
of consequences. The ow of new capital goods into the inventories of
capitalists is an asset effect that alters their demand and supply functions
for all types of commodities. For example, it changes the quantities of
old capital-goods services that they want to offer at each price at the
beginning of each reprise.
Even if it is arbitrarily assumed that the supply functions for the ser-
vices of old capital goods do not change and that no services of new
P1: KsF
capital goods are offered, the extensive disequilibrium behavior that
Walras built into the course of the tatonnement of the model would have
precisely the consequence that his supposed equilibrium equations do
not give the equilibrium values. Disequilibrium transactions alter the
distribution of a given amount of a commodity. Disequilibrium pro-
duction of a commodity alters the total amounts of it and therefore the
amounts possessed by particular individuals. The changes in the amounts
of goods held by individuals alter their demand functions for goods and
services and their supply functions for goods and services, thus alter-
ing the equilibrium values of the variables. With respect to new capital
goods, for example, during each phase of the tatonnement, in the course
of which savings occur and therefore additional new durable capital
goods are produced and sold, and consequently at the beginning of each
newreprise, capitalists have different amounts of those goods and there-
fore have different demand functions for additional capital goods. That
leads to a set of prices at which their production is stabilized that differs
from the set during each reprise. The equilibrium of the system is there-
fore path dependent for that reason, in addition to others that have been
discussed.
Capital use and accumulation give rise to other sources of difculties
in Walrass model. Let it be supposed for a moment as Walras supposed,
disregarding the illogical aspects of the model, that only the old capital
goods are used during the tatonnement and that the new capital goods
are put into use only when the tatonnement is ostensibly nished, that
is, as soon as an ostensible static equilibrium obtains. A rst problem is
that Walras did not take account of the depreciation of old capital goods
properly speaking during the tatonnement. Some of them reach the end
of their productive life during the tatonnement; others are partially used.
There is no owof maintenance of xed capital goods properly speaking
in use, so there cannot be the same amounts of old capital goods at
the beginning of each reprise. There must be less. To insure that there
are the same amounts, he should have assumed that investment equals
depreciation. Alternatively, he could have assumed, contrary to reason,
that the xed capital goods do not wear out and are indestructible, as he
did in part of his treatment of the maximum utility of new capital goods
(176, 1889, p. 300). Even those assumptions would not have eliminated,
from a purely logical point of view, the difculty posed by the use of
primary materials. The variable capital that is not held in inventories
but that is used and therefore used up during the course of a reprise
must also be exactly replaced if the services of variable capital goods
are to be the same at the beginning of each reprise. That is a necessary
condition, but Walras did not provide any market mechanism to ensure
P1: KsF
160 Walrass ideas
that the amounts of newly produced variable capital goods equal the
disequilibrium amounts that are bought and used.
Whether or not the xed capital goods in use depreciate, by assuming
that the particular batch of capital goods in use during the tatonnement
does not change, Walras created a second problem for his model.
Demands and relative prices of inputs and outputs change during
the tatonnement. The changing demands for nal goods shift derived
demands to some capital goods in the existing stock, and away fromoth-
ers. The problem, rightly pointed out by Fabio Petri (2004, Chapter 5),
is that the demand for some or many capital goods would decrease,
but lowering their prices could not lead to their employment for the
production of different commodities because they are technologically
incapable of being adapted. Most capital goods have specic uses. If
the form of the capital stock in use cannot be changed to respond to
changing demands for the services of particular capital goods, the prices
of the unused capital goods would fall toward zero, but their owners still
could not tempt producers to use them. There would therefore be unem-
ployment to some degree of some or many capital goods. The absurdity
of the situation is increased by the fact that the response to the changing
pattern of demand for capital goods would be to increase the production
of some new capital goods, namely of the kinds of capital goods for
which prices have increased. Those new goods, however, according to
Walrass assumption, are nevertheless not put on the market during the
tatonnement to augment the supply of the old capital goods the prices
of which have increased.
Walrass assumption of xed coefcients of production poses a third
problem. The amounts of the inputs released from the production of
commodities that are less in demand may not t the proportions in which
they are required in the industries for whose products demand increases.
Once again, there would be involuntary unemployment of resources, a
condition that Walras did not envisage.
A fourth problem is that Walras was in error to have believed that
capital accumulation is compatible with the type of equilibrium that he
envisaged in his discussion of tatonnement. The equilibriumto which he
referred is factitious, existing only transitorily while his model is held
in a state of arbitrarily suspended animation by the postulate that addi-
tions to the capital stock are not used a deus ex machina that interrupts
the incomplete working of its endogenous processes. The instant the
postulate is removed, the equilibrium is ruptured . . . (Walker 1987b,
pp. 86061). As soon as the additions to the stock of capital are put into
use, which Walras stated occurs at what he supposed is the end of the
tatonnement, relative prices, outputs of many different commodities,
P1: KsF
and seeming parameters continue to change, and it is revealed that a
static equilibriumdid not exist. Thus the mature comprehensive model is
inherently dynamic in two respects. First, the tatonnement is a dynamic,
time-consuming process. Second, the model turns into a growth model
because the use of the additions to the capital stock propels it beyond
a stationary equilibrium and onto a path of economic growth. Indeed,
Walras recognized as much when he stated, as noted, that, after the taton-
nement is over, Production and capitalization will be able to continue,
but, of course, with the changes resulting from the existence of the
new capital goods (176, 1889, p. 280), and when he analyzed the
transition that occurs to what he called a progressive society a
dynamic system in the sense of one in which there is capital accumu-
lation and an increase in the labor force, and in which aggregate out-
put grows (123, 1877, pp. 31223; 176, 1889, pp. 31829; 32324,
pp. 58498).
6
Nevertheless, there is nothing contradictory in the model created by
Walrass assertion that the (maximum possible) amounts of the services
of land and of workers are the same at the beginning of each reprise and
thus during the entire course of the equilibrating process in the model,
as he believed was substantially true in reality. In the model, the amount
of land is xed (123, 1877, pp. 224, 278; 176, 1889, pp. 207, 266; 236,
pp. 277, 352), and so is the technology used on it (123, 1877, p. 310;
322, p. 580). The size of the work force (123, 1877, pp. 27879; 176,
1889, p. 266; 237, pp. 35253), its efciency, the workers preferences,
and the work week (205, 1898, p. 275) are xed. Changes in those con-
ditions do not occur as an endogenous consequence of the tatonnement
process in the model, so they are truly parametric constants with respect
to that process.
7
Moreover, services cannot be accumulated. The max-
imum possible ow of them that could be provided at any moment by
the land and human capital is therefore unchanged during the whole of
the tatonnement process, and, in particular, is the same at the beginning
6
WilliamJaff e believedthat Walrass writings on a progressive society constitute merely
a coda, an addendum rather than an integral part of his central body of theorizing
(Jaff e 1978). Michio Morishima (1996) and other economists (for example, Currie
and Steedman 1990) disagree with that contention. They argue that Walras intended
his general equilibrium model to lead to and to be the point of departure for the
development of a model of economic growth. In fact, this chapter shows that the
implications of Walrass mature comprehensive model, his central body of theorizing,
are that it becomes a growth model when the use of new capital goods occurs.
7
Similarly, there is no contradiction in Walrass model occasioned by his assuming
that the preferences of the economic agents for consumer goods and for saving are
unchanged by disequilibrium transactions and disequilibrium production (176, 1889,
p. 280).
P1: KsF
162 Walrass ideas
of each reprise of tatonnement. The supply functions for the services
of land and labor will change, however, even if the amounts of those
factors and preferences for work and income do not change, because the
functions also have the holdings of consumer durables and capital goods
properly speaking as arguments, and those arguments change during the
course of the tatonnement.
Uniqueness
Walrass study of uniqueness was limited to the consideration of pure
exchange in a model of an isolated market and in a multi-market model.
Regarding the former, in which one commodity is exchanged for another,
he used graphs and verbal reasoning to arrive at the conclusion that an
isolated market may have more than one set of values at which ex ante
supply and demand are equal (106, 1874, p. 74; 68, p. 99). Regarding
a multi-market system, he concluded on the basis of incomplete verbal
reasoning that generally, in the sense of ordinarily, multiple equilib-
ria do not occur. That was an opinion, not a proof (106, 1874, p. 156;
156, pp. 24142). If he had tried to use his equation system to furnish a
proof he would have been unsuccessful, because, as has been explained,
it does not describe the model. Furthermore, the proof, if it had been
forthcoming, could not have established that there are multiple equilib-
ria. It would have established only that there are, from the mathematical
point of view, more than one set of solutions to the equations. If the
initial conditions and the dynamics of the model were to be specied in
sufcient detail, the solution set to which the model actually converges
and that is therefore the unique equilibrium set would be known (see
Walker 1997b, pp. 132141).
IV. Mathematical treatment of phenomena
Mathematical treatment of existence
Walras developed a second analytical approach to the characterization
of the economy and of his model in order to supplement and add pre-
cision to his verbal reasoning in the demonstration of their outcomes.
It consisted of attempts to use equations to represent the phenomena in
his model that were amenable to mathematical treatment (for example,
123, 1877, pp. 24565; 200220, pp. 30130). Of course, he had
to use verbal reasoning to form the equations. Thus he gave economic
meanings to mathematical symbols and reected his conviction of the
equilibrating nature of markets in the real economy and hence in his
P1: KsF
model by constructing equations that express the condition of economic
equilibrium. He claimed to have proved the existence of equilibrium in
his model because the number of unknowns in his equations equals the
number of independent equations (for example, 123, 1877, pp. 25051;
176, 1889, pp. 23334; 205, pp. 3067). As is well known, that equality
does not establish existence, for three reasons.
A contradiction between the model and the equations
The rst reason is that the model to which he had reference the mature
comprehensive model is path dependent but the equations he used
do not take account of that feature. Starting with an assumed state of
disequilibrium, Walras tried to use equations in which the initial holdings
of consumer goods and capital goods are parametric constants in order
to deduce a different set of holdings of commodities the equilibrium
set reached through a series of changes of the set of holdings. That is
illogical. Walrass attempt to establish the existence of equilibrium in
the model with the use of the equations was an impossible task. Many
of the constants in the equations are actually endogenous variables in
the model, changing during the course of the equilibrating process. Thus
once the adjustment process has begun, the symbols in the equations no
longer have referents in the model; or, to put it the other way around,
the changing values of the variables in the model are not symbolized in
the equations. The equilibrium values of the model are therefore not the
ones indicated by the solutions to his equations.
To determine whether static equilibrium exists in a variation of the
mature comprehensive model, Walras would have had to exclude capital
accumulation, and, in recognition of the fact that many of the phenom-
ena that he treated as parameters in his equations are really variables
in his model, he would have had to use equations that are appropriate
for determining whether a path-dependent system has solutions, namely
a set of differential or difference equations, or a mixture of the two.
Walrass failure in those regards was surely facilitated by his habit of
writing the equations for supply and demand quantities with only prices
between the parentheses on the right-hand side, never introducing sym-
bols to show what he assumed were the constants in those functions
(see Van Daal and Walker 1990). Had he done so, he would probably
have recognized much earlier than 1899 that the model did not behave
in the manner that the equations supposed. As it was, it took him until
that year to recognize the implications of the contradiction between his
comprehensive model and the equations.
P1: KsF
164 Walrass ideas
The second reason is that, even if the model were not path depen-
dent, or even if Walras had correctly described it with a set of dynamic
equations, his assertion would have been erroneous because he neglected
to consider free goods, which are used in positive amounts but have zero
prices.
The third reason is that he did not introduce the requirement that
the solutions to his equations all be non-negative, which is necessary
inasmuch as negative prices or quantities of commodities are nonsense,
except in accounting balances and concepts such as a negative rate of
interest, whichhave norole inthe questionof existence. Nevertheless, his
studies of existence went far beyond those of other nineteenth-century
economists, inasmuch as he dealt with the question of the existence of
general rather than particular equilibrium and tried to do so with the use
of mathematics.
Stability and the equation system
Walrass attempts to treat stability mathematically do not deal with his
model but with his equation system. He tried to develop a technique
of mathematical iteration whereby he could determine the direction of
change of the variables in the equations. That he was conducting an exer-
cise in mathematical iteration is plain from his statements that various
conditions are selected and imposed on the equations by the theoretician
who is contemplating them. In that capacity, Walras picked a random
initial disequilibrium set of prices and determined the associated excess
demand quantities. He then selected new prices either greater or smaller
than the initial ones in accordance with the Walrasian pricing rule, and
evaluated the absolute excess demand quantities to see if they displayed
a tendency to decrease; that is, to see if the prices and quantities con-
verged to a solution set (for example, 123, 1877, p. 29398; 25254,
pp. 38789).
In one exercise of mathematical iteration, for instance, referring to
the conditions of the tatonnement that we apply at the present, Walras
arbitrarily assumed that the quantities produced of the numeraire and of
capital goods are constant, changed their prices, and then examined the
successive sets of disequilibriumquantities produced (176, 1889, p. 286;
254, p. 388). He made particularly clear his role as a theoretician who
uses arbitrary postulates in the study of equations by commenting that
the assumption of the xity of the prices of the other commodities is an
essentially momentary and provisional assumption which plays a part
only in a partial phase of the general tatonnement and takes the place of
P1: KsF
the assumption of the variation of those prices in the other phases (239,
1895, 2, p. 631). Again, signaling that he was conducting a mathemati-
cal exercise in iteration regarding a hypothetical situation, he noted that
in the conditions of tatonnement that we apply at the present, the prices
of services are xedanddonot change (176, 1889, p. 239; 212, p. 317).
That is a condition he could put into the equations, thereby implicitly
changing the structure and manner of functioning of the situation he
contemplated, but that cannot be true of the endogenous processes that
take place in the model. If the model is allowed to generate its path with-
out the theoretician introducing modications, those prices will not be
constant during the tatonnement. There are many pages in the El ements
on which Walras made similar arbitrary assumptions regarding the vari-
ables instead of tracing out the endogenously determined variations in
the model.
Walras assertedthat the mathematical iterations inhis equations reveal
that the prices in his model converge to an equilibrium (123, 1877,
pp. 291, 300; 252, p. 379; 258, p. 401). He contended that the
direction of change of the variables in the equations induced by his pos-
tulated changes of prices are the same as occur in the model, describing
the iterative changes as variations of disequilibrium transactions and of
disequilibrium production (176, 1889, p. 292):
In these conditions, it can be accepted that the change in the quantity manufactured
of each new capital good has brought the price and the average cost of the good
closer than the changes in the quantities manufactured of other newcapital goods,
supposing them all to have changed in the same direction, have caused the price
and average cost to differ. The system of newquantities manufactured and of new
prices and average costs is thus closer to equilibrium than the previous set, and
it is necessary only to continue the tatonnement to approach it more and more
closely (176, 1889, p. 293; with a change, 258, p. 399).
It is clear yet once again that Walras was not discussing a virtual system.
Inthat passage, as throughout his account of the functioningof the mature
comprehensive model, the activity has occurred in the recent past; the
supplies and demands are ex post. The different quantities of goods
have been manufactured; they are produced in successively changing
disequilibrium amounts and sold at disequilibrium prices during the
course of the tatonnement.
In his mature phase of theorizing, Walras believed that the real econ-
omy, ceteris paribus, was characterized in a general sense by his static
system of equations and therefore would end up with the same solutions
as those equations if they had the necessary numerical coefcients: The
result of the ow of the interrelated phenomena of reality is truly the
P1: KsF
166 Walrass ideas
empirical solution of the equations (123, 1877, p. 365; 370, p. 651).
After nishing his iterations with his equations, the indicated taton-
nement, he expressed that opinion rst with respect to the market for
consumer goods (123, 1877, p. 255; 212, p. 319) and then with respect
to capital goods:
Nowthe indicated tatonnement is exactly that which generates itself on the capital
market when the newcapital goods are sold according to the mechanismof raising
and lowering prices . . . (176, 1889, p. 287; with changes, 254, p. 390).
Walras also evidenced his belief that his mathematical exercises demon-
strate the stability of real markets by asserting that the price changes that
he discussed in relation to his model occur in a real economic institution,
the Paris Bourse (210, 1900, p. 267; 254, p. 390; and see 176, 1889,
p. 289; 255, p. 391).
Comparative statics
Walrass belief that the real economy and his mature comprehensive
model are stable equilibrating systems was his justication for compar-
ative static analysis. He undertook it in order to analyze the impact on
equilibrium of changes of parameters, such as preferences, the quan-
tity of money, and the quantity of a commodity held by the participants
(for example, 106, 1874, pp. 13437; 176, 1889, pp. 15963; 137
38, pp. 20914. 176, 1889, pp. 12728; 103, pp. 149, 151. 106, 1874,
p. 104; 103, p. 151. 123, 1877, p. 294; 176, 1889, p. 309; 265, p. 430).
Walrass verbal analyses of the directions of change of the variables in
his model resulting from changes of the parameters were plausible, but
the defect of his comparative statics, as well as of his mathematical
iterations, was that he undertook his analyses with equations that could
be used only to determine the stability of a virtual model without net
investment, that is, one unlike his model and, of course, unlike the real
economy.
To have identied the true new equilibrium values resulting from
parametric changes in the mature comprehensive model, Walras would
have had to take account of the changes generated by the course of the
tatonnement itself and deal with the irrevocable movement to the true
new equilibrium, which would be a path of growth. The path would
be generated, as has been seen, because, Production and capitalization
will be able to continue (176, 1889, p. 280) after the tatonnement is
over. His iterations and his comparative statics do not relate to that
model.
P1: KsF
V. Other processes in the model
Monetary processes
8
One of the fundamental aspects of Walrass legacy was ignored by most
economists until attention was drawn to it by Arthur W. Marget (1931,
1935), namely Walrass analysis of the role of monetary behavior in
the real economy and his incorporation of that role into his mature
comprehensive model (see Walker 1996, Chapter 12). Even when that
aspect has not been ignored, it has often been misunderstood.
9
The task
of understanding his views is complicated by the fact that he constructed
or tried to construct more than one model of monetary behavior and, as
a related matter, by his espousal of three different theories of money
during the course of his career.
10
In his mature comprehensive model,
he used a cash-balance approach (168, 1886; 176, 1889),), stating that
money is held because it will be needed as a medium of exchange. Was
that approach integrated into the model Walras associated it with in
such a fashion that the model functions as an equilibrating monetary
economy?
Takashi Negishi (1989) asserted that the integration is not possible
because he believed that there is no uncertainty in Walrass models and
that money has no place in that type of model. Negishi does not make
clear whether he refers to subjective beliefs about the future magnitudes
of variables in states of disequilibrium and equilibrium, or only the
latter. Some economists, like Martin Currie and Ian Steedman (1990),
contend that there is no uncertainty in any state of Walrass mature
comprehensive model. Toothers, however, it seems clear that uncertainty
in disequilibrium is one of its features. That position and hence an
implied rejection of Negishis argument is taken by, for example, A.
van Witteloostujin and J. A. H. Maks (1990).
8
Walrass treatment of money in the mature comprehensive model has been exhaus-
tively described, explained, and analyzed elsewhere (Walker 1996, Part I, Section II,
and especially Chapter 11 of that book).
9
H. C. Hilton (1995) has rightly observed that many economists have formed their ideas
about Walrass views on money and credit from reading inaccurate interpretations of
them.
10
There are differences between the monetary aspects of the model he presented in his
creative phase, in his mature comprehensive model, and in his last comprehensive
model. The latter name is explained in the introduction and in Chapter 6, and the
writings to which that name is given, including the monetary aspects of the model,
are analyzed in Walker 1996, Chapters 17 and 18. In his creative phase, Walras
expounded the idea of a circulation to be cleared and used an equation of exchange.
In his nal phase of theorizing, he contended that money is held for the purpose of
obtaining its service of availability.
P1: KsF
168 Walrass ideas
Walras did not state that there is no uncertainty in his mature com-
prehensive model (as distinct from his written pledges sketch), and he
constructed the mature model in such a fashion that there has to be uncer-
tainty in order for it to function. It would be absurd to suggest that the
participants in that model are certain at the beginning of a tatonnement
or during it about the dates of future transactions and the magnitudes
of future prices, offers, demands, money balances, other asset holdings,
and other economic variables. If they had certain knowledge of the future
equilibriumvalues of the variables, no tatonnements would be necessary
or possible. There would be no series of disequilibrium prices. Every-
thing that Walras wrote about disequilibrium behavior in his mature
comprehensive model would have been nonsensical. Furthermore, after
the equilibrium conguration of variables is reached, there is no reason
in Walrass model why the participants would believe, contrary to all
their experience, that there will be no future changes in asset holdings,
prices, and dates of transactions. They are uncertain of the duration of the
equilibrium variables. As long as that conguration obtains, however,
they either would not want to change their behavior or would not be able
to change it, so their realization that the situation is not going to remain
unchanged forever is not a cause of change. Change would be initiated
by an exogenous shock that alters the specic elements entering into the
structure of anticipations and the degree of uncertainty of beliefs. Walras
did not state that the participants are certain of the future; he stated that
they are clairvoyant (see Chapter 4). As such they are sagacious and they
reason lucidly, but they do not have perfect foresight in the mature model.
Walras analyzedthe role of true moneyandspecicallythe demandfor
cash balances in the context of an economy in which there are irrevocable
disequilibrium transactions and disequilibrium production; thereby he
successfully integrated the use of money into the mature comprehensive
model (see Marget 1931, 1935).
11
Due to the manner of functioning of
markets in that model, true money is an essential part of the exchange
process, and hence it is inextricably linked with and necessary for the
attainment of equilibrium. To summarize Walrass contributions in this
regard: He dened desired real and nominal cash balances and identied
their role; devised a dynamic period analysis and used it in application to
a model of the money market; developed metallic-money and at-money
models; introduced the activities of banks; integrated the rate of interest
11
One scholar argues that money is not compatible with any of Walrass general equilib-
rium models (Bridel 1997, pp. 14145). Discussing a virtual Walrasian model, other
economists, such as Antoine Rebeyrol (1998), present at length a demonstration that
money is fully integrated into it, or rather into their interpretation of it.
P1: KsF
into the dynamics of his model, making the demand for cash balances a
function of that rate; distinguished circulating and savings balances and
identied their role; and analyzed the causes of variations in the velocity
of money.
Thus Walras used his perceptions of the behavior of the real economy
to develop the most sophisticated and realistic monetary model that had
been achieved by any economist by 1889, anticipating central features of
the analyses of monetarymatters made byJ. M. Keynes, D. H. Robertson,
and J. R. Hicks during the 1920s and 30s.
Consumer sovereignty and the allocation of resources
Like other neoclassical economists, Walras bequeathed to later eco-
nomists a concern with the allocation of economic resources among dif-
ferent economic activities. In that regard, he described economic behav-
ior and processes that have come to be known as the action of consumer
sovereignty. Explaining the way that it operates in the real economy to
determine the set of commodities that are produced, Walras noted that
workers are hired to produce the commodities that consumers demand,
and are laid off in industries in which consumer demand decreases.
Therefore, the consumer is in the last analysis, the true demander of
labor (97, 1872, booklet, p. 35). Identifying, in an early exposition,
the entrepreneur as being only an intermediary between the worker
and the consumer (ibid., p. 36), Walras explained that the entrepreneur
transmits the desires of the latter to the production side of the market,
thus allocating resources so that the set of commodities produced is in
accordance with the structure of consumer demands and hence reects
consumer preferences and purchasing power (123, 1877, p. 231; 176,
1889, p. 214; 188, pp. 28384).
That implies that the structure of economic activities responds to
changes in consumer demand. Walras explained that the reason why the
price of a consumer commodity becomes greater than its average cost is
because consumers shift their demand in favor of it. They bid up its price,
thus making its production protable, and that induces entrepreneurs to
produce more of it. The reason the price becomes less than the average
cost is because consumers decide that they do not want to consume as
much of the commodity as before at each possible price. The suppliers
are forced to lower the price, the rms incur losses, and they there-
fore produce less of the commodity (176, 1889, p. XVIII; pp. 1516).
Changes of consumer demand are also transmitted to the capital goods
industries throughincreasingor decreasingthe demandfor the goods that
they supply to consumption goods industries; and changes of consumer
P1: KsF
170 Walrass ideas
demand inuence the markets for the productive services that are used
to produce the consumer goods. Walras incorporated all those matters
into his model.
Welfare economics
Closely related to the concept of consumer sovereignty, Walras devel-
oped a thesis that became a central issue in the study of welfare eco-
nomics, namely that free competition, contrary to the denials of the
socialists (176, 1889, p. 306; 264, p. 425), tends to generate a
maximumof well-being for society (123, 1877, pp. 266, 305; 176, 1889,
pp. 251, 306; 221, p. 334; 264, p. 425). He emphasized that this
results from actions by economic agents to maximize their utility, from
the existence of a set of prices that would equalize supply and demand,
from the sovereignty of the consumer, and from the other features of a
competitive economy that constitute his model. Mixing up his equations
and his model once more, he declared that the theory makes clear that
the mechanism of free competition leads precisely to the solution by
tatonnement of this system of equations; from which it follows that the
mechanism creates maximum satisfaction (239, 1889, 2, pp. 36465;
and see 176, 1889, p. 306; 264, p. 424).
That implicit confusion can be neglected in the interests of identifying
Walrass central point, which is that the maximization of satisfaction
tends to occur in markets for both consumer commodities and capital
goods:
The mechanismof free competition is, under certain conditions and within certain
limits, a self-driving mechanism and automatic regulator of the transformation of
savings into capital goods as well as the transformation of services into products.
And thus . . . free competition in regard to exchange and production procures
the maximum utility of services and of products . . . , [and] free competition in
regard to capital formation and credit procures the maximumutility of newcapital
goods . . . (176, 1889, p. 287; 264, p. 424).
The maximumis a relative one, because it depends upon the distribution
of income and wealth and the dynamic characteristics of a freely com-
petitive economy, features that result in it moving toward a particular set
of equilibrium values. As will be seen in Chapters 9 and 10, subsequent
economists tried to nd the proof of the proposition that eluded Walras.
VI. The legacy of the mature comprehensive model
The chapters in this part of the book have set forth the legacy of Walrass
period of maturity as an economic theoretician, a legacy based upon a
P1: KsF
remarkable grasp of the features of the economic life of his time. By
1889, he had a fully formed conception of the interrelatedness of eco-
nomic phenomena, incorporated into a model of general equilibration
and equilibrium that was almost complete and that was in most respects
well constructed. It dealt with the important parts of a freely competi-
tive economic system exchange, consumption, production, and capital
formation, all undertaken with the use of money. His account of non-
virtual economic behavior was a laudable effort to provide a model that
explained the real economy on an abstract level. He identied the chang-
ing phenomena in disequilibrium such as prices, the variations in hiring
of services in particular rms, and related variations of sales that occur
as entrepreneurs adjust levels of output in order to maximize prots.
The phenomena include the participants holdings of money and the
amounts of each type of capital good. The amount and composition of
investment vary with changes in the costs and prices of capital goods,
in the incomes generated by them, and in the rate of interest on money
savings.
A competitive system of production and exchange is vast and com-
plicated in its many details, but Walras divined its underlying order.
He found simplicity in the motivation of consumers, namely their
desire to maximize utility; simplicity in the driving force that makes
the system responsive to changes in tastes and technology, namely the
entrepreneurs desire to maximize prots; and simplicity in the unifor-
mity of the mechanism whereby markets undergo adjustments, namely
the functioning of competitive pricing. Walrass ideas on these matters
have become standard parts of modern models of competitive market
systems.
P1: KsF
172
P1: KsF
SECTI ON I I
The written pledges sketch
173
P1: KsF
174
P1: KsF
CHAPTER 6
Rationale for the written pledges sketch
and its characteristics
I. The problem and Walrass solution
A review of the problem
It has been seen in Chapter 5 that Walras explained, in reference to reality
and to his mature comprehensive model (see, for example, 123, 1877,
pp. 253, 255; 176, 1889, pp. 235, 238, 294; 209, p. 315; 212, p. 319;
258, p. 401), that production and exchange occur at disequilibrium
prices, varying as prices change during the course of the adjustment of
the markets toward their equilibriumset of variables. In the model, hold-
ings of goods and money are parameters for individual static supply and
demand functions. Disequilibrium transactions alter the distribution of
goods and money, resulting in changes in those functions. Disequilib-
rium production alters the amounts of goods and hence the holdings of
goods, and thus also changes supply and demand functions. Walras came
to realize with increasing clarity that those disequilibrium phenomena
create a problem for the logical structure of his work, considered in
relation to his equation system. Inasmuch as the equations of supply
and demand have parameters that are actually endogenous variables in
the model, their solutions prices, rates of employment of resources,
incomes, outputs, and quantities traded are not the values toward which
the model actually converges. The mature comprehensive model is there-
fore non-virtual in disequilibriumas well as in regard to what takes place
when the equilibrium prices are quoted. Walras realized that the equa-
tions he represented as relating to the mature model of general equilib-
rium do not in fact describe it. They relate to a different type of model.
The written pledges sketch
1
Walras did not try to formulate the equations that could be used to nd the
general equilibriumvalues of the variables of his mature comprehensive
1
The rst part of this chapter summarizes briey the relevant parts of Walker 1996,
Part II, in which the written pledges sketch and Walrass last comprehensive model in
general (see below) are examined in detail.
175
P1: KsF
176 Walrass ideas
model. He decided instead to try to construct a model of which his exist-
ing equations would be a description. He therefore had the objective of
replacing his mature comprehensive model with a virtual one (see Chap-
ter 5, note 2), intending the latter to be the underpinning of and justi-
cation for his virtual system of equations of general equilibrium. In that
failed effort, Walras abandoned the philosophical and methodological
guidelines of his creative and mature phases. He discarded the procedure
of drawing ideal types and general facts from (his perception of) reality.
He put ctional elements into that effort, made fundamental assumptions
that were not only contradictory to fact but also that resulted in a distor-
tion of reality, sketched a scenario of hypothetical situations that had no
relation to the real economy, and presented as conclusions various condi-
tions and outcomes that could not be deduced from his assumptions and
that contradicted the conditions and functioning of the real economy.
Walras rst presented the idea of written pledges in a note published
in 1899 (206, 1899, p. 103). Written pledges is the translation of
the words engagements ecrits or engagements par ecrit that were used
in French securities markets at the time Walras was writing (see, for
example, Marinitsch 1892, pp. 64, 234) and are used today. To refer to
engagements ecrits, he chose the word bons, but because that means
engagements ecrits in the use he made of the word in 1899,
2
it is best
translated as written pledges. That describes the item and practice in
question and is historically authentic.
2
It is interesting that, in the context of discussing a partially planned economy in 1896,
Walras used the word bons with a meaning that is totally different fromwritten pledges.
In that economy, the bons are tokens, denominated as either numeraire-work units or
numeraire-silver units, distributed to would-be buyers by the state. Unlike written
pledges, they have nothing to do with price formation. They are not used to convey
desired supply quantities at different suggested prices prior to production; they are
a medium of exchange, used to pay, at prices that are xed by the state, for goods
that have already been produced (194, 1896, pp. 22830). Some economists contend
that Walrass general equilibrium model, by which they mean their conception of
his 1899 written pledges sketch, is literally a blueprint for a planned economy or is
adaptable for that purpose. One scholar offers the bafing notion that Walras presented
a model economythat is simultaneouslybothplannedandunplannedthe astonishing
paradox of a decentralized economy that turns out to be a planned system (M enard
1990, p. 109; and see De Vroey 1987). Most economists (see, for example, Morishima
1996), however, indicate that it is clear that Walrass models were intended by him
to enable him to understand the capitalist economy of his day, which he regarded on
an abstract level as a private enterprise system of decentralized markets. In fact, the
mature comprehensive model has nothing to do with a planned economy, and there
is nothing in the 1899 written pledges sketch that suggests one. It is true that Walras
considered aspects of a partially planned economy, but, as noted at the beginning of
this footnote, he did that in an explicit treatment of precisely that topic (194, 1896),
not in a development or application of his written pledges sketch.
P1: KsF
Rationale for the written pledges sketch 177
In that rst note and in the sketch that he introduced into the fourth
edition of the El ements (1900, 210), Walras asserted that he wanted to
achieve the virtual property in all aspects of the model and in all its
phases by assuming that suppliers of services and suppliers of products
do not produce or sell disequilibrium amounts (210, 1900, p. VIII;
pp. 67). Instead, they make written pledges to provide the amounts that
they write down on slips of paper, provided that the price that is currently
quoted is found to be the equilibrium price. Consumers write pledges to
supply productive services, and those pledges are supposed to be handed
to producers of products; producers write pledges to supply products
and those are supposed to be handed to consumers (210, 1900, p. 224;
214, p. 323). Walras stated that suppliers vary the amounts offered at
suggested prices until the desired supply and demand of every commod-
ity become simultaneously equal in every market, and that only when
that condition is fullled are trade and other economic activities allowed
to take place (210, 1900, pp. 215, 224, 260, 298, 302; pp. 57; 207,
p. 309; 21314, p. 323; 251, p. 377; 273, p. 441; 274, p. 447).
Walras rst assumed that there is a virtual tatonnement, naturally
without the equilibrium prices being known in advance; that is, with
uncertainty. Then he dealt with a period of time during which transac-
tions and the delivery of commodities go on at the equilibrium prices,
with no uncertainty, just as J. R. Hicks (1939, pp. 12227) subsequently
(without acknowledging Walrass construction) did with the days of his
week following the Monday on which prices are determined. In his
discussions of uncertainty, Walras had reference to constancy of asset
holdings and to knowledge of prices and the timing of transactions. He
modeled behavior regarding given consumer and business inventories
and holdings of money and acquisitions of them over a period of time
starting at the present in this way: By assuming [the] given conditions
invariable during a certain period of time, and by assuming the prices of
products and of services, and the dates at which they are bought and sold
known regarding the entire period, we do not leave open the possibility
for any uncertainty (210, p. 300; 273, pp. 443, 445). He made a paral-
lel construction about given business inventories of primary material and
products, and of money, and the acquisition of inputs during the period.
Once more he wrote that there could be uncertainty resulting from the
possibility of changes in the given conditions of the problemand the dif-
culty of foreseeing them. But, here also, by suppressing that possibility
for a certain period of time, and by assuming [the economic variables
are] known for that entire period, we suppress all causes of uncertainty
(ibid., p. 301; ibid., p. 445). Walrass treatment in that regard is typical
of the character of his last comprehensive model.
P1: KsF
178 Walrass ideas
Those were Walrass assertions, and commentators on the sketch have
accepted them at face value. They have persuaded themselves that there
is, in the fourth edition, a formulation of a complete tatonnement process
conducted by means of written pledges and eventuating in general equi-
librium in which actual production and sales of services and products
occur. That, however, is not the case. Of course, Walras asserted that
there is a tatonnement and that general equilibrium supply and demand
quantities exist in the model and are found, but those were just unsup-
ported postulates, not a consequence of his assumptions, not an outcome
of the structure of the sketch and the behavior of its participants.
Indeed, Walras was claiming results that, for three reasons, are impos-
sible for the model that he wanted to construct that is, impossible given
his particular formulation of the characteristics of written pledges and
given howhe imaginedthey would be used. First, he assumed that would-
be demanders do not make written pledges, in the mistaken belief that
it is unnecessary for them to do so. Nor did he provide the would-be
demanders with any alternative means of expressing their desires. Con-
sequently, no demands are expressed in the markets by either would-
be purchasers of services or would-be purchasers of products. Poten-
tial suppliers of services and potential suppliers of products therefore
have no way of knowing the quantity demanded nor therefore whether
demand and supply are equal or not (Walker 1996, pp. 38283). So
strong, however, is the conditioning of the economists mind in regard
to the supposition that markets always have demands as well as supplies,
that describers of the sketch, including the present writer on two occa-
sions, have read into it something that is not there, namely buyers who
make written pledges, and have furnished their version with that feature
(Schumpeter 1954, p. 1008; Patinkin 1965, p. 533; Jaff e 1967, pp. 12,
18; Jaff e 1983, p. 241; Morishima 1977, p. 55; Walker 1987a, p. 859;
Walker 1987b, p. 765).
3
Second, Walras did not nish his written pledges sketch in some
other major respects, presenting instead a hodge-podge of situations
that, together with the incomplete written pledges markets, have been
called his last comprehensive model. That model is briey mentioned
in the next section of this chapter. This matter of the incompleteness of
the sketch has been discussed elsewhere (Walker 1991; 1996, Part II) so
here it is sufcient to mention some examples of its deciencies. Walras
neglected to provide for the equalization of price and average cost in
3
That idea, totally without foundation, has now been disseminated at the level of text-
books on the history of economic thought, as witness the statement: In the fourth
edition he added the notion of bons or tickets: when a set of prices is tried, producers
and purchasers issue tickets showing what they would buy and sell at those prices
(Staley 1989, p. 174; emphasis added).
P1: KsF
each rm by means of variations in the supplies that producers pledge
to produce. Also, although he mentioned old capital goods and land in
the fourth edition of the El ements, he did not provide markets for them
in the last comprehensive model and therefore, obviously, neglected to
assume that written pledges are used in connection with them. He dis-
cussed markets for circulating capital and money, but he assumed that
written pledges are not used in them. Walras did not assume, regard-
ing the exchange of consumable services among consumers, that con-
sumers write pledges to supply consumable services to consumers who
are would-be purchasers of them; nor that the latter write pledges to pur-
chase consumer services (210, 1900, p. 224; 215, p. 324). Therefore,
neither would-be suppliers nor would-be demanders express their trad-
ing desires with the use of written pledges. In the written pledges sketch,
there is no exchange mechanism on either side of the markets for the
consumable services that Walras asserted are sold; or, more accurately,
those markets do not exist in that sketch.
Third, Walras gave the written pledges process certain features that
would make it impossible to allocate inputs among suppliers, and impos-
sible to allocate outputs among demanders even if demanders were able
to make written pledges. The result of the foregoing problems is that the
device of written pledges is so deeply awed that it cannot serve as a
tatonnement mechanism (Walker 1996, pp. 39394).
Nor is Walrass sketch of written pledges markets realistic in any sense
of that word or useful as an explanatory device or thought experiment.
It cannot be used as the basis of an applied model because of its lack of
demand functions and its incompleteness in other major respects. Even
if it were to be changed and completed, the amended version would
describe behavior that has no counterpart in reality and would fail to
make reference to variables that exist in reality. No data could be drawn
from the real economy to provide the numerical coefcients of an eco-
nomic version of it, and no real economic problem could be analyzed
with it, because all ex post variables production, consumption, savings,
and so forth in the real economy are disequilibrium magnitudes from
the viewpoint of general equilibrium. Inasmuch as Walras assumed that
the written pledges markets are virtual, there is no place for such mag-
nitudes even in an amended version of his sketch.
A surprise discovery
As I wrote to a number of Walrasian scholars on May 15, 2002,
4
during
the previous week I had happened upon an article that dealt with the
4
I addressed my letter to professors Richard Arena, Jan van Daal, Jean-Pierre Potier,
and Antoine Rebeyrol.
P1: KsF
180 Walrass ideas
subject of one of my own articles, Walrass Theories of Tatonnement
(Walker 1987a), and that had almost the same title. My discovery was
La th eorie des t atonnements chez Walras, by Joseph de Zatarin (1972).
Just as William Jaff e had done before him (1969) and as I did fteen
years after him, De Zatarin noted the obvious and well-known fact that
written pledges are used in the fourth edition of the El ements but not
in the rst three, identied non-virtual tatonnements without written
pledges and virtual tatonnements with them, and presented a view of
Walrass treatment of both exchange and production.
De Zatarins introduction to his article seems like some of the ideas in
my article, and his laudable concern was, like mine, the issue of whether
or not the solutions to Walrass equations identify the equilibriumvalues
of the variables in his various models. It becomes clear in the body
of De Zatarins text, however, that our views of the characteristics of
Walrass writings on the subject are not at all the same. For example, De
Zatarin believes that, in the case of production without written pledges,
the equations are invalid because of le r esidu, a concept that he did
not explain, and la dynamique. In actuality, it is the changes in the
parameters of Walrass equations resulting from disequilibrium trade
and production that render their solutions invalid. As an example of
indefensible reasoning, De Zatarincontends that because writtenpledges
are not used in Walrass model of pure exchange, the equations do not
give the equilibrium magnitudes. Apparently explaining his reference
to dynamics, De Zatarin argues that the equations do not do so for
the reason that the variations of the prices quoted by the agents during
the tatonnement change the values of the commodities possessed by the
agents. If that were true, it would also, contrary to De Zatarin, occur if
written pledges were used in exchange, and would occur in a complete
written-pledges model of production also. In each case, the values would
change during the tatonnement. It is not, however, true that the equations
are invalidated by changes of the values. Walrass exposition of supply
and demand functions indicates that a trader has an ex ante demand or
supply quantity that is a function of each possible price of the commodity
and of other variables, and for which the traders stock of the commodity,
money, and preferences are parameters. Each possible valuation of the
traders stock is therefore uniquely related to and determined by each
possible price; the trader takes that valuation into account in determining
his supply or demand quantity at any given possible price. Thus each
valuation is implied by the possible price and the stock, and is therefore
automatically one of the implicit arguments of the Walrasian demand or
supply function. Changes of the valuations no more change the functions
than do changes of the price.
P1: KsF
The functions would be changed if the changes of the price directly, or
via the change of the valuation, change the traders preferences. That is
a possibility that Walras was entitled to assume did not happen, for oth-
erwise static supply and demand functions would not exist; the quantity
would not be a function of an independent variable the price along
with other independent variables. Preferences would not be a parametric
condition but would change with each price. A demand curve could be
dened for given preferences, but the trader could not move along it. If
the price changed, the curve would shift.
Moreover, De Zatarin bases much of his analysis upon considerations
drawn from the modeling of the results of asymmetric information and
from game theory, which are both inappropriate for the understanding
of the market situation that Walras contemplated.
The foundation, substance, and raison d etre of De Zatarins brief
article is his acceptance, without even the most cursory examination of
Walrass written pledges sketch, of the notion that it is a complete func-
tioning model. That means that, as is the case with other past writers
on the subject, the characteristics that De Zatarin attributes to it, ana-
lyzes, and criticizes are in fact his own inventions. Long before writing
Walrass Market Models (1996), I rejected the idea that Walras had a
model of tatonnement with written pledges, for the reasons explained in
this chapter; that is, briey, because his sketch is not only incomplete but
also fundamentally awed as a conception. That is why I did not intro-
duce into my 1996 book the ideas on that subject that I had expressed
in my 1987a article. There is therefore no point in my dissecting the
many problematic aspects of De Zatarins brief and cryptic contentions
about Walrass model of exchange and his treatment of the production
and pricing of consumer commodities without and with written pledges.
I am in disagreement not only with the part of my 1987a article that
deals with the tatonnement with written pledges, but also with Jaff e, De
Zatarin, and other scholars who base their analyses on the supposition
that Walras presented a model of such a process.
The last comprehensive model
The collection of various submodels in the fourth edition, dating from
different editions and differing in structural and behavioral ways, will
be called Walrass last comprehensive model of general economic equi-
librium (Walker 1996, Chapters 17, 18) because he presented them as
such, and for the sake of brevity of expression, although they do not
in fact constitute a coherent model (see ibid.). He featured suppliers
written pledges in what has just been noted are incomplete sketches of
P1: KsF
182 Walrass ideas
some markets, but he did not in any respect carry out his plan (210,
1900, p. VIII; pp. 5, 7) to convert into written-pledges markets all the
older submodels non-virtual components of his mature comprehen-
sive model that he brought over from the third edition to the fourth.
During the course of preparation of the fourth edition, he mentioned
written pledges only a few times and with decreasing frequency and
nally abandoned the subject before he nished the revision, never to
mention it again in any of his writings. Thus he introduced some mar-
kets in the fourth edition such as the markets for long- and short-term
loans in which written pledges are not used. This was the case in his
entire exposition of markets for circulating capital and money. In some
instances he presented, in a particular place in the El ements, a commod-
ity as being traded in a market in which written pledges are used, and,
in another place in the El ements, one and the same commodity as being
traded in a market in which they are not. Walras therefore presented
disequilibrium production and exchange as occurring in some markets
but not in others, a situation contradicted by his (unsupported) assertion
in the fourth edition that exchange and production occur only when all
equilibrium prices are found, contradicted by his virtual equation sys-
tem, and exacerbated by the circumstance that there is no pricing process
nor exchange or production in the written pledges markets. Thus the
chaotic collection of non-virtual and incomplete written pledges markets
do not constitute a functioning general equilibrium model.
Also symptomatic of Walrass increasing difculty in undertaking
theoretical construction is the confused state of his efforts to introduce
a revised theory of money in the fourth edition. He incorporated into
Le con 29 on circulating capital and money in that edition approximately
twenty-seven lines about phases of tatonnement, at the beginning of
whichhe speciedthat writtenpledges are used(210, 1900, p. 302; 274,
p. 447; see Chapter 8, entry 206). He made no reference whatsoever to
them, however, in his immediately following discussion of the phases,
nor elsewhere in Le cons 29 and 30, thus once more leaving his written
pledges sketch incomplete and the functioning of written pledges trading
unexplained. The twenty-seven lines were the reset lines of the last
two-thirds of the Note (206, 1899, p. 103) that he wrote after he
nished his article on the Equations de la circulation (ibid., pp. 85
102) and that he appended to it without in any way recognizing the
content of the note in the article itself; that is, without changing his
modeling of markets for circulating capital and money to reect the
use of written pledges. As the Equations de la circulation became
the almost verbatim exposition of monetary phenomena in the fourth
edition, that exposition is therefore substantially devoid of reference to
P1: KsF
written pledges. Walras did not alter his treatment of these matters in the
nearly identical posthumously published fth edition of the El ements
(224, 1926).
Moreover, the models of markets for circulating capital and money
in the fourth edition are incomplete and indeed nonsensical for a variety
of other reasons (Walker 1996, Chapter 18). As Pascal Bridel rightly
indicated, there is in the fourth edition no genuine monetary theory
(Bridel 1997, p. 133). That was necessarily the case, given the incom-
patible and dysfunctional mixture of nonvirtual and incomplete written
pledges markets in that edition. Money could not be a functioning aspect
of markets in such a context. There is an important lesson to be drawn
from the analysis of the incompleteness of Walrass sketch, which is
why that analysis was worth undertaking (Walker 1996). It is that what
are presented as being economic models should not be accepted as being
logically complete if the results that they are alleged to have are not gen-
erated by institutions, procedures, technologies, rules, and other struc-
tural and behavioral underpinnings.
Various scholars have, of course, written articles that purport to estab-
lish the existence of equilibrium in one or another of Walrass models,
using equations that he set forth in the El ements (for example, Walker
2001, 1, ch. 22; 2, chs. 3, 5, 20, 22). In fact, however, the equations do
not describe the mature comprehensive model, or the written pledges
sketch, or the last comprehensive model, for the reasons just explained.
Those articles on existence are therefore instead examples of how ele-
ments of Walrass constructions have been used by modern scholars in
new elds of study.
II. Accounting for the poor quality of the revisions
in the fourth edition
Most of Walrass theorizing after 1895 was of inferior quality. In par-
ticular, his revisions in the fourth edition of the El ements reveal his
incapacity, in his nal phase of theoretical activity, to formulate the-
oretical constructions that necessitated extended chains of logical rea-
soning, and to develop a coherent comprehensive model. This judgment
that the quality of most of his theorizing after 1895 is poor is indepen-
dent of whether or not it was worse or better before that date, whether
his output increased or not, whether his work was like or unlike that
of other economists, whether his health deteriorated or not, and indeed,
whether or not any explanation of the defective characteristics of his
work is made. In recognition of this, contrary to the allegations of Pascal
Bridel (1998, p. 232), Michel De Vroey (1999), and Fabio Petri (2004,
P1: KsF
184 Walrass ideas
p. 149), in my studies of Walrass last phase of theorizing (Walker 1996,
pp. 321419), I did not mention his health as evidence supporting my
contentions about the quality and quantity of his work. His adoption
of the device of written pledges, for example, was not an aberration
occasioned by the state of his health. Walras had theoretical reasons,
however misguided, for wanting to revise the El ements in the way that
he attempted, and he was perfectly clear about wanting to change his
model of markets to one in which written pledges are used (210, 1900,
p. VIII; pp. 67). It was by means of detailed economic analysis that I
showed that the quality of his work was poor after 1895, and that his
execution of his plan for the revision was awed and incomplete (Walker
1996, pp. 321419), as regards both the written pledges sketch and the
treatment of the markets without written pledges that he introduced in
the fourth edition. I also asserted that the quantity of his theoretical
output diminished (ibid., p. 322).
As a matter of some interest, however although of no consequence
for my economic analysis I sought to explain why the quality and quan-
tity of Walrass work diminished. The most probable explanation I could
think of was that his health deteriorated. In contrast, Bridel contended
that the apparent worsening of Walrass health between 1890 and 1892
has probably more to do with a teaching load he found unbearable. As a
matter of fact, soon after his retirement in 1892, his publishing rate went
up quite drastically hardly a sign of a sudden softening of the brain
(Bridel 1998, p. 232). Bridel thereby imputed to me a diagnosis that I
never made and that I explicitly disavowed when I stated that my attri-
bution of Walrass difculty with theoretical construction late in life to
his poor health was not to suggest that Walras was then or ever became
senile (Walker 1996, p. 322).
Thus Bridel alleged that Walrass health improved after he was free
of teaching duties, and that his productivity consequently increased.
5
Those assertions will be examined. It will rst be seen whether Walras
felt better after he retired, whether he believed that his health affected
his analytical abilities, and then whether his publishing rate went up.
The case for believing that Walrass health deteriorated was made by
him, not by me; I simply recorded his statements in my book. It was
not necessary to quote all his many reiterations of those circumstances;
instead, from the mass of evidence, I chose representative and fully
descriptive assertions (ibid., pp. 32122, 400401), and I repeat them
here, with some additional material. Walras indicated that his health
became worse shortly after he retired in early July 1892 and that it
5
The same contentions are made in De Vroey 1999.
P1: KsF
continued progressively to deteriorate. On August 1, 1892, he wrote that
walks were no longer therapeutic and that he had begun to intersperse
his studies with manual work to prevent himself frompursuing extended
and fatiguing chains of thought (239, 1892, 2, pp. 499500). In July 1893
he noted that he continued to abstain from all intellectual work and in
consequence felt better, leading him to form the intention of abstaining
similarly for another year (239, July 1, 1893, 2, p. 563). His physical
forces and his intellectual forces (239, October 10, 1893, 2, p. 594),
he reported, became tolerable only if he did not work very much and
became intolerable if he did. Nevertheless, he generally did not feel
well. A year and four months after retiring he wrote to Wicksell that
he had attentively read a book of Wicksells despite the state of my
brain, which prohibits me from undertaking that sort of work (239,
November 10, 1893, 2, p. 598).
Six and a half months later about two years after ceasing teaching
Walras stated that he had not been getting better and complained once
more that he was physically and intellectually exhausted: My brain is
still in quite a pitiful state; it is not probable that I will be able to under-
take new work (239, May 27, 1894, 2, p. 603). He explained that he
had to abandon his plan of writing a book on applied economics and
one on normative economics (ibid.). He wanted to write an exposition
of doctrine linking those two books but felt he would be unable to do
that; the effort to do so, he wrote, caused me painful cerebral troubles
(ibid.). He abandoned it also in favor of restating some of his ideas in an
essay (205, 1898, section VII), the rst draft of which took himeighteen
months to complete. Walras lamented that he found sustained analyti-
cal work involving chains of reasoning particularly difcult. Thinking
through the aspects of the equations of marginal productivity is a point
on which I must reect very calmly, because my head is very tired
and refuses all work that is even remotely intense and sustained (239,
October 30, 1895, in 239, 2, p. 651). In February 1896, he was unable
to make more than a few minor changes, all of them purely editorial,
to produce the third edition of the El ements d economie politique pure
(1896). He explained that it differed from the second (1889) only by the
elimination of the lessons on the applied theory of money and by the
addition of three appendices of already published material. That proce-
dure, he stated, enabled him to avoid trying to undertake a revision, a
task, he wrote, which I fear I am no longer capable of doing (192,
1896, p. X). He wrote to a friend in November 1896: I am no longer
involved with economic theory and normative economics. I am tran-
quilly going to take two years to have the volume of Applied Economics
printed and my work will be nished. During the boring days of winter,
P1: KsF
186 Walrass ideas
he explained, my head can no longer work and even distractions exhaust
me (239, November 11, 1896, 2, p. 702).
Walras began to prepare the fourth edition of the El ements in the
rst part of 1899, and in the course of doing so altered his treatment
of money. That task, he reported, has made me very tired and even
quite indisposed (239, May 6, 1899, 3, p. 71). He nished it only with
innite difculty and fatigue which have made me ill and left me in
that condition all summer (239, November 19, 1899, 3, p. 93). A few
days later, he wrote, initially referring to 189899: In 1899 (after
41 years of work and strife, being 65 years old and completely at the end
of my forces), I had to retire denitively fromactive political economy,
although he thought he would be able to nish denitive editions of his
works (249.XIII, November 26, 1899, p. 513; and see p. 517). That was
his condition during the rst few months of 1899 when he wrote the
Equations de circulation, which, as noted above, became the revision
of the monetary theory in the El ements, and that was his condition during
the ensuing summer when he made most of the rest of the revisions to
the El ements to produce the fourth edition.
The revision took him longer than he expected. He was not quite
nished with it by October 1899, but he ceased work on it, writing that
I have renounced two things which fatigue and irritate me [overexcite
my nerves]: the problems of mathematical economics and attempts to
formulate contemporary propaganda
6
(239, October 6, 1899, 3, p. 90).
He began a period of rest lasting about ve weeks that, he reported on
November 19, enabled himto recover (239, 1899, 3, p. 93). He evidently
felt that the revision was incomplete because he then began again to try
to make some changes, but he found it so difcult to concentrate and
construct that his sporadic efforts dragged on for three more months,
until the end of March 1900, by which time he was dead tired (239,
April 7, 1900, 3, p. 114). It had taken himeleven months, fromMay 1899
to the end of March 1900, to write some half-lines, lines, and pages that,
added together, total a little less than 34.5 pages, and his modications
had dwindled in the course of the revision until nally he abandoned the
task and left it unnished (as shown in Walker 1996, Part II).
A few months later, he wrote that he could not read or write or
talk very long without having a crisis of cerebral congestion during the
night. In short, my brain, which for a long time has asked for mercy
has openly revolted and refuses me its service (239, October 27, 1900,
3, p. 125). He could not follow chains of reasoning, noting in 1901
that my cerebral and nervous fatigue is extreme and my attention span
6
He meant persuasive policy proposals.
P1: KsF
is very limited (239, December 21, 1901, 3, p. 182). His condition
continued to worsen, and by 1903 it was very bad: My nerves and
my brain are extremely weakened and fatigued. I can hardly any longer
write or converse without having an attack [a nervous crisis] during the
night (239, April 6, 1903, 3, pp. 22324). I am actually suffering very
much, he wrote in 1904. My strength is totally exhausted. I cannot
converse, read or write for half an hour without having an attack during
the night (239, February 28, 1904, 3, p. 246). His daughter read to him
and he dictated letters to her, but, he explained, I do not have her read
economics, and I do not dictate to her any letters except those that are
as short as possible (ibid.). In October 1904, he was invited to write
a piece on A.-A. Cournots contributions to economics, but he could
not manage to do the work so it was assigned instead to Albert Aupetit
(249.VII, p. 444). He explained: I suffer froma cerebral neurosis which
forbids me to do any work at all, even that of writing a letter unaided
(239, November 3, 1904, 3, p. 256). What he produced after ten months,
with Cournots name in the title as a formality, was six pages that had
the objective of referring to some of his own past writings, and that in
actuality did little more than list them (213, 1905).
Thus Walras stated that his health continued to deteriorate after his
retirement and that it worsened with each passing year. His contention
is credible. A condition is properly described as a deterioration of a
scholars mental and physical state if it troubles his waking hours and
disturbs his sleep, leads to his incapacity to concentrate for very long,
initially limits his efforts to write, then forces him to abandon totally
his revisions of his writings and theoretical research in general, and
ultimately renders him unable to read or write or talk for more than half
an hour.
III. Walrass productivity
The purpose of the following account is to establish the facts and to
draw attention to the need for careful investigation before making pro-
nouncements about Walrass productivity. I was in a position to assess
it because I spent a great deal of time during six years compiling and
annotating the list of his writings (Walker 1987c), and I am in an even
more advantageous position to judge his output after having spent a great
deal more time rening the bibliography (Appendix, Part 2) during the
nineteen years since I prepared the rst edition. Inasmuch as I reported
that by 1896 Walras was publishing very little theoretical work, and
occupied himself, on a greatly diminished scale, primarily with applied
topics and with his old papers (Walker 1996, p. 322), and inasmuch
P1: KsF
188 Walrass ideas
Table 1. Walrass Productivity: Number of Pages Newly
Published or Written
Economics
Theory Other Total Other Grand Total
18581871 260 1,310 1,570 501 2,071
18721877 448 218 666 135 801
18781895 479 271 750 245 995
18961900 59 98 157 0 157
19011910 0 50 50 3 53
as he and I both asserted that he became incapable of undertaking new
theoretical constructions, it is reasonable to ask whether the objective
record supports those contentions. Are those statements true or (as has
been alleged) false (Bridel 1998, De Vroey 1999)?
Walrass productivity is measured by the number of pages he wrote,
as rst shown in Table 1. The details of the calculations are given in
the Appendix to this chapter. The periods selected are those to which I
referred in Walrass Market Models (1996, pp. 713) as the phases of
his journalistic and experimental period (18581871), of great creativity
(18721877), of intellectual maturity (18781895), and of decline. That
latter phase is divided into the periods 1896 through 1900, the year of
the publication of the fourth edition of the El ements, and 1901 thorough
1910, the year of Walrass death. The data are used to calculate Table 2,
whichshows the average number of pages per year. The results are robust.
Changes in the years included within each period and even changes in
the dating or classication of one or another publication do not affect
the general picture.
The results are also quite clear. It is strikingly evident that Walrass
average annual theoretical output was much greater in his creative
phase, 18721877, than in any other. The economics-other column
Table 2. Walrass Productivity: Average Number of Pages Per Year
Economics
Number
of Years Theory Other Total Other Grand Total
18581871 14 18.57 93.57 112.14 35.78 147.93
18721877 6 74.67 36.33 111.0 22.5 133.5
18781895 18 26.61 15.06 41.67 13.61 52.28
18961900 5 11.8 19.6 31.4 0 31.4
19011910 10 0 5.0 5.0 0.3 5.3
P1: KsF
for the years 18581871 records the pages in his three early economics
books and his voluminous output as a young journalist during the years
18581871. Despite that output, which was on a level of economic
sophistication lower than the applied studies of his creative phase, his
total economics average productivity was as great during the latter phase
and his total productivity did not decrease greatly. His average output in
every category was much less in every phase after 1877. It is also evi-
dent that after 1895 his average annual output of economic theory, total
economics, and other writings (and therefore his total output) declined,
and that his output in every category was small in absolute terms.
Let us examine even more closely Bridels contention that it is a
matter of fact as distinct, he meant, fromwhat he wished to portray as
mistruths that I concocted that Walrass productivity increased quite
drastically shortly after his retirement in 1892 (Bridel 1998, p. 232,
emphasis added). Counting new writings, as is done throughout this
examination, the facts are that in 18881892, the ve years prior to
1893, Walras published a total of 147 pages of economics (and zero of
other) writings, or 29.4 pages per year, and that in the next ve years,
18931897, he published 126 pages of economics (and zero of other)
writings, or 25.2 per year that is, not a drastic increase, not a medium-
sized or even a small increase, but in fact a decrease.
What about Walrass contention that he was running out of mental
energy in regard to his economic theorizing after 1893? Walrass out-
put of theoretical writing was an average of 28.4 pages per year during
18881892. Then, after his retirement, reecting his admitted decline
in mental powers in constructing chains of abstract reasoning, his the-
oretical writings fell to an average of 1.2 pages per year in the period
18931897 a drastic decrease. Let 1890 be selected as an initial year
so as to exclude Walrass revision of the El ements published in 1889.
That will eliminate the El ements augmentation of his pre-retirement
average output and will thus present the most favorable possible setting
for Bridels contention that Walras became very much more productive
after he retired in 1892. Walrass total number of pages of new theo-
retical writings was four in 1890, eight in 1891, and eight in 1892. The
average for those three years was 6.7 pages. The average then fell to zero
for the next three-year period. Specically, the facts are that shortly after
his retirement, the number of pages of new theory that he published was
zero in 1893, zero in 1894, and zero in 1895. By the way, it is not that
Walras was writing theoretical material that would later be published.
He was not revising the El ements during those three years: he was not
writing any new theory at all.
P1: KsF
190 Walrass ideas
Walrass theoretical productivity, total and per year, continued to
diminish after 1895, and it was virtually exhausted during the sum-
mer of 1899, when he was trying to revise the El ements. Moreover, if
theoretical creativity were the criterion by which his output was judged,
it would have to be recognized that there was little of that after 1889.
The tables show this clearly, but it is evident even by taking averages
of long periods. During the fteen-year period 1874 through 1889, he
wrote 708 pages of theory, an average of 44.25 pages per year. In con-
trast, during the eleven-year period 1890 through 1900, he wrote 79.5
pages of theory, 53.5 of which were the defective material that he put
into the fourth edition of his El ements (19 pages of the Equations de
circulation including the Note (206) plus 34.5 additional lines). That
is an average of only 7.2 pages per year. The facts of the matter are the
contrary of what Bridel and De Vroey asserted.
Apart fromthose theoretical writings, what Walras did after 1895 was
to work with old papers and reect on the past, as shown by the publica-
tions listed in Chapter 8. He assembled collections of his old essays. He
developed plans for the publication of the collected economic writings
of his father and himself. He compiled an autobibliography and wrote a
brief autobiographical note. He wrote a few pages of applied economics
that were elaborations of ideas that he had conceived many years previ-
ously restatements of methodological and philosophical views, and of
his favorite economic policy schemes.
What about Walrass contention that his health in the period 1900
to 1907 was continuing to deteriorate so that he found theoretical con-
struction impossible? What about his assertions that he became virtually
incapable of writing anything unaided, even a letter? Bridel and De
Vroey have inexplicably pointed to my quotations of Walrass letters as
evidence of what the two scolars represent is my inability to recognize
truth, and evidence of the ridiculousness of the notion that Walras had a
phase of intellectual decline manifested in decreased productivity. It has
just been shown that their contentions are contrary to the facts regarding
years prior to 1900. Is their view supported by the data for subsequent
years? Was Walrass last decade fruitful? If he was accurately repre-
senting his mental and physical condition, the record should show that
he became unproductive. Table 3 presents the evidence. Virtually all the
revisions to the 1900 edition of the El ements were made in 1899, so a row
could be added for 1900 showing little or no theoretical writing in that
year also. That row would also show only three pages of economics-
other writing and zero other writings, for a grand total of about three
pages.
P1: KsF
Table 3. Pages Newly Published or Written,
19011910
Economics
Theory Other Other
1901 0 0 0
1902 0 0 3
1903 0 0 0
1904 0 0 0
1905 0 6 0
1906 0 0 0
1907 0 0 0
1908 0 14 0
1909 0 28 0
1910 0 2 0
The data speak for themselves regarding the quantitative facts. They
reveal clearly the continuing and intensied cruel effects of Walrass
ill health on his productivity. As he indicated in his letters and as the
record shows, he made no more attempts at writing economic theory
after about the end of 1899, and he wrote a total of only nine pages
of new economics (economics-other) and other professional writing
(other) during the seven-year period 1901 through 1907. The six pages
he published in 1905 (213) have not a single novel idea in them; they
concern brief descriptions of his major publications, as explained ear-
lier. In 1906 he assembled La paix par la justice sociale et le libre
echange (215), published in 1907. It was excluded from the count of
his pages because it consists of summaries of and extracts from his ear-
lier publications, Recherche de lid eal social (90, 1868), the Esquisse
dune doctrine economique et sociale (205, 1898 Part VII), and a ver-
batim reproduction of almost all of the Th eorie du libre echange (197,
1895). The fourteen pages recorded in the table for 1908 constitute an
article (216) on his fathers economics. That is a descriptive piece that
made no demands upon Walras and that contains no newideas. Similarly,
the twenty-eight pages he wrote in1909are not reallynewresearch. They
consist of two articles. One is an undemanding descriptive article (218)
that contains not a single new idea. In it he quoted Louis Ruchonnet, the
Swiss statesman, at considerable length, summarized very briey some
of his own publications using phrases that he had often used before,
advocated yet again his proposal that the government purchase all land,
praised Switzerland for providing him with intellectual freedom, and
P1: KsF
192 Walrass ideas
that is all. The other article (219) is one in which he sketchily repeated
economic formulas and discussions found in the El ements, and restated
briey his decade-old notions about the parallels between economics
and mechanics.
7
Walras did not accomplish any original research after
March 1900.
APPENDIX
The source for the data in the tables is the denitive bibliography of
Walrass writings (Chapter 8). The numbers in parentheses are those
assigned to the entries in the bibliography. The following procedures
were followed.
The items that appear in the economics-theory column are those
that Walras classied as theory, or that are economic models or theoret-
ical constructions. If Walras classied a text as applied economics, then
the relevant pages were put into the economics-other column, except
for some items that were, nevertheless, credited to him as economic
theory. This was true in the case of entries 148, 153, 154, 158 and 169.
His article on H.-H. Gossen (163) is history of economic thought, but
its analytical and critical approach merits it being counted as theory. A
dubious classication was the placement of one of his early treatments of
taxation, Th eorie critique de limp ot (27) in the theory column. Walras
classied a note on the quantity theory of money (199) as applied eco-
nomics but it is here classied as theory. A small part of the material
in some of the entries classied as economics-other is theoretical in
nature, but in most cases they are truly applied economics and not works
of theory, much less of original theorizing. Moreover, there are many
pages in the economics-other column that could be put into the other
(non-economics) column because they deal with philosophy or method-
ology or thoughts about human nature or the desirable way of organizing
society or other such topics. This is true of entry 189, Th eorie de la
propri et e, which has some unoriginal theoretical statements but deals
principally with social questions in a normative way. The other col-
umn records Walrass output of literary productions such as his novel,
short story, literary criticism, and reviews of paintings.
Excluded from the statistics were republications of Walrass writ-
ings. Also excluded, because they were not research or were not his
own writings, were his personal note (183); his assembly of his fathers
7
Jean-Pierre Potier describes the article well by saying that it does not constitute
new research, but is instead a summary of Walrass past contributions (Potier 1994,
p. 277).
P1: KsF
bibliography (187), his lecture notes (252, 253) which did not do more
than present the ideas and frequently very nearly the same language
as are found in his publications; his mainly undated Notes dhumeur
(251); his translation of W. S. Jevonss treatise (144); his contribution to a
translation of H.-H. Gossens treatise (260); his abridgement (230) of the
El ements; his autobiographical note (212); and his autobibliographies
(214, 240, 262). The few pages of his undated and unpublished essays
(250, 25459, 261), all empirical in content, were not counted precisely
because it is not known in what year they were written, although Walras
evidently wrote them all during his early years. He prepared the lecture
notes mainly during the 1870s. Adding the undated essays and the lec-
ture notes among Walrass new writings would therefore have the effect
of augmenting the contrast between his earlier and his later productivity,
thereby increasing the imbalance pointed out in this study.
The Recherche de lid eal social was published as a series in colum-
nar form in Le Travail in 1867 and 1868. Three of the parts were pub-
lished in 1867 and three in 1868. The series was republished in book
formin 1868 (90), so instead of using the number of newspaper columns,
the 226 book pages that they required are credited partially to 1867 and
partially to 1868. The lengthy introduction to the book is credited to
1868.
Unless there were exceptional circumstances, as was the case regard-
ing the m emoires discussed below, if it is known that Walras wrote an
entry in a year earlier than the date of its publication, the entry is cred-
ited to the earlier year. For example, he wrote entry 51, published in
December 1866, in January 1863, so it credited to that year; he wrote
entry 197, published in July 1897, in March 1895, so it is added into the
total for 1895. This procedure was not applied to Walrass four theoreti-
cal m emoires (103, 118, 119, 122). They were not counted inasmuch as
all the ideas and much of the identical language in those m emoires are in
the rst edition of the El ements, which was counted. It would have been
too complicated to do otherwise, and then to try to pick out the parts of
the El ements that do not duplicate the m emoires. Walras read one of them
(103) in Paris in August 1873, before the publication of the rst part of
the El ements (106), and read another (118) in Lausanne in December
1875 after the publication of that part, but both m emoires were included
in it. He wrote much of Part 2 of the El ements d economie politique
pure (123) before 1875, but not all, because he continually revised the
text during the period 1875 to September 1877, even altering the proofs.
Part 2 of the El ements includes the ideas and most of the language of
the two m emoires (119 and 122) that were read to the Soci et e vaudoise
des Sciences naturelles in 1876, one of which was published in that year
P1: KsF
194 Walrass ideas
and the other in March 1877. It is not possible, however, to determine
specically the date on which Walras wrote many of the words of Part 2,
so all of it is credited to 1877, the year of its publication. This treatment
does not inuence the outcome shown in the tables because the four
memoires and the rst edition of the El ements were all published in the
range of years of Walrass creative period.
Walras wrote the manuscript entry 95 in 1871. It comprises 37 auto-
graph pages. As published in entry 249.XI, pp. 385409, it takes a little
more than 24 pages, which have been duly added into the count of the-
oretical writings.
Walrass essays on economics education are a mixture of normative,
philosophical, and methodological analyses of economics studies, dis-
cussions of free competition as a policy, other economic policies, other
applied economic topics, and non-analytical references to the views
of other economists on such topics, so they have been classied under
economics-other. This is true of entries 97, 141, and 152. Walras wrote
entry 141, published in 1879, partly during 1878 and partly during 1879,
so it is divided between those two years. Walrass publications in 1891
(179) and 1892 (180) were credited to those respective years and not to
the third edition of the El ements (192, 1896), in which they were repub-
lished as Appendixes I and II, respectively. Entry 181 was not counted
because it is a translation into English of entry 180. For the second
edition of the El ements (176), the number of newly written pages was
calculated as the length of the second edition minus the length of the
rst edition, namely 116 pages. They were attributed to 1889. No pages
were counted from the third edition (192) because it is simply a reprint
of the second edition minus the lessons on applied monetary topics plus
the addition of the appendixes. The 18941895 Note sur la Refutation
de la Th eorie anglaise de Fermage de M. Wicksteed (entry 191, with
lines that add up to 8 pages of print), became Appendix III. In the pre-
sentation in entry 249.VIII, pp. 715, 71722, that appendix has lines that
add up to 6.25 pages. Entry 191 was not counted, however, because it
does not contain new research. As noted in the Appendix, Part 2, Walras
simply patched it together by using passages and equations from the
second edition of the El ements, by reproducing Wicksteeds equations
and quoting or paraphrasing his work, and by presenting an analysis
authored by Enrico Barone.
The fourth edition of the El ements (210) contains only 34.5 pages of
newly published writing, calculated by counting precisely the lines and
pages that Walras added to that edition, but excluding his partial reword-
ing of the short subtitle indexes preceding each lesson and excluding
lessons 29 and 30 on the theory of money except as indicated below,
P1: KsF
because those lessons are a republication of entry 206, the Equations
de la circulation, with a few slight changes. The 19 pages of entry 206
are credited to 1899, the year Walras wrote it. When he put it into the
El ements, he eliminated some lines in the two Le cons, and added 29
lines of new writing (see entry 206), which equal about 0.8 of a page
in the fourth edition. Those lines are duly added into the total of new
writings for 1900.
Walras wrote a version of many or all of the 34.5 net new pages in
the fourth edition during 1899, so many of them should be credited to
that year, but he continued to make small changes until the end of March
1900, so some of the words could be credited to 1900. All the pages are,
in any case, allotted to the period 18961900, and the grouping of years
makes the allocation as between 1899 and 1900 unimportant.
Entry 188 was published in 1896, but its twenty-two pages are not
credited to that year but to 1868, when Walras wrote them. The expansion
of the ideas in entry 188 to ninety pages (entry 99), although dated
18721873 by Walras, appeared in 1872, and is credited to that year.
Walrass publications in newspapers were quantitatively of impor-
tance in some of the years during the 1860s. Where possible, newspa-
per columns were counted as equaling the number of pages that were
required to reprint themin book form, and where possible the pagination
of the republication of the article in entry 249 was used. A newspaper
column was otherwise counted as being equal to two pages.
The Etudes d economie sociale (193) is excluded, except for the mate-
rial not previously published, namely a Note of two pages in length
(see 193). The Etudes d economie politique appliqu ee (205), is excluded,
with two exceptions. The rst is the Note sur la th eorie de la quantit e,
which was not previously published. The second relates to group VII of
the collection, titled Esquisse dune doctrine economique et sociale.
Part of the Esquisse had been published, namely Politique fran caise.
La pri` ere du libre penseur (202, 203); those two essays were origi-
nally published in the same year as the Etudes were published. Most
of the substance of the previously unpublished parts of the Esquisse
had already been published. Walras wrote several drafts of them during
the period September 1893 to February 1895, one of which amounted
to twenty-ve pages. He continued to add to the text, nishing a revi-
sion shortly before March 1897. He then kept working on the text, not
completing the last version, thirty-one manuscript pages in length, until
April 1898 (see 249.X, pp. XIXXX). Walrass increasing difculty of
composition is well illustrated by his taking four years and four months
to write those thirty-one pages. The article also reveals two characteris-
tics of most of his work after 1895: the repetition of ideas that he had
P1: KsF
196 Walrass ideas
expressed long before the 1890s and the treatment of themes that were
not on the same level of difculty as economic theory. Nevertheless, the
28 previously unpublished pages of the last version of the Esquisse,
plus the 9 pages of the Politique fran caise and the Pri` ere, as pub-
lished in entry 249.X, pp. 40532, are credited to himas newwritings.
In an effort to reect the dates when he wrote them, they have been
divided into six parts. Seventeen pages are attributed to 1893 and the
remaining twenty pages are divided up by allotting four pages to each
year during the period 1894 through 1898. Only two other entries for
the year 1898 were newwritings, namely numbers 200 and 201, totaling
thirty-ve pages.
P1: PIG
SECTI ON I I I
Walrass writings
197
P1: PIG
198
P1: PIG
CHAPTER 7
Some bibliographical remarks
I. Description of the denitive bibliography
I nished the preparation of the rst edition of the bibliography of L eon
Walrass writings in 1986 (Walker 1987c). These remarks and the bib-
liography that follows are revised versions, made necessary by new
information and publication events during subsequent years. I compiled
the rst edition of the bibliography in the conviction that it would be a
valuable research aid to scholars who are interested in his work, and that
has proved to be the case. Articles are written on Walrass ideas each
year and many other papers refer to his ideas and cite his publications,
and the authors are no longer at the disadvantage of not knowing of the
existence of many of his writings. Formerly, scholars were familiar only
with the part of his writings that appears in his El ements d economie
politique pure and in the two volumes of his collected papers (194 and
205), although that is a very important part. Especially in need of the
information provided in this bibliography is the historian of thought who
wishes to explore the development of Walrass ideas.
Since the time that I prepared the rst edition, the Centre Auguste et
L eon Walras (recently renamed TRIANGLE, Unit e Mixte de Recherche
5206 du Centre National de la Recherche Scientique) at the Universit e
de Lyon 2 began the project of publishing Auguste and L eon Walrass
uvres economiques compl` etes (OEC, entry 249), and all the volumes
in that series that are devoted to L eons writings have nowappeared. This
bibliography and those volumes are complementary. Chronological and
annotated, this bibliography will be useful to the Walrasian scholar not
only because of whatever it possesses in the way of comprehensive-
ness, precision, and wealth of detail, but also because it draws to the
researchers attention economic entries that can be readily located in the
OEC. It also lists and annotates many items that are not in the OEC. In
the following remarks, the publication of Walrass writings in the OECis
excludedfromconsideration unless specicallyindicatedtothe contrary.
Walras published an autobibliography in 1897 (see entry 196) that is
incomplete in many respects other than the obvious one of not including
199
P1: PIG
200 Walrass ideas
writings that date from years after its composition. The other bibli-
ography that has thus far been published, presented by Georges-Henri
Bousquet in 1964 (238), is Walrass manuscript autobibliography of eco-
nomic writings compiled by him in 1906 (214), with some additions.
It contains the following entries listed by Walras: 125 numbered items,
ten unnumbered items that were unpublished at the time that Walras
compiled the list, and one unnumbered item (199 in the present bibli-
ography) that was published for the rst time in 1898 (see 205). It also
contains three entries published after 1906 that were added by Bousquet
(see 238). Thus it has a total of 139 entries. That autobibliography is
unsatisfactory. It is seriously incomplete regarding both published and
unpublished items. Some entries are assigned to the wrong year. Some of
the titles are erroneous and many of them are incomplete. The citations
are all incomplete because Walras neglected to give the numbers of the
journal issues and the pagination of his articles, the journal volume for
most of his articles, and the place of publication and publisher of his
books. He briey annotated some of the entries, as the complete tran-
script of his autobibliography in my possession shows, but none of his
annotations was included by Bousquet. In his autobibliography Walras
classied the entries into six subject-matter categories such as economic
policy, applied economics, and economic policy and justice. His cate-
gorization has not been copied in the present bibliography because its
taxonomic accuracy and value is debatable with respect to some of the
items, and because he did not list and hence did not categorize many of
the entries that appear in the present bibliography.
Every effort has been made to make the bibliography a useful research
tool. It corrects all the deciencies of Bousquets presentationof Walrass
autobibliography and contains a great deal more information. It provides
a much more complete listing than has been given before. With respect
to the initial publications of Walrass writings, it is probably complete,
with the exception that references have not been included to publications
in journals of Walrass letters if they also appear in Correspondence of
L eon Walras and Related Papers, edited by William Jaff e (239). There
are a few exceptions to this rule. For example, entry 101 appears in the
bibliographybecause it was anopenletter, publishedina journal before it
was mailed; entry 164 appears because it is a letter addressed to the editor
of the journal in which it was published; and entry 211 was similarly
an open letter published in the month that it was written. I use the
term probably complete because I have inferred from my experiences
during the years in which I compiled the bibliography and revised it
that no amount of diligence can exclude the possibility that minor items
have escaped my search. Be that as it may, this bibliography includes
P1: PIG
Some bibliographical remarks 201
all the entries that appear in Walrass autobibliography, and gives many
additional economic entries plus the bibliographies themselves (214 and
238). In contrast to Bousquets presentation, the economic entries that
date from before Walrass death in 1910 total 191, not counting his
autobibliography or his autobiography in French (212) and in Italian
(217), or his bibliography of his fathers writings (187), although he so
included it. These gures include reprints and translations. Microforms
are not included.
With respect to unpublished economic manuscripts, the present bibli-
ography lists 36. In this discussion, an unpublished manuscript means
one that was unpublished during Walrass lifetime. Some of them are
difcult to classify as such, however, because, like entry 151, they were
used as the basis of a text that was published during Walrass lifetime.
The 36 entries probably include every manuscript that Walras intended
to be a research paper or extended statement, as distinct from jottings
and brief notes. Despite the publication of the OEC volumes since the
preparation of the rst edition of this bibliography, in the present edition
the entries for economic manuscripts have been presented as they orig-
inally appeared, so that the reader can learn their titles and the fact that
they were unpublished during Walrass lifetime, and can then turn to the
OEC for the texts. The bibliography indicates if they were published
prior to their appearance in the OEC, and gives the entry number of the
published version.
The bibliography includes the compilations of Walrass jottings and
brief notes that have been published (242; 249.XIII, pp. 509622). The
compilations are unclassiable entries that are counted only in the grand
total. Furthermore, the bibliography is complete in the respect that it
includes Walrass 26 literary (non-economic) writings, some of which
have relevance for his economic ideas, and his Bibliographie litt eraire
(263), of which I possess a complete transcript, whereas his autobib-
liography lists only his economic writings. Thus the total number of
entries is 263. Of course, that is not an indication of the amount of orig-
inal material in his texts. His practice of republishing his articles and
memoires, and of including verbatim copies or the substance of already
published material in texts with a new or similar title, greatly increases
the number of entries in the bibliography, as do reprints, translations, and
the separate listing of both unpublished and published forms of texts.
The present bibliography establishes the correct title of each entry and
presents the titles that Walras, in his autobibliography and Bibliogra-
phie litt eraire, gave to items published without a title. Unlike the lists of
Walrass writings prepared by other scholars, the present one indicates
the dimensions and location of unpublished manuscripts. It provides the
P1: PIG
202 Walrass ideas
date of composition of 53 of the unpublished manuscripts. It describes
the contents and pagination of the contents of volumes of collected
papers. It gives complete bibliographical information on republications,
indicating whether changes were made in the entry when it was repub-
lished. With respect to articles, it species whether they were reprinted
as offprints and whether they were republished in a collection. It gives
the location in the Correspondence of the many summaries of entries
provided in that source. It includes Walrass annotations, indicated by
the phrase, Walras explained that . . . or by a similar phrase, if they
add information about the entries.
Regarding published materials, which constitute the vast majority of
Walrass writings and are by far the most important part, I have been able
to establish almost all the bibliographical information by using my col-
lection of originals or copies of Walrass publications. I supplemented
the information in my collection by using the papers and other mate-
rials in the William Jaff e collection that is now in the York University
library in Canada, by trips and telephone calls to libraries, and by com-
municating with Ms. Marlyse Vernez and Mr. Christian Graf, reference
librarians at the Biblioth` eque Cantonale et Universitaire in Lausanne, to
whom I am very grateful for clear, detailed, and accurate information.
I wish to thank Ms. Carol E. Connell and Mr. Blaine Knupp, bibliog-
rapher and reference librarian, respectively, at the University Libraries,
Indiana University of Pennsylvania, for their skilled assistance in track-
ing down obscure texts in a variety of languages. I also wish to thank
Kunio Nakakubo of Himeji Dokkyo University, Japan, for supplying
some of the information that appears under entry 234 and all of the
information for entries 225, 246, and 247. For information that appears
under entry 226 and especially under entry 234, I am indebted to the
editorial section of Iwanami Bunko. I am grateful for nancial support
from Indiana University of Pennsylvania for my preparation of the rst
edition of the bibliography. Since preparing it, I resigned my positions
at IUP in order to devote myself to research, and the denitive edition
is one of the fruits of my current independent way of life.
Regarding Walrass unpublished economic and non-economic manu-
scripts, I have used a variety of sources to identify them. Those materials
were either listedbyWalras inhis autobibliographyof economic writings
or in his Bibliographie litt eraire, listed by WilliamJaff e, together with
their archival location and classication number, in the notes to the
letters in his edition of Walrass correspondence, identied through my
examination of microlms and transcripts collected by Jaff e, or noted by
him in his papers, together with their archival location and classication
number. With respect to the unpublished manuscripts, I am therefore
greatly indebted to Jaff e.
P1: PIG
It should be noted that, with a few exceptions, Walrass unpublished
professional papers constitute part of the Fonds Walras, the archival
materials deposited in the D epartement des manuscrits of the Bib-
lioth` eque Cantonale et Universitaire de Dorigny, on the university cam-
pus near Lausanne, Switzerland. The materials in the Fonds Walras have
been reclassied, but the old classication numbers, referring to folders
(in roman numerals) and items within those folders (in arabic numerals)
in the archives, are still valid and are given in the present bibliography.
The Lyons Collection, to which reference is made in entry 8, was called
the Collection Walras de la Facult e de Droit de Lyon. That collection,
which had not been classied at the time that I listed that entry, is now
under the care of the research unit 5206.
II. Jaff es plans
Jaff es work on Walrass writings leads one to wonder why he did not
publish a bibliography of them. The answer is an interesting story that
deserves to be told in this context. Jaff e did not prepare and publish
Walrass bibliography for the same reasons that he did not write Walrass
biography, because the bibliography was to be part of that biography.
Specically, it was to be the last part, in the manner, he told me, of the
bibliography at the end of Lionel Robbinss book on Robert Torrens.
The reasons are made evident in the following excerpts from our corre-
spondence. I rst quote froma letter to Jaff e that I sent himon January 7,
1980, in response to a letter in which he had referred to the project that
he titled the Life and Writings of L eon Walras as the other thing.
I think that you may be incorrigible in your tendency to stray from the straight
and narrow path. . . . Im forced to conclude that you have tended to avoid THE
OTHER THING for the last 25 years . . . because you really dont want to work
on it. You dont want to go through the effort of immersing yourself in all the
source material so as to recapture the perfect intimacy and uency with the source
material that would facilitate the documentation of the account of LWs life. I think
other subjects interest you more than what you have come to regard as a chore. I
will therefore conclude this topic by exhorting you to take up the task once more,
although I must admit, in compassion, that Im reminded of the lines by Ogden
Nash in his Ode to Duty. They go something like this: Oh Duty, why doest
thou not have the visage of a sweetie or a cutie! We must embrace Duty despite
her austere countenance!
Jaff es response, in a letter to me dated January 28, 1980, was as follows:
What you say about my notorious delay in getting out the Life and Writings
book may be true. May I be pardoned for seeing the matter differently? I do not
think I really did not want to work on it, as you suppose. From the very start I
did not want to do anything else; but I was inhibited. I doubted whether I could
P1: PIG
204 Walrass ideas
do justice to the materials I was quarrying, not only because of their mass and
complexity, but also because I was afraid I lacked the literary capacity to turn
these materials to readable account, to penetrate their psychological and historical
import, and to reveal their analytical signicance, at least in a manner acceptable
to modern economists. If it was principally for economists that I was writing,
I felt technically inadequate. I tried then to make up for these inadequacies by
turning my attention to recent developments in theoretical economics, but found I
had neither the training nor the ability to keep up with it. Moreover, biographical
writing as such did not appear to me to have any standing in the guild of which I
wanted desperately to be a member in good repute.
How could I hope for the professional recognition I craved from economists,
as I spent year after year in archival research, learning to decipher diabolically
obscure manuscripts . . . ? Surely cost-benet analysis would have demonstrated
that if the same time and effort were spent on mastering matrix algebra, topology,
set theory, etc., I should have fared better in the eyes of my colleagues. And yet
I stubbornly, if not wisely, persisted in my thankless grubbing.
Jaff e should have added that he directed his attention for many years to
his translation of Walrass El ements, and of course, his archival research
resulted in his superb edition of Walrass correspondence and related
papers.
Jaff e went on to explain that despite the inhibitions and gnawing self-
doubt, I never abandoned the idea of writing a properly documented Life
and Writings, but that, nevertheless, his efforts to begin the project
. . . were interrupted by the shock of retirement and nearly a decade was lost in
pursuit of a stable post sufciently remunerative to afford the peace and quiet for
the accomplishment of my task. . . . [Then] the upheaval of the move to York was
shattering. And here at York, in order to consolidate as well as I could a non-
tenured position, I felt it would be too much of a risk to retire to the biography
without turning out small publication after small publication for whatever they
might be worth. And so progress with the biography suffered. That, in brief, is
my side of the story. Perhaps I protest too much and perhaps you are right in your
surmise, that au fond, I really dont want to work on it. How can I say? Only
time will tell, if enough is allotted to me.
Unfortunately, that was not to be. He was already terminally ill when
he wrote those words, and had become incapable of sustained detailed
work. For the last fteen years of his life he had devoted himself to
writing articles, and then his intellectual mission was over. Neverthe-
less, he left a legacy of work on the Life and Writings project, namely
notes and preparatory writings dealing with Walrass antecedents and
early life, and bibliographical notes. I used the biographical material
and wove it in with long extracts from French sources to prepare an arti-
cle (Jaff e 1984), and I have incorporated his bibliographical notes into
the following bibliography, so that what he accomplished has not been
lost.
P1: PIG
III. Comparison with the rst edition
The compilation of the rst edition of the bibliography absorbed many
hours of my time for six years, so many that I sometimes felt that it had
taken on an evolving and demanding life of its own. In November 1986,
I wrote in the introduction to that edition that,
. . . it is so long and the history of many of the entries is so complex that the
probability of new pieces of information about them turning up is very high.
I could therefore probably spend several more years rening it. Moreover, as
time passes, additional items are published or republished, thus requiring that the
bibliography be extended. At last, however, I have called a halt to my efforts.
Believing that the bibliography is complete in all important and most minor
respects as of the present date, I have decided it is time to give the results as they
now stand to those who need them.
In fact, the rst edition has thus far proved to be almost complete
regarding the listings of Walrass writings. In the present edition, there
are only ve new entries regarding original writings, as distinct from
new entries regarding reprints and translations of publications that were
listed in the rst edition of the bibliography. The ve entries are: an
early article (40); a short note (81) that was implicitly included in the
rst editions entry 75; an early economic manuscript (95); some early
philosophical notes (251); and a more complete presentation of jottings
(252) than was given in the rst editions entry 226. Normally, it would
not be expected that a scholars scribblings would be included in his
bibliography, but I have judged that entries 251 and 252 are of sufcient
interest to merit it.
The present edition has nevertheless required a great deal of time and
effort, again, because of the complexity of the subject. I never halted my
efforts to improve the bibliography, carefully accumulating a plethora
of additional information. For example, the lessons that were assembled
to constitute entry 90 have been noted in a total of 6 entries (Le Travail
in 1867 and 1868), and the annotations now indicate whether a lesson in
that series or part of the lesson appeared in an earlier publication (see 34
and 51). Other annotations have been added and many of the annotations
have been claried or augmented. The reprints indicated in entries 221,
222, and 232 have been added. Bibliographical information has been
added to many of the entries. Typographical errors, such as errors of
pagination, have been corrected.
The bibliography has also been brought up to date by presenting
important reprints up to the present, and translations of exceptional
interest uptothe present. The publicationof Walrass complete economic
writings in nine volumes of the OEC renders superuous any additional
P1: PIG
206 Walrass ideas
reprints of those writings and any additional listings of reprints in this
bibliography. Each volume of the OEC has, of course, a table of con-
tents, and now that the index (249.XIV) has been published, researchers
will be able readily to nd Walrass economic writings in that denition
collection.
The form used for the bibliographical entries differs from that of the
entries in the general list of references. This is because the entries of
Walrass writings are in many cases quite complex, partially resulting
from their intrinsic character and partially because more information
is required about them than for general references. For Walrass hand-
written manuscripts, for example, the size of the pages in centimeters
is given. For books, the number of pages is given. In order to make a
long and complex entry clear, therefore, the information is broken up
into short parts by full colons and periods, unlike the general references
for which there are no periods in the listing. The short references in the
bibliography do not need the indicated measures, but it would be incon-
sistent to adopt one form for the long entries and another for the short
ones. Readers of the bibliography will have no difculty understanding
the meaning of the numerical parts of the entries.
P1: IWV
CHAPTER 8
The denitive bibliography of
the writings of L eon Walras
1858
1 Notes de voyage. Unpublished.
This title is a penciled addition by Walras to his Bibliographie
litt eraire (entry 260). The entry was assigned by him to 1858.
2 Francis Sauveur. Paris, E. Dentu, 1858. xxxv, 260.
This is a novel.
3 Pr eface sur LArt et la litt erature dans la d emocratie. Unpub-
lished.
This title is a pencilled addition by Walras to his Bibliographie
litt eraire (entry 260). The entry was assigned by him to 1858. On
the cover of the copy of Francis Sauveur (entry 2) in D. A. Walkers
possession, Walras wrote and initialed the words Avertissement et
corrections pour une 2
` eme
edition. Among the corrections, Walras
wrote a superscript (1) before the word PREFACE (page VII),
and at the top of the page he wrote Lart et la d emocratie.
(1)
It is
probable, therefore, that entry 3 is actually his renamed preface to
Francis Sauveur and has therefore been published.
1859
4 Philosophie de lart. 1` ere partie: Ontologie. Autograph manu-
script, Fonds Walras V 6, unpublished (1859), 39 leaves, 25 cm. by
19 cm.
Walras added the title in pencil to his Bibliographie litt eraire
(entry 260). The entry was assigned by him to 1859. Some of the
sentences in entry 4 appear word for word in entry 9.
5 Salon de 1859. Le Bonhomme, May 29, 1859.
The periodical was published in Bordeaux. The entry is in the
form of a letter dated May 24, 1859. Its length is 116 cm. on 5
columns of print.
6 De la propri et e intellectuelle. Position de la question economique.
Journal des Economistes, Revue de la Science Economique et de la
Statistique, 2nd series, 24, no. 12 (December 15, 1859): 392407.
207
P1: IWV
208 Walrass ideas
This is a reviewof De la propri et e intellectuelle, etudes par MM.
Fr ed eric Passy, Victor Modeste et P. Paillottet, avec une pr eface par
M. Jules Simon. Paris, E. Dentu, 1859.
The Journal was then in its 6th year, it having been founded in
1853.
7 La Lettre. Revue Francaise, 16, no. 4 (1859): 193206; no. 5
(1859): 27585.
This is a short story.
8 Du sens esth etique; de lart; et des opinions en mati` ere dart.
Autograph manuscript, Fonds Walras, unpublished (1859): 48 p.,
22.3 cm. by 16.8 cm.
For a summary, see entry 239, letter 9, n. 4, where the title is
given the variation les opinions, instead of des opinions, and
where a version of the manuscript is described as preserved in the
Lyons Collection, 35 foolscap pages in length, signed and dated
Paris, 1859, 3 rue de lAbbaye.
1860
9 Philosophie des sciences economiques. Journal des Economistes,
Revue de la Science Economique et de la Statistique, 2nd series, 25,
no. 2 (February 15, 1860): 196206.
This is an advance publication of the material appearing on pages
415 and 2938 of entry 15.
The Journal was in its 20th year, it having been founded in 1841.
10 Des octrois, ` a propos de la loi belge. La Presse, 25th year, July
12, 22, and August 3, 1860.
11 De la mise en valeur des biens communaux. La Presse, 25th year,
September 20, 21, and 23, 1860.
For a summary, see entry 239, letter 41, n. 2.
12 De la chert e des loyers ` a Paris. La Presse, 25th year, October 19,
26, 29, and November 6, 1860.
13 La Bourse et le d eveloppement du capital. La Presse, 25th year,
December 25 and 26, 1860.
14 Paradoxes economiques. I Que le sens commun nest point
le crit erium de la science en g en eral, ni en particulier celui de
l economie politique. Journal des Economistes, Revue de la
Science Economique et de la Statistique, 2nd series, 28, no. 12
(December 15, 1860): 37391.
For a brief summary, see entry 239, letter 34, n. 7.
P1: IWV
The denitive bibliography of the writings of L eon Walras 209
15 L economie politique et la justice. Examen critique et r efutation
des doctrines economiques de M. P.-J. Proudhon, pr ec ed es dune
Introduction ` a l etude de la question sociale. Paris, Librairie de
Guillaumin & C
ie
, 1860. LXIV, 255, plus an unnumbered Table
des Mati` eres.
This was reprinted, New York, Burt Franklin, 1970.
16 Application des math ematiques ` a l economie politique (1
` ere
Tenta-
tive, 1860). Autograph manuscript, Fonds Walras V, unpublished
(1860): 8 p., 26.5 cm. by 19.6 cm.; 3 leaves, on one of which there
is a graph.
1861
17 De l el evation du taux de lescompte. La Presse, 26th year,
January 23 and 26, 1861.
18 A M. F elix Solar, r edacteur en chef de la Presse. La Presse,
26th year, February 15, 1861.
19 Lavenir economique de lItalie (1861). Autograph manuscript,
Fonds Walras V 1, unpublished (1861): 5 p., 26.4 cm. by 19.5 cm.
A notation in Walrass hand reads: Utilis e pour la bibliographie
Errera.
20 Lexpos e de la situation de lEmpire. (1861). Corrected draft
manuscript, Fonds Walras V 6, unpublished (1861): 2 leaves, 26.3
cm. by 19.6 cm.
21 Lindustrie moderne et l economie politique. Autograph
manuscript, Fonds Walras VII X, unpublished (1861): 15 p., 26.1
cm. by 20.1 cm.
22 M emoire de M. Fould. (novembre 1861). Autograph manuscript,
Fonds Walras V 6, unpublished (November 1861): 5 p., 26.3 cm.
by 19.9 cm.
23 Paradoxes economiques. II. De la richesse naturelle ou gratuite.
De la richesse sociale ou economique. De la richesse intel-
lectuelle. Rough draft manuscript, partly in pencil and partly in
ink, Fonds Walras V 1, unpublished (1861): 5 leaves, 25.9 cm. by
19.7 cm.
This manuscript appears to be the second installment of the
Paradoxes economiques (entry 14). It is related to the rst install-
ment (entry 14) in style and subject matter. Two manuscripts in
folders labeled Paradoxes economiques are in the Fonds Walras
at Lausanne under the classication mark F. W. V 1. A pencilled
P1: IWV
210 Walrass ideas
note included with those folders lists the following as the third and
fourth items among seventeen in edits utilis es: 3. 2
e
paradox[e]
economique (1861) and 4. 3
e
id. It is probable that those descrip-
tions refer to the manuscripts, though neither of them is numbered.
See entry 24.
24 Paradoxes economiques. (1861) II. Que le point de vue econo-
mique est le point de vue social par excellence. Autograph manu-
script, Fonds Walras V 1, unpublished (1861): 20 p., 26.2 cm. by
19.6 cm.
Walras numbered this II although he gave the same number to
entry 23.
The text was used for Recherche de lid eal social (entry 90).
25 Des r eformes economiques. Autograph manuscript, Fonds
Walras V 1, unpublished (1861): 12 p., 26 cm. by 20 cm.
This appeared in substance as entry 46.
26 Le nouveau r egime commercial. Autograph manuscript, Fonds
Walras VII X, unpublished (1861): 5 p., 26.3 cm. by 19.6 cm.
27 Th eorie critique de limp ot, pr ec ed ee de Souvenirs du Congr` es de
Lausanne. Paris, Guillaumin et C
ie
, 1861. 8
o
, XXXVI, 119.
This is the basis of a paper delivered at the International Congress
on Taxation, July 1860. See entry 194, section IV of the contents,
where pages 95111 were republished. Walras put the substance of
part of entry 27 into the El ements d economie politique pure, and
reprinted large parts of it in entry 28.
28 De limp ot dans le canton de Vaud. M emoire auquel un quatri` eme
accessit a et e d ecern e ensuite du concours ouvert par le Conseil
dEtat du canton de Vaud sur les questions relatives ` a limp ot.
Lausanne, Imprimerie de Louis Vincent, 1861. 8
o
, 100.
This was submitted by Walras to the committee of the canton in
1860. The Vaud council paid for the publication of 500 copies. See
entry 194, section IV of the contents, for Walrass later use of the
text.
1862
29 Note sur la suppression des octrois en Belgique et la r eforme
de limp ot dans le canton de Vaud. Corrected draft manuscript,
Fonds Walras V 6, unpublished (1862): 8 leaves, 20.7 cm. by
13.5 cm.
P1: IWV
1863
30 Principes de la th eorie des richesses, par M. Cournot.
LInd ependant de la Moselle, July 13, 1863.
This review was republished in Augustin Cournot, Recherches
sur les principes math ematiques de la th eorie des richesses.
Nouvelle edition avec des compl ements de L eon Walras, Joseph
Bertrand et Vilfredo Pareto, publi ee avec une introduction et des
notes par Georges Lutfalla. Paris, Marcel Rivi` ere & C
ie
, 1938,
22532.
For a summary, see entry 239, letter 86, n. 4, part 1.
31 La crise cotonni` ere et les textiles indig` enes, par J. E. Horn.
LInd ependant de la Moselle, July 27, 1863.
32 De lesprit-communal et de la routine administrative, par M. de
Labry. LInd ependant de la Moselle, August 19, 1863.
33 De la constitution de la propri et e en Alg erie. LInd ependant de
la Moselle, September 2 and 21, and October 12, 1863.
34 Du mat erialisme et du spiritualisme en economie politique et
sociale. Autograph manuscript, Fonds Walras V 1, unpublished
(1863): 32 p., 28.3 cm. by 22.9 cm.
This was used for entries 74, 76, and 79, which are the rst three
lessons of entry 90.
1864
35 Latmosph` ere est un engrais complet, par M. le Dr. Schneider.
LInd ependant de la Moselle, January 13 and 16, 1864.
36 De lenseignement de l economie politique dans les Facult es de
Droit. Jellygraphed copy of autograph manuscript, Fonds Walras
V 1, unpublished, undated (probably October 1864): 13 p., num-
bered 6477, 28.5 cm. by 23.7 cm.
37 Recherche de lId eal Social en mati` ere de propri et e et dimp ot.
Autographmanuscript, Fonds Walras V1, unpublished(1864): 17p.
plus 10 p., 26.5 cm. by 19.5 cm.
Walras envisioned the imminent publication of this entry, writing
on the title page: Paris, Librairie de Guillaumin et C
ie
Editeurs,
P1: IWV
212 Walrass ideas
1863. He wrote 1864 at the top of the title page, but at least some
of the text was written in 1863.
38 Le S enatus-consulte du 2 d ecembre 1861 et les cr edits suppl e-
mentaires des exercices 1862 et 1863. Autograph manuscript,
Fonds Walras V 6, unpublished (1864): 18 p., 22.7 cm. by 17.4 cm.
1865
39 De lorganisation nanci` ere et de la constitution l egale des associ-
ations populaires. Journal des Economistes, Revue de la Science
Economique et de la Statistique, 2nd series, 45, no. 135 (March,
1865): 36181.
This was republished in entry 43.
For summaries, see entry 239, letter 96 and the enclosure to
letter 99.
40 Projet de loi sur les soci et es ` a responsabilit e proportionelle.
Moniteur Financier, March 5, 1865, 916.
An offprint of this was published in Paris, Imprimerie de Dubuis-
son et C
ie
.
41 Les soci et es coop eratives et la l egislation. Lettre au R edacteur de
la Presse. La Presse, 30th year, April 20, 1865.
The editors name was Rouy.
42 Le cr edit et les nances, par Victor Bonnet. Journal des
Economistes, Revue de la Science Economique et de la Statistique,
2nd series, 48, no. 144 (December 1865): 48083.
43 Les associations populaires de consommation, de production et de
cr edit. Lecons publiques faites ` a Paris en janvier et f evrier 1865.
Paris, E. Dentu, 1865. xxiii, 222.
This was reprinted under the editorship of Oscar Nuccio,
Rome, Edizioni Bizzari (R. Pioda), Ristampe Anastatiche de Opere
Antiche e Rare, 1969.
1866
44 Le mouvement dassociation et la politique lib erale. La Presse,
31st year, April 2, 1866.
45 LAssociation ` a la fran caise. La Presse, 31st year, April 20, 1866.
46 Programme economique et politique. Le Travail, 1, no. 1 (July
31, 1866), cols. 17. Unsigned.
This is, in substance, entry 25.
P1: IWV
The complete reference to the journal is: Le Travail. Organe
international des int er ets de la classe laborieuse. Revue du mou-
vement coop eratif. Publi e par MM. L eon Say et L eon Walras.
Bruxelles, V
e
Parent et Fils; Paris, Librairie des Auteurs et Com-
positeurs, 18671868, 2 vols. In the rst volume, labeled Premi` ere
Ann ee, are bound no. 1 (July 31, 1866) through no. 12 (June 30,
1867). In the second volume, labeled Deuxi` eme Ann ee, are bound
no. 1 (July 31, 1867) through no. 12 (June 30, 1868).
47 Soci et e coop erative immobili` ere. [1] Type dhabitation ` a bon
march e. [2] Combinaison nanci` ere. Le Travail, [1] 1, no. 2
(August 31, 1866), cols. 4446; [2] 1, no. 3 (September 30, 1866),
cols. 7678. Unsigned.
48 A propos dun article de M. Horn. Le Travail, 1, no. 4 (October
31, 1866), cols. 12122. Unsigned and without title.
The above title was given to the article by Walras in his autobib-
liography (entries 216 and 239).
49 De la chert e du pain et de l etablissement de boulangeries
coop eratives. Le Travail, 1, no. 4 (October 31, 1866), cols. 100
103. Unsigned.
50 Discussion sur les associations coop eratives ` a la Soci et e
d economie politique de Paris. Le Travail, 1, no. 5 (November
31 [sic], 1866), cols. 13238. Unsigned.
51 Socialisme et lib eralisme. Lettres ` a M. Ed. Scherer. Le Travail, 1,
no. 4 (October 31, 1866), cols. 11619; no. 6 (December 31, 1866),
cols. 18083; no. 8 (February 28, 1867), cols. 24447. Unsigned.
This was written in January 1863. It was republished in entries
90 and 93.
52 Des doctrines en mati` ere dassociation coop erative. Le Travail,
1, no. 6 (December 31, 1866), cols. 16466. Unsigned.
53 Les obligations populaires, lecons publiques faites ` aParis enf evrier
et mars 1866 par MM. L eon Say et L eon Walras, Administrateurs
de la Caisse dEscompte des Associations Populaires. Paris, Guil-
laumin et C
ie
, Editeurs, 14 rue Richelieu et E. Dentu, Editeur, 17 et
19, Galerie dOrl eans (Palais Royal), 1866. 8
o
, 59.
From the Table of Contents:
Des Op erations de la Caisse descompte des Associations populaires, par
M. L eon Walras . . . . . . 35
Les obligations populaires . . . . . . 57
The latter brief note was presumably authored by both Walras and
Say.
P1: IWV
214 Walrass ideas
The Caisse dEscompte failed. The accountant A. Rousseau gave
what he believed to be the reasons for that failure in a lengthy report
to the Governor of the Bank of France and a liquidator. Walras
defended the administrators of the Caisse in a Contre-m emoire
completed October 1870, for a summary of which see entry 239,
letter 143, n. 2.
54 Des op erations de lacaisse descompte des associations populaires.
Lecon publique faite ` a Paris en mars 1866. Paris, Guillaumin et
Dentu, 1866, 3348.
This is an offprint of pages 3556 of entry 53, with original
pagination.
1867
55 Enonc e de principes relatifs aux associations populaires
coop eratives pour servir dexpos e des motifs dun Projet de loi sur
les Soci et es ` a responsabilit e proportionnelle. Le Travail, 1, no. 7
(January 31, 1867), cols. 200194. Unsigned.
56 Le cr edit gratuit r eciproque. Le Travail, 1, no. 7 (January 31,
1867), cols. 21617. Unsigned and without title.
The above title was given to the article by Walras in his autobib-
liography (entries 214 and 238).
57 La libert e des soci et es. Le Travail, 1, no. 8 (February 28, 1867),
cols. 22728.
58 Le futur parti. Le Travail, 1, no. 8 (February 28, 1867), cols. 229
31.
59 De la gratuit e par la r eciprocit e dans les banques d echanges. Le
Travail, 1, no. 9 (March 31, 1867), cols. 25961. Unsigned.
60 Entrave au libre exercice du travail. Quatre pr evenus, impri-
meurs sur etoffes. Le Travail, 1, no. 9 (March 31, 1867), cols.
26465.
This appears under the title TRIBUNAUX Tribunal de 1
re
Instance de la Seine (6
e
Chambre).
The title given to this entry by Walras in his autobibliography
(entries 214 and 238) is Les soci et es de r esistance.
61 Les Id ees de Mme Aubray, com edie en 4 actes, par M. Alexandre
Dumas ls. Le Travail, 1, no. 10 (April 30, 1867), cols. 31114.
Unsigned.
The book that was the subject of this review was published in
Paris, Michel L evy, 1867. 8
o
, 119.
P1: IWV
62 Discussion sur les coalitions et les gr` eves ` a la Soci et e d economie
politique de Paris. Le Travail, 1, no. 10 (April 30, 1867), cols. 291
97.
63 Le projet de loi sur les soci et es ` a capital variable. Le Travail, 1,
no. 11 (May 31, 1867), cols. 32425. Unsigned.
64 Les erreurs du syst` eme mon etaire fran cais. Le Travail, 1, no. 11
(May 31, 1867), cols. 32529. Unsigned.
65 La Bourse. Le Travail, 1, no. 11 (May 31, 1867), cols. 33841
This appeared as part I of entry 70, pp. 173135.
66 Les Soci et es Coop eratives. Le Travail, 1, no. 11 (May 31, 1867),
cols. 34145.
This appeared as part III of entry 70, pp. 174851.
67 Sur les soci et es ` a capital variable. Le Travail, 1, no. 12 (June 30,
1867), cols. 35563.
This appeared in a section titled La discussion du corps
l egislatif.
68 Syndicat du cr edit ` a Paris. Le Travail, 1, no. 12 (June 30, 1867),
cols. 36367.
69 Conf erence faite par M. Horn sur le Cr edit, la Mutualit e Syndicale,
LAssurance et la Contre-Assurance. Le Travail, 2, no. 1 (July 31,
1867), cols. 2024.
70 La Bourse et le cr edit. InParis Guide, par les principaux ecrivains
et artistes de la France. Paris, Librairie internationale; Bruxelles,
Leipzig, and Livourne, A. Lacroix, Verboeckhoven et C
ie
, 1867;
vol. 2, pp. 173151.
The Paris Guide was published in two volumes, called par-
ties, but the covers, title pages, and tables of content were put
on the wrong set of contributions for some or perhaps all of the
copies. Thus, for at least some copies, Walrass article appears
in the volume incorrectly stated to be PREMI
`
ERE PARTIE.
LA SCIENCELART, instead of DEUXI
`
EME PARTIE. LA
VIE.
71 Congr` es international coop eratif. Le Travail, 2, no. 2 (August 31,
1867), cols. 3536.
72 Soci et e d economie politique de Paris. Le Travail, 2, no. 3
(September 30, 1867), cols. 7085.
73 Les syndicats de garantie mutuelle. Le Travail, 2, no. 3 (Septem-
ber 30, 1867), cols. 8587.
P1: IWV
216 Walrass ideas
74 Recherche de lid eal social. Le cons publiques faites ` a Paris en
1867 et 1868. Premi` ere Le con. Concurrence du principe de
lInt er et et du principe de la Justice dans les questions economico-
sociales. Le Travail, 2, no. 4 (October 31, 1867), cols. 11122.
The source of the substance of this entry is entry 34.
75 De l education des lles. Le Travail, 2, no. 5 (November 30,
1867), cols. 13335.
1867 et 1868. 2
e
Le con. Intervention des doctrines philosophi-
ques. Lutte actuelle du mat erialisme et du spiritualisme sur le terrain
de l economie politique et de la science sociale. Le Travail, 2, no. 5
(November 30, 1867), cols. 14151.
77 La science et le socialisme. Le Travail, 2, no. 6 (December 31,
1867), cols. 16465.
This was reprinted in 1918 with the title Le Socialisme Scien-
tique (entry 222).
78 La s ecurit e g en erale. Le Travail, 2, no. 6 (December 31, 1867),
cols. 16668.
79 Recherche de lid eal social. Le cons publiques faites ` a Paris en1867
et 1868. 3
e
Le con. Critique du mat erialisme. Critique du spiritu-
alisme. Nouveau point de vue de la morale sociale. Le Travail, 2,
no. 6 (December 31, 1867), cols. 17787.
1868
80 Le socialisme scientique. Le Travail, 2, no. 7 (January 31, 1868),
cols. 198205.
81 Observations. Le Travail, 2, no. 7 (January 31, 1868), cols. 2034.
This is a reply to Jean Mac e, Charles K uss, Boulogne, and L eon
Say, whose remarks were published in the same issue of Le Travail.
For excerpts from entry 81, see entry 239, letter 124, n. 2.
1867 et 1868. 4
e
Le con. De lhomme et de la destin ee humaine
au double point de vue physiologico- economique et psychologico-
moral. Le Travail, 2, no. 7 (January 31, 1868), cols. 20616.
1867 et 1868. 5
e
Le con. De la concordance de lint er et et de la
justice. Le Travail, 2, no. 8 (February 29, 1868), cols. 24050.
84 Le mouvement dinstruction populaire. Le Travail, 2, no. 9
(March 31, 1868), cols. 25960.
P1: IWV
85 Recherche de lid eal social. Le cons publiques faites ` a Paris en1867
et 1868. 6
e
Le con. De lindividu et de l
Etat. Formule g en erale de

constitution de la science sociale. Le Travail, 2, no. 9 (March 31,
1868), cols. 27384.
86 De la sp eculation, etude d economie nanci` ere. Le Travail, 2,
no. 10 (April 30, 1868), cols. 29295; 2, no. 12 (June 30, 1868),
87 Les r eunions publiques. Le Travail, 2, no. 12 (June 30, 1868),
This was published without a title in the section titled Nou-
velles. The above title was given to the article by Walras in his
autobibliography (entries 214 and 238).
88 Histoire de Napol eon I
er
, par P. Lanfrey, tomes I et II. Paris. Charp-
entier . . . 1867. Le Travail, 2, no. 12 (June 30, 1868), cols. 37880.
Unsigned.
89 M ethode de conciliation ou de synth` ese. Manuscript.
This was rst published in 1896 (entries 188 and 194). Walras
used entry 89 as the basis of entry 99.
90 Recherche de lid eal social. Lecons publiques faites ` a Paris.
Premi` ere s erie (186768) Th eorie g en erale de la soci et e. Paris,
Guillaumin &C
ie
et Ag
ce
G en
le
des Auteurs et Compositeurs, 1868.
XXXII, 194.
This included entries 74, 76, 79, 82, 83, and 85, entries that, in
turn, used entries 24 and 34. It was republished in entry 194.
1869
90 Application des math ematiques ` a l economie politique (2
` eme
Tentative 186970). Autograph manuscript, Fonds Walras V 5,
unpublished (186970): 47 p., 27 cm. by 21.6 cm., plus 2 smaller
unnumbered sheets, plus 4 sheets of formulas and graphs.
For a summary, see entry 239, letter 148, n. 33, where the three
major parts of the manuscript are described.
1870
92 Int er ets des comptes courants. Manuscript, jellygraphed copy,
Fonds Walras V 6, unpublished (1870): 5 leaves, 26.9 cm. by
21 cm.
P1: IWV
218 Walrass ideas
1871
93 Correspondance sur lentr ee ` a Lausanne des troupes de larm ee de
lEst. Journal des D ebats, 8 (February 1871).
This title and the publication information were added in pencil by
Walras to his Bibliographie litt eraire (entry 260), and the title is
the one he gave to the manuscript of the text in the Lyons Collection.
See entry 239, letter 177, n. 4.
94 Economie domestique. Conf erences faites ` a Lausanne (aux
internes fran cais, f evrier-mars 1871). Manuscript, jellygraphed
copy, Fonds Walras V 6, unpublished (1871): 36 p., 23.2 cm. by
15.4 cm.
The Lausanne copy is faded and completely illegible.
95 Application des math ematiques ` a l economie politique 3
e
tenta-
tive 1871 la bonne (v. p. 15). Autograph manuscript, 37 p.
96 Des billets de banque en Suisse. Biblioth` eque Universelle et
Revue Suisse, 76th year, new period, 41, no. 163 (July 1871): 321
42.
1872
97 Discours dinstallation en qualit e de Professeur ordinaire
d economie politique ` a lAcad emie de Lausanne prononc e dans
la s eance acad emique du 20 octobre 1871. S eances acad emiques
du 23 octobre 1869, du 31 octobre 1870 et du 20 octobre 1871, Dis-
cours dinstallation. Lausanne, Imprimerie Sim eon Genton, 1872:
6993.
This was also published in a booklet, S eance acad emique du 20
octobre 1871, Discours dinstallation, Acad emie de Lausanne, no
date: 1842.
98 Syst` eme des ph enom` enes economiques. 10 Le cons publiques
faites ` a lHotel de Ville de Gen` eve (1872). Autograph manuscript,
Fonds Walras V 1, unpublished (1872): 24 leaves, 28.9 cm. by 21.7
cm.
This was described by Walras as 10 Le cons publiques faites ` a
Gen` eve 187172, and placed by him under 1871 in his autobibli-
ography (entries 214 and 238).
99 Exposition et conciliation des doctrines sociales. Corrected draft
manuscript, Fonds Walras V 1, unpublished (1872): 90 p., 28.8 cm.
by 21.8 cm.
P1: IWV
This was described by Walras as 6 Le cons publiques faites ` a
Gen` eve 187273 (in edit). Walras stated (entry 194, p. 175) that he
wrote the text (entry 89) on which this entry is based in 1868. Much
of the substance of entry 99 appears in entry 194.
100 Cours el ementaire ` a lEcole Industrielle. Autograph manuscript,
Fonds Walras V 6, unpublished course (18721873): 21 leaves,
22.4 cm. by 17.7 cm.
1873
101 Sur la th eorie math ematique de l echange. Bulletin de la Soci et e
Vaudoise des Sciences Naturelles, 12, no. 70 (September 1, 1873):
3172l.
This is an open letter to Paul Piccard, dated October 25, 1873, and
read to the Soci et e vaudoise des sciences naturelles on November 5,
1873.
102 Le cadastre et limp ot foncier. Biblioth` eque Universelle et Revue
Suisse, 78th year, new period, 48 (November 1873): 47090;
(December 1873): 61029.
This was republished, with very minor changes and with section
titles added, in entry 194, 387421, and in entry 228, 41044. A
long footnote on pages 477478 of the original article was omitted
fromthose entries. See entry 239, letter 236, n. 4, for the text of that
footnote.
1874
103 Principe dune th eorie math ematique de l echange. S eances et
Travaux de lAcad emie des Sciences Morales et Politiques (Institut
de France), Extrait du Compte-rendu r edig e par M. Ch. Verg e.
Collection, new series, 33rd year, 101, tome I, Paris, Alphonse
Picard (January 1874): 97116, 1 plate.
Walras read this paper to the Acad emie des sciences morales et
politiques in Paris on August 16 and 23, 1873. The paper is listed
in the table of contents on page 848 under Communications des
savants etrangers. It is followed by a summary of observations on
it by Pierre-Emile Levasseur, Valette, and Louis Wolowski (117
20). The paper was republished in the Journal des Economistes,
Revue de la Science Economique et de la Statistique, 3rd series,
34, no. 100 (April 1874): 521, 1 plate, without the observations;
republished as an offprint with newpagination, Orl eans, Imprimerie
P1: IWV
220 Walrass ideas
Ernest Colas, 1874, pp. 24, 1 plate; bound in entry 124; republished
in entry 160; and republished in entry 244. It was translated into
Italian by Gerolamo Boccardo as Principio duna teoria matem-
atica dello scambio in entry 137, and translated into German by
Ludwig von Winterfeld as Prinzip einer mathematischen Theorie
des Tausches in entry 157.
104 De linuence de la communication des march es sur la situation
des populations agricoles. M emoire lu ` a la Soci et e Vaudoise dutilit e
publique dans la s eance du 29 avril 1874. Journal de la Soci et e
Vaudoise dUtilit e Publique, no. 5 (May 1874): 10316; no. 6 (June
1874): 12135.
This was republished in entry 205, pp. 23964.
105 Correspondance entre M. Jevons, Professeur ` a Manchester et M.
Walras, Professeur ` a Lausanne. Journal des Economistes, Revue
de la Science Economique et de la Statistique, 3rd series, 34, no. 102
(June 15, 1874): 41722.
This consists of a letter from W. S. Jevons to Walras, May 12,
1874, in French translation, and a letter from Walras to Jevons,
May 23, 1874, in French. The entry was republished as an offprint,
Paris, Guillaumin. 8
o
, 8; bound in entry 124; published in Italian
translation in entry 137; published in German translation in Math-
ematische Theorie der Preisbestimmung (entry 157) with Jevonss
letter in bad English; and republished in entry 160 with Jevonss
letter in French translation.
106 El ements d economie politique pure ou Th eorie de la richesse
sociale. 1st edition, 1st part. Lausanne, Imprimerie L. Corbaz &
C
ie
; Paris, Guillaumin & C
ie
; B ale, H. Georg, 1874. 8
o
, VIII, 208,
2 plates.
This includes lessons 134; 1207. The title onthe paper cover
reads:

EL
EMENTS D
ECONOMIE POLITIQUE PURE, and bears

as subtitles in roman font the titles of sections I, II, and III: Objet et
divisions de l economie politique et sociale. Th eorie math ematique
de l echange. Du num eraire et de la monnaie.
The words OUTH
EORIEDELARICHESSESOCIALE appear
on the title page, but not on the cover.
107 El ements d economie politique pure, par L eon Walras. Revue de
Th eologie et de Philosophie et Compte Rendu des Principales Pub-
lications Scientiques, Lausanne, Georges Bridel Editeur (October
1874): 62832.
This is an unsigned account of the El ements that appeared under
the title Philosophie. It refers in a footnote on page 628 to entry
P1: IWV
106. It was listed by Walras as number 61 under 1874 in his auto-
bibliography (entry 214), and was erroneously listed under 1873 in
Bousquets presentation of the autobibliography (entry 238).
108 LItalia industriale, stud. j. del Prof. Alberto Errera, con parti-
colare riguardo allAdriatico superiore (regno dItalia ed impe-
rio Austro-Ungarico). Roma-Torino-Firenze. Ermanno Loescher
libraio-editore, 1873. Le nuove istituzioni economiche nel secolo
XIX, di Alberto Errera, professore titolare di economia politica.
Milano. Fratelli Tr` eves editori, 1874. Journal des Economistes,
Revue de la Science Economique et de la Statistique, 3rd series, 36,
no. 107 (November 1874): 32934.
The meaning of stud. j. is juridical studies.
For a summary of this review, see entry 239, letter 318, n. 2.
1875
109 Economie politique. La Revue, Journal Politique, Agricole,
Industrial et Commercial, Paraissant le Mercredi et le Samedi, 7,
no. 2 (January 9, 1875). Unsigned.
The journal was published in Lausanne.
110 R esum e des principaux faits statistiques du Danemark. Publi e
par le Bureau royal de statistique. Copenhague, imprimerie de
Bianco Luno, 1874. Journal des Economistes, Revue de la Sci-
ence Economique et de la Statistique, 3rd series, 37, no. 111 (March
1875): 492. This review is signed J. C.
111 La loi f ed erale sur l emission et le remboursement des billets de
banque. Gazette des Tribunaux Suisses, Journal Hebdomadaire
de Jurisprudence et de L egislation, 11, no. 32 (August 12, 1875):
24143.
112 Biblioteca dellEconomista. Raccolta delle pi ` u pregiate opere
moderne italiane e straniere di economia politica, diretta dal profes-
sore Gerolamo Boccardo. Terzia serie. Torino, Unione tipograco-
editrice, 1875. Journal des Economistes, Revue de la Science
Economique et de la Statistique, 3rd series, 40, no. 118 (October
1875): 15759.
113 Rapport sur le calcul des r eserves de la C
ie
La Suisse au 31
d ecembre 1874. Autograph manuscript, Fonds Walras VII, unpub-
lished (April 14, 1875): 62 p., 25.3 cm. by 20 cm.
For a brief description of the contents, see entry 239, letter 393,
n. 3.
P1: IWV
222 Walrass ideas
114 Une branche nouvelle de la math ematique. De lapplication
des math ematiques ` a l economie politique. Corrected draft
manuscript, Fonds Walras V 1, unpublished in French (1875): 37
p., 29.2 cm. by 21.7 cm.
This was published in Italian translation (entry 117).
115 LEtat et les chemins de fer. Manuscript, 1875.
This was rst published in 1897 (entry 196), and appears in
English translation in entry 245.
1876
116 La loi f ed erale sur le travail dans les fabriques. Gazette des
Tribunaux Suisses, Journal Hebdomadaire de Jurisprudence et de
L egislation, 11, no. 5, February 10, 1876.
117 Un nuovo ramo della matematica. Dellapplicazione delle matem-
atiche alleconomia politica. Giornale degli Economisti, 3, no. 1
(April 1876): 140.
This is a translation by Gerolamo Boccardo of entry 114. It was
republished as an offprint, Padua, Premiata tipograa alla Minerva,
1876. 8
o
, 40.
118 Equations de l echange. Bulletin de la Soci et e Vaudoise des Sci-
ences Naturelles, 14, no. 76 (October 1876): 36794.
Walras read this paper to the Soci et e vaudoise des sciences
naturelles on December 1 and December 15, 1875. It is preceded
by the general title, Th eorie math ematique de la richesse sociale,
and a prefatory note (36566) signed L. W., Chateau de Gl erolles,
par St-Saphorin, Vaud (Suisse). 8 ao ut 1876. The note discusses
the four m emoires: Principe dune th eorie math ematique de
l echange (entry 103), Equations de l echange (entry 118),
Equations de la production (entry 119), and Equations de la
capitalisation (entry 122). Walras explained that entries 118 and
119 were to be published in the present number of the Bulletin,
that is, in BULL. SOC. VAUD. SC. NAT. XIV. 76., as printed
at the top of the page on which the prefatory note begins. The off-
print was bound together with an offprint of the Equations de la
production, Lausanne, Imprimerie Ed. Allenspach Fils, 1876, 3
30. The title page of the pamphlet offprint erroneously ascribes the
paper to volume XV, for it reads: Extrait du Bulletin de la Soci et e
vaudoise des sciences naturelles. 2
e
S. Vol. XV. N
o
76, the 2
e
S
referring to the second series of the journal. Entry 118 was bound
in entry 124; and republished in entry 160, 3353. In the latter, the
P1: IWV
prefatory note became the Preface des quatre premiers m emoires,
56. Entry 118 was translated into Italian by Gerolamo Boccardo as
Equazioni dello scambio in entry 137, with the prefatory note as
a Prefazione, and translated into German by Ludwig von Winter-
feld as Gleichungen des Tausches in entry 157, with the prefatory
note as a Vorwort.
The substance of this entry appears in entry 123 and subsequent
editions.
119 Equations de la production. Bulletin de la Soci et e Vaudoise des
Sciences Naturelles, 14, no. 76 (October 1876): 395430.
naturelles on January 19 and February 16, 1876. Entry 119 was
republished as an offprint and bound together with an offprint of
the Equations de l echange (entry118), Lausanne, Imprimerie Ed.
Allenspach Fils, 1876, pp. 3166. Although the offprint indicates
that it was published in BULL. SOC. VAUD. SC. NAT. XIV. 76.
the title page of the pamphlet offprint reads as in the case of entry
118, because the two entries shared the same title page. The remark
regarding the erroneous volume printed on the title page of the
offprint of entry 118 therefore applies also to entry 119. Entry 119
was bound in entry 124; and republished in entry 160, 5582. It was
translated into Italian by Gerolamo Boccardo as Equazioni della
produzione in entry 137, and translated into German by Ludwig
von Winterfeld as Gleichungen der Produktion in entry 157.
The substance of this entry, including most of the equations,
appears in entry 123 and subsequent editions of the El ements
d economie politique pure.
120 Note sur limp ot progressif. Rough draft manuscript, Fonds
Walras V1, unpublished (November 1876): 4 p., 25.1 cm. by 19 cm.
Walras indicated on the title page that it was Utilis ee pour le
Probl` eme scal (Et. Ec. soc.) (see entry 194, 43141; entry 228,
45464). The rst two pages of this paper are virtually the same as
the beginning of the second part of Le probl` eme scal (entry 190),
entitled Critique de limp ot comme fait normal et d enitif, page
454 to the end of the top paragraph on page 458: . . . , la science
pour lempirisme. Some of the other sentences in entries 120 and
190 are identical.
121 Note sur le 15 1/2 l egal. Journal des Economistes, Revue de la
Science Economique et de la Statistique, 3rd series, 44, no. 132
(December, 1876): 45457.
This was followed by R eponse de M. Cernuschi, 45758.
Entry 121 was republished as part of a pamphlet entitled Th eorie
P1: IWV
224 Walrass ideas
math ematique du bim etallisme, Paris, Guillaumin, 1881, 36 (see
entry 153); and in entry 160, 11923.
1877
122 Equations de la capitalisation. Bulletin de la Soci et e Vaudoise des
Sciences Naturelles, 14, no. 77 (March 1877): 52561.
This was preceded by the title: Th eorie math ematique de la
richesse sociale (525). It was followed by an Observation (562
64) that dealt with the four m emoires listed under entry 124. Walras
read entry 122 to the Soci et e vaudoise des sciences naturelles. In
the Bulletin the date of delivery was given as July 6, 1876, but
Walras represented it as July 5 in three places: on the title page
of the offprint mentioned below, in the rst footnote (page 82) of
the version published in entry 160, and in his autobibliography
(entries 214 and 238). Walrass explanatory note described under
entry 118 continues by indicating that entry 122 was to be published
in the following number of the Bulletin, namely in number 77,
the one that was to follow the number in which entries 118 and 119
appeared. Nevertheless, the title page composed for the offprint
reads Extrait du Bulletin de la Soc. Vaud. des Sciences Naturelles.
2
e
S. Vol. XV. N
o
77. That is erroneous because entry 122 appeared
in volume XIV, which comprises numbers 75, 76, and 77 published
in February 1876, October 1876, and March 1877, respectively.
Walras correctly listed entry 122 in his autobibliography (entries
214 and 238) as appearing in volume XIV. The top of the rst page
(525) of the offprint of entry 122 states that the entry is part of
BULL. SOC. VAUD. SC. NAT. XIV. 76, the same volume and
number as in the case of entries 118 and 119, but the 76 is an
error. Moreover, the offprint of entry 122 produced by L. Corbaz is
dated 1876, regarding which Mr. Christian Graf (see Chapter 7) has
informed D. A. Walker that it is quite possible that the tir e ` a part
[separate printing or offprint] was published before the appearance
of the article in the Bulletin. A further complication is that in
Bousquets presentation of Walrass autobibliography (entry 238),
entry 122 appears as item71 under 1876, whereas in D. A. Walkers
transcript of Walrass autobibliography, the paper appears as item
71 under 1877. As Bousquets presentation also wrongly puts the
second part of the rst edition of the El ements d economie politique
pure (entry 123) under 1876, and lists no items as appearing in 1877,
it is clear that Bousquet or the printer neglected to insert the date
1877 before the items that Walras listed under that year.
P1: IWV
In his autobibliography, Walras listed entry 122 with the title
Equations de la capitalisation et du cr edit, but the last three
words were not added to the title until 1883 (entry 160). Entry
122 was, as indicated, republished as an offprint, Lausanne, L.
Corbaz et Comp., 1876, 67103, followed by the Observation,
1046; and bound in entry 124. It was republished in entry 160,
with a number of changes, as Equations de la capitalisation et
du cr edit, 83112, followed by the Observation, 11315. The
revised version, as presented in entry 160, was translated into Italian
by Gerolamo Boccardo as Equazioni della capitalizzazione e del
credito in entry 137, followed by the Osservazione, and trans-
lated into German by Ludwig von Winterfeld as Gleichungen der
Kapitalisirung und des Kredites in entry 157, followed by the
Schloss that is, Observation). The Observation appears with
some changes in the second part of the rst edition of El ements
d economie politique pure (entry 123), 36466, and in subsequent
editions.
The substance of entry 122 appears in entry 123 and subsequent
editions.
123 El ements d economie politique pure. 1st edition, 2nd part.
Lausanne, Imprimerie L. Corbaz & C
ie
ie
;
B ale, H. Georg, 1877. 8
o
, 209407, 1 plate.
This includes lessons 3564, 208385. The paper cover bears
as subtitles in roman font the titles of sections IV, V, and VI: Th eorie
naturelle de la production et de la consommation de la richesse.
Conditions et cons equences du progr` es economique. Effets naturels
et n ecessaires des divers modes dorganisation economique de la
soci et e.
The words ou Th eorie de la richesse sociale (see entry 106)
do not appear on the cover.
There is no title page, for the reason that entry 123 is the contin-
uation of entry 106.
124 Th eorie math ematique de la richesse sociale. Quatre m emoires
lus ` a lAcad emie des sciences morales et politiques, ` a Paris, et
` a la Soci et e Vaudoise des Sciences Naturelles, ` a Lausanne, Paris,
Guillaumin & C
ie
, 1877. 8
, 145, 1 plate.
In this entry are bound offprints of the following papers in
their original published versions: (1) Principe dune th eorie
math ematique de l echange (entry 103); (2) Correspondance
entre M. Jevons . . . et M. Walras . . . (entry 105); (3) Equations
de l echange, preceded by a prefatory note (entry 118), 36566;
(4) Equations de la production (entry 119); (5) Equations de
P1: IWV
226 Walrass ideas
la capitalisation (entry 122), followed by the Observation (see
entry 122).
125 Analyse de la Th eorie math ematique de la richesse sociale.
Manuscript, Fonds Walras V b 21, unpublished (March 1877).
1878
126 Exposition de tableaux et d etudes de M. Bocion. Gazette de
Lausanne, no. 1 (January 2, 1878). Unsigned.
This is a commentary on the work of Fran cois Bocion.
The full name of the newspaper was Gazette de Lausanne et
Journal Suisse. Its issues were numbered sequentially within each
calendar year. 1878 was the 79th year of that journal, it having been
founded in 1799.
127 Chronique de la quinzaine, under the title Feuilleton. Gazette
de Lausanne, no. 10 (January 12, 1878). Signed Paul.
Walras listed this entry in his Bibliographie litt eraire
(entry 260) as Gleyre et Courbet. R ealisme. Walras probably
wrote the entry on January 10, the day in the calendar of saints
of the little-known hermit Paul, whose name Walras assumed.
de Lausanne, no. 22, (January 26, 1878). Signed Paul.
(entry 260) as Victor-Emmnanuel. Le g en eral Cousin-Montauban.
F.-V. Raspail. Le th e atre de campagne.
de Lausanne, no. 34 (February 9, 1878). Signed Paul.
(entry 260) as La Correspondance de Sainte-Beuve.
130 Chronique de la quinzaine under the title Feuilleton. Gazette
de Lausanne, no. 46 (February 23, 1878). Signed Paul.
(entry 260) as Discours de M. Du Bois-Raymond sur lHistoire
de la civilisation et la Science de la nature. Lam ericanisme.
de Lausanne, no. 58 (March 9, 1878). Signed Paul.
(entry 260) as Rodogune. La trag edie. Les Femmes Savantes.
de Lausanne, no. 70 (March 23, 1878). Signed Paul.
P1: IWV
(entry 260) as M. Victor Hugo: Histoire dun Crime; La L egende
des Si` ecles, nouvelle S erie, T. II.
133 Chronique de la quinzaine under the title Feuilleton. Gazette
de Lausanne, no. 82 (April 6, 1878). Signed Paul.
(entry 260) as Le tour du monde en 80 jours. Le Th eatre de Mari-
onnettes de M. Marc-Monnier.
de Lausanne, no. 93 (April 20, 1878). Signed Paul.
(entry 260) as Le Dictionnaire de lAcad emie fran caise.
de Lausanne, no. 105 (May 4, 1878). Signed Paul.
(entry 260) as Charlotte Corday et Vera Zassoulitch.
de Lausanne, no. 117 (May 18, 1878). Signed Paul.
(entry 260) as Exposition des Beaux-Arts ` a Lausanne. Exposition
des Amis des Arts ` a Neuch atel.
137 Teoria Matematica della Ricchezza Sociale. Quattro Memorie lette
allAccademia delle scienze morali e politiche a Parige ed alla
Societ ` a delle scienze naturali a Losanna, da Leone Walras, Pro-
fessore di economia politica nellacademia di Losanna. Translated
by Gerolamo Boccardo, Bibliotheca delleconomista, 3rd series, 2
(1878).
These are translations into Italian (1) of the prefatory note
described in entry 118: Prefazione, 129192; (2) of entry 103:
Principio duna teoria matematica della scambio, 12931310,
1 plate; (3) of entry 105: Corrispondenza tra il Signor Jevons
professore a Manchester ed il Sig. Walras Prof. a Losanna,
131113; (4) of entry 118: Equazioni della scambio, 131432; (5)
of entry 119: Equazioni della produzione, 133356; (6) of entry
122: Equazioni della capitalizzazione e del credito, 135783; and
(7) of the Osservazione in entry 122, 138485.
138 Bibliographie des ouvrages relatifs ` a lapplication des math e-
matiques ` a l economie politique, encollaborationavec le Professeur
Jevons. Journal des Economistes, Revue de la Science Economique
et de la Statistique, 4th series, 4, no. 12 (December 1878):
47077.
P1: IWV
228 Walrass ideas
139 De lassurance sur la vie. La Suisse, Soci et e dAssurances sur
la Vie ` a Lausanne, Almanach pour 1879, Lausanne, Imprimerie L.
Corbaz & Comp., 1878. 8
o
, 512.
The title on the front cover of this booklet is Almanach de La
Suisse 1879.
140 Rapport sur la r eserve correspondant ` a une fraction dann ee dans le
cas de lassurance au d ec` es. Autograph manuscript, Fonds Walras
V II, unpublished (1878): 6 leaves, 25.8 cm. by 20.5 cm.
The title was recorded by Walras in his autobibliography (entry
214) with the following difference: l` a fraction, instead of ` a une
fraction, and the entry was marked: Lausanne, 2 d ecembre 1878.
The manuscript was also described by Walras as having 8 pages. It
is a sequel to entry 113.
1879
141 De la culture et de lenseignement des sciences morales et poli-
tiques. Biblioth` eque Universelle et Revue Suisse, 84th year, 3rd
period, 3, no. 7 (July 1879): 532; no. 8 (August 1879): 22351.
142 De l emission des billets de banque. Gazette de Lausanne, no.
285 (December 2, 1879) and no. 286 (December 3, 1879).
This is a summary of Th eorie math ematique du billet de
banque (entry 148), which Walras read to the Soci et e vaudoise
des sciences naturelles on November 19, 1879.
143 Am elie Lasaulx en religion Sur Augustine. Traduction autoris ee
des souvenirs dAm elie de Lasaulx. Lausanne. Imer, editeurs.
1880. Gazette de Lausanne, December 29, 1879. Unsigned.
This is a review of a biography, 369 pages in length, that was
written anonymously and translated anonymously from German
into French. In fact, the translator was Walrass friend, Charles
Secr etan. Walras evidently had an advance copy of the book.
144 Th eorie de lEconomie politique, par W. Stanley Jevons, L.L.D.
(Ed.) M.A. (L.), F.R.S., Professeur d economie politique au
Coll` ege de lUniversit e, ` a Londres, Examinateur de Sciences
philosophiques et morales ` alUniversit e ` aLondres. 2
e
Editionrevue
et augment ee avec une pr eface nouvelle et des appendices. Auto-
graphmanuscript, Fonds Walras 1966, unpublished(Autumn1879):
79 leaves, 25.1 cm. by 19 cm., and one large sheet of unnished
copied graphs, 43 cm. by 29.5 cm.
P1: IWV
This is a translation of W. Stanley Jevonss Theory of Political
Economy, 2nd ed., London, Macmillan, 1879.
See entry 239, letter 465, second paragraph, for Walrass plan,
never carried out, to publish his translation together with entries
124 and 250 and an introduction.
145 Note sur lorganisation de lenseignement de l economie poli-
tique ` a lEcole pratique des Hautes Etudes. Autograph manuscript,
Fonds Walras V 1, unpublished (1879): 2 leaves, 26.4 cm. by 20.9
cm.
Walras added to the title: remise ` a M. Albert Dumont.
A copy of the manuscript given to William Jaff e by Etienne
Antonelli is reproduced in entry 239, letter 455, n. 6. That copy is
not complete. The transcript in the possession of D. A. Walker of the
manuscript in the Fonds Walras contains in addition three introduc-
tory paragraphs on the organization of instruction in economique
et politique.
1880
146 Deux M eprises. Nouvelles, par Mme Bonzon de Gardonne,
Paris, Sandoz et Fischbacher; Neuchatel, Jules Sandoz; Gen` eve,
Desroges: 1880. Gazette de Lausanne, January 30, 1880.
Unsigned.
147 La Bourse, la sp eculation et lagiotage. Biblioth` eque Universelle
et Revue Suisse, 85th year, 3rd period, 5 (March 1880): 45276; 6
(April 1880): 66107.
This was republished in entry 205, 40145, with section titles
added.
148 Th eorie math ematique du billet de banque. Bulletin de la Soci et e
Vaudoise des Sciences Naturelles, 2ndseries, 16, no. 83(May1880):
55392, 1 plate.
naturelles on November 19, 1879. It was republished, with some
minor changes, in entry 160, 14575, 1 plate; and, with some addi-
tional changes, in entry 205, 33975, 1 plate.
149 De la propri et e intellectuelle. Gazette de Lausanne, no. 136 (June
10, 1880), no. 137 (June 11, 1880), and no. 138 (June 12, 1880).
150 Opuscules et pens ees, par Giacomo Leopardi. Traduit de
litalien et pr ec ed e dune pr eface par Auguste Dapples. Paris,
Germer-Bailli` ere, 1880. Gazette de Lausanne, September 24 and
25, 1880. Unsigned.
P1: IWV
230 Walrass ideas
The book (in 8
o
, XVI, 198) was published in the series Bib-
lioth` eque de philosophie contemporaine.
151 D efense des salaires. Unpublished, manuscript, 1880.
A note by Walras reads: Conf erences faites ` a Lausanne (in edit)
utilis ees pour lEconomie politique appliqu ee [entry 205] et la
d efense des Salaires en 1897 (entry 198).
152 De lEnseignement de l economie politique enFrance. Autograph
manuscript, Fonds Walras V 1, unpublished (1880): 16 p., 27.8 cm.
by 21.5 cm.
1881
153 Th eorie math ematique du bim etallisme. Journal des Econo-
mistes, Revue de la Science Economique et de la Statistique, 4th
series, 14, no. 41 (May 1881): 18999, 1 plate.
This was republished as part of a pamphlet entitled Th eorie
math ematique du bim etallisme, Paris, Guillaumin, 1881, 616,
1 plate. In the pamphlet, it followed a reprinting of Walrass Note
sur le 15 1/2 l egal (entry 121). Entry 153 was republished, with
minor changes, in entry 160, 12334.
154 Th eorie math ematique du prix des terres et de leur rachat par
lEtat. M emoire lu ` a la Soci et e vaudoise des sciences naturelles
` a Lausanne, dans la s eance du 17 novembre 1880. Bulletin de la
Soci et e Vaudoise des Sciences Naturelles, 2nd series, 17, no. 85
(June 1881): 189284.
This was republished in entry 160, 177253; in entry 194, 267
350; and in entry 244.
155 Lettera del prof. L eone Walras al presidente del Circolo Univer-
sitario prof. Alberto Errera. Letter dated January 26, 1881 pub-
lished as a preface (12) to Teoria matematica della Moneta, Con-
ferenza fatta al Circolo Universitario A
o
, Genovesi in Napoli del
Marchese Pasquale del Pezzo, dottore in legge, studente del 4
o
anno di matematiche. Estratto dagli Atti al Circolo universitario
napolitano Antonio Genovesi. Napoli, Tipograa A. Trani, 1881.
23 p.
156 Exposition de tableaux et d etudes de M. Bocion. Gazette de
Lausanne, December 3, 1881. Unsigned.
This is a commentary on the work of Fran cois Bocion (see entry
126).
157 Mathematische Theorie der Preisbestimmung der wirthschaftli-
chen G uter, Vier Denkschriften gelesen vor der Akademie der mor-
alischen und politischen Wissenschaften zu Paris und vor der
P1: IWV
naturwissenschaftlichen Gesellschaft des Waadt-Landes zu Lau-
sanne. Von L eon Walras, Professor der Wirtschaft an der Akademie
zu Lausanne. Vorwort, signed by L eon Walras, Schloss Glerolles
bei St-Saphorin, Waadt (Schweiz), August 1876. TranslatedbyLud-
wig von Winterfeld. Stuttgart, Verlag von Ferdinand Enke, 1881.
8
o
, VII, 96, 1 plate.
This is a translation of entry 124. Entry 157 was reprinted,
Glash utten im Taunus, D. Auvermann, 1972.
Table of contents: Vorbemerkung des Uebersetzers, dated
February 1881, VVI, signed: Lausanne im Februar 1881, Ludwig
von Winterfeld; Walrass Vorwort, dated August 1996, VIIVIII;
I. Prinzip einer mathematischen Theorie des Tausches, 117, 1
plate at end of book; Correspondenz zwischen Herrn Jevons und
Herrn Walras, with Jevonss letter of May 12, 1994 in English and
Walrass letter of May 23, 1994 in French, 1322. II. Gleichungen
des Tausches, 2340; III. Gleichungen der Produktion, 4165;
IV. Gleichungen der Kapitalisirung und des Kredites (the ver-
sion that appears in entry 160), 6691; Schluss (Observation),
92107.
1882
158 De la xit e de valeur de l etalon mon etaire. Journal des Econo-
mistes, Revue de la Science Economique et de la Statistique, 4th
series, 20, no. 10 (October 1882): 513, 1 plate.
This was reprinted separately, no date, 8 pp., 1 plate; and repub-
lished in entry 160, 13544, as part III of Th eorie math ematique
du bim etallisme. The article was reprinted with only one sub-
stantial change as Le con 35 of the second edition of the El ements
d economie politique pure (entry 176), numbered Le con 32 in the
4th edition (entry 210).
In the second edition, the passage from line 11 of p. 408 to line
31 of p. 409 was substituted in place of the following passage in the
article: La xit e remarquable de valeur de l etalon bim etallique
dans notre example tient toutefois ` a ce que, dans cet exemple, les
variations dans la quantit e dor et de largent, qui sont les seules
dont nous ayons tenu compte, se contrarient le plus souvent. Quand
ces variations sont dans le m eme sens, ainsi que cela arrive au
commencement et ` a la n de la p eriode consider ee, les variations
de la courbe de prix de l etalon bim etallique sont sensiblement
egales aux variations des courbes de prix de lun ou lautre des
deux etalons monom etalliques.
P1: IWV
232 Walrass ideas
1883
159 A M. le R edacteur du Figaro. Figaro, June 17, 1883.
This was described by Walras as Entrevue L. Say (a et e reproduit
par le Journal des D ebats du 18 Juin ou 19 Juin).
160 Th eorie math ematique de la richesse sociale. Lausanne, Corbaz
& C
ie
ie
; Rome, Ermanno Loescher e
Co.; Leipzig, Verlag von Duncker & Humblot, 1883. 4
, 256,
6 plates.
This was reprinted, Osnabr uck, O. Zeller, 1964.
This is a collection of the following previously published papers:
(1) Principe dune th eorie math ematique de l echange (entry
103), 725, 1 plate, and the Correspondance entre M. Jevons, Pro-
fesseur ` a Manchester et M. Walras, Professeur ` a Lausanne (entry
105), 2631; (2) Equations de l echange (entry 118), 3353; (3)
Equations de la production (entry 119), 5582; (4) Equations de
la capitalisation et du cr edit (entry 122), 83123. Those four papers
were preceded by a Pr eface des quatre premiers m emoires, dated
August 1876, 56, that is identical, except for the opening words
and the addition in 1883 of two footnotes, to the prefatory note
to entry 118. The papers were followed by an Observation (see
entry 122), 11315; (5) Th eorie math ematique du bim etallisme
(entry 153), 11944, 3 plates; (6) Th eorie math ematique du billet
de banque (entry 148), 14575, 1 plate; (7) Th eorie math ematique
du prix des terres et de leur rachat par lEtat (entry 154), 177253,
1 plate. The last three papers were preceded by a Pr eface des trois
derniers m emoires, dated November 1882, 11718. In one of the
above-mentioned footnotes, Walras pointed out that the rst four
papers had been translated into Italian (entry 137) and German
(entry 157).
1884
161 Monnaie dor avec billon dargent r egulateur. Principes propos es
` a la conf erence mon etaire pour la prorogation de lUnion latine.
Revue de Droit International et de L egislation Compar ee, 16, no. 6
(December 1, 1884): 57588.
This was republished as a pamphlet, Bruxelles et Leipzig,
Librairie Europ eenne C. Muquardt; La Haye, Belinfante Fr` eres;
Paris, Durand et Pedone-Lauriel, 1884. 16; and in entry 205,
319.
P1: IWV
1885
162 Un syst` eme rationnel de monnaie. Gazette de Lausanne, no. 9
(January 12, 1885).
163 Un economiste inconnu, Hermann-Henri Gossen. Journal des
Economistes, Revue de la Science Economique et de la Statistique,
4th series, 30, no. 4 (April 1885): 6890.
This was republished in Etudes d economie sociale (entry 194),
35174. A very slightly abridged translation into English is pre-
sented in Henry William Spiegel, The Development of Economic
Thought, Great Economists in Perspective, New York, Wiley and
London, Chapman, 1952, 47188.
164 Rectication de M. L eon Walras ` a propos dun article sur
H. H. Gossen. Journal des Economistes, Revue de la Science
Economique et de la Statistique, 4th series, 30, no. 5 (May 1885):
26061.
This is a letter, dated April 29, 1885, addressed to the editor of
the Journal. It was republished in entry 239, letter 647.
165 Primi Elementi di Economia Politica, di Luigi Cossa, professore
nella R. Universit` a di Pavia. Sesta edizione. Milano, Hoepli, 1883.
Primi Elementi di Scienza della Finanze, del dottor Luigi Cossa.
Terza edizione. Milano, Hoepli, 1882. Biblioth` eque Universelle
et Revue Suisse, 90th year, 3rd period, 27, no. 8 (August 1885):
44547.
For a brief summary, see entry 239, letter 656, n. 2.
166 Dune m ethode de r egularisation de la variation de la valeur de la
monnaie. Bulletin de la Soci et e Vaudoise des Sciences Naturelles,
2nd series, 21, no. 92 (August 1885): 7192.
naturelles on May 6, 1885, and it was reported in the Bulletin,
no. 93, p. XXV. It was republished as is indicated in entry 167; and
in entry 205, 2649.
167 Contribution ` a l etude des variations des prix depuis la suspension
de la frappe des ecus dargent. Bulletin de la Soci et e Vaudoise des
Sciences Naturelles, 2nd series, 21, no. 92 (August 1885): 93103,
1 plate.
This paper was written in collaboration with Alfred Simon. It
was read to the Soci et e vaudoise des sciences naturelles on June 3,
1885, and reported in the Bulletin, no. 93, pp. XXVIIXXVIII. It
was republished together with entry 166 in a pamphlet with the title
of entry 166 plus the words par M. L eon Walras, and the title of
P1: IWV
234 Walrass ideas
entry 167 plus the words par MM. Alfred Simon et L eon Walras,
Lausanne, Corbaz et Comp., 1885. 8
, 22, 11; and in the X, 4959,

1 plate.
1886
168 Th eorie de la monnaie. Revue Scientique (Revue Rose), 3rd
series, 11, no. 15 (April 10, 1886): 44957; no. 16 (April 17, 1886):
493500. Collection, 37.
This paper was reprinted as an offprint, Paris, Bureau des Revues,
1886. 4
o
, 24; and republished in entry 205 as part of entry 169. The
article is constituted of the Introduction; Part I, Exposition des
Principes, 113; and Part II, Critique des syst` emes, 1426
of Walrass complete Th eorie de la monnaie (entry 169).
169 Th eorie de la monnaie. Lausanne, Imprimerie Corbaz & C
ie
; Paris,
L. Larose & Forcel; Rome, Loescher & C
ie
; Leipzig, Verlag von
Duncker & Humblot, 1886. 8
o
, XII, 123, 4 plates.
This was republished, with changes, in entry 205, 63152, and
Plates IIIII andIVV. The Introduction, 322, Part I, Exposition
des Principes, 2555, and Part II, Critique des syst` emes, 57
85, were part of entry 168. A preface and Part III, Desiderata
statistiques, 2740, were added to them to form entry 169.
170 La question sociale, par Ch. Secr etan. Gazette de Lausanne,
no. 171 (July 22, 1886).
1887
171 L eon Walras, Th eorie de la Monnaie, Lausanne et Paris, 1887.
Revue dEconomie Politique, 1, no. 1 (JanuaryFebruary 1887):
91101.
This is a synopsis by Walras of his Th eorie de la Monnaie (entry
169), so the date in the title is incorrect. The synopsis was unsigned.
It was published in the section titled Bulletin Bibliographique.
See entry 239, letter 762, n. 3 for an editorial footnote to the
review.
172 La livre etalon et la livre Sterling, projet pour rendre la mesure de
la valeur ind ependante du prix de lor et pour asseoir le syst` eme
mon etaire sur une base solide par W. Cross. 4 lettres de lAberdeen
Herald du 29 d ecembre 1855 et des 5, 12 et 19 janvier 1856.
Autograph manuscript, Fonds Walras V III, unpublished (March
1887): 16 leaves, 22.5 cm. by 18.1 cm.
P1: IWV
This is a translation by Walras of William Crosss A Standard
Pound versus the Pound Sterling, AProject for Rendering the Mea-
sure of Value Independent of the Price of Gold, and Establishing the
MonetarySystemona Secure Foundation, inFour Letters Reprinted
from the Aberdeen Herald of 29th December 1855 and 5th, 12th
and 19th January, 1856, Edinburgh, Sutherland and Knox; Glasgow,
Grifn; Aberdeen, L. and J. Smith, 1856, 32 pp.
See entry 239, letter 793, second paragraph, for Walrass evalu-
ation of Crosss proposal.
173 Note sur la solution du probl` eme mon etaire anglo-indien com-
muniqu ee ` a la Section economique de lAssociation britannique
pour lavancement des Sciences, r eunie ` a Manchester. Revue
dEconomie Politique, 1, no. 6 (NovemberDecember 1887):
63336.
This was published in English translation (entry 174); and repub-
lished in entry 205, with an additional note, 15961.
1888
174 Onthe Solutionof the Anglo-IndianMonetaryProblem. Report of
the Fifty-seventhMeetingof the BritishAssociationfor the Advance-
ment of Science held at Manchester in August and in September
1887, London, John Murray, 1888.
This is a translation of entry 173.
1889
175 Th eor` eme de lutilit e maxima des capitaux neufs. Revue
dEconomie Politique, 3, no. 4 (MayJune 1889): 31015.
This was republished as 257 and 258 of the El ements
d economie politique pure, 2nd edition (entry 176).
For unpublished additional ideas of Walrass on the subject, see
entry 239, letter 851, n. 5.
sociale. 2nd edition. Revue, corrig ee et augment ee. Lausanne, F.
Rouge; Paris, Guillaumin & C
ie
; Leipzig, Verlag von Duncker &
Humblot, 1889. 8
, XXIV, 524, 6 plates.

This edition contains an introductory chapter, Des fonctions
et de leur repr esentation g eom etrique. Th eorie math ematique de
la chute des corps, 321, which does not appear in subsequent
editions.
P1: IWV
236 Walrass ideas
177 Opinions emises ` a la commission dexperts touchant la r evision
de la loi f ed erale sur l emission et le remboursement des billets de
banque. Unpublished manuscript (1889).
This is item 97 in Bousquets presentation (entry 238) of
Walrass autobibliography of 1906 (entry 214), but is unnumbered
in Walrass original manuscript, in which he gave the number
97 to his preceding entry (176 in the present bibliography). The
manuscript, communicated to William Jaff e by Etienne Antonelli,
was published by Jaff e in entry 239, letter 925, n. 12.
1890
178 Observations sur le principe de la th eorie du prix de MM.
AUSPITZ et LIEBEN. Revue dEconomie Politique, 4, no. 3
(MayJune 1890): 32023.
This article appeared as Appendix II in the third and subsequent
editions of El ements d economie politique pure, and was translated
into English in entry 235, 48388.
1891
179 De l echange de plusieurs marchandises entre elles. M emoires et
Compte Rendu des Travaux de la Soci et e des Ing enieurs Civils, 5th
series, 44, no. 1 (January 1891): 4249.
Walras read this paper to the Soci ete des ing enieurs civils, Octo-
ber 17, 1890. It was reprinted as an offprint, Paris, Cit e Rougemont,
1891. 8
o
, 9 including the cover and a following blank page, plus
a page with 4 diagrams. It appeared, after revision, as Part I of
Appendix I to El ements d economie politique pure in the third and
subsequent editions. A translated version of entry 179 appeared
in entry 181. A r esum e in 21 lines of print, erroneously reporting
that Walras explained how buyers and sellers act to minimize (sic)
their utility, appeared in R esum e paraissant les premier et troisi` eme
vendredis de chaque mois, S eance du 17 octobre 1890, Soci et e des
Ing enieurs civils, 31415.
1892
180 Th eorie g eom etrique de la d etermination des prix. De l echange
de produits et services entre eux. De l echange d epargnes contre
capitaux neufs. Recueil Inaugural de lUniversit e de Lausanne,
P1: IWV
Travaux des Facult es. Lausanne, Imprimerie Ch. Viret-Genton,
1892. 16976, 1 plate.
In slightly revised form this entry became Parts II and III of
Appendix I to the El ements d economie politique pure in the third
and subsequent editions. Atranslated version of entry 180 appeared
in entry 181. See entry 192 for an offprint of the appendix.
181 Geometrical Theory of the Determination of Prices. Annals of
the American Academy of Political and Social Science, 3, no. 63
(July 1892): 4564.
This is a translation of a version of entries 179 and 180. It was
translated under the supervision of Irving Fisher and contains a
Translators Note by him, 4547.
1893
182 Le probl eme mon etaire anglo-indien. Gazette de Lausanne, no.
172 (July 24, 1893).
This was reprinted in entry 205 in section II of Le Probl` eme
Mon etaire, 16267.
1894
183 Personal Note. Annals of the American Academy of Political and
Social Science, 4, no. 5 (January 1894): 16568 [65760].
This is a biographical and bibliographical note.
184 Le probl` eme mon etaire. Gazette de Lausanne, no. 49 (February
27, 1894).
This was reprinted in entry 205 in section III of Le Probl` eme
Mon etaire, 16874, with the words en Europe et aux Etats-Unis
added to the title.
185 La monnaie de papier. Gazette de Lausanne, no. 286 (December
3, 1894).
This was incorrectly listed under 1893 by Walras in his autobib-
liography (entry 214).
1895
186 Le p eril bim etalliste. Revue Socialiste, 22, no. 127 (July 15,
1895): 1425.
This was republished, with corrections, in entry 205, 175190.
187 Bibliographie de [mon] p` ere. Unpublished, manuscript, 1895.
P1: IWV
238 Walrass ideas
This was listed by Walras as an unnumbered item under 1895
in his autobibliography (entry 214). In entry 238, Bousquet added
to it a note referring the reader to the entry numbered 216 in this
bibliography.
1896
188 M ethode de conciliation ou de synth` ese. Revue Socialiste, 23,
no. 136 (April 15, 1896): 385406.
This is the publication of entry 89. Entry 188 was republished
in entry 194, 175202, where Walras stated (p. 175) that he wrote
entry 89 and therefore entry 188 in 1868, and used it in the writing
of entry 99.
189 Th eorie de la propri et e. Revue Socialiste, 23, no. 138 (June 15,
1896): 66881, 1 plate; 24, no. 139 (July 15, 1896): 2335, 1 plate
(plate II).
This was republishedwithminor alterations inentry194, 20539.
The principal modication is that the mathematics in the footnotes
in the version in the Revue Socialiste were placed in the body of the
article.
190 Le probl` eme scal. Revue Socialiste, 24, no. 142 (October 15,
1896); 386400; 24, no. 154 (November 15, 1896): 53751.
This was republished, with minor changes, in Etudes d economie
sociale, 1896 edition (entry 194), 42262; 1936 edition (entry 228),
45585. Entry 120 is the source of much of the substance of entry
190 and of many of its sentences.
191 Note sur la r efutation de la th eorie anglaise du fermage de M.
Wicksteed. Recueil publi e par la Facult e de Droit de lUniversit e
de Lausanne ` a loccasion de lExposition nationale suisse. Geneva,
Imprimerie Ch. Viret-Genton, 1896, 311.
Walras dated the Note proper (pp. 38) Septembre 1894. The
galleys are stamped 16 Jan. 96 and the page proofs are stamped
21 Jan. 95, which misstates the year. In each case, the stamp is
of the Imprimerie Typographique Ch. Viret-Genton, Lausanne.
Before the Recueil version appeared, Walras added a postscript
(pp. 911) and dated it Octobre 1895. Walras pencilled 1895
on the cover of an offprint of the Recueil version. The title of the
note appears on the front cover of the offprint; the inside of that
cover is blank; and the title is repeated and the text begins on the
following page. The offprint has the same pagination as the Receuil.
The entry was republished in 1896 with some alterations as
Appendix III to the third edition of El ements d economie politique
P1: IWV
pure (entry 192), 48592, and was translated into English in entry
235, 48995. Walras drew the substance and most of its details of
the Note proper from the 31st Le con of the second edition of the
El ements (entry 176), the rst seven pages of which were, with a
fewchanges on the seventh page, the 55th Le con of the rst edition.
He simply quoted or paraphrased passages and copied equations
from the 31st Le con, quoted passages from the 22nd (p. 254) and
32nd (pp. 367, 369), reproduced equations from other lessons in
the second edition ( 199, 274), quoted or paraphrased passages
written by Philip Wicksteed and reproduced his equations, and, in
the postscript, presented an analysis authored by Enrico Barone.
The entry does not appear as such in the 1900 and 1926 editions
of El ements d economie politique pure, but Walras incorporated
the part of it that was directly concerned with marginal productivity
into Le con 36 of those editions.
For probable additions that Walras intended to make to entry 191,
see entry 239, letter 1223, n. 6.
192 E1 ements d economie politique pure ou Th eorie de la richesse
sociale. 3rd edition. Lausanne, F. Rouge; Paris, F. Pichon; Leipzig,
Verlag von Duncker & Humblot, 1896. 8
o
, XXIV, 496, 6 plates.
The introduction and the lessons that this edition shares with the
second edition differ in only very minor respects from those in the
latter, and have the same pagination. To create the third edition,
Walras eliminated the introductory chapter (see entry 176) and four
lessons on the applied theory of money that were in the second
edition, and added three brief appendixes consisting of material
written during 1890 through 1895, as explained under entries 179,
180, and 191.
193 Appendice aux E1 ements d economie politique pure (Th eorie
g eom etrique de la d etermination des prix). Lausanne, F. Rouge;
Paris, F. Pichon; Leipzig, Verlag von Duncker & Humblot, 1896.
20, 1 plate.
See entries 179 and 180.
194 Etudes d economie sociale (Th eorie de la r epartition de la richesse
sociale). Lausanne, F. Rouge; Paris, F. Pichon, 1896. 8
o
, VIII, 464,
3 plates.
This edition was reprinted under the editorship of Oscar Nuccio,
Rome, Edizioni Bizzari (R. Pioda), Ristampe Anastatiche de Opere
Antiche e Rare, CLVII, 1969, and a new edition was published in
1936 (entry 228).
Entry 194 is a collection of the following materials, of which all
but two pages were already published. In section I. RECHERCHE
P1: IWV
240 Walrass ideas
DE LIDEAL SOCIAL: Socialisme et lib eralisme. Lettres ` a
M. Ed. Sch erer; Th eorie g en erale de la soci et e ; Note
(not previously published); M ethode de conciliation ou de
synth` ese. In section II. PROPRIETE: Th eorie de la pro-
pri et e; La question sociale; De la propri et e intellectuelle. In
section III. REALISATION DE LIDEAL SOCIAL: Th eorie
math ematique du prix des terres et de leur rachat par l etat;
Un economiste inconnu: Hermann-Henri Gossen. In section IV.
LIMPOT: De limp ot sur le revenu et de limp ot sur le capital,
constructed from 4 of Th eorie critique de limp ot (entry 27) and
from De limp ot dans le Canton de Vaud (entry 28); Le cadastre
et limp ot foncier; Le probl` eme scal.
195 Prospectus.
This was undated but assigned by Walras to 1896 in his auto-
bibliography (entries 214 and 238). The prospectus, distributed by
the Paris publisher, F. Pichon, announces the publication of the
third edition of El ements d economie politique pure and outlines its
contents.
For a long extract from this entry and a summary, see entry 239,
letter 1237, n. 7.
1897
196 LEtat et les chemins de fer. Revue du droit public et de la science
politique en France et ` a letranger, 7, no. 3 (May-June 1897): 417
36; 8, no. 1 (July-August 1897): 4258.
This article is the publication of entry 115. It was republished in
entry 205, 193236 and translated into English in entry 245. In the
rst installment, a footnote to the title translates as follows: This
memoir was written during 1875 at the request of two members
of the Council of State of Vaud, at a time when the question of
the purchase of the railroads [by the state], being considered again
in Switzerland at the present time, had been raised in the canton
of Vaud. . . . I have used it myself since then as assigned reading
in my course in applied economics, adding to it the advances of
thought provided to me by various publications that have succes-
sively appeared in which the authors have been more or less in
agreement with my point of view. I leave this piece just as it was
originally written, conning myself to analyzing the works with
which it deals in a nal Note. The rst installment was followed by
ve lines listing Walrass professional positions and a short bibliog-
raphy of principales publications and m emoires scientiques
(43638) signed, The Editorial Staff, but evidently written by
P1: IWV
Walras (see entry 238). The second installment was followed by a
Note (5861) that cited and summarized books that conrmed or
completed Walrass views on the subject.
197 Th eorie du libre echange. Revue dEconomie Politique, 11, no. 7
(July 1897): 64964.
This was written in March 1895. It was republished in entry 205,
286304.
198 L economique appliqu ee et la d efense des salaires. Revue
dEconomie Politique, 11, no. 12 (December 1897): 101836.
This was republished in entry 205, 26585.
199 Note sur la th eorie de la quantit e. Manuscript (1897).
This was published in entry 205, 15358.
1898
200 Th eorie du cr edit. Revue dEconomie Politique, 12, no. 2 (Febru-
ary 1898): 12843.
This corresponds to sections I, II, and V of Th eorie du cr edit
in entry 205, 30718 (sections I and II), and 32936 (section V).
Section III in entry 200 corresponds to section V in entry 205.
201 La caisse d epargne postale de Vienne et le comptabilisme
social. Revue dEconomie Politique, 12, no. 3 (March 1898):
20220.
This appears in form and substance in entry 205, 37698.
202 Politique fran caise. Gazette de Lausanne, no. 162 (July 14, 1898).
This was reprinted in entry 204 with the indicated new title;
and in entry 205, 48590, with small and unimportant changes in
wording.
203 La pri` ere du libre penseur. Gazette de Lausanne, no. 165 (July 18,
1898).
This was reprinted in entry 205, 49195, with small and unim-
portant changes in wording.
204 Esquisse dune doctrine economique et social. LAssociation
Catholique Revue des questions sociales et ouvri` eres, 23rd year,
46 (December 15, 1898): 57176.
This was published in the section entitled Documents et Faits
sociaux. It is a retitled reprint of entry 202. It was reprinted in entry
205, 48590, as the rst part of the last section of group VII.
205 Etudes d economie politique appliqu ee (Th eorie de la production
de la richesse sociale). Lausanne, F. Rouge; Paris, F. Pichon, 1898.
8
o
, 499, 4 plates (I, IIIII, IVV, and VI).
This edition was reprinted under the editorship of Oscar Nuccio,
Rome, Edizioni Bizzarri (R. Pioda), Ristampe Anastatiche de Opere
P1: IWV
242 Walrass ideas
Antiche e Rare, CLVIII, 1969; and a new edition was published in
1936 (entry 229). The edition indicated by entry 249.X has been
translated into English (entry 250).
Entry 205 is a collection of the following materials, most of
them already published. In section I. MONNAIE: Monnaie dor
avec billiondargent r egulateur; Mesure et r egularisationdes vari-
ations de la monnaie; Th eorie de la monnaie, reproducing the
entire Th eorie de la monnaie (entry 169) with changes that took
this route: Parts of entry 169 were introduced, with alterations, into
entry 176, and those changes were then introduced into Th eorie
de la monnaie in entry 205. The other articles in section I are:
Note sur la th eorie de la quantit e (not previously published);
Le probl` eme mon etaire, which contains Note sur la solution du
probl eme mon etaire anglo-indien; Le probl` eme mon etaire anglo-
indien; Le probl` eme mon etaire en Europe et aux Etats-Unis;
Le p eril bim etalliste. In section II. MONOPOLES: LEtat et
les chemins de fer. In section III. AGRICULTURE, INDUSTRIE,
COMMERCE: Linuence de la communication des march es sur
la situation des populations agricoles; L economie politique et la
d efense des salaires; Th eorie du libre echange. In section IV.
CREDIT: Th eorie du cr edit. In section V. BANQUE: Th eorie
math ematique du billet de banque; Caisse d epargne postale de
Vienne et le comptabilisme social. In section VI. BOURSE: La
Bourse, la sp eculation et lagiotage. In section VII. ESQUISSE
DUNE DOCTRINE ECONOMIQUE ET SOCIALE: Distinc-
tion entre la science pure, la science morale, la science appliqu ee et
la pratique; Science pure de lhomme et de la soci et e; Science
pure de la richesse sociale; Th eorie morale de la r epartition de la
richesse sociale. Rachat des terres par lEtat; Th eorie appliqu ee
de la production de la richesse sociale. R egularisation des varia-
tions de valeur de la monnaie; Politique fran caise. La pri` ere du
libre penseur. In group VII, all the parts except the one between
the last set of quotation marks (see entries 202 and 203) were pre-
viously unpublished, although Walras had already published most
of their substance. He put together the rst drafts of the rst ve
parts of group VII during the period September 1893 to February
1895.
1899
206 Equations de la circulation. Bulletin de la Soci et e Vaudoise des
Sciences Naturelles 35, no. 132 (June 1899): 85103.
P1: IWV
This paper was read to the Soci et e vaudoise des sciences
naturelles on May 3, 1899, and incorporated, with changes, into
the fourth edition of the El ements d economie politique pure (entry
210). It was translated into Italian by Vilfredo Pareto (entry 207).
In a Note appended to the article (p. 103), Walras introduced
the device of bons with the objective of constructing a virtual
model.Walras reproduced pages 85 to half-way down page 94 as
Le con 29 of the fourth edition of the El ements (210) except that
he omitted about 7 lines from p. 91, and incorporated the last 26
lines of the Note as 274 of the Le con. He inserted the rst 11
lines of the Note into 207 of that edition. He reproduced the
remaining pages of entry 206 as Le con 30 of the fourth edition,
with the following changes: He eliminated 13 lines from p. 98,
and added 17 lines of new writing into 279 and 280. To nish
Le con 30, he then added 12 lines of new writing at the end of the
Equations de circulation material, and then about three pages
that he took from the second (entry 176) and third (entry 192)
editions.
See entry 239, letter 1401, n. 3, and letter 1407, n. 3 for additional
bibliographical information.
207 Sulle equazioni della circolazione (Comunicazione a la Societ` a di
Scienze naturale di Vaud, Seduta del 3 maggio 1899). Giornale
degli economisti, 2nd series, 19, no. 2 (August 1899): 11016.
This is a translation by Vilfredo Pareto of an abridged version of
entry 206.
See entry 239, letter 1407, n. 3 for additional bibliographical
information.
208 Sur les equations de la circulation. Unpublished manuscript
(1899).
This entry is the same in substance as entry 206.
1900
209 Note sur l equation du taux du revenu net. Bulletin Trimestriel
de lInstitut des Actuaires Francais, 11, no. 43 (December 1900):
16264.
See entry239, letter 1454, n. 8, where it is explainedthat this entry
is in substance the same as 242 of entry 210 with the exception
of a difference in the last equation of entry 210.
sociale. 4th edition. Lausanne, F. Rouge; Paris, F. Pichon, 1900.
8
o
, XX, 491, 5 plates.
P1: IWV
244 Walrass ideas
This was reprinted, Paris, Librairie G en erale de Droit et de
Jurisprudence, 1976.
1902
211 Lettre ` a M. P. Pic. Lettre de M. L. Walras en r eponse ` a l etude
bibliographique de M. P. Pic. Questions Pratiques de L egislation
Ouvri` ere et dEconomie Sociale, 3, no. 4 (April 1902): 1057.
212 Autobiographie. Originally unpublished until 1908 (entry 240).
This was published in an Italian translation in 1908 (entry 217).
The French original was rst published in 1965 (entry 240). A
corrected version of entry 212 appears in entry 249. V, pp. 1127.
For a denitive account of Walrass autobiographical efforts see
entry 239, 1, p. 1, n. 1.
1905
213 Cournot et l economique math ematique. Gazette de Lausanne,
no. 163 (July 13, 1905).
This was republished almost in its entirety in entry 237.
1906
214 Bibliographie economique. Autograph manuscript, Fonds Walras
IV a 7 (1906).
This was unpublished until presented by G.-H. Bousquet in entry
238. Entry 214 is described in Chapter 7 of the present book.
1907
215 La paix par la justice sociale et le libre echange. Questions Pra-
tiques de L egislation Ouvri` ere et dEconomie Sociale, 8, no. 6 (June
1907): 16978; nos. 78 (JulyAugust 1907): 22132; nos. 910
(SeptemberOctober 1907): 27986.
In this article, Walras summarized and copied passages from a
number of his previous writings, in particular from entries 82, 154,
189, 197, and part VII of entry 205.
Entry 215 was republished, with an introduction (pp. III) in
which Walras noted contributions by his father and himself, in Lyon,
A. Storck & C
ie
, Paris, F. Pichon & Durand-Auzias, 1907. II, 30.
For a brief description of the contents of entry 215, see entry
239, letter 1623, n. 4.
P1: IWV
1908
216 Un initiateur en economie politique, A.-A. Walras. Revue du
Mois, 6, no. 32 (August 10, 1908): 17083.
This was republished as an offprint, Paris, Editions de la Revue
du Mois, 1908. 14.
217 Leone WalrasAutobiograa. Giornale degli Economisti, 2nd
series, 37, no. 6 (December 1908): 60310.
This is a translation by Maffeo Pantaleoni of entry 212. See entry
239, 1, p. 1, n. 1, for information about the translation, and see entry
240 for the original French autobiography.
1909
218 Ruchonnet et la socialisme scientique. Gazette de Lausanne,
supplement, April 13, 1909.
Walras read this paper on the occasion of his jubilee on July 10,
1909. The paper was republished in Revue Socialiste, 50, no. 295
(July 1909): 57789; and as an offprint with a cover differing from
the journals cover by giving the date of publication as Juillet-
D ecembre 1909, and by stating Paris, R edaction et Administra-
tion instead of Paris, Marcel Rivi` ere & C
ie
.
219 Economique et m ecanique. Bulletin de la Soci et e Vaudoise des
Sciences Naturelles, 5th series, 45, no. 166 (June 1909): 31327.
This paper was republishedas anoffprint, Lausanne, Imprimeries
R eunies, S. A., 1909, 15 p. The article and its accompanying letter
to Walras from Henri Poincar e were republished in entry 236.
1910
220 Histoire des doctrines economiques depuis les physiocrates
jusqu` a nos jours, par Charles Gide et Charles Rist, Paris, Larose
et Tenin, 1909. Gazette de Lausanne, January 6, 1910.
1918
221 De la mise en valeur des biens communaux. La Clairi` ere, Revue
Syndicaliste Bi-Mensuelle, 1st year, no. 14 (February 15, 1918):
66375.
This is a reprint of entry 11. An introductory note was authored
by A. Daud e-Bancel.
P1: IWV
246 Walrass ideas
222 Le Socialisme Scientique. La Clairi` ere, Revue Syndicaliste
Bi-Mensuelle, 1st year, no. 18 (April 15, 1918): 88391.
This is a reprint of entry 80.
1922
223 Th eorie des Geldes. Die Stabilisierung des Geldwertes als das
Problem von heute und vor f unfzig Jahren. Translated by Richard
Kerschagl and Stephan Raditz. Jena, Verlag von Gustav Fischer,
1922. 8
, 115, 4 plates.
This entry is preceded by a Vorwort by the editors (56), and an
Einleitung/Das Problem der besten Geldschopfung; theoretische
Problemstellung und historische Entwicklung (726), primarily by
Kerschagl. The translation of Th eorie de la Monnaie (entry 169),
primarily by Raditz, is on pages 27115.
1926
sociale. Edition d enitive revue et augment ee par lauteur. Paris,
R. Pichon et R. Durand-Auzias; Lausanne, F. Rouge, 1926. 8
, XX,
491, 5 plates.
This is the fth edition of the El ements, differing from the fourth
in only a very few minor respects. It has been reprinted (entry 233)
and translated into Japanese, English, Chinese, Italian, and Spanish.
Section I has been translated into Portuguese.
1931
225 Reon Warurahsu Junsui-Keizaigaku Nyumon. Translated by Miyoji
Hayakawa. Tokyo, Nihon-Hyoron-sha, 1931. 204, 15 cm., table.
This is a translation into Japanese of entry 124, in part via entry
157. Its title translates as Introduction to Walrass Pure Economics.
As Hayakawa indicated in his preface, he used both the original
French text and the German translation, abridged the text, modied
some parts, and omitted the correspondence between Walras and
Jevons and the mathematical appendix.
1933
226 Junsui keizaigaku y oron. Volume 1. Translated by Sumio Tezuka.
Tokyo, Moriyama shoten, 1933.
P1: IWV
This is a translation into Japanese of entry 224. Tezuka divided
the text into two volumes. The rst volume was published with a
foreword dated March 1933. It includes the rst four sections of
entry 224 (Le cons 122). The second volume was not published
in the form in which Tezuka left it (see entry 234). When he used
the Roman alphabet, as in his letters to Europeans, the translator
represented his given name as Juro and spelled his last name
Tedzuka.
1934
227 De lunit e de la valeur (Correspondance entre L eon Walras et Henri
Laurent). Revue dEconomie Politique, 48 (JulyAugust 1934):
125578.
This correspondence between Walras and Hermann (not Henri)
Laurent was presented by Etienne Antonelli. Two letters of dubious
relevance are missing, for which see entry 239, letters 1410 and
1529. See entry 239, letter 1374, n. 2 for a brief discussion of the
entry and of the missing letters.
1936
228 Etudes d economie sociale (Th eorie de la r epartition de la richesse
sociale). Edition d enitive par les soins de Gaston Leduc. 2nd edi-
tion. Lausanne, F. Rouge et C
ie
; Paris, R. Pichon et R. Durand-
Auzias, 1936. 8
o
, VIII, 488, portrait.
This is an edition of entry 194. The content differs from that
of entry 194 in that Leduc, following Walrass wishes, inserted
Souvenirs du Congr` es de Lausanne on pages 379400, preceded
by a preface by Walras dated June 1899 (37778). After page 400,
the content is the same as appears on pages 377462 of entry 194.
229 Etudes d economie politique appliqu ee (Th eorie de la production
de la richesse sociale). Edition d enitive par les soins de Gaston
Leduc. 2nd edition. Lausanne, F. Rouge et C
ie
; Paris, R. Pichon et
R. Durand-Auzias, 1936. 8
o
, 499, portrait, 6 plates.
This is an edition of entry 205. The content and pagination of
entries 205 and 229 are the same.
1938
230 Abr eg e des el ements d economie politique pure par L eon Walras.
Pr ec ed e dun avertissement et r evis e par les soins de Gaston Leduc.
P1: IWV
248 Walrass ideas
Paris, R. Pichon et R. Durand-Auzias; Lausanne, F. Rouge et C
ie
,
1938. 8
o
, 399, 4 plates.
Walras prepared this abridged version of entry 210 for use
in a course on mathematical economics to be taught by Albert
Aupetit. Walrass daughter, Aline, gave the abridgement to Eti-
enne Antonelli, who, in 1911, became the rst person to use it in a
classroom in France.
This was reprinted in 1953.
1940
231 Objecto e divis oes da economia poltica e social. Translated by
Eduardo Salgueiro. Lisbon, Editorial Inquerito, no date. 75,
19 cm.
This is a translation into Portugese of Section I of entry 224. The
date of publication is indicated in the introductory text.
1950
232 Oul` es, Firmin, editor. LEcole de Lausanne. Textes choisis de L.
Walras et V. Pareto, Paris, Librairie Dalloz, 1950.
This entry includes excerpts froma variety of Walrass published
writings. The excerpts are too fragmentary to list separately.
1952
sociale. Edition d enitive revue et augment ee par lauteur. Paris,
Librairie G en erale de Droit et de Jurisprudence, R. Pichon et R.
Durand-Auzias, 1952. 8
o
, XX, 491, 5 plates.
This is a reprint of entry 224, differing from it in only the fol-
lowing respect, apart fromcorrections sometimes defectively exe-
cuted of typographical errors. On page IX of the fourth edition of
the El ements (entry210), the rst line of the secondparagraphreads:
Mais cest surtout la th eorie de la monnaie qui a et e sensible- and
the second line reads in part: ment modi ee par suite des etudes que
jai poursuivies. On page IX of entry 224 a defect of printing left a
blank at the end of the rst line, so that it reads: Mais cest surtout
la th eorie de la monnaie qui. The word qui is mainly obliterated
and the remaining 2 cm of the line are blank. The second line reads
as in entry 233. D. A. Walkers copies of the corrections that Walras
P1: IWV
made to the fourth edition that were incorporated into the 1926 edi-
tion, and of the corrections that Walras wanted to be made but that
were not carried out in print, reveal that he did not change anything
on page IX, so he evidently wanted it to continue to appear as it
had in the fourth edition. Nevertheless, when the 1952 reprint was
prepared, the typesetter or an editor lled in the blank in the 1926
edition so as to make it read: Mais cest surtout la th eorie de la
monnaie qui a et e enti ere-. Walras was therefore mistakenly repre-
sented as writing enti` erement modi ee instead of sensiblement
modi ee.
Entry 233 was reprinted in 1976 by the same publisher.
1953
234 Junsui keizaigaku y oron. Revised by Masao Hisatake. Tokyo,
Iwanami-Shoten, 19531954. 2 vols. Vol. 1, 357; vol. 2, 344.
This is a revision of entry 226 and of the unpublished (second)
volume of Tezukas 1933 translation. The volumes are in Bunko
style, namely pocketsize. The rst volume was published in 1953
and the second in 1954. In volume 1, Tezukas given name is printed
Sumio. Involume 2, it is printedJuro. Volume 1is introducedby
Ichiro Nakayama. Hisatake wrote in his postscript that he respected
Tezukas translation insofar as possible; that he corrected the mis-
translations, of which there were not many; and that his major con-
tribution was to update the terminology and expressions. Hisatake
placed the diagrams in the text instead of at the end, and translated
Walrass lesson summaries and inserted them at the beginning of
each lesson. Five thousand copies of volume 1 were printed and
4,000 of volume 2, all of which were sold by 1963.
1954
235 Elements of Pure Economics or The Theory of Social Wealth. Trans-
lated by WilliamJaff e. Published for the American Economic Asso-
ciation and the Royal Economic Society. Homewood, IL., Richard
D. Irwin; London, George Allen and Unwin Ltd, 1954. 620.
This is a translation into English of entry 224. It is annotated by
Jaff e with Translators Notes, 497558, and collated with previ-
ous editions by him in Table, 55963, and in Collation of Edi-
tions, 565610. It was reprinted, New York, Augustus M. Kelley,
1969; Faireld, NJ, Augustus M. Kelley, 1977; Philadelphia, Orion
Editions, 1984; and London and New York, Routledge, 2003.
P1: IWV
250 Walrass ideas
1960
236 Economique et m ecanique. Metroeconomica, 12, fasc. 1 (April
1960): 313.
This is a republication of entry 219.
1962
237 Cournot et l economique math ematique. Revue dEconomie Poli-
tique, 72, no. 1 (JanuaryFebruary 1962): 6164.
This is a republication, almost in its entirety, of entry 213.
1964
238 LAutobibliographie in edite de L eon Walras (1906) (Les Deux
Autobibliographies de L eon Walras). Presented by G.-H.
Bousquet. Revue Economique, no. 2 (March 1964): 295304.
This entry is described in the Introduction to the present bib-
liography. Bousquet added the items that appear as entries 216,
218, and 219 in the present bibliography. The parenthetical title
given by Bousquet was in recognition of the bibliography of
Walrass works published in 1897 (see entry 196), the entries
in which are included within the 1906 autobibliography (entry
214). Bousquet noted that the 1897 bibliography was signed, The
Editorial Staff, but explained that it was obviously written by
Walras.
1965
239 Correspondence of L eon Walras and Related Papers. Edited and
annotated by William Jaff e. Published for the Royal Netherlands
Academy of Sciences and Letters. Amsterdam, North-Holland Pub-
lishing Company, 1965. 3 vols. Vol. 1: XLIII, 799; vol. 2: XXVII,
763; vol. 3: XXIII, 538.
Each volume has a portrait of Walras.
240 Notice autobiographique. In entry 239, 1, 115.
WilliamJaff e wrote: The text of this Autobiographical Note was
established by collating a manuscript copy presented to me in [the]
1930s by L eon Walrass daughter, Aline Walras, and written in her
hand, with another manuscript copy written in the same hand and
preserved among the late professor Henry Ludwell Moores papers
at Columbia University (239, 1, p. 1, n. 1). Jaff e then summarized
the history of Walrass writing of the autobiography. A version that
P1: IWV
takes account of variations in other drafts of the autobiography
appears in 249.V, pp. 1127.
1966
241 Li lun ching chi hs ueh yao i. Translated by Wang tso-ts e. Tai-pei,
Tai-wan, yin hang. Publisher: ching hsiao che Chung hua shu ch u,
no date. 574, 2 vols.
This is a translation into Chinese of entry 224. The date of pub-
lication, assumed from internal evidence, is almost certainly 1966.
1967
242 Pens ees et r eexions. Cahiers Vilfredo Pareto, 11 (1967): 104
40.
This entry consists of previously unpublished short notes written
by Walras to express his opinions privately on a variety of personal,
social, political, and institutional matters. The notes were preserved
in manuscript form by having been copied by Aline, his daughter.
The manuscript was found by G. H. Bousquet in the Lyons Col-
lection and presented by him with an introduction (1034) and
annotations.
1974
243 Elementi di economia politica pura. Translated by Anna Bagiotti.
Torino, Unione TipogracoEditrice Torinese, 1974. 656, 5 plates.
This is a translation into Italian of entry 224. The introduction is
by Giuseppe Palomba.
Biographical and bibliographical notes were authored by
Giovanni Busino.
244 LApplication des math ematiques ` a l economie politique et sociale.
New York, Burt Franklin Reprints, [1974].
This is a reprint of Principe dune th eorie math ematique de
l echange (entry 103), and of Th eorie math ematique du prix des
terres et de leur rachat par lEtat (entry 154).
1980
245 The State and the Railways. Journal of Public Economics, 13,
no. 1 (February 1980), 81100.
This is a translation of entry 115, which was published in entry
196.
P1: IWV
252 Walrass ideas
1983
246 Junsui keizaigaku y oron. Translated by Masao Hisatake. Tokyo,
Iwanami-Shoten, 1983. XXVII, 531.
This is a translation of entry 224.
Because of the extensive changes in the style of written Japanese
since the publication of entries 226 and 234, Iwanami Publishing
asked Hisatake to make this new translation. He also provided a
27-page introduction and a 3-page postscript.
1984
247 Warurasu, Shakaiteki Tomi no Sugakuteki Riron. Translated by
Toshinosuke Kashiwazaki. Tokyo, Nihon-Keizai-Hyron-Sha, 1984.
130.
This is a translation into Japanese of the texts listed before part
(5) under entry 160; that is of entry 124. The title of entry 247
translates as Walras, The Mathematical Theories of Social Wealth.
1987
248 Elementos de economapolticapura(oTeorade lariquesasocial).
Edited and translated by Julio Segura. Madrid, Alianza Editorial,
1987. 818.
This is a translation into Spanish of entry 224, with an introduc-
tion and commentary by Segura.
249 Auguste and L eon Walras, uvres economiques compl` etes. Edited
by Pierre Dock` es, Pierre-Henri Goutte, Claude H ebert, Claude
Mouchot, Jean-Pierre Potier, and Jean-Michel Servet, under the
auspices of the Centre Auguste et L eon Walras. Paris, Economica,
19872005.
The volumes of the writings of L eon Walras are:
V. L economie politique et la justice. Edited by Pierre-Henri
Goutte with the collaboration of Jean-Michel Servet, 2001.
VI. Les associations populaires coop eratives. Edited by Claude
H ebert and Jean-Pierre Potier, 1990.
VII. M elanges d economie politique et sociale. Edited by Claude
H ebert and Jean-Pierre Potier, 1987.
VIII.

El ements d economie politique pure ou Th eorie de la richesse
sociale, comparative edition. Edited by Claude Mouchot,
1988.
IX.

El ements d economie sociale (Th eorie de la r epartition de la
richesse sociale). Edited by Pierre Dock` es, 1990.
P1: IWV
X.

Etudes d economie politique appliqu ee (Th eorie de la pro-
duction de la richesse sociale). Edited by Jean-Pierre Potier,
1992. Translated into English (entry 250).
XI. Th eorie math ematique de la richesse sociale et autres ecrits
d economie pure. Edited by Claude Mouchot, 1993.
XII. Cours. Cours d economie sociale. Cours d economie poli-
tique appliqu ee. Mat eriaux du Cours d economie politique
pure. The rst of these texts was edited by Pierre Dock` es,
and the second by Jean-Pierre Potier; they edited the third in
collaboration with Pascal Bridel, 1996.
XIII. uvres diverses. Edited by Pierre Dock` es, Claude Mouchot,
and Jean-Pierre Potier, 2000.
XIV. Tables et index. Prepared by Roberto Baranzini, Claude Mou-
chot, and Jean-Pierre Potier, 2005.
250 Studies in Applied Economics Theory of the production of social
wealth. Translated and introduced by Jan van Daal. 2 vols. Vol. 1,
lxviv,193, 1 photograph; vol. II, xii, 223. London and New York,
Routledge, Taylor & Francis Group, 2005.
This is a translation into English of entry 205. It was prepared
under the auspices of the Centre Auguste et L eon Walras, Univer-
sit e Lyon-2. There is an extensive introduction by Jan van Daal
(volume 1, xvlxvii). Most of the annotations are translations of
those made by Jean-Pierre Potier in entry 249.X. In volume 1, the
main text, including Potiers notes and notes added by the trans-
lator, is on pages 1193. In volume 2, the main text, including
Potiers notes and notes added by the translator, is on pages 195
394; the index of persons for both volumes, with biographical notes,
is on pages 395407; and the subject index for both volumes is on
pages 40817.
UNDATED OR WITH SEVERAL DATES
251 Notes sur la M etaphysique et la science dEtienne Vacherot,
notamment sur les chapitres relatifs ` a lanalyse et ` a la critique
de lintelligence. Two autograph manuscripts, 18591861, unpub-
lished, Fonds Walras IS 1927, V/16/20.
These are notes on a work of Vacherots (see 249.X, p. 503). The
rst is a summary of Vacherots ideas; the second is constituted of
quotations culled from his book.
252 Notes dhumeur, in the Collection Walras de la Facult e de Droit
de Lyon, Fonds Leduc, and Collection Antonelli. Autograph
jottings and copies by Aline Walras, various dates and undated.
The collections are described in 239, 1, p. XII and in 249. XIII,
pp. 5038. Many of the jottings are presented in entry 242, and
P1: IWV
254 Walrass ideas
a complete presentation of the jottings that are in the collections
listed above is given in 249.XIII, pp. 50962.
253 Cours dEconomie Sociale profess e ` a lUniversit e de Lausanne.
(Th eorie de la r epartition de la richesse sociale). Autograph
manuscript, Fonds Walras V 6, unpublished course, pages 1240,
25376, 3016, 31336; 28.7 cm. by 21.7 cm.
254 Cours dEconomie Politique Appliqu ee profess e ` a lUniversit e
de Lausanne (Th eorie de la production de la richesse social).
Autograph manuscript, Fonds Walras V 6, unpublished course, 167
leaves, 29.3 cm. by 21.7 cm.
255 Sur les d ecrets du 19 janvier 1867. Autograph manuscript, Fonds
Walras V 6, unpublished, 5 p. and 3 leaves, 20.2 cm. by 15.3 cm.
256 Les p etitions au S enat. Autograph manuscript, Fonds Walras V
1, unpublished, 4 p., 26.4 cm. by 19.7 cm.
257 De la viande ` a bon march e. Autograph manuscript, Fonds Walras
V 6, unpublished, 2 leaves, 21.2 cm. by 16 cm.
258 La situation politique en 1868. Autograph manuscript, Fonds
Walras V 6, unpublished, 5 p., 20.3 cm. by 15.8 cm.
259 Voies et moyens pour la suppressionde loctroi ` a Paris. Autograph
manuscript, Fonds Walras V 6, unpublished, 10 leaves, 26.3 cm.
by 19.6 cm.
260 Etudes sur la monnaie, par Victor Bonnet, 1 vol. in 8
o
, Paris, Guil-
laumin, 1870. Autograph manuscript, Fonds Walras V 1, unpub-
lished, 6 p., 27 cm. by 21.7 cm. 27 cm. by 21.7 cm.
261 Hermann Heinrich Gossen, Expositions des lois de l echange et des
r` egles de lindustrie qui sen d eduisent. Translated by L eon Walras
and Charles Secr etan. Manuscript, unpublished.
This is a translation into French of Hermann Heinrich Gossen,
Entwicklung der Gesetze des menschlichen Verkehrs. It was edited
by Jan van Daal, Albert Jolink, Jean-Pierre Potier, and Jean-Michel
Servet and published in Paris, Economica, 1995. 346.
In a letter to W. S. Jevons, Walras indicated that the translation
was made during January and February 1879 (239, 1879, 1, p. 597).
It is included in this bibliography because Walras may have made
some contribution to it. Not all the information about who did the
work of translation is in agreement. The editors state without qual-
ication, regarding Walras and Secr etan, that the translation was
the fruit of their collaboration (259, 1995, p. 19), and indeed in
1879 Walras referred to the translation that I made with the aid of
one of my colleagues and that will soon be nished (239, 1879,
1, p. 597). The decision to put Walrass name rst on the title page
of entry 259 is nevertheless more than questionable, for Walras
P1: IWV
also wrote that, in Secr etans ofce, we read and put into French
entire books in German and English, him dictating, me writing,
interrupting each other to reach agreement on an idea or an expres-
sion of which I needed to have an exact and complete translation
(242, 1967, p. 128). In 1885, he wrote that Secr etan offered to trans-
late a book in German for me, like he translated Gossens (239,
1885, 2, p. 49). In yet another place, Walras indicated clearly that
Secr etan wished to take the trouble to read that work with me and
to dictate to me, during the reading, a complete translation, which
was made extremely valuable by his admirable knowledge of the
two languages French and German (194, 1896, p. VII).
262 Etude de deux courbes r eelies de variation de prix moyen dun
certain nombre de marchandises. Autograph manuscript, Fonds
Walras V, circa 1887, unpublished: 6 p., 22.2 cm. by 17.3 cm.
263 Bibliographie litt eraire. Autograph manuscript, Fonds Walras
1966, unpublished.
This consists of 25 entries, all included in the present bib-
liography.
P1: IWV
256
P1: PIG
PART I I
Walrass inuence
257
P1: PIG
258
P1: PIG
CHAPTER 9
Models constructed by Walrass
contemporaries and immediate successors
I. Vilfredo Pareto
Background
In 1848, Europe was ablaze with revolutionary activity fueled by oppo-
sition to monarchy and by enthusiasm for democracy, socialism, and
liberal reform. In Italy, the authorities retained their positions and
punished the dissidents. One of the persons whose life was disrupted by
the situation was the Marquis Raphael Pareto, the head of an old Genoese
family. His republican sentiments put him into opposition with the poli-
cies of the government so he took refuge in France, where he married
a French woman. Thus it was that their son, the second most important
economist in the early development of the theory of general equilibrium,
was born in Paris on July 15, 1848, and christened Federico-Vilfredo-
Damaso. From 1852 to 1858 the family lived in Genoa, Liguria; then
an amnesty in 1858 made it possible for the marquis to move to Casale
Monferrato, Piedmont. The Paretos spoke French at home, and Vilfredo
spent thirty years of his life in the French-speaking part of Switzerland.
His second companion, with whom he lived for the last twenty years
of his life and whom he married shortly before his death, was French.
Pareto therefore drew upon both French and Italian cultures. It has been
said that three-quarters of his heart was Italian and one-quarter French
(see Walker 2003).
Pareto learned as a youth to love the Greek and Latin classics, and
beganthe studies that were tomake himprofoundlyerudite as a classicist.
He also loved mathematics. During his high school years he rst attended
the Istituto Tecnico Leardi and then the R` egio Istituto Tecnico in Turin,
concentrating on physics and mathematics and graduating fromthe latter
in 1864 at the age of 16. He entered the University of Turin in that same
year, graduating with a degree in mathematics and physical sciences in
1867. He then entered the Scuola di Applicazione per Ingegneri and
obtained a Dr. Ingegneria diploma in January 1870. Another inuence
upon Pareto was the work of H. T. Buckle (1872), a historian whose work
he read when he was twenty years of age, and who impressed upon him
259
P1: PIG
260 Walrass inuence
the idea that social sciences could be developed on the same level of rigor
as the physical sciences (Pareto to Antonio Antonucci, December 7,
1907, in Antonucci 1938, p. 17; and in Busino 1973, p. 613). Many
passages in his writings also reveal the inuence of mathematics, his
studies of physical science, and his classical studies, which he used to
support his views on modern problems.
In 1870, Pareto began work as a railway engineer in Tuscany, head-
quarteredin Florence. In1873 he became deputy head of an iron and steel
foundry, then technical director in 1875, and director general in 1880.
Frequently he dressed in workmans clothing, but his casual appearance
did not imply that he was not a perfectionist. He demanded excellence
from himself and from the engineers he supervised. In 1886 he began to
give occasional lectures on economics at the University of Florence. He
gave up engineering in 1890 actually he was red for losing large sums
of money in iron commerce speculation and began to concentrate on
publicizing his public policy views and on his academic studies.
Although Pareto succeeded to his fathers title and was proud of
the ancient origins of his family, as noble and venerable as those of
Savoy (Einaudi 1935, p. 338), he never used his title. He accepted
being addressed as professor, but thought it was even preferable to be
called mister (ibid.); a person who knew him well said that he some-
times made gentle fun of his nobility. Nevertheless, there was more
than a hint of an aristocratic manner in his scathing denunciations of
his adversaries. He married the Countess Allesandrina Bakounine in
1889, which has led to the notion that he was linked by marriage to the
Russian anarchist Mikhail Bakunin. In fact, Bakunin belonged to a
totally different family. Neither party entered into the marriage for mate-
rial considerations. Pareto declared that he was never interested in money
in his entire life, not any more on the occasion of his marriage than on any
other, and it is true that Allesandrina was impoverished. Nevertheless,
his declaration was contradicted by the vigor of his efforts to protect his
considerable inheritance from the depredations of socialists, and by his
moving to C eligny to avoid the taxes of the Canton of Lausanne (ibid.,
p. 342). The Paretos did not have any children, which pleased Pareto.
He believed that the divine Malthus, as he referred to the Englishman,
had correctly identied problems created by overpopulation. Preferring
cats to children, Pareto exhorted his friends to make science rather than
babies.
During the years that he lived in Florence and Fiesole, Pareto read
voraciously, gathering together an abundance of information on a wide
variety of topics, information that he was to use in his subsequent cre-
ative endeavors. Because he had to spend ten hours a day working for
P1: PIG
Models constructed 261
a living, much of his reading had to be done at night, and as he was
aficted with insomnia, he would study far into the early hours of the
morning. It was at this time that he was inuenced by the French sociol-
ogist Auguste Comte, by Charles Darwin, and by Herbert Spencer, the
English philosopher who applied Darwinist ideas to an explanation of
the development of civilizations even before Darwin had published his
treatise on the origin of species.
In the 1870s and 1880s, Pareto was much occupied with practical
economic policies. During those years, he was a very active pacist and
humanitarian, opposing the oppression of the weak by the strong. His
characteristic reaction upon detecting such oppression was to launch a
verbal attack on the oppressors. He believed passionately in individ-
ual freedom, a sentiment that differed from his familys on that topic.
From this I can say . . . that my feelings were not acquired, but were
the effect of a character that I had from the time of my birth (Pareto to
Antonio Antonucci, December 7, 1907, in Antonucci 1938, p. 18; and
in Busino 1973, p. 613). The convictions that animated him, he wrote,
were that the sovereignty of the people was an axiom. Liberty was the
universal panacea. . . . Militarism and religion were the major scourges
of humanity (ibid., p. 19). His studies of the economic thought of the
period 1776 to 1870 led him to become ardently in favor of laissez-faire
economic liberalism. Classical economics, he declared, was a perfect
or almost perfect science (ibid.). Even for those people who are most
rmly attached to an ideal of economic liberty . . . , it is almost impossi-
ble to understand exactly the mentality of the extreme liberals, such as
Pareto was in this epoch. Liberalism was for them an ideal, almost a sort
of religion . . . (Bousquet 1960, pp. 3940; and see Bousquet 1928b).
Pareto thought that a purely competitive private enterprise economy, free
of government intervention and regulation, was the best possible system,
on both the domestic and the international levels. He believed strongly
in the free-trade movement, and, during his career in business, published
many articles and pamphlets in both Italian and French, protesting bit-
terly that the tariffs and other foreign trade restrictions of the Italian
government constituted, in effect, a tax on consumers (Pareto 1898a).
Eager to have an impact on government policy, he stood for a position
on the town council of San Giovanni in 1877 and was elected. He ran
for public ofce again in 1880 and 1882, but without success.
Pareto felt deceived when a left-wing government came into power
in 1876, because it did not support the principles of liberalism that he
espoused, and particularly because it continued protectionist policies.
The intervention of the state into the economic life of a country, he
argued, gives rise to much corruption. He conded to Walras that he was
P1: PIG
262 Walrass inuence
thinking of leaving Italy in order to devote himself exclusively to pure
science. He felt that he had achieved nothing in Italy. He wanted, for
example, to lecture on economics, but, he complained to Walras:
Here I am prohibited from doing that. I wanted to teach a course in mathematical
economics, without pay, you understand. The government opposed it. Any citizen
is permitted to give as many lectures as he wants, but they must not be in a series
so as to constitute a course! It doesnt suit me to buy the good will of the Italian
government by prostituting science in its service. Thus, I cannot hope to express
my ideas except in a foreign county (239, 1893, 2, p. 546).
His bitterness became extreme. To live well in this country, one has to
be a thief, or a friend of thieves. Therefore, I very much want to leave
it (ibid., p. 547).
Obviously, Pareto disagreed very strongly with socialist reasoning,
and especially with the idea of state intervention or control of the econ-
omy, and with Karl Marxs ideas about surplus value and wages funds
(Pareto 19021903). Nevertheless, Pareto accepted the concept of the
class struggle. He believed that socialism could enable the underpriv-
ileged to have a voice in social decisions, and he supported the rights
of socialists to freedom of speech and political action. In the spring of
1898 there were violent riots in Italy, resulting from increases in the
price of bread. Almost 90 people died and nearly 800 republicans and
socialists were arrested. Pareto was angered by the governments policy
and welcomed into his home in Switzerland a number of socialists who
had to ee from Italy. In recognition of that and of his eminent stature as
a scientist, the socialists called him the Karl Marx of the bourgeoisie,
a label that Pareto did not appreciate.
Although he had displayed great energy in pursuing liberal causes, as
late as his fortieth year Pareto had shown no evidence of being capable
of contributing to economic science. He was enormously learned, but
he had not used that knowledge to fashion new ideas. If he had died
at forty-two years of age, nobody would have suspected that he would
someday be a thinker, and I suspect that, even in the most detailed Italian
works on the history of economic or political thought, his name would
have appeared solely as part of a footnote, at the bottom of a page. That
silence would have been justied (Bousquet 1960, p. 46).
Walrass inuence
During the 1880s, Pareto began to become interested in pure eco-
nomic theory. A chance meeting in a train with the well-known Italian
economist Maffeo Pantaleoni was a decisive event. It led Pareto to read
Pantaleonis book on the principles of economics, which showed Pareto
P1: PIG
the importance of economic theory. In particular, it pointed the way to
mathematical economics, and therefore to the economics of L eon Walras
and the concept of general economic equilibrium. He was also urged to
study Walrass positive economics by Georges de Molinari, a prominent
Italian economist. Toward the end of the 1880s, Pareto studied Walrass
mature comprehensive model appreciatively, and he started to apply
mathematics to economic theory and to economic policy formulation.
Pareto noted that,
Pantaleoni has worked steadily to make the scientic study of economics familiar
in Italy. I can deny this less than any other person, since it is to his inuence
that I owe my studying of mathematical economics. I had read Walras, but had
neglected the gold in his work in favor of seeing nothing but the sterile vein-stone
of metaphysical reasonings. I was disgusted by them. Since they seemed absurd
to me then and have continued to seem to me to be absurd, I had no condence
that such theories could have a place in an experimental science. But after having
read the Principles of Pantaleoni, I modied my opinion of Walras. I started again
to study him, and this time I could nd the gold in his work, that is to say, the
concept of economic equilibrium (Pareto quoted in Bousquet 1928, p. 18, and
Ricci 1924, p. 10).
Walrass conception of the interconnections, equilibrating processes,
and general equilibriumof a multi-market competitive economy, and his
use of mathematics, were central to Paretos economic reasoning. The
study of his works, Pareto wrote, initiated me into the theories of math-
ematical economics and was at the origin of my own research (Pareto
to Maffeo Pantaleoni, June 16, 1909, in Pareto 1960, 3, appendix 38,
p. 429):
It is no exaggeration to say that the idea of general equilibrium formed the sci-
entic thought of Pareto, and that all his work . . . is founded on that idea. . . . The
inuence of the ideas of Walras in regard to economic equilibriumwas . . . decisive
on the mind of Pareto. . . . Moreover, his work in economic theory was founded
directly on Walrass system, and would be inconceivable without its existence.
The linkage of the thought of the two great scientists is total (Bousquet 1960,
pp. 4849).
Pareto emphasized the importance of the legacy of Walrass equation
system, declaring that economists are indebted to him for the general
representation of economic equilibrium in equations, and asserted that
the discovery of this highly generalized formulation marks the begin-
ning of a very important epoch for the development of new theories
(Pareto 1898b, p. 320, translated in Jaff e 1977, p. 199; Jaff e 1983, p. 79).
He assured Walras that no person who studies mathematical economics
should forget to accord you the very great honor of having been the rst
to establish the systemof equations that governs economic equilibrium
(239, Pareto to Walras, 1901, 3, p. 154). Applying himself to the subjects
P1: PIG
264 Walrass inuence
of Walrass mature comprehensive model, Pareto made it more general
in some important respects, replaced some of its questionable assump-
tions, rened some of its parts, and made original contributions in the
course of improving it.
Many of Walrass contemporaries and successors found his El ements
difcult to read, even when he conned himself strictly to verbal rea-
soning. Georges-Henri Bousquet noted regarding himself that he cer-
tainly belonged to the group of economists who regarded Walrass
El ements as a great treatise, but he also thought that the exposition
in it is detestable: off-putting, dull, prolix, of such a character as
to discourage even readers who would be predisposed to appreciate
it (Bousquet 1964, p. XI). One of Paretos achievements was to set forth
the Walrasian theory of general equilibrium in ways that were compre-
hensible to those who could read Italian and French, and he presented
some of his ideas in articles in English. Displaying the gift of felicitous
expression, his exposition was very effective in the seventy-ve pages
of his Cours d economie politique (1896/1897) that were specically
devoted to economic theory, and in his Manuale deconomia politica
(1906), subsequently translated into French (1909). His denitions and
explanations of economic phenomena were necessarily verbal, and his
analyses of complex economic processes in the body of the text in
those treatises were also carried out by means of verbal reasoning in
the manner of Walras. He therefore made it much easier for the the-
ory of general equilibrium to be understood by those who were not
mathematically well equipped. Another effective aspect of Paretos pre-
sentation was his abundant use of statistical data, rendering it crystal
clear that he conceived of economic theory as properly dealing directly
with the real world. The profusion of facts that he was able to present
and to organize in order to illustrate and substantiate his theoretical posi-
tions, and the soundness and clarity of his statistical reasoning, reveal
that he would have been a great statistician if he had chosen that spe-
cialty. Nevertheless, Paretos economic theories did not become widely
inuential outside of Italy and France until the 1930s, when there was
a Pareto revival, centered primarily on his contributions to welfare
economics.
Unlike Walras, who put equation systems into his texts and often inter-
spersed them with symbols and equations, Pareto conned his mathe-
matics to footnotes in the Cours, and to an appendix in the Manuale. He
remarked that many people think that the advantage arising from the
use of mathematics consists in making demonstrations more rigorous.
This is an error. A demonstration well constructed by the method of
ordinary logic is just as rigorous as one made by the application of that
P1: PIG
other kind of logic which bears the name of mathematics (Pareto 1897,
pp. 49091). He pointed out that he employed mathematics to deal with
a certain type of problem: The advantage of mathematics lies chiey in
this, that it permits us to treat problems far more complicated than those
generally solved by ordinary logic, especially to demonstrate the condi-
tions of general equilibriumin detail regarding the different components
of the economy (ibid., p. 491).
Pareto met Walras in Clarens in 1891 and they entered into an exten-
sive correspondence with each other. In the early 1890s, Pareto published
a series of articles on economic theory and the use of mathematics in
the formulation of economic policies, convincing Walras that he would
develop the Walrasian line of general equilibrium analysis. Thus it was
that, because of his approach to economics, Pareto was offered (after
Walrass retirement, and recommended by Walras and Pantaleoni) the
position that Walras had held at the University of Lausanne (239, 1893, 2,
note 4, pp. 55354). He began his duties there in 1893 and, unlike Walras,
he was a popular teacher. The students applauded when he entered the
classroom and when he left it. Eventually he decided to retire to devote
himself exclusively to research. His request to do so was refused twice
by the Canton of Vaud, and then accepted in 1911, when the onset of
heart disease lent urgency to his request.
Pareto distinguished pure economic theory and applied economics,
and he agreed with Walras that the scope of pure economic theory is
limited to economic facts and relationships regarding which free will
does not play a part. Like Walras, he believed that there are economic
laws, similar in their validity to the laws discovered by natural science.
An anecdote illustrates how that opinion was an integral part of his
thinking. His remarks at a conference prompted Gustav Schmoller, the
German economist of the historical school, to contradict him by declar-
ing that there are no economic laws. At the close of the meeting Pareto
asked him, seemingly in need of advice, whether he knewof a restaurant
where a free meal could be had that evening. Schmoller replied that of
course there was no such place. Pareto allowed a signicant pause to
occur and then remarked: There you see the action of the natural laws
of economics, thereby, he believed, effectively answering Schmollers
comment. Paretos irony and propensity to engage in biting repartee,
often ferocious, did not diminish even in his last years.
Empiricism
In his exposition of the components that he featured in his economic
model, Pareto was, like Walras, writing about his understanding of the
P1: PIG
266 Walrass inuence
behavior of the real economy of his day. It is patently obvious that he
was analyzing real rms, consumers, entrepreneurs, workers, capitalists,
and landlords, and real technology and institutions. His identication of
their nature and his explanations of their behavior was his model. Like
Walras, he discussed that matter explicitly, stating that he was analyz-
ing the real economy and asserting that assumptions and hypotheses
should be realistic (Pareto 1916/1963, pp. 2830). He likewise argued
that inferences from them should be evaluated by empirical studies.
Theories, their principles, their implications, are altogether subordi-
nate to facts and possess no other criterion of truth than their capacity
for picturing them (ibid., p. 30). I am a believer in the efciency of
experimental methods to the exclusion of all others. For me there are no
valuable demonstrations except those that are based on facts (Pareto
1897, p. 491). Regarding economics specically, political economy is
conceived as a natural science. Therefore it must be studied only in the
light of the experimental method . . . (Pareto 1912, pp. 46768). The
equations of general equilibrium can be evaluated only by reference to
concrete facts (ibid., p. 493):
All the conclusions to which deductive studies founded on the general equations
of the economic equilibrium can lead us must nally be veried by a careful
scrutiny of facts, both present and past that is to say, by statistics, by close
observation, and by the evidence of history. This is the method of all the material
sciences. Deductive studies in political economy must not be opposed to the
inductive; these two lines of work should, on the contrary, supplement each other,
and neither should be neglected (ibid., p. 500).
Pareto also espoused the method of successive approximations of the-
ory to the real economy, by which he meant the progressive introduction
into economic theories of empirically derived considerations so as to
achieve progressively detailed degrees of realism (Pareto 1896/1897, 1,
pp. 1617; 2, pp. 15, 78). The situation in economic theorizing is of the
same nature as in physical science:
If we simplify our problem by supposing a heavy body to be falling in a vacuum
and attracted by the earth alone, we may infer the well-known theory of falling
bodies. We thereupon complicate the problem by introducing the fact of the airs
resistance, and in this way arrive at theories approaching more and more closely
to reality.
Such are the considerations which lead to the method of successive approxi-
mations in political economy (Pareto 1897, p. 490).
In other respects, there were contrasts between Walras and Pareto.
Walras never gave up his interest in practical economic problems,
whereas Pareto eventually turned his back upon them. Walras advocated
the intervention of the state in certain respects, whereas Pareto rejected
such intervention. Pareto did not agree with Walras on all theoretical
P1: PIG
matters. If there are some points on which I think you are wrong, I have
told you that frankly he wrote to Walras (239, 1901, 3, p. 154). Unlike
Walras, Pareto believed that the methods of positive science should not
be conned to the realm of pure economic theory (where free will is
absent) but should be used in the study of all aspects of economics and
of human behavior generally, including therefore those in which free will
does play a part. Unlike Walras, as has been seen, Pareto did not think
that moral philosophy was part of economics, and he detested the specic
content the stupidity (Pareto 1909, in Pareto 1960, 3, letter 596)
of Walrass social economics, which he considered to be metaphysical
(Pareto 1897, p. 491). I do not accept in the slightest, he wrote, his
metaphysical way of treating science (Pareto 1908, in Pareto 1960, 3,
letter 590). He lamented: Although, I have always freely acknowl-
edged my obligation to . . . Walras . . . , Walras nevertheless became my
enemy, because I would not lend myself to his metaphysical notions
(Pareto 1911, in Pareto 1948, p. 62).
II. Paretos contributions to economics
Theory of demand
Paretos theory of consumer demand is a central pillar of his model of
general equilibrium. He used Walrass mathematical form of expression
of supply and demand, and was similarly contemptuous of literary dis-
cussions of those functions. The terms supplyanddemand, he exclaimed,
like all the terms of non-mathematical economics, have been employed
in an unrigorous, equivocal, ambiguous manner, and the considerable
number of purposeless, vain, and disorganized discussions of which they
have been the subject is truly unbelievable (Pareto 1909, p. 220). He
adopted some of the features of Walrass theory, but reformulated it in
two major ways.
First, he replaced Walrass assumption that it is possible to specify the
number of units of satisfaction obtained from an amount of a commod-
ity with the assumption of ordinal utility. Pareto thought that it is not
impossible in principle to measure utility objectively (see Kirman 1987,
p. 805), but he observed that no one has been able to discover how one
could go about measuring it (Pareto 1909, p. 664). He therefore dropped
the assumption that the levels of utility can be measured cardinally.
Pareto then modied the indifference analysis of preferences (Pareto
1896/1897, 1, p. 35; 1909, pp. 16870, 18384) that had been initiated by
Francis Y. Edgeworth (1881), retaining indifference curves but assuming
that utility is ordinally measurable. The consumer, he asserted, is able
to specify that he prefers one batch of commodities to another without
P1: PIG
268 Walrass inuence
being able to state by how much it is preferred or to specify that
he is indifferent to them (Pareto 1909, p. 540). Because utility is not in
fact objectively cardinally measurable, this desirable theoretical advance
replaced a restrictive assumption with one that is less restrictive but that
nevertheless facilitates the derivation of demand functions.
Second, Pareto replaced Walrass assumption that the utility obtained
from a commodity depends only on the amount of that commodity con-
sumed. Edgeworth had recognized that the utilities of different com-
modities are not independent, and by putting that consideration into an
ordinal utility framework, Pareto made his second major contribution
to consumer demand theory. He argued that the marginal utility of a
commodity depends not only on the amount of it consumed but also on
the amounts consumed of other commodities (Pareto 1893b, pp. 299,
3067), and that some commodities are substitutes for each other and
others are complements. He explored these matters verbally and mathe-
matically, showed the related forms of indifference maps, and deduced
some of the consequences for demand functions (ibid.; Pareto 1896/
1897, 1, pp. 1011; 1909, pp. 24959).
Pareto then formulated a theory of consumer demand based on the
further assumptions, made by Walras, that the consumer wants to maxi-
mize his utility and knows how to do so (see Marchionatti and Mornati
2003), and that the quantity he demands of a commodity is a function, in
principle, of the prices of all consumer commodities, given his income
and preferences (Pareto 1896/1897, 1, p. 35). Through the use of cal-
culus, in the manner of Walras, Pareto repeated Walrass demonstration
that changes in quantities supplied and demanded can be shown to be
determinate functions of changes in prices. Pareto also afrmed Walrass
conclusion that the consumer achieves maximum utility by purchasing
the amounts of any two commodities for which the ratio of their marginal
utilities is equal to the ratio of their prices, although Pareto expressed
that condition in the way appropriate for an ordinal indifference analysis
(Pareto 1909, p. 559). He thereby derived a consumer demand function
for commodities that was more general than Walrass. Pareto then fol-
lowed Walras in demonstrating howchanges in the quantity demanded of
one commodity are determinate functions of changes in the price of that
commodity and of other commodities, and how the responsiveness of
consumer demand to a change in any price can be measured as a function
of that change. Paretos reformulation of Walrass account of consumer
behavior was adopted by many Continental economists shortly after his
Cours appeared, and, together with the work of Irving Fisher (1892)
and Eugen Slutsky (1915), was developed into the modern theory of
consumer demand.
P1: PIG
Existence, uniqueness, and stability of equilibrium
Like Walras, Pareto used a set of simultaneous equations in an effort
to describe the characteristics of his model of general equilibration
and equilibrium of a competitive economy, with the difference that
he constructed a completely disaggregated version (Pareto 1896/1897,
1, pp. 4461). That is, instead of using market supply and demand
functions in the manner of Walras, Pareto used the individual supply
and/or demand functions of each economic agent of each consumer,
of each resource supplier, and of each rm a completely microeco-
nomic approach. Paretos Walrasian model gave rise to the disaggrega-
tive approach used by many economists.
Adopting the same procedure Walras did, and thus committing the
same error, Pareto asserted that equilibrium exists in his model because
the number of independent equations equals the number of unknowns,
namely the number of prices and quantities of commodities (ibid., pp. 26,
4446, 61; 1909, p. 658). Moreover, he was oblivious to the fact that his
virtual equations, like Walrass, could not provide the equilibriumvalues
of the variable in his non-virtual model. As for uniqueness, adding his
authority to the error that persists to this very day, Pareto asserted that
multiple equilibria exist (Pareto 1909, p. 197).
Stability
Walrass attempts to show that his mature comprehensive model is
stable were studied by his contemporaries and immediate successors.
The starting point of their work was naturally that particular model,
with all of its irrevocable disequilibrium processes and phenomena
naturally, because that is the one presented in the editions of the El ements
they studied, Walras having not yet devised the written pledges sketch
when they began to learn his theories.
1
Furthermore, the inuence of
Walrass elaboration of non-virtual disequilibrium behavior continued
to be manifested in certain academic quarters by H. L. Moore, J. A.
Schumpeter, and Henry Schultz, for example long after the publication
of the written pledges sketch in 1899 and 1900.
Pareto, in particular, studied the mature comprehensive model in the
years that preceded his writing of the Cours. Like Walras, he treated sta-
bility by giving a verbal analysis of how a freely competitive economy
moves toward equilibrium. In his studies of economic adjustments, he
1
See Paretos acknowledgement of receipt of the second edition of the El ements that
Walras sent him in 1891 (239, 1891, 2, p. 465) and Fishers acknowledgement of
receipt of the almost identical third edition (239, 1896, 2, p. 676).
P1: PIG
270 Walrass inuence
simply assumed that a freely competitive economy is stable. With respect
to exchange, Pareto declared that the tatonnement process featured in
Walrass mature comprehensive model accurately described the dis-
equilibrium behavior of the real market system: Walras has shown that
the bargaining that takes place in free competition is the means of solv-
ing the equations of exchange by repeated attempts (ibid., pp. 2425).
Mr Edgeworth has objected that that is only one means by which mar-
kets move toward equilibrium. He is right, Pareto declared, but the
way indicated by Mr Walras is truly the one that describes the largest
proportion of markets (ibid., p. 25) exactly what Walras had con-
tended. With respect to production, Pareto argued that Walrass idea of
tatonnement in that aspect of economic activity should also be adopted
and for the same reason that Walras had espoused it, namely that it
was an accurate description of what happened in the real economy:
Mr Walras has shown that the competition of entrepreneurs and traders
is a means of solving the equations of the equilibrium of production
through successive attempts. This idea, in general, seems very fruitful
for economic science (ibid., pp. 4546). Pareto therefore used Walrass
mature concept of tatonnement in all his formulations of competitive
economic adjustments in the 1890s and subsequently, not paying any
attention to the written pledges sketch that Walras devised in 1899.
2
Thus in Paretos model of competitive general equilibrium, as in
Walrass, prices, transactions, and production undergo successive irre-
vocable changes until equilibriumof the entire market systemis reached.
Pareto and all subsequent modelers of competitive general equilibrat-
ing processes assumed that prices are changed iteratively in the same
direction as the sign of the market excess demand, which was described
above as the essence of Walrass idea of tatonnement in exchange; and
2
Expressing an opinion contrasting with these pronouncements of Paretos, and with
Chapter 5 of this book, Emeric Lendjel contends that Walras arranged an account
of the disequilibrium behavior of freely competitive markets in such a fashion that
it was in accordance with his scheme of mathematical iteration and that Pareto dis-
sented from that view. Walras wanted to arrange matters so that the behavior of
competitive markets corresponds to a mathematical method of numerical solution of
equations. The second approach [Paretos] emphasizes the organizational coherence
of the pricing process. That approach tends to neglect the mathematical dimension of
the tatonnement, thus attaining a greater descriptive richness (Lendjel 1999, p. 312).
It is certainly true that Pareto did not want mathematical techniques to be used to deter-
mine how markets are represented, but neither did Walras. If Lendjel was contending
that Walras constructed his mature comprehensive model so as make it justify his
equations, that is not an accurate opinion. Lendjel probably meant that Walras decided
to construct the written pledges model to achieve that purpose. Lendjel represents
Pareto as arguing against that model but in fact, as noted, he did not discuss it, and his
work was squarely based on the mature model.
P1: PIG
Pareto and all others who constructed competitive models with disequi-
librium production assumed that output is changed iteratively in each
rm in the same direction as the sign of price minus average cost, which
was described above as the essence of Walrass idea of tatonnement in
production.
Pareto also claimed to have treated stability mathematically, believing
that he used his equations to study dynamics in a way similar to the
one Walras had tried to develop in his mathematical iterations. Pareto
believed that his own treatment of dynamics resulted in a highly realistic
model:
I have endeavored to extend to dynamic questions the use of the equations given
for the static equilibrium. The most accurate description possible of the economic
phenomenon is to be reached in this way. Is it not a most remarkable fact that a
systemof equations should thus be able to express not only the general character of
economic phenomena, but everysingle detail as far as we mayhave anyknowledge
of them (Pareto 1897, p. 492).
Reminiscent of passages in Walrass writings (205, 1898, p. 467), Pareto
declared that the entire body of economic theory is henceforth bound
together in this way and knitted into an integral whole (ibid.).
The entrepreneur
Walrass contemporaries and successors praised his concept of the
entrepreneur and were inuenced by it. Francis Y. Edgeworth, who dis-
agreed with the idea that the entrepreneur makes zero prots in equilib-
rium, nevertheless wrote in recognition of the importance of Walrass
theory that Professor Walras is one of the rst who correctly con-
ceived the entrepreneur as buying agencies of production . . . and selling
nished products in [different] markets, which thus become interde-
pendent (Edgeworth 1889, p. 435). Enrico Barone based his work on
the entrepreneur entirely upon it, afrming how profound and correct
is Walrass conception of an entrepreneur who, under the conditions
postulated, makes neither gain nor loss. . . . It is absolutely astounding
that the conception should have been made the subject of criticism. . . . I
frankly must confess myself absolutely incapable of understanding how
any difculty whatever can arise as to the validity of this conception,
which is indeed most simple (Barone 1896, p. 145).
Pareto likewise adopted Walrass theory of the entrepreneur and
extended it. For the case of free competition, he treated the cases of
speculation and production. Accepting Walrass analysis of the activity
of speculators (Walras 1880, pp. 370, 379), Pareto explained that, by
P1: PIG
272 Walrass inuence
responding to price changes by buying or selling, speculators transmit
information about the state of demand and supply to the production side
of the economy, and they facilitate the process of transforming savings
into new capital goods (Pareto 1896/1897, 2, pp. 24245). The social
function of speculators, insofar as they do not act directly on prices, is to
solve the equations of economic equilibrium in the best and promptest
manner possible (ibid., p. 245).
Regarding entrepreneurs in rms that produce commodities, Pareto
followed Walrass account to the effect that, by buying and selling,
entrepreneurs also cause the economy to move toward an equilibrium
in which their prots are zero (Pareto 1909. p. 197). He extended the
analysis of how entrepreneurs behave in the phase of disequilibrium,
however, in two major ways. First, he recognized that they make errors
in their production decisions:
It is necessary to produce commodities a certain time and sometimes a very long
time before they are consumed. In order for there to be a perfect adaptation of
production to consumption it would be necessary: 1
that consumer demand be

predicted; 2
that the results of the process of production be accurately predicted.

It is impossible to do these two things with precision.
In the present state of economic organization, it is the producers and the mer-
chants who try to make those predictions. If they guess correctly, they make
money; if they are mistaken, they ruin themselves (Pareto 1909, p. 530).
Second, Pareto realized that entrepreneurs keep changing their prot
goals and thereby repeatedly modify the path taken to equilibrium, and
that as a result the equilibrium values of the variables change. Thus
disequilibrium transactions and disequilibrium production in the real
economy not only change the total amount of commodities in the system
and their distribution, and therefore the supply and demand functions
(the causes of path dependency in Walrass and Paretos model), they are
also associated with errors of expectations and consequent revisions of
plans on the part of real entrepreneurs, which also cause path dependency
in Paretos model (Pareto 1896/1897, 1, pp. 1819; Steiner 1999).
Pareto also examined the case of a monopolistic entrepreneur (Pareto
1896/1897, 1, pp. 6269, passim). The prots he makes, Paretoobserved,
are not reduced to zero by the force of competition. He is able to restrict
output and thereby to charge indenitely a price for his product that is
greater than its average cost, which he would not be able to do if he
were in a freely competitive industry. Like Walras, Pareto argued that
private monopolies are obstacles to an optimum allocation of resources
and to efcient rates of their use. It is easy to see that in all cases the
monopolists prot is obtained only by harming others (ibid., p. 69).
Pareto wanted to integrate monopoly into a general equilibrium setting,
P1: PIG
but, like Walras, had to content himself with a model containing a system
of freely competitive markets and unrelated instances of monopoly.
Economics of the rm
Pareto took Walrass theory of production in the rm as the point of
departure for his own treatment of that subject. As was the case in
Walrass analysis, entrepreneurs are the central agents in Paretos theory
of production in the rm. They choose the technology that is used and
hence select either xed or variable coefcients. It will be remembered
that, in the terminology used by Walras, the coefcients of produc-
tion are the quantities of each of the kinds of productive services that
are used in the production of one unit of the output of a rm. Pareto
introduced the consideration that in some rms the proportions in which
economic resources are combined are xed, and that in other rms they
can be varied. Drawing the consequences of those phenomena, Pareto
formulated a version of the theory of marginal productivity that was a
signicant extension of Walrass work (Pareto 1896/1897, 2, pp. 8490).
An implication that Pareto drew from his own account is that, in some
rms, the proportion of capital goods services to labor services can be
altered in accordance with changes in the relative prices of capital and
labor, while in others it cannot be (Pareto 1909, pp. 32728). He analyzed
the combination and amounts of inputs hired by a rmand explained the
distribution of its revenue among the owners of the economic resources
that contribute to producing it.
In all of those reasonings, Pareto was setting forth his perception of
what happened in reality. Thus his line of argument when he considered
variable coefcients was not that their variability was a purely logical
possibility, but that they are actually variable in many real rms. Failure
to recognize that situation is to misrepresent the facts:
We have already noted the error that consists in believing that the coefcients of
production depend only on the technical conditions of production.
Another completely erroneous theory is the theory of xed proportions. . . .
Now, the majority of economists who use the theory of xed proportions seem
to believe that there exist certain proportions in which the factors of production
are combined, independently of the prices of those factors. That is false (ibid.,
pp. 32627).
Pareto strengthened his exposition by providing examples of how the
proportions are varied (ibid.).
Cecil G. Phipps (1954), in a neglected but brilliant contribution, eval-
uated the controversy between Walras and Pareto over the respective
merits of their treatment of minimization of costs of a given output and
P1: PIG
274 Walrass inuence
maximization of prot. Phipps remarked that both authors used so much
verbiage that it is difcult, if not impossible, to determine precisely
what they wished to say, that some of their notation was incomplete,
and that they unnecessarily restricted themselves to a linear homoge-
neous production function. He showed that the notion that they reached
different solutions is incorrect, that while their approaches to the prob-
lem were different, their solutions are equivalent (ibid., p. 31). Phipps
concluded that:
In sum, the assumptions of Pareto appear to be more complicated, more limited
and incomplete, and therefore less useful than those of Walras as emended here.
However, since the two approaches lead to equivalent results, it is a matter of
personal choice which method is to be followed in the solution of the problem of
production (ibid., p. 38).
Welfare economics
Pareto studied Walrass thesis that free competition generates a relative
maximum of welfare for a society, and based his analysis on Walrass
work. Walrass assertions of that condition do not mean that he had a
sound proof of it. Pareto explained the shortcomings of Walrass anal-
ysis, and rened the analysis of the conditions for maximum efciency
in exchange, in production, and in the capital goods market (Pareto
1896/1897, 2, 90104). The members of a group, Pareto contended,
enjoy, in a certain state of the economy, a maximumof well-being when
it is impossible to nd a means of altering that state a little in such a way
that the well-being of every one of the members increases or that the
well-being of every one of the members decreases. That is to say, a small
change . . . is agreeable to some but disagreeable to others (Pareto 1909,
p. 354). In other words, in an optimum situation it is impossible to make
anyone better off without making someone worse off. In recognition of
his contribution, the condition that he identied is known as a Pareto
optimum.
Pareto also developed an analysis of the welfare aspects of produc-
tion and consumption of consumer commodities that was patterned on
Walrass theorem on the maximum utility of new capital goods (176,
1889, pp. 3017; 26364, pp. 41725). One of Paretos notable con-
tributions in this regard was to distinguish between the conditions for
maximizing individual welfare, and the conditions for maximizing the
welfare of society as a whole. He showed that earlier writers had com-
mitted a fallacy of composition, assuming that because one individual
may improve his welfare if he acts alone, all individuals can necessarily
do so if they act simultaneously to try to achieve that goal. For example,
P1: PIG
if national income is constant, one individual may be able to acquire
more income, but all individuals cannot do so.
Neither Walras nor Pareto, nor anyone else, has proved that pure
competition necessarily achieves a social welfare maximum, but Pareto
showed that pure competition provides part of the situation that would
be necessary to achieve such a maximum. He became the rst theorist
to demonstrate with a high degree of rigor that, subject to various con-
ditions, a state of maximum efciency can be achieved by an economy
of the type that Walras described in his model. Paretos formulation and
his disaggregative approach became the foundation of the new welfare
economics, the study of maximumefciency and well-being developed
during the 1930s and 40s. Thus Walrass legacy in this regard was, via
Paretos work, tostimulate the development of twentieth-centurywelfare
economics.
Another interesting feature of Paretos welfare economics is his
demonstration that pure economic theory does not give us a truly deci-
sive criterion for choosing between an organization of society based
on private property and one based on socialism (Pareto 1909, p. 364).
He believed that, in principle, socialism can achieve a maximum of
well-being if it follows policies that have the same results as pure com-
petition (Pareto 1896/1897, 2, 9293). Evidently, however, he thought
that, in practice, socialism would not follow those policies. He did not,
of course, have the benet of the analysis by the Austrian school of
economic thought of the inherent aws and inevitable failure of central
planning, but he seems to have come to the same conclusion: The abuse
and the scandals of the system of protection . . . are well made for giving
us an idea of what awaits us when socialism will reign in all its glory.
Then many people will recognize, a little late, how poorly advised they
have been to abandon the defense of economic liberty (Pareto 1893a,
p. 36). We will be with the socialists when they aid us in resisting the
present oppression; we will say goodbye to them when they decide to
substitute another for it (Pareto 1899, p. 216).
Paretos law
The personal distribution of income in many different economies and
at many different times appeared to Pareto to have a general form in
common, an idea that he did not owe to Walras. He devised an equation
to describe that form and analyzed its goodness of t to statistical data
drawn from a number of countries and relating to different eras. Believ-
ing that the results validated his hypothesis, Pareto concluded that the
distribution of income is not the effect of chance (Pareto 1896/1897, 2,
P1: PIG
276 Walrass inuence
p. 315). He contended that there are underlying laws of production and
of the use of economic resources that cause the income distribution to
be of that general form. Paretos law was therefore interpreted as cast-
ing doubt upon the possibility of altering the distribution of income by
government policies. That implication was not palatable to many social
reformers, and, undeterred by their lack of scientic method or evi-
dence, they harshly criticized his conclusions. Pareto was not perturbed
by that. He was concerned with discovering the realities of the matter:
The investigator who is in search of the lawof the distribution of wealth
does not aim to encourage or dissuade anyone; he is simply intent on
discovering the truth (Pareto 1897, p. 487).
It should be noted that even if Paretos law were true, that would
not imply that the personal distribution of income cannot be affected,
for example by a socialist government, by taxes, and by transfer pay-
ments, as Pareto was well aware (ibid., p. 500). In any case, the law is
an interesting hypothesis: Few if any economists seemed to have real-
ized the possibilities that such invariants hold out for the future of our
science. . . . Invariants of this type might lay the foundations of an entire
novel type of theory (Schumpeter 1949, in 1951, p. 121 and note).
General socioeconomic equilibrium
Pareto came to believe, like most economists, that an understanding
of economic behavior necessitates an understanding of the functioning
of society in general, so after 1905 he concentrated on the latter. He
contended that political economy is but a part and a small part of
the more general science of sociology. It considers only one aspect of
the things we see around us. To complete our knowledge we must also
consult other branches of social science (Pareto 1912b, p. 468). Just
as there are interconnections between economic variables, he argued,
so also are there interconnections between non-economic variables, and
between them and the economic ones.
The economic aspects of Paretos sociological studies dealt with
what he called subjective behavior, particularly the behavior considered
in applied economic studies and normative principles and judgments.
Whereas Walrass social economics was the expression of his own value
judgments about how society and the economy should be organized and
conducted, the value judgments that Pareto was interested in studying,
whatever their character and consequences, are those made by the mem-
bers of a society. He believed that many aspects of life must be taken
into consideration in the formulation of economic policies and normative
economics, and therefore doing so requires bringing to bear information
P1: PIG
and laws developed in many branches of knowledge. That idea had been
espoused in a more limited form by Walras in regard to the use of eco-
nomic theory in the formulation of policies to solve practical problems
(Chapter 4).
Paretos sociology was not an abandonment of economic analysis. It
was an attempt to achieve a theory of the general equilibrium of society
as a whole (Pareto 1916).
3
Two thinkers that inuenced his adoption
of that goal were Auguste Comte, who wanted to see the development
of a unied social science, and Herbert Spencer. Pareto was not able to
achieve a satisfactory synthesis of the diverse materials that enter into
the problem, not only because the task is too great for one individual to
accomplish, but also because it is probably impossible, for it amounts
to a theory of the equilibrium of every aspect of human life. Moreover,
it is by no means clear, and certainly has never been demonstrated, that
society as a whole is an equilibrating system. It is indeed highly unlikely
that many of its parts, such as criminal activity, sports, and artistic,
theatrical, and musical life, are equilibrating phenomena, or even what
equilibrium could possibly mean in those aspects of behavior. In any
event, Pareto was unable to show that a society tends to move toward a
certainequilibriumcongurationinits class relations, its judicial system,
its political system, and so forth, as well as toward an equilibrium of the
economic variables in the manner described by Walras. This should not
diminish our appreciation of the magnitude of Paretos achievement in
pointing out the many respects in which economic activities and the
other aspects of private and social life are interrelated.
Paretos conservativism
There was a difference between Paretos attitude in his last years toward
intellectual system building and politics on the one hand and toward his
economic theory on the other. He never abandoned the latter. He never,
for example, declared himself against his theory of demand or against
general equilibrium analysis in economics. As he grew older, however,
he became critical of the ideas of Comte and Spencer, declaring that they
were pseudoscientists; he continued to afrm a variation on the idea of
the class struggle while continuing his vehement opposition to socialist
doctrines; and he rejected the political beliefs of his youth. Concluding
that foolish leaders were elected and that elected governments obeyed
the will of the majority no matter how ignorant it was, he questioned the
3
See Busino 1987, pp. 8013 for an instructive brief discussion and appreciation of
Paretos sociological ideas.
P1: PIG
278 Walrass inuence
results of democracy. By the time of his death on August 19, 1923, he
had become an embittered conservative. When I was young, he wrote,
I made light of, or at least excused, the evils of democracy. The French Terror
[I believed] was nothing more than a slight stain on the luminous picture of the
French Revolution. . . . I did not realize that my reasoning was only an attempt to
give a logical appearance to what my emotions led me, in every respect, to want to
believe (Pareto 1907, in Bousquet 1960, p. 26; Pareto to Antonucci, December 7,
1907, in Antonucci 1938).
The attacks that Pareto made in his old age on reformers, democracy,
and humanitarianism, and his approval of Mussolinis early reforms
views that are painfully embarrassing to those who recognize his genius
and have learned from his works cannot obscure the greatness of
his contribution as an economist and social scientist. It should also be
remembered that he refused many of the honors that the fascists heaped
upon him, and wrote that they should not continue on the course of
restricting freedomof academic expression, nor seek to erect a corporate
state.
Pareto was an exceptional thinker, a highly creative scholar, able to
discern what was sound in previous science and to nd new truths.
The range of his intellectual vision was enormous. He was uniquely
endowed with the ability to deal with theoretical and empirical aspects
of science, to analyze and to synthesize. His economic theorizing was
unobstructed by metaphysics, sophistical epistemology, or the haze of
error that normative convictions can interpose between reality and the
apprehension of it. He had a clear, direct, and sensible understanding that
the relation that should exist between theory and its subject matter is that
the former should describe, explain, and predict reality. His powers of
reasoning, and the crystalline clarity of his methodological views on how
hypotheses should be obtained, veried, and used, greatly contributed to
the solidity of his achievements. His was a truly seminal intellect, blazing
paths that have been fruitfully followed by generations of subsequent
scholars. Social scientists will always be indebted to him.
4
III. The second generation
Some economists of the school of Lausanne
Walrass methods and theories inuenced a number of members of
the generation of economists that succeeded him. In France, Walras
did not have many followers during his lifetime, with the notable
4
For an expression of that indebtedness, see Allais 1968, p. 408.
P1: PIG
exceptions of Albert Aupetit, who attempted to elaborate the theory
of money in a general equilibrium setting (1901) and made other contri-
butions to Walrasian economics (1914), and Etienne Antonelli (1939),
who delivered lectures on Walrass theories at the University of Lyons
(1914) and contributed in other ways to the dissemination of Walrass
ideas (for example, Antonelli 1939; and see Potier and Walker 2004).
Hermann Laurent (1902), Wladyslaw Zawadzki (1914), and Jacques
Moret (1915) gave expositions of mathematical economics and gen-
eral equilibrium theory. There were a few non-Continental adherents
of Walrass approach and theories. With respect to much of his work,
Irving Fisher (1892, 1896) adopted a Walrasian outlook and methods
(see Introduction, note 10). Arthur Bowley (1924) was the rst impor-
tant English academic who supported Walrasian tenets. Henry Ludwell
Moores (1929) ideas will be discussed below.
Many of those who expounded and elaborated upon the ideas of
Walras and Pareto were Italian. Among themwas Enrico Barone (1896),
who developed fundamental aspects of the theory of marginal produc-
tivity that was begun by Walras and improved by Pareto. Barone also
contributed to the theory of public nance, and used general equilib-
riumtheory in a model of a socialist economy (1908). Maffeo Pantaleoni
should also be mentioned, not so much for contributions to general equi-
librium theory, although he certainly adhered to Walrasian tenets, but
because, as has been seen, his 1889 book on economics persuaded Pareto
that Walrass work was worthy of being studied. Another economist who
certainly beneted from Walrass legacy was Pasquale Boninsegni. He
taught at the University of Lausanne after Paretos departure, making
contributions to the theory of general equilibrium until the 1930s. He
wrote to Walras, whom he called cher Matre: I know you through
your contributions, of which I am a modest popularizer, and through
the devotion of your former direct students (239, 1908, 3, pp. 364,
36869).
A complication in the history of Walrasian ideas is that Pareto was
thought by a number of economists to have founded a school of his own
(Schumpeter 1954, pp. 829, 855, 858; Walker 1997a). Theywere inspired
directly by him, so to them the school of Lausanne meant the Paretian
school rather than the Walras-Pareto school. Scholars of that persua-
sion included the civil engineer Giovanni Battista Antonelli; Antonio
Osorio, whose book (1913) was introduced by Pareto; Luigi Amoroso
(1909, 1921); Alfonso de Pietri-Tonelli (1924, 1927); Umberto Ricci
(1924); and economists of the generation after Pareto, such as Con-
stantino Bresciani-Turroni, Gustavo Del Vecchio, Luigi Einaudi, and
Marco Fanno.
P1: PIG
280 Walrass inuence
Arthur W. Marget observed that extravagant praise has been
bestowed on the monetary theory of Pareto by members of the Lausanne
school who have had no such words of praise for the monetary theory
of Walrass (Marget 1935, p. 152), in no small measure because Pareto
wrote as though no such theory existed. Guido Sensini (1929, p. 215),
for example, stated that everything that is true regarding the theory of
money is found in Paretos work. These are strange notions, considering
that Pareto can hardly be said to have had a monetary theory and did not
even regard the question of money as being on a theoretical level (Bridel
1997, pp. 15458), whereas Walras had brilliant insights on the role of
money. Among the French, Roberto A. Murray (1920) ignored some of
Walrass main contributions to economic theory and praised Paretos,
and Fran cois Divisia (1928) believed that Pareto was the formulator of
the true theory of price determination. One of Paretos obituarists, his
former student Pierre Boven, wrote that:
In economic theory, he has not only continued L eon Walrass work; he has pro-
foundly modied the theories of his predecessor, and has greatly advanced the
School of Lausanne, not only by new discoveries, but also and above all by the
direction he gave to that school, by extracting it little by little fromextra-scientic
preoccupations (Boven 1923).
5
The viewthat Pareto was not greatly indebted to Walras has its modern
exponents. Roberto Marchionatti and Enrico Gambino (1997, p. 1322),
for example, contend that Joseph Schumpeters widely accepted
judgment that Paretos work is completely rooted in Walrass system
constitutes a misreading of Pareto. In fact, already during the period
18921900, Pareto traces the methodological outlines of an economic
science profoundly different from that of Walras (see also Bruni 2002).
Nevertheless, the underappreciation, depreciation, or disregard of
Walrass writings by those scholars does not obviate the facts that in fol-
lowing Pareto they were beneting fromWalrass legacy, that in praising
Pareto they were praising the Walrasian heritage upon which he admit-
tedly based his economics, and that their research solidied the position
of Walrass work as the basis of Continental neoclassical economics.
6
5
While Bousquet agreed with that opinion in a general sense, he confessed that he
found certain shortcomings in Paretos work (Bousquet 1971).
6
Here is an interesting assessment: The stock of Marshall declined on the market
of reputations relative to those of L eon Walras, Stanley Jevons, Knut Wicksell, and
Francis Edgeworth. After being underestimated as an eclectic, John Stuart Mills
terms of trade steadily appreciated vis-` a-vis David Ricardos. After Vilfredo Pareto
succeeded to Walrass Chair at Lausanne, Paretos reputation at rst tended to eclipse
that of his master; but Schumpeters championship of the unique and stellar worth of
the system of general equilibrium helped engineer a Kondratieff wave that brought
Walras to the top of the microeconomics pole (Samuelson 1998, p. 329).
P1: PIG
B. Johan Gustav Knut Wicksell
Walrass legacy
A major economist of the second generation that developed Walrass
theory of general equilibriumwas Knut Wicksell, a Swede who was born
in1851. Like Pareto, he didnot beginhis contributions toeconomics until
middle age. Like the young Pareto, he was deeply committed to liberal
causes, writing and lecturing with a reformist attitude about topics like
prostitution, birth control, and prison reform. His interest in the latter was
stimulated by his imprisonment, at the age of 57, for being excessively
outspoken against organized religion. It was concern over such issues
that began his interest in economics, and led to his seeking and obtaining
a position at the University of Lund in 1900. He retired in 1916 and died
in 1926.
Wicksell gladly admitted his indebtedness to Walras. On the occasion
of sendinghima copyof one of his books (1893), he beganbymentioning
that he had a rather complete collection of Walrass writings, and that
their used covers showed how intensively he had studied them. He had
tried in his book to emphasize on all the topics your manner of reasoning
and to set forth the price equations, much superior to those of Jevons
(239, 1893, 2, p. 596). He had, he continued, defendedWalrass doctrine
against the objections of critics, one of whom had accused Walras of an
error that he had not committed, and another of whom had evidently
only supercially read his work. He nished by writing:
I have not been able to follow you completely on the subject of production, of
capitalization, and of credit, but that does not prevent the fact that, fundamentally,
it is always your general method of treating the problemin question that I employ,
with the sole modication of having introduced the newconception of capital and
interest. . . .
My debt of gratitude to you is therefore very great, and I hope that I have not
failed to mention it in the course of my book (ibid.).
7
Contributions
Wicksell was concerned with the real economy and hence with non-
virtual behavior, just as Walras had been in developing his mature
7
Schumpeter justly observed of Wicksell that no ner intellect and no higher character
have ever graced our eld. If the depth and originality of his thought do not stand out
more clearly than they do, this is only owing to his lovable modesty, which lead himto
present novelty semi-hesitatingly, as little suggestions for the improvement of existing
pieces of apparatus, and to his admirable honesty, which pointed incessantly to his
predecessors, Walras, Menger and B ohm-Bawerk (Schumpeter 1954, p. 862).
P1: PIG
282 Walrass inuence
comprehensive model. Most of Wicksells expositions consist of verbal
accounts of economic behavior and verbal reasoning to analyze it and
deduce its consequences, but he also, as has just been intimated, used
Walrass mathematical method of symbolizing some of the features that
he identied, representing their interrelationships, and discovering some
of their consequences. Wicksells greatest contribution to general equi-
librium theory is his analysis of many of the features of the economy
that result from the use of money (Wicksell 1898).
8
Disagreeing with
the quantity theorists idea of the neutrality of money, he described how
it affects the real parts of the system the processes of production, con-
sumption, saving, and investment. Drawing upon Walrass intellectual
legacy (123, 1877, pp. 278311; 176, 1889, pp. 261312; 23171,
pp. 345436), Wicksell followed Walras in constructing a model of the
determination of the rate of net income from the use of capital goods,
and of the determination of the market and equilibrium rates of inter-
est (Wicksell 1898/1936). Exploring the interrelationships between the
quantity of money, the interest rate, and the level of prices, he intro-
duced his well-known distinction between the market and the natural
rates of interest, the origins of which can be found in Walrass writ-
ings. According to Wicksell, This natural rate is roughly the same
thing as the real interest of actual business. A more accurate, though
rather abstract, criterion is obtained by thinking of it as the rate which
would be determined by supply and demand if real capital were lent
in kind without the intervention of money (ibid., p. xxv). The market
rate of interest, on the other hand, is the rate of interest charged by
banks to lend money. In equilibrium, he explained, the natural rate of
interest equates the supply and demand for savings. Wicksells distinc-
tion enabled him to see that it is not a high or low rate of interest in
the absolute sense which must be regarded as inuencing the demand
for raw materials, labor, and land or other productive resources, and so
indirectly as determining the movement of prices. The causative factor
is the current rate of interest on loans as compared with . . . the natural
rate of interest on capital (ibid., pp. xxivxxv). It was, in fact, Walras
who had explored some of these matters in his mature comprehensive
model.
8
An opinion shared by many economists of Wicksells Interest and Prices is that few
books can have contained such a wealth of fundamental ideas calling for further eluci-
dation and development, and though probably even now there would be no complete
or even close agreement as to which were the best of Wicksells ideas, or the line of
development to which they pointed, it is safe to write down Interest and Prices as
one of the two or three outstanding theoretical works of our period (Hutchison 1953,
p. 245).
P1: PIG
Examination of the relation between those two rates also enabled
Wicksell to identify some important aspects of changes in the price
level and business uctuations. For example, if the natural rate is above
the market rate of interest, the returns to physical capital exceed the cost
of borrowing, and an expansion of investment occurs, nanced through
an expansion of bank credit. Prices rise, reducing the purchasing power
of consumers. Because they are forced to buy less, the amount of eco-
nomic resources used to make consumer goods is reduced, thus making
available a greater amount of economic resources for the production of
capital goods. The price level will be stable if the market rate equals the
natural rate of interest.
Wicksells contribution to the theory of capital drew upon Walrass
theoryof capital, but supplementedandextendedit byintroducingEugen
von B ohm-Bawerks ideas about time and capital. Wicksell analyzed
how the structure of capital goods depends not only on the number of
units of investment, but also upon the length of time during which the
inputs are invested. He asserted that the importance of the time-element
in production was never properly appreciated by Walras and his school.
The idea of a period of production or of capital-investment does not, as
we have said, exist in the Walras-Pareto theory; in it capital and interest
rank equally with land and rent; in other words, it remains a theory of
production under essentially non-capitalistic conditions, even though the
existence of durable, but apparently indestructible instruments, is taken
into account (Wicksell 1901/1934, 1, p. 171). Wicksells analysis made
it possible for him to explore such concepts as the investment of capital
for longer periods of time as distinct from larger amounts of investment,
and to examine the effect of technological change on the distribution of
income.
Wicksell combined the marginal productivity theory of distribution
with Walrass marginal utility theory of value. He showed that in a state
of competitive equilibrium, each unit of each type of economic resource
employed in a rm would be paid the value of the amount of output
that a unit of the resource contributes to production. He proved that, if
the rms production function is homogeneous of the rst degree, the
sum of the payments to the owners of the economic resources deter-
mined in that way would equal the total value of the output (ibid.,
pp. 12433).
Wicksell, like Walras, viewed the market economy and free com-
petition favorably, but believed that social welfare and the desire to
maximize prots are not necessarily compatible. He therefore disagreed
with Walrass and Paretos contention that a competitive economy gen-
erates a maximum of utility in a society. That proposition, he declared,
P1: PIG
284 Walrass inuence
was the Lausanne school version of harmony economics. A strong argu-
ment can be made that Wicksell and that school disagreed because they
held different views of the possibility of making interpersonal com-
parisons of utility and had different social and political philosophies
(Syll 1993). It should be added, however, that Wicksell did not suf-
ciently recognize the qualications (noted in Potier 1998) that Walras
made to the doctrine that pure competition results in maximum satisfac-
tion (see Chapter 5).
C. Joseph A. Schumpeter
Schumpeter made substantial contributions to the analysis of disequilib-
rium exchange and disequilibrium production. Indeed, his life work can
be seen to be an investigation of non-virtual economic disequilibrium: of
what causes it, of what happens during its phases, of howit follows what
he believed to be cycles. His early theorizing on this matter (1911/1926)
manifested the inuence of Walrass theory of the entrepreneur, and
toward the end of his life he emphasized that Walrass idea that the
entrepreneur earns prots only in disequilibrium is essential for clear
thinking on prots (Schumpeter 1954, p. 893). Schumpeter identied
the innovating entrepreneur. That economic agent introduces new
commodities and business methods, including production processes,
capital goods that incorporate new technology, and the exploitation
of newly discovered natural resources. Businesses that use outmoded
methods and technologies and the products associated with them are
displaced by the enterprises that the innovators establish new com-
binations mean the competitive elimination of the old (Schumpeter
1911/1926, p. 67). The process of innovation therefore creates disequi-
libriumand leads either to monopoly or to an adjustment process involv-
ing prices, disequilibrium transactions, and disequilibrium production
and consumption.
To explain the latter, Schumpeter adopted the adjustment mechanism
that Walras had identied whereby entrepreneurs channel resources into
the production of protable commodities. That enabled Schumpeter to
explain the movement toward a competitive equilibrium that ensues
after a secondary wave of investment by imitative entrepreneurs ends
the monopoly created by the innovating entrepreneur. If that invest-
ment is not prevented by laws, regulations, and patents that maintain the
monopoly, or by the economic power of the monopolist, as new rms
are created the output of the industry will increase until prot becomes
zero, thus eliminating the incentive for further investment. Thus, through
the work of Schumpeter, Walrass model of the entrepreneur became a
fundamental strand of Continental thought on that subject.
P1: PIG
D. Henry Moore and Henry Schultz
Moore was another important economist of the generation immediately
following Walrass who beneted fromWalrass legacy. He closely stud-
ied and built upon Walrass general equilibrium model, even insisting
upon using Walrass nomenclature and notation (Moore 1929). Moore
was convinced that Walras had correctly delineated the interconnect-
edness of a market economy and the method necessary for studying
that feature. Moore afrmed that the central problem to the solution
of which Walras gave his life was the interdependence of all economic
quantities . . . (Moore 1929, p. 2). Moore also asserted that Walrass
legacy in that regard was seminal, that his solution of the problem
of the mathematical conditions of equilibrium in a static state inspired
an ideal conception of the goal toward which future investigators must
work (ibid.). A comprehensive treatment of economic questions in a
changing society must take cognizance of the interdependence of all
types of economic change, and the only kind of treatment that will lead
to rational forecasting and control is mathematical in character (ibid.).
Regarding economic variables, it was Walras, he wrote, who had seen
the necessity of expressing in simultaneous mathematical equations the
conditions of their common determination. To facilitate his inquiry he
created a hypothetical static state having the properties that are familiar
to all students of economic theory (ibid.).
Moore tried to develop a dynamized version of Walrass model and to
test it empirically. He believed that the assumption of pure competition
and the static features of the model, which resulted in an hypothetical
static state in a stable equilibrium, should be replaced (ibid., p. 111). In
this connection, it should be recognized that one type of dynamic system
is a model that moves from a disequilibrium state to a static equilibrium
in which aggregate output and its composition are constant. That would
be a model without capital accumulation in which the dynamics are a
tatonnement process such as the one Walras had in mind. Another type
of dynamic motion is that of a model that not only has an equilibrating
process like the tatonnement but also has a positively (or conceivably
negatively) sloped path of change of aggregate output over time. In that
case the tatonnement leads the economy to that path and the economy
then expands (or contracts) along it. Of course, there are a variety of
other possibilities, such as uctuations around the path, displacements
of the path, and so forth.
Moore wanted, rst of all, to develop a growth-path model of the
type just described. He argued that Walrass model cannot be made
to resemble a real, moving equilibrium because the approach to reality
would destroy the network of implications by means of which it exists
P1: PIG
286 Walrass inuence
(ibid., p. 110). Publishing his work in 1929, he justly claimed priority for
this undertaking: No mathematical economist, as far as I amaware, has
ever attempted to pass from[my] or any similar presentation of a statical,
hypothetical equilibrium to a realistic treatment of an actual, moving
general equilibrium (ibid., p. 106), applicable tothe real economy(ibid.,
pp. 11011).
Moores way of trying to accomplish that task was to construct a
system of dynamic general equilibrium equations that determine the
equilibrium trend values of the variables, and to use statistically derived
functions. General equilibrium theorists should take partial elastici-
ties of demand, partial elasticities of supply, and partial relative efcien-
cies of organization, andweave these concrete functions intoa network
of continuous relations describing the solidarity of exchange, produc-
tion, capitalization, and distribution as a moving general equilibrium
(ibid., p. 92). The next part of Moores plan was to use his modied
version of Walrass model to study business cycles. The economy, he
contended, follows an oscillatory path about a moving position of nor-
mal equilibrium, and an explanation of that feature requires an inves-
tigation of the way that recurrent and non-recurrent exogenous shocks
are translated into economic oscillations (ibid., pp. 14674). Moore also
undertook empirical tests of general equilibrium models, and he tried to
add to general equilibrium theory by introducing imperfectly competi-
tive market structures into a model of a dynamically growing economy.
Although he did not nish any of the tasks he had set for himself, Moore
was able to add constructively to the general equilibration and equilib-
riumapproach employed by Walras in his mature comprehensive model,
and he had a clear vision of some of the directions that economic theory
and statistical implementation needed to take.
As has beenmentioned, HenrySchultz (18931938) was alsoa careful
and enthusiastic student of Walrass theories and, through his publica-
tions and teaching, was important in the transmission of Walrass legacy
to other economists. His writings express his interest in the theory and
methods of Walrass mature comprehensive model, and have nothing to
do with virtual general equilibrium modeling. He also developed math-
ematical economic methods and applied them in econometric studies
(Schultz 1938), furthering the type of empirical statistical studies advo-
cated by Moore, who was his teacher at Columbia University (Mosak
1987, pp. 26162). Schultz stimulated William Jaff e to read Walrass
El ements, and inuenced his decision to translate that work (Jaff e in
235, 1954, p. 8; Walker 1983, p. 29) and thereby to make Walrass ideas
accessible to Anglophone economists. Schultz wrote two articles (1929,
P1: PIG
1932) on marginal productivity in relation to general equilibrium, and
two articles on the interrelations of demand, price, and income (1933,
1935). As it happened, the type of work that Moore and Schultz hoped
would be done was for many years largely displaced by concern with
other issues relating to virtual general equilibrium models, as will be
seen in Chapter 10.
P1: JZZ
CHAPTER 10
Models drawing upon the heritage of the
written pledges sketch, 1930 to 1971
I. Introduction
The previous chapter has explored the rst phase of the realization of
Walrass prediction that subsequent economists would draw upon his
mature comprehensive model, the second phase being considered in
Chapter 11. The present chapter examines the aspects of his written
pledges sketch accepted by general equilibrium theorists after 1929,
and that, in accordance with their interpretations of that sketch, formed
the starting point of their own contributions and inuenced their ongoing
research.
There is a need for the clarication undertaken in this chapter of
aspects of the history of general equilibrium theorizing, as a brief pre-
view of some of those aspects demonstrates. Mark Blaug remarked that,
It was (Henry) Schultz, Hicks, Hotelling, Lange and Samuelson who
were responsible in the golden decade of the 1930s in bringing about this
revival of GE theory (Blaug 1997, p. 253; and see Arrow 1968, p. 380).
That statement should be qualied. In the rst place, it does not take
account of most of the major developments in general equilibrium theo-
rizing in the 1930s, namely of those that took place outside of the United
States and England, a matter that will be discussed shortly. Second, a
number of qualications can be made about the precise contributions and
dates of the work of the scholars Blaug mentioned. It was in the 1920s
that Henry Schultz sustained the vitality of the Walrasian non-virtual
tradition and was joined by others. It may be that Oskar Lange (1904
1965), in his little book On the Economic Theory of Socialism (Lange
193637)
1
taught an entire generation to appreciate the supposed practi-
cal relevance of Walrasian GEtheory (Blaug 1997, p. 253). Lange made
1
This is the publication in book form of two articles (Lange 1936, 1937). For bio-
graphical information about Lange, see Kowalik 1987. Abba Lerners work on the
same type of topic will not be considered here because it concerns the application of
general equilibrium principles to socialism and the control of economies, rather than
adding to economic knowledge on such questions as price formation, the existence of
equilibrium, and stability.
288
P1: JZZ
Models drawing upon heritage of written pledges sketch 289
his major contributions, however, in the 1940s, and they were mainly
inspired by questions posed by John Maynard Keynes. Lange did draw
on Walrass legacy, albeit indirectly, especially in one research program
in which he relied on mathematical tools of general economic equi-
librium as developed and modied by Henry Schultz, R. G. D. Allen
and Paul Samuelson, but especially by J. R. Hicks (Kowalik 1987,
p. 124). Harold Hotelling (18951973), a mathematical statistician, did
not contribute anything to the analysis of purely competitive general
equilibrium. It is true that one of his very few papers on economics bore
the words stability and competition in the title (Hotelling 1929).
Nevertheless, that paper was concerned, not with general equilibrium
nor even with purely competitive markets, but with insights given by
game theory to the understanding of spatial competition in oligopolis-
tic market structures. Paul Samuelson (1915), who had just begun his
career toward the end of the 1930s, wrote nothing on the subject of gen-
eral equilibrium theory during that decade. His rst analysis of stability,
made in 1941 (see below), was noted by J. R. Hicks (1946, pp. 33637),
but Hicks did not do more than make the brief suggestion that it would
open a most promising line of investigation (ibid., p. 336). In spite
of its importance, Samuelsons work remained isolated and gave rise to
no signicant developments (Ingrao and Israel 1990, p. 335) until 1958
(Arrow and Hurwicz 1958).
Claude M enard, disagreeing with Blaug, states that a revival of inter-
est in Walrasian general equilibrium did not start until after 1940. He
writes that there was not much more than the germ of a school of
Lausanne before that date and it was to bloom only after 1940. . . . If
there is such a thing as a Lausanne school, its expansion is mostly a con-
temporary phenomenon (M enard 1990, pp. 99, 104). M enards view is
as difcult to appreciate as is Blaugs. The historical record shows that
neither the 1930s nor the 1940s can properly be described as a period
of revival of general equilibrium theory, because there was no period
in which it was not studied. After Walras, to mention only the major
gures in the history, the theme was rst carried forward by a num-
ber of French theoreticians during Walrass life and subsequently (see
Zylberberg 1988, 1990), in numerous publications by Enrico Barone
until 1924 and Vilfredo Pareto until 1927, by many Italian followers
of Pareto during his lifetime, and by Knut Wicksell until 1926. Gustav
Cassel was productive on the topic of general equilibrium during the
period 1918 to the early 1930s. Henry Schultz wrote inuential articles
on marginal productivity and the Lausanne school in 1928. H. L. Moore
was studying general equilibrium in the two decades after 1910, pub-
lishing his major contribution in 1929, and the Ukrainian-born Jacob
P1: JZZ
290 Walrass inuence
Marschak and the Danish Frederik L. B. Zeuthen were examining Wal-
rasian systems in 1928. There was intense interest in general equilibrium
theory during the 1930s on the part of some economists in Austria, as
is well-known (see, for example, Weintraub 1983; Brems 1986) and as
this chapter fully recognizes. The Germans Hans Neisser (1932) and
Heinrich von Stackelberg (1933), and Zeuthen (1933) examined the
existence of equilibrium in Walrasian systems in additional writings
that predated and therefore obviously owed nothing to the economists
mentioned by Blaug. Thus it was always true, during and since the time
of Walras, that leading theoreticians cultivated general equilibrium the-
ory with unabated enthusiasmand in an unbroken chain (see also Walker
1990, pp. 13738; 1997a).
To limit the scope of this chapter in a reasonable way, only the general
equilibrium theorists who made signicant original contributions will
be considered. The discussion does not deal with the theorists philoso-
phies and methodological views. Nor does it present a comprehensive
history of modern general equilibrium modeling, because much of that
work has no real connection with Walrass ideas. Modern theory has
gone far beyond his constructions, incorporating concepts and devel-
oping in directions of which he had not the slightest idea. Nor would
there be in any event a useful purpose served by repeating the details
of the contents of the theorists relevant papers and books, for they are
well presented in the primary sources and explained and analyzed many
times in secondary sources. The purpose of the remainder of this book
is to point out whatever specic links to Walrass economic ideas there
are in economic models developed by his contemporaries and succes-
sors. The models are therefore classied and briey examined strictly
from the special points of view and interests of this study. A few facts
about the careers of some of the economists involved before 1936 are
given, if the information is useful to establish the importance of the per-
son, the lines of transmission, and the geographical reach of Walrass
inuence.
2
II. Karl Gustav Cassel
Background
Gustav Cassels work on general equilibrium theory is a direct descen-
dant of Walrass written pledges sketch. Cassel was born in Sweden
2
Biographical information about all the economists that are mentioned is available in
a variety of publications, notably Eatwell et al. 1987.
P1: JZZ
on October 20, 1866, attended Uppsala University and the University
of Stockholm, and was appointed professor of economics and nancial
science at the latter institution in 1904. He had a wide variety of inter-
ests. His career began with the publication of a long article in 1899
in which he expressed almost all the ideas about price formation that
he was to espouse in later years (Cassel 1899). Shortly thereafter, he
argued innovatively in favor of marginal cost pricing for railways and
other enterprises for which rates are regulated (Cassel 1900). In Nature
and Necessity of Interest (Cassel 1903), he repeated and extended in
some respects Walrass theory of capital. He published the rst edition
of The Theory of Social Economy, his major economic treatise, in 1918.
He rejected Eugen von B ohm-Bawerks ideas about time-preference and
capital, and pointed out that B ohm-Bawerk did not grasp the fact that
the prices of factors of production and of commodities are mutually
determined, so that it is pointless to suggest that the prices of the for-
mer are determined by those of the latter or vice versa (Cassel 1932,
p. 195). In his own theory of capital, Cassel resurrected Nassau Seniors
notions about abstinence, to which he added his original idea of the
importance of life-cycle motivations in saving behavior (Cassel 1903,
1932).
Cassel became an authority on international monetary affairs. He was
an important analyst of the German reparations question after World
War I. He advised the League of Nations and many governments on
monetary matters, expressing his ideas in, notably, The Worlds Mone-
tary Problems (Cassel 1921). In that work, on the basis of his reasoning
that the ratio of the rates of exchange of the currencies of two countries
should be the same as the ratio of their general price levels, he devel-
oped an inuential purchasing-power parity theory of exchange rates.
He adhered to the quantity theory of money and to an overconsumption
theory of business uctuations (Cassel 1918, 1932). Unlike those of his
contemporaries who wanted the gold standard to be an automatic regula-
tor of the money supply and prices, Cassel contended that the monetary
authority could and should regulate the bank rate of interest, thus obtain-
ing the benecial consequence that the conditions for the development
of trade cycles would be radically altered, and that indeed our familiar
trade cycles would be a thing of the past (Cassel 1928, p. 529). Highly
critical (Cassel 1937) of J. M. Keyness General Theory of Employ-
ment, Interest and Money (1936), he opposed efforts to stimulate the
economy by means that would directly increase income, such as public
works. Among his achievements was the book On Quantitative Thinking
in Economics (Cassel 1935), which, together with his many empirical
studies and keen insight into statistical economic questions, established
P1: JZZ
292 Walrass inuence
his position as an exceptionally capable early econometrician. He died
on January 15, 1945.
Approach and theoretical foundations
In his rst important essay (1899), Cassel had the sole purpose of giving a
clear exposition of Walrass ideas; he consistently used Walrass general
equilibrium approach as a basis for his work during his entire career;
and he carried a number of the specic constructions of Walras and
Pareto forward in the stream of economic studies. Nevertheless, he did
not acknowledge the extent of his debts to those economists. One of
Walrass correspondents wrote to him about that essay as follows:
With respect to Mr. Cassel, I would say . . . he wishes to contribute something new
and independent in the domain of [pure economic theory] and that he is unable
to. Basically he accepts, with some paraphrases, all of your theory of exchange,
but to differentiate his work, however, he rejects average marginal utility which
is in a manner common to several persons, but retains subjective marginal utility,
measuring it by the price of the commodities. . . . He obtains the equations of
exchange which basically result from your fundamental equations. With respect
to the theory of production, he accepts all of your theory almost without alteration,
but without admitting it, proclaiming you to be his predecessor. The Gentleman
has very great pretensions, but very little original force (239, 1899, 3, pp. 9091).
Cassel did not even mention Walras or Pareto in the later editions of his
Theory of Social Economy (for example, Cassel 1932).
Cassel disagreed with a number of the components of the models that
those two economists had developed. He rejected the marginal utility
theory of value, whether based on cardinally or ordinally measurable
utility:
This purely formal [utility] theory, which in no way extends our knowledge of
actual processes, is in any case superuous for the theory of prices. . . . This deduc-
tion of the nature of demand from a single principle, in which so much childish
pleasure has been taken, was only made possible by articial constructions and a
considerable distortion of reality (Cassel 1918, p. 81).
Cassel argued that because the amount of commodities consumed could
not be known until the set of equilibrium prices and quantities is deter-
mined with the use of a general equilibrium system of equations, the
marginal need that is satised and therefore the marginal utility to each
consumer could not be known until that set is determined. He concluded
that, What we call marginal utility if we now wish to introduce this
conception thus occupies exactly the same place as an unknown in
the problem as does price, and it is therefore obviously absurd to cite
marginal utility as a factor explaining price (1932, p. 147).
P1: JZZ
In place of that concept, Cassel wanted to substitute the principle of
scarcity, in response to which Knut Wicksell commented that something
is scarce
. . . only in relation to wants, or to the extent it becomes an object of demand. And
the degree of scarcity is measured in exactly the same way as marginal utility,
by the strength of the next unsatised need, which rst causes the commodity
to be recognized as scarce. In other words, scarcity and marginal utility are
fundamentally one and the same thing (Wicksell 1919, in 1934, p. 221).
Instead of using utility theory explicitly, Cassel assumed in his model
of general equilibrium that demand functions are primitive construc-
tions, which has led many commentators to declare that he anticipated
revealed preference theory. Paul Samuelson (1993, pp. 515, 517) has
rmly rejected that view, arguing that Cassel instead had a revealed
demand approach, which seems a reasonable assertion inasmuch as
Cassel denied the value of regarding demand functions as being expres-
sive of underlying preference functions.
Cassel did not accept the Walras-Pareto doctrine that maximum sat-
isfaction is obtained by a perfectly competitive economy, arguing that
large-scale enterprises are much more efcient than small-scale ones,
but are absolutely incompatible with free competition (Cassel 1932,
p. 129). Moreover, competition generates monopoly; it brings into
being its own antithesis, so to take free competition as the starting-
point for a general theory of prices is of very little use (ibid., p. 129).
That was an implicit criticism of the procedure adopted by Walras and
Pareto.
Cassel thought that in most production processes the factors of pro-
duction are indivisible, or that it is very often impossible to use less of
one factor without throwing the others out of work (ibid., p. 179). In his
equation system, he consequently adopted the assumption of techno-
logically xed coefcients of production, as Walras had done in much
of his work. It has been seen, however, that Walras also treated the
case of variable coefcients, which is a necessary assumption for the
Walras-Pareto theory of marginal productivity. Perhaps in order to avoid
completely rejecting that theory, Cassel suggested in his literary account
that substitution among factors is possible (ibid., pp. 17981).
Cassels three models
Cassels general equilibrium models were inferior to those of Walras
and Pareto, for reasons that will be made clear, but he nevertheless made
an original contribution to the subject. He presented three models based
P1: JZZ
294 Walrass inuence
upon Walrass general equilibrium equations and his virtual approach
that consequently have become known as Walras-Cassel models. The
use of that name is understandable, but unless it is interpreted with a full
recognition of the facts of the matter, it is, strictly speaking, mislead-
ing. Walrass major contribution, his mature comprehensive model, is
incompatible with Cassels models, and the construction of Walrass to
which the name refers is just an incomplete sketch, not a proper model.
Cassel rst constructed a virtual model of pure exchange with xed
available amounts of the commodities (ibid., pp. 13840). He dened a
demand function aggregated over all demanders for each commodity, the
quantity demanded being a function of all prices. For each commodity,
he set the demand function equal to the xed supply and declared that
there is an equilibrium set of prices because there are as many equations
as unknowns. Inasmuch as the money expenditures of the consumers are
given, absolute prices are determined.
In the second model, which he called a model of the stationary state
because the quantities of the commodities produced are constant, Cassel
assumed once more that the amounts of money to be spent by consumers
are given. The quantities of the factors of production are given constants,
their supply functions are perfectly price-inelastic, and the factors are
always fully employed. In accordance with Walrass assumption, the
technical coefcients are given. With great clarity and simplicity, Cassel
then constructed a system in which the demand and supply for each
commodity produced and for each factor of production are equal, and
in which the output and the input sides of markets are linked. He then
dropped the assumption that the incomes and expenditures of the con-
sumers are given, and introduced equations that result from his identi-
cation of them as the owners of the factors of production. Their prices
and quantities, and hence the incomes of their owners, are determined
as part of the general equilibrium of an expanded system.
In the third model, Cassel presented a dynamized virtual Walrasian
general equilibrium system.
3
He was concerned not with a position of
equilibrium but with a path of growth. His interest in the theory of
3
Claude M enard(1990), evidentlyreferringtoWalrass static equationsystem, contends
that the structure of his model makes its dynamization inherently impossible. The
discussion of this matter by M enard and others, like that of so many issues, is difcult
to followbecause the participants mix together Walrass equations and his models, and
mix together the mature comprehensive model and their interpretation and elaboration
of the written pledges sketch. In fact, M enards opinion is unconvincing because Cassel
did what M enard states is impossible and because John von Neumann and other
theoreticians similarly constructed models that dynamized Walrasian-type general
equilibrium equation systems.
P1: JZZ
economic growth was probably inspired by Walrass investigation of
some of the properties of a growing economy (123, 1877, lessons 51
52; 176, 1889, lesson 28; 249.VIII, lessons 3536). Cassels reasoning
and his exposition of the model were verbal. Continuing to use Walrass
general equilibrium approach, and continuing to build on a restatement
of Walrass equations, he discussed the modications that were neces-
sary to the equations of his second model and briey outlined a model
that is always in equilibrium. He called it a model of the uniformly pro-
gressing state. He assumed that the amounts of the reproducible factors
of production increase at a xed rate (Cassel 1918; 1932, pp. 15255),
amounts that Walras had treated as parameters, and verbally deduced
the consequences. As in Cassels second model, the factors are always
fully employed and the technical coefcients of production are xed.
Cassel showed that the economy expands uniformly at a xed rate. The
prices of the factors and of consumer commodities remain unchanged
as the economy grows. The production of each commodity increases at
that xed rate, as do money incomes, demands, supplies, savings, invest-
ment, and consumption (ibid., p. 153). The model thus introduced the
concept of steady-state growth. Anticipating the Harrod-Domar model,
it is an original early formulation of a multiplier-accelerator process.
Character and limitations of the models
A serious limitation of Cassels general equilibrium models is that they
are devoid of behavioral content, and are therefore devoid of features,
plausible or otherwise, to which his equations could have reference.
Walrass and Paretos models are vastly richer. They took pains to draw
from the real economy the institutions, procedures, technology, rules,
and pricing processes in their models, and only then tried to describe
many of those characteristics or their outcomes with equations. Wal-
ras, for example, used 116 pages for his theory of exchange alone. In
contrast, Cassel used only 16 pages to present all three of his models.
He made no mention of the characteristics of markets or the behav-
ior of suppliers and demanders, offering no explanation of how prices
are formed. His models are even more highly idealized systems than
Walrass written pledges sketch. That was a sketch of a virtual model,
but Walras described it as having a disequilibrium state and, indeed,
he introduced the device of written pledges with the intention that the
model would be virtual but would also have a process of adjustments in
disequilibrium that would lead it to general equilibrium. Cassel, on the
other hand, constructed models that he simply assumed are always in
equilibrium; they have no tatonnement process. That greatly simplied
P1: JZZ
296 Walrass inuence
them, and therefore Cassels theoretical task, because he had no reason
to examine the questions of the existence of equilibrium or of stability
or whether or not there are multiple solution sets of the variables. It
also means, however, that he did not provide any explanation of why
his models should be considered theoretically interesting or empirically
applicable.
Nevertheless, Cassels models were very inuential. He widely dis-
seminated knowledge of a type of formalistic general equilibrium sys-
tem. One reason for his success was that his exposition was readily
comprehensible, a feature that it owed to its simplistic character. Another
reason was the languages in which his book appeared. He originally pub-
lished it in German (Cassel 1918), which was the language spoken and
written by the important Continental general equilibriumtheorists in the
1930s. Through the English versions of his books, he was instrumen-
tal in communicating some of Walrass economic ideas to Anglophone
economists. The work of Walras, in French, and of Pareto, principally in
Italian and French, had a relatively limited readership, whereas Cassels
text, in German or English (Cassel 1932), was used in many European
universities and was considered to be the denitive statement of general
equilibrium theory. It was the standard theoretical textbook for students
and for the mathematicians and economists who were members of the
Vienna Colloquium in the 1930s (Weintraub 1983, pp. 45, 7). It thus
became the starting point of the investigations of the existence of equi-
librium undertaken by Karl Schlesinger and Abraham Wald. It was a
stimulus to the work of John von Neumann on general equilibrium, and,
although of far less importance than the ideas of Walras and Pareto to
J. R. Hicks, it probably had a role in the development of Hicks views
on static and dynamic models (Hicks 1965, pp. 13, 39).
4
III. Virtual Walrasian modeling
Modeling in the 1930s
During the 1930s, in addition to Cassel, the economists who were most
important in the study of virtual Walrasian general equilibrium systems
were Karl Schlesinger (18891939), Hans Philipp Neisser (18951975),
AbrahamWald(19021950), JohnvonNeumann(19031957), andJohn
R. Hicks (19041989). Paying no attention to the rst, second, or third
editions of the El ements and without examining the fourth (or fth)
4
Unlike his monoglot Anglophone contemporaries, Hicks had the advantage of being
able to read Italian, French, and German.
P1: JZZ
edition with sufcient attention to discern its problems, those economists
followed Walrass 1899 suggestion that a virtual general equilibrium
model be devised and analyzed. They interpreted that to mean a sug-
gestion to construct an equation system alongside terminology indi-
cating that it relates to a virtual purely competitive model. To those
economists, that (together with some of Walrass particular assump-
tions) was Walrass legacy regarding the study of general equilibrium.
They were concerned with the existence, uniqueness, and, to a lesser
extent, stability of equilibrium matters they conceived of as being
properties of their equation systems.
Neissers contribution to the question of interest here was a trenchant
critique of Cassels models (Neisser 1932), a critique that undoubtedly
stimulated the Viennese circle of economists to reexamine them. One
member of that circle was Schlesinger, a Hungarian who had settled
in Vienna. In the late 1920s and in the 1930s, that city was a center
of intellectual developments in many elds, a stimulating environment
for the development of scientic thought that attracted scholars from all
over Europe. Schlesinger wrote a dissertation (Schlesinger 1914) that
extended Walrass theory of money in a general equilibrium setting,
and later took up the theory of general equilibrium proper. He has a
place in this history primarily because of his inuence on Wald and
because he publicized an idea of Zeuthens (18881959). The latter
considered problems rst detected by William Lexis in 1881, namely
that Walras did not introduce conditions into his equation system that
would ensure that their solutions are real and non-negative (Lexis 1881;
see Jaffe 1971, p. 105, n. 58). Zeuthen realized that the total amount of
a commodity used in all economic processes must be smaller than or
equal to the total amount produced. If there is a positive excess supply,
the price of the commodity must be zero (Zeuthen 1928, p. 27; Zeuthen
1933, pp. 23; Brems 1986, p. 250; Brems 1987, 4, p. 944). Apparently
independently conceiving of those conditions, but crediting Zeuthen for
having done so, Schlesinger noted that Walras and Cassel had dealt
only with scarce economic resources and therefore assumed that all
commodity prices are positive (Schlesinger 19331934). He recognized,
in contrast, that there are free inputs, like air, sunshine, and sometimes
water, and accordingly modied the equations that Cassel had based on
Walrass work. Schlesinger did that by recognizing zero prices and by
introducing complementary slack conditions via the use of inequalities.
His procedure made it clear that it is impossible to say that equilibrium
exists in a model simply because there are as many equations as there are
unknowns. In reality, prices are positive or zero, none of a commodity
may be produced in a particular period, and quantities produced are
P1: JZZ
298 Walrass inuence
obviously positive, so equations with any pretense to realismmust reect
those conditions by having solutions that are equal to or greater than
zero. Recognizing that necessity did not, however, constitute a proof of
existence.
The research of Zeuthen and Schlesinger drew attention to the fact
that no one had rigorously established whether or not an equilibriumfor a
perfectly competitive virtual economy exists. If not, a virtual competitive
model would not behave in the way that general equilibrium theorists,
such as Walras in his last phase and Cassel, had asserted. It would not
be an equilibrating system. Its variables suggested prices and desired
demands and supplies would not tend, in disequilibrium, to move in the
direction of equilibriumvalues, because such values would not exist. The
entire system of equations would have to be rejected. In the subsequent
study of this matter, the techniques used for examining existence, and,
to some extent, stability became very sophisticated, and the models
actually equation systems became not only far removed from the
behavior of any past or present real economy. but also unattached to
an underlying hypothetical structure of institutions and behavior.
Abraham Wald (19021950) was a Rumanian mathematician and
statistician. He visited the Mathematical Institute of the University of
Vienna in 1927, where he met Karl Menger, who was professor of math-
ematics at the University and a circumstance that contributed greatly
to his interest in economics the son of the Austrian economist Carl
Menger. Karl Menger referred Wald to Schlesinger, who persuaded Wald
to consider the question of existence. The model that Wald chose and
modied to do that was Walrasian, in the form presented by Cassel
and altered by Zeuthen and Schlesinger (Wald 19341935, 1936; see
Menger 1973, p. 51). Like those economists, therefore, Wald followed
the tradition of Walrass 1899 sketch by constructing a virtual model.
Walras, however, had perceived very clearly the necessity of postulating
some hypothetical market characteristics the written pledges proce-
dure that he hoped would justify his virtual equation system. Wald,
in contrast, simply made the postulate that transactions, production, and
consumption occur only at a set of prices at which the entire system is in
equilibrium. He made no attempt to explain by means of a hypothetical
market model how that could possibly happen. Consequently, he did not
give his readers any reason to believe that his equation systemwas linked
to anything in the sense of having referents, either real or imaginary.
Wald adopted Walrass assumption that markets are purely compet-
itive, and that utilities are regarded by consumers as independent of
each other. He departed from Walrass procedures by assuming that
P1: JZZ
prices are functions of quantities, rather than the other way around,
5
and that individual and market desired consumption, which he did not
clearly distinguish, do not depend on the distribution of income. He also
made the assumption that the quantities demanded of all commodities
are positive, not recognizing that some bundles of commodities are not
demanded at any positive price (Dorfman, Samuelson, and Solow 1958,
p. 367). Wald was able to provide a proof that a set of solutions that
satised his criteria for economic acceptability exists for the equation
systemthat he constructed. The result of those special restrictions on the
equations, however, naturally had the consequence that his proof of the
existence of general equilibrium applies to a highly special case. It was,
moreover, an unrealistic case, precisely because Wald was not dealing
with a conception of the real economy but responding to the exigencies
of building a virtual equation system, just as Walras had tried to do in
1899. He did not actually prove existence in an economic model because
he did not construct one that had behavior in it that would result in the
properties that he postulated for his equations.
The issue of uniqueness of the solutions to a system of general equi-
librium equations relating to exchange had been touched on by Walras
(Chapter 5), and Wald followed his example by studying that question
in reference to his own system of equations. He assumed that either all
commodities are gross substitutes, or that the weak axiom of revealed
preference with reference to market demand functions is true. The rst
of these assumptions is that, for any two commodities, a rise in the price
of one of them increases the quantity demanded of the other one; the
second is that if at a rst set of prices a bundle of commodities x
2
costs
no more than the bundle x
1
associated with the rst set of prices, and
the consumers as a whole choose x
1
, then it has been revealed to be
superior to x
2
. Therefore at a second set of prices at which x
2
is actually
bought, that bundle must cost less than the rst bundle because other-
wise the preferences revealed by the purchase of the rst bundle would
be contradicted (see Dorfman, Samuelson, and Solow 1958, p. 368).
Wald determined that, with either of those assumptions, the system has
only one set of prices at which supply and demand quantities would be
simultaneously equal in all markets.
Thus Wald initiated the era of proofs of existence and uniqueness of
solutions in virtual general equilibrium theory, and if the scope of his
work was limited by his extremely restrictive assumptions, it was nev-
ertheless an ingenious achievement and it drew attention to the question
5
He assumed a strictly pseudomonotone inverse demand function.
P1: JZZ
300 Walrass inuence
of those proofs. Karl Menger accurately predicted the kind of research
agenda that general equilibrium mathematicians would be inuenced to
follow:
I wish to remark in conclusion that with Walds work we bring to a close the
period in which economists simply formulated equations, without concern for the
existence or uniqueness of their solutions, or at best made sure that the number
of equations and unknowns be equal (something that is neither necessary nor
sufcient for solvabilityanduniqueness). Inthe future as the economists formulate
equations and concern themselves with their solution (as the physicists have long
done) they will have to deal explicitly with the deep mathematical questions of
existence and uniqueness (Menger in Baumol and Goldfeld 1968, p. 288).
It has been pointed out elsewhere that the deep questions of general equi-
librium theorizing are not those that Menger identied and that Wald
explored, but are instead the construction of models that are genuine
functioning systems that is, with institutions, rules, technology, par-
ticipants, and market pricing procedures (Walker 1997b, p. 142 and pas-
sim). Applying Mengers advice to incomplete virtual models incom-
plete because they are actually equation systems substantially without
economic referents has led to research that has no economic interpre-
tations.
John von Neumann (19031957) was the next person to develop a
virtual general equilibrium construction. He was a Hungarian, schooled
in Budapest and Zurich, who very early showed an exceptional apti-
tude for mathematics. In 1927, when he was twenty-four, he accepted
a position on the faculty at the University of Berlin and two years later
at the University of Hamburg. During the next three years he became
well-known for his contributions to algebra, set theory, and quantum
mechanics. In 1930 he began teaching at Princeton, where he remained
for the rest of his academic career. Probably stimulated by his visits to
Vienna, where he met Karl Menger in 1932, von Neumann presented
his general equilibriummodel at Mengers mathematical seminar in that
city in 1936 and published it in 1937.
Von Neumann wanted to go beyond the study of the stationary equi-
librium of a Walrasian system without capital accumulation, a system
in which production and consumption proceed at constant rates in equi-
librium. He therefore attempted to dynamize virtual general equilibrium
modeling, which Walras had contended was a desirable goal, which
Moore had set out to do, and which Cassel had succeeded in doing. Von
Neumanns point of departure was the Walrasian equation system as it
had been modied and extended by Cassel, Zeuthen, and Schlesinger.
Having assimilated the Walrasian legacy of 1899, he assumed pure
P1: JZZ
competition and that the economy he invented has no disequilibrium
transactions or disequilibriumproduction, and he proceeded to construct
a microeconomic model of moving equilibrium.
Von Neumanns model had a variety of aspects that made it a special
and very peculiar case, such as the assumption that there are unlimited
amounts of natural resources so that primary factors of production other
than labor are free, that every rm hires some of every input or pro-
duces some of every output, that overproduced goods are always totally
free, that entrepreneurs and capitalists do not consume anything, that
there are no consumer preferences, that there are no demand functions,
and that consumption per worker is always at the subsistence level.
Von Neumann followed the example of Cassel and Wald in neglecting
to provide structural features, technology, institutions, and market pro-
cesses that would result in a virtual system, thus solidifying that neglect
into a tradition that has lasted to this day in the construction of virtual
models. Like Cassel, he simply assumed that his model is always in
equilibrium, without explaining how that could be possible, and there-
fore did not consider the stability or uniqueness of equilibrium.
6
There
is a major difference, however, between Cassels and von Neumanns
models, namely that whereas the rate of growth of the economy is exoge-
nously determined in the former, it is endogenously determined in von
Neumanns. He assumed there is positive net investment equal to the
excess of income over the amount of consumption that is a necessary
subsistence minimum. The result is that the stock of capital goods, the
total output of the economy, and the income of its members grow con-
tinuously through time. He proved that, in his model, equilibriumexists,
and that the equilibrium rate of interest equals the equilibrium rate of
growth.
Von Neumann has been extravagantly praised for constructing his
model, but a historical perspective makes it clear that the major durable
contributions of his paper were his development in it of activity analysis,
and, for some economists, his use of the xed-point theoremtechnique of
determining the existence of equilibrium.
7
The sheer unreasonableness
of his assumptions and the absence of content in his model regarding the
functioning of markets, economic institutions, characteristics of rms,
6
See La Volpe 1936 for a general equilibrium model that is also dynamic and always
in equilibrium.
7
Tjalling Koopmans observed that the paper contains the rst explicit statement . . . of
what has been subsequently called the activity analysis model of production. . . . [It
displays] a model of competitive equilibrium. . . . [T]he paper contains the rst rigor-
ous, formal, and fully explicit model in non-aggregative capital theory . . . (Koopmans
1964, p. 356).
P1: JZZ
302 Walrass inuence
consumers, workers, and entrepreneurs, soon led to another attempt to
develop a dynamic model.
8
J. R. Hickss model
That attempt was made by John R. Hicks (1939, 1946; see Bliss 1987,
p. 641). His purpose was to extend the Continental tradition on general
equilibrium theory and to convey his ideas to those who read English.
He drew upon Walrass ideas in several major respects.
First, he recognized the crucial importance of the interrelatedness
of economic variables. It turns out, on investigation, Hicks wrote,
that most of the problems of several variables, with which economic
theory has to concern itself, are problems of the interrelation of mar-
kets. . . . What we mainly need is a technique for studying the interrela-
tions of markets (Hicks 1939, p. 2). The technique that he used was,
of course, in a general sense the one that Walras had initiated, namely
general equilibriumanalysis. Hicks acknowledged the value of Walrass
legacy and its development by his immediate successors in this way:
When looking for such a technique we are naturally impelled to turn to the works
of those writers who have specially studied such interrelations that is to say,
the economists of the Lausanne school, Walras and Pareto, to whom, I think,
Wicksell should be added. The method of General Equilibrium, which these
writers elaborated, was specially designed to exhibit the economic system as a
whole, in the form of a complex pattern of interrelations of markets. Our own
work is bound to be in that tradition, and to be a continuation of theirs (ibid.).
Second, Hicks followedWalras inassumingthat all markets are purely
competitive (ibid.). He arrived at that decision in the belief that his
only choices in the selection of market structures for his model were
either ubiquitous perfect competition or ubiquitous monopoly:
9
a gen-
eral abandonment of the assumption of perfect competition, a universal
adoptionof the assumptionof monopoly, must have verydestructive con-
sequences for economic theory . . . (ibid., p. 83). Let us, then, return
to the case of perfect competition (ibid., p. 85).
Third, like Walras, Hicks counted equations and unknowns to try to
establish the existence of equilibrium, although he wanted to do more
than that. He used
8
It has been contended (Kurz and Salvador 2004) that all the salient features of the
von Neumann model are classical in spirit, but surely the classicals assumptions had
closer connections to the economic realities of their time than von Neumanns did to
those of any time in world history.
9
See the variety of models with mixed market structures listed in the bibliography for
Chapter 11.
P1: JZZ
. . . the results of our revised theory of Subjective Value to rework the General
Equilibriumanalysis of Walras and Pareto. Most important here is the opportunity
thrown open to us to transcend the mere counting of equations and unknowns, and
to lay down general laws for the working of a price-system with many markets.
This is the main thing which needed to be done in order to free the Lausanne
theory from the reproach of sterility brought against it by Marshallians (ibid.).
Fourth, in his effort to transcend the counting procedure, an effort
that took the form of developing what he believed are the Foundations
of Dynamic Economics, Hicks noted yet again Walrass heritage. The
study of those foundations, he wrote, is concerned particularly with
that setting-out of problems which, as we saw, was the main concern of
General Equilibrium analysis in its Walrasian stage. I shall go into the
matter in much greater detail than Walras did in his sketch of a theory
of capital (ibid., p. 6).
Fifth, Hicks drew upon Walrass legacy of 1899 by constructing a
model that is so close to being virtual as to be in effect that type of
system. He used the technique devised by Walras in 1899, without men-
tioning Walras in that connection, whereby all the equilibrating activity
takes place in an initial stage, followed by equilibrium behavior dur-
ing the subsequent stage (206, 1899, p. 103; 210, 1900, p. 302; 274,
p. 447). Hicks featured disequilibrium trading as occurring in his model
on the rst day of each analytical week, and then assumed that during
the ensuing days of the week all economic activity takes place at the
equilibrium prices established on the rst day. For simplicity, this will
be discussed with reference to a market for a single commodity, but the
process occurs in all the markets of his model.
During the rst day, transactions occur at a series of disequilibrium
prices. After trade takes place at such a price, the participants have
different holdings of money and the commodity, and that results in a new
set of individual and hence of market supply and demand curves. Unless
a special assumption is made, Hicks noted, such as Alfred Marshalls
assumption that the marginal utility of money is constant, the newcurves
intersect at a price different from the solution price of the preceding
pair. The path of the unrealized solution price traced out by the series of
market curves falls or rises. When the current actual price reaches the
value at which the current pair of market curves intersect, that price is the
equilibrium value. The remaining trade on Monday takes place at that
price as does all trade during the rest of the week. The price at which the
initial pair on Monday intersect is not the same as the price at which the
last pair of curves intersect. The static equations describe only the former
because they contain the initial distribution of the assets as a parameter,
and they therefore do not furnish the equilibrium price. To discover its
P1: JZZ
304 Walrass inuence
value, it is necessary to have a dynamic model that traces the path of
the actual price and transactions, reveals how the changing holdings of
assets alter supply and demand, and determines the equilibrium price.
Walras, as has been seen, chose to try to deal with this problem by
eliminating disequilibrium transactions in his written pledges sketch.
Hicks, to his credit, was not prepared to construct a model that would
have such an extreme result, nor to adopt Marshalls assumption. He
followed a different line of reasoning and procedure to accomplish, as
he thought, the same objective. He stated, on two grounds, that the
equilibrium price in a non-virtual market is not appreciably different
from what it would be if the market were virtual.
The rst is that disequilibrium prices are, to use his word, likely to
be close to the equilibrium price. He asserted that, If any intelligence
is shown in price-xing, they will be (ibid., 129). He thus introduced
an implication about the knowledge of the market supply and demand
functions by someone or other, and an implication about the institutions
and procedures of price formation (perhaps that there is a market author-
ity that suggests and sets prices?) that are alien to the impersonal, purely
competitive markets that were the avowed subject of his model. Hicks
meant that not much trading would occur at prices far from the one
that turns out to be the equilibrium value: If very extensive transac-
tions take place at prices vastly different from equilibrium prices, the
disturbance will be serious. But I think we may reasonably suppose that
the transactions which take place at very false prices are limited in
volume (ibid.). That reasoning is mistaken regarding the normal case
in which the demand quantity is an inverse function of the price, and
the supply quantity is a direct function of it. In that case, the volume of
trade would be greater at a price close to the current solution price than
if the price were further away from that value.
The second grounds for Hickss contention is that the gains to the
buyers mean losses to the sellers, and vice versa. Thus, whenever the two
sides are at all similar in their distribution of increments of expenditure
among different goods, a shift in demand will be partially offset by
corresponding shift in supply (ibid.). That does not make sense. Trading
a commodity is not a game in which buyers or sellers gain or lose money
or commodity. Both buyers and sellers gain utility from trade. Buyers
obtain amounts of the commodity and sellers obtain money. The amounts
that buyers obtain are not losses borne by the sellers, nor is the money
they receive a loss to the buyers. Furthermore, the situation to consider
is not the distribution of increments of expenditure among different
goods. The situation to consider is what happens within each given
market to buyers demand functions as they give up money and acquire
P1: JZZ
the commodity at differing prices, and to sellers supply functions as
they acquire money and give up the commodity at different prices. On
a market day, after trade at any price, the market demand curve always
shifts to the left, and the market supply curve also always shifts to the
left, but it is not meaningful to say that the shifts partially or wholly offset
each other. The question is: How far to the left do each of the market
curves shift, and what happens to their slopes. The well-known answer
is that what happens to the individual and hence the market functions
depends on the form of the preference functions. Given the initial stocks
of money and commodity and the sequence of prices, it is ultimately
the latter functions that determine by how much the equilibrium price
in each market differs from the price at which the initial market desired
supply and demand would be equal (Walker 1971a, 1971b).
Thus, on the basis of what Hicks presented as reasoning but is really a
number of arbitrarypostulates imposeduponhis model andunfoundedin
economic behavior, he asserted that the equilibrium price in any market
in which irrevocable disequilibrium transactions occur is not different
or not appreciably different from the equilibrium price that would result
if there were no such transactions (Hicks 1939, pp. 12829).
10
In other
words, he believed there is no appreciable difference between his model
and a purely virtual model in regard to the determination of equilibrium
prices. Hicks then postulated that the equilibrium prices found on the
rst day rule during the subsequent days of the week. In other words,
there is no disequilibriumactivity on those days. Thus he did not employ
Walrass device of pledges to trade but only at equilibriumprices, believ-
ing that he achieved the virtual effect without it.
It has just been indicated that Hicks did not establish that disequi-
librium transactions are unimportant, and he could not have done so
without making very special assumptions about the forms of the prefer-
ence functions. In actuality, the equilibrium price in any market will in
most cases be affected signicantly by disequilibrium transactions. In
addition to that matter, there are a variety of relevant phenomena other
than the type of effects that Hicks had in mind that result from changes
in the amounts of money and consumer and capital goods held by indi-
viduals. Economic activity at disequilibrium prices includes the hiring
of factors of production at disequilibrium prices and the payment of
disequilibrium incomes. The amounts hired and the incomes paid differ
10
It is also to Hickss credit that he avowed, in the second edition of Value and Capital,
about his assertion: I endeavored, inthe note onpp. 1279toprovide that justication,
but I did not pretend to be very satised with the results. However this was all I could
do with the technique which was at my disposal (Hicks 1946, p. 336).
P1: JZZ
306 Walrass inuence
from and may well be quantitatively much larger than the amounts hired
and incomes paid at equilibrium prices. Likewise, the occurrence of dis-
equilibriumrates of saving and investment affects many variables. Those
phenomena have signicant impacts upon desired and actual supply and
demand quantities and hence upon the path to equilibrium and the set of
equilibrium values.
The fact is that Hicks did not expend much effort on the matter in
question because he was not really concerned with the equilibrating
process: Since we shall not pay much attention to the process of equi-
libration which must precede the formation of the equilibrium price,
our method seems to imply that we conceive of the economic system as
being always in equilibrium (ibid., p. 131). Today it would be thought
strange to presume to treat the Foundations of Dynamic Economics
(Hicks 1939, Part III) and the stability of equilibrium (ibid., pp. 24572,
31525) without considering very extensively the question of what hap-
pens in disequilibrium.
Post-Hicksian virtual Walrasian modeling
In the thirty years after Hickss work, the constructors of virtual models
wove new themes and techniques together with those that had had their
origins in Walrass ideas. Hickss work stimulated Paul A. Samuelson
to make important contributions to the understanding of the different
types of stability in the early 1940s. In that period, Samuelson did not,
however, even mention the behavior of economic agents that could cause
the variables in a model to move toward or away from equilibrium.
He did not construct a model with participants, commodities such as
consumer goods, services and capital goods, and pricing institutions and
procedures, and then examine it to determine if equilibriumexists in it or
if it is stable. Rather, he constructed various equation systems to illustrate
different types of adjustment processes, some of them physical, that are
stable or unstable (Samuelson 1941a; 1941b; 1947, pp. 257355).
In the 1950s and 60s, Kenneth Arrow, G erard Debreu,
11
and Frank
Hahn, among others, developed aspects of virtual general equilibrium
theory. Maurice Allais constructed both virtual and non-virtual mod-
els in the line of descent from Walras, Fisher and Pareto (Belloc and
Moreaux 1987, p. 78). Arrow wrote that, Of course Hicks Value
and Capital made the biggest impression on me . . . (Arrow to E. Roy
Weintraub, November 19, 1981, quoted in Weintraub 1983, p. 29), thus
11
Autobiographical information provided by Debreu is published in Bini and Bruni
1998.
P1: JZZ
indicating the most important conduit by which he was indirectly inu-
enced by Walrass legacy. Arrows attention was thereby drawn to the
issue of competitive equilibrium. Believing that Wald had dealt with an
unsatisfactorily special case, and that Hicks had not pronounced the last
word on the matter, Arrowdeclared that the existence of a solution to the
equations of general equilibriumwas an open question (ibid.). In tack-
ling that question, both Arrow and Debreu independently turned away
from the specic type of mathematics that had been used by Walras,
Pareto, Cassel, and Wald. Arrow and Debreu used set theory instead of
calculus. This enabled them to deal with the complexities of the prob-
lem, and to ensure that the conditions they deemed necessary prevailed
in the equilibriums of their models.
They also used game theory, developed by von Neumann and his
collaborator, Oskar Morgenstern, an Austrian who had also joined the
faculty at Princeton (von Neumann and Morgenstern 1944). It cannot
be said that they owed any of their ideas on that topic to Walras. Rather,
their dealing with the strategies of economic competitors and the results
of their interactions enlarged the scope of theorizing about competition.
John Nash (1950, 1951), showed that game theory is applicable to the
study of the existence of a competitive equilibrium in a virtual type of
game played by many persons. He dened the conditions of equilibrium
that exist in games in which the participants cooperate and in those in
which they do not.
The non-cooperative equilibrium that Nash presented was developed
and generalized to some extent by Arrow and Debreu in their explo-
ration of a model with more general properties than Walds (Debreu
1952; Arrow and Debreu 1954). Following in the tradition of the model
Walras had envisioned in 1899, it was perfectly competitive and also
virtual, and was therefore similarly a special case. They demonstrated
that the equilibria generated by a virtual competitive economy are Pareto
optimums, giving maximum efciency and well-being. Optimum dis-
tributions of income and resources, they showed, are characterized by
sets of prices that equate supply and demand quantities, and are there-
fore competitive equilibria. The existence theorempresented here gives
general conditions under which there is, for [a certain type of] social
system, an equilibrium, i.e. a situation where the action of every agent
[is constrained by choices made by other agents] and no agent has incen-
tive to choose another action (Debreu 1952, p. 887). Arrowand Debreu
explained the conditions that must prevail if equilibrium exists, such as
the necessity that prices be non-negative, and they proved that equilib-
rium exists in their model. They also established, with reference to their
own model, the accuracy of Walrass conjecture that the participants in
P1: JZZ
308 Walrass inuence
a competitive model attain a constrained and non-unique maximum of
well-being in equilibrium.
Subsequently, Arrow and Frank H. Hahn, delving selectively back
into the history of the virtual Walrasian tradition, constructed a purely
competitive model of that type. To do so, they employed the device of
to use their term a super-auctioneer (1971, pp. 264, 266, 325),
one that, as has been seen, has subsequently unjustly been attributed
to Walras by other economists and become a rmly established mythic
aspect of the Walrasian tradition. That ctional personage, who does
not conduct any auctions and does not in any respect represent exclu-
sively sellers, controls an undescribed and unimaginable information-
collection system. He also controls an undescribed and unimaginable
information-dissemination technology that he uses to quote prices in all
markets and to which participants have access to learn those prices in
at least related markets. Arrow and Hahn indicate no more about these
matters than that he calls a given set of prices p and receives transac-
tion offers from the agents in the economy. If these do not match, he
calls another set of prices, following the Walrasian pricing rule, but
no transactions are allowed to take place (ibid., p. 264). Arrow and
Hahn should have explicitly assumed that Walrass device of written
pledges is used by the participants to register their desired supply and
demand quantities. Without such pledges, it is impossible to see how
the auctioneer could be informed of market conditions. Like Walrass
sketch, Arrows and Hahns model is lacking a means by which the
individual supply and demand quantities are totalized in each market.
Their model also needs some means by which that information is trans-
mitted to the auctioneer. In any event, he somehow learns the market
supply and demand quantities for every commodity, and he changes the
prices, somehow transmitting them to each market, until he nds the
set that would put them all simultaneously into equilibrium, whereupon
he allows economic activities to take place at that set. Arrow and Hahn
proved the existence of equilibrium with reference to the equation sys-
tem ostensibly relating to their principal model and to a variety of other
equation sets, none of them justied by an account of market features
(ibid., pp. 10728).
The relative simplicity of virtual Walrasian models relative to the
non-virtual variety and ones with imperfect competition has occasion-
ally attracted the interest of other scholars. They have devised proofs of
existence in models of that type with minor variations in their charac-
teristics, or novel proofs regarding the standard Arrow-Debreu model.
Christian Bidard and Reiner Franke (1987), for example, furnished
it with a simpler proof. Yvan Lengwiler (1998) dealt with atomless
exchange economies that do not fulll the usual autarky assumption,
P1: JZZ
meaning that the endowments may lie outside the consumption set. He
established the existence of a Walrasian equilibrium, namely one in
which, briey, the supply and demand quantities of each commodity
are equal, provided the economy is heterogeneous. Antonio DAgata
(2001) provided a new proof of the existence of a Walrasian equilib-
rium in virtual pure exchange economies under standard and general
assumptions. The proof, which employs a xed-point theorem, is shorter
and simpler than previous ones. Nils Hauenschild and Peter Stahlecker
(2002) assumed that individual preferences are strictly monotone. In
this case, the continuity of the excess demand functions that is usu-
ally assumed to show the existence of a Walrasian equilibrium does
not hold for price vectors in which at least one component is equal to
zero. Equilibrium prices will not be equal to zero, however, if prefer-
ences are strictly monotone. A simple proof of existence in the latter
case, very similar to the non-monotone case, is provided by those the-
orists. A nal example is the virtual general equilibrium model with
indivisible commodities examined by Takuya Iimura (2003). He proved
a xed-point theorem on a discrete set; using the property of contiguous
convexity of the set and the direction-preserving character of Walrasian
correspondence. He gave a set of sufcient conditions, in terms of the
characteristics of excess demand functions, for the existence of Wal-
rasian equilibrium prices.
12
The present state of virtual Walrasian modeling
For many years, most theorists chose to elaborate upon the concept of
a virtual purely competitive model that had been introduced by Wal-
ras. Cassel followed that approach, added the assumption that the sys-
tem is always in equilibrium, reverted to Walrass use of supply and
demand functions aggregated at the market level, and developed a
model of steady-state growth. Subsequently, John von Neumann fol-
lowed Cassels lead by assuming that his model is always in equilib-
rium and by developing a virtual purely competitive steady-state growth
model. General equilibrium modelers in the years following 1930, like
Wald, von Neumann, and Hicks, wanted to determine whether equilib-
rium exists in a virtual purely competitive model and whether there is a
unique solution set of prices.
Kenneth Arrow, G erard Debreu, Frank Hahn, and other mathemat-
ical economists used Walrass ideas about a virtual purely competi-
tive model, Paretos ideas about efciency, concepts taken from game
12
The work of other modelers of virtual Walrasian systems is collected in Debreu 1996,
Walker 2000a, 2, and Arrow and Debreu 2001.
P1: JZZ
310 Walrass inuence
theory, and the notion of a central price setter to develop further the vir-
tual Walrasian strand of theorizing. Because Walrass notion of the use
of written pledges is manifestly awed and unworkable, general equi-
librium theorists did not mention it, even though their models cannot
function without written pledges.
The scholars who developed and analyzed virtual Walrasian models
have proved the existence of equilibrium in the equation system relating
to them and its uniqueness if one or the other of Walds assumptions
about demand is made. They have established that very little else can be
said about uniqueness and, regarding Walds assumptions, it is quite
clear that uniqueness theorems can only be obtained on assumptions so
restrictive as to appear unacceptable (Ingrao and Israel 1990, p. 360).
They have also established that very little can be said about the models
stability. HugoSonnenscheinandothers showedthat the characterization
of preferences functions, the assumption of utility maximization, and the
resulting demand functions at the microeconomic level do not result in
any meaningful necessary features of aggregate demand functions. An
aggregate excess demand function can be specied completely arbitrar-
ily and it can nevertheless be an excess demand function for some
commodity in a general equilibrium economy (Sonnenschein 1972,
p. 549). An arbitrary function can be constructed that has second partial
derivatives that indicate the local instability of a solution position of the
market and hence the impossibility of global stability. The consequence
is that, Sonnenscheins theorem (together with its developments and
implications) seems to be an increasingly bulky obstacle blocking all
developments of the theory of general equilibrium in the critical sectors
of uniqueness and stability (Ingrao and Israel 1990, p. 319).
Sufce it to say, general equilibrium theory . . . had a very large bur-
den to bear. It proved unequal to this task. Such became clear in a spec-
tacular series of impossibility results that might be called Sonnenschein-
Mantel-Debreu theory after its main promulgators (Rizvi 2003, p. 384).
Their work showed that formalist general equilibrium theory had
reached a dead end: no general results beyond existence of equilibrium
were possible (ibid.). Strictly, the arbitrariness results put an end to
neoclassical general equilibriumtheory of the Arrow-Debreu-McKenzie
variety (ibid., p. 385), that is to say, the virtual line of purely compet-
itive general equilibrium models that were the descendents of Walrass
written pledges sketch.
In addition to the internal difculties of that type of model, there
was also the problem of its lack of realism. Walras would have said,
in his mature phase of theorizing, that the virtual model was a triumph
of pure reasoning without recourse to experience, or, more accurately, a
P1: JZZ
triumph of pure reasoning over experience. It became increasingly clear
to many economists that general equilibrium theory could not fulll a
promise of over 30 years. . . . The results had an epoch-ending impact.
Erstwhile champions of general equilibriumtheory have had to abandon
the eld. Christopher Bliss thus wrote, The near emptiness of general
equilibrium theory is a theorem of the theory (ibid., p. 38485). One
of the principal participants in the virtual Walrasian research agenda
describes matters in this way:
I have always regarded Competitive General Equilibrium analysis as akin to the
mock-up an aircraft engineer might build. . . . Theorists all over the world have
become aware that anything based on this mock-up is unlikely to y, since it
neglects some crucial aspects of the world, the recognition of which will force
some drastic re-designing. Moreover, at no stage was the mock-up complete;
in particular, it provided no account of the actual working of the invisible hand
(Hahn 1981, p. 1036).
The line of researchwas unrealistic ina number of important respects, the
most important of which was its virtual character. This means that the
principal achievement of the research the proof of the existence of
equilibrium in the model was one of logical deduction, not one that is
applicable to real purely or approximately purely competitive economic
processes, because they are characterized by irrevocable disequilibrium
behavior. If those processes have an equilibrium, it is path dependent,
and hence its existence is not proven by the virtual approach. Thus,
We are at a turning point in economic theory. Much of the elegant
theoretical structure that has been constructed over the last one hundred
years in economics will be seen over the next decade to have provided
a wrong focus and misleading and ephemeral idea of what constitutes
an equilibrium (Kirman 1999, p. 8).
13
Alan Kirman was referring to
the idea of equilibrium in the virtual purely competitive model; that is,
equilibrium based on the constancy of the initial conditions during the
process of nding it.
The many contributions of virtual purely competitive general equi-
librium theorists resulted in a largely completed research agenda. Their
extremely clever work has shown what can be achieved in that frame-
work, and what cannot be achieved. The consequence of the latter reve-
lations (that is, the negative results of the Sonnenschein-Mantel-Debreu
analyses for the virtual purely competitive models) has divided those
who had worked in the eld into two groups: those concerned only to
13
This quotation and the one fromHahns work are also to be found in Fabio Petris ne
analysis (2004) of the problems of the virtual line of purely competitive modeling
(see pp. 12 for his use of the quotations).
P1: JZZ
312 Walrass inuence
mourn the loss of global stability and those who see that no progress is
possible in that direction and so a change is called for (Ingrao and Israel
1990, p. 347).
14
The outlook of the latter group has been reinforced
by criticisms such as those of Hahn and Kirman. Their contention is
that whatever results the virtual line of research may have been able to
achieve, they are not relevant for understanding or inuencing the real
economy. General equilibrium theorists have therefore chosen to heed
the counsel calling for a change and have moved on to the consideration
of other types of models.
14
There are some questionable aspects (Walker 1991) of Ingraos and Israels book
(1990). Nevertheless, they provide an excellent survey (ibid., pp. 31662) of the
paucity of the results of purely competitive modeling of the virtual type, and of the
impossibility of making further progress in that eld of research.
P1: PIG
CHAPTER 11
Concluding comments: Walrass ideas
in modern economics
This book has told the story thus far of the past inuence of Walrass
mature comprehensive model and his written pledges sketch. The latter
gave rise to the virtual line of general equilibrium research that domi-
nated the agenda of economists working in that eld for more than forty
years in the twentieth century, and aspects of that line of research con-
tinue to be pursued in some modern theorizing. The constructions and
functioning of the mature comprehensive model have a secure place in
the history of general equilibrium modeling as providing its beginning
and as furnishing ideas that have been used throughout its history.
1
In
that model, Walras developed a non-virtual model in which economic
decisions are made in private markets by entrepreneurs, suppliers of eco-
nomic resources, consumers, and savers, and in which the markets are
purely competitive. It has four major characteristics. First, the markets
in it are interrelated, so the values of the economic variables are mutu-
ally determined and the impact of a change in a parameter is manifested
throughout the economic system. Second, irrevocable transactions, pro-
duction, consumption, savings, and investment occur in disequilibrium.
Third, when the economy is in disequilibrium it moves toward an equi-
libriumposition. Fourth, that position is on a path of growth. The position
changes as the economy grows, or possibly contracts, tracing out a path
through time. Of course, Walras recognized that in the real economy
the movement to equilibrium under a given set of basic conditions is
frequently interrupted because changes in those conditions occur fre-
quently. The conditions include the development of new technologies,
changes of tastes, changes in the size of the labor force, and changes
of government policies. He also recognized, although he expressed the
1
In the light of the vast body and continuing vitality of non-virtual Walrasian mod-
ern research it is inexplicable why one author advanced the idea that the Non-
T atonnement Line of Research Died Out (Busetto 1995). A research project with a
valid relation to the historical record would be: Why the non-virtual line of research
continues, and why the line of research into virtual purely competitive general equi-
librium models has reached some dead ends, the latter being a matter addressed at
the end of Chapter 10.
313
P1: PIG
314 Walrass inuence
matter in different words, that the position and slope of the growth path
are affected by those changes.
Historians of economic thought sometimes exhort modern theorists
to study Walrass work, because they would nd ideas in it that could
be used to advantage in modern theoretical constructions. That exhor-
tation is unnecessary. His legacy is a vital part of modern theorizing
and of modern empirical applications of it. Indeed, an exposition of all
of modern scholarship that benets from his work would be impossi-
ble, as it would encompass much of modern microeconomic theory
not just general equilibrium models and some macroeconomics. Eco-
nomic theorists use every day, as a matter of course, many of Walrass
special constructions as modied by his successors. These include the
reciprocal nature of demand and supply, general equilibrium demand
and supply functions, the rigorously expressed idea of optimization
and constrained maxima in economic connections, the individuals bud-
get equation, Walrass law, the Walrasian equilibrium conditions, the
Walras correspondence relationships, Walras optimum allocations, the
concept of the production function, Walrass contribution to the theory of
marginal productivity, the concept of a numeraire, the price-dependent
and production-dependent analyses of stability in particular markets, the
theorem of equivalent distributions, what is called the Fisher equation,
and a cash balances theory of money not just a cash balances equation.
The constructors of manyrecent models explicitlyacknowledge Walrass
inuence
2
as enriching elds of economic research in addition to the
purely competitive situation that he envisaged. There are adaptations and
extentions of his ideas in game theory, Cournot-Walras models and other
models combining Walrasian features and imperfect competition, Nash-
Walras models, bargaining models, x-price models, Walras-Keynes
models (such as ones that blend failures of coordination with multi-
market general equilibrium features), and matching and search models.
As has been noted in Chapter 9, the empirical testing of aspects of
Walrass mature comprehensive model was initiated during his life by
H. L. Moore and continued shortly after Walrass life by Moore and
Henry Schultz. Another early development of importance for applied
general equilibrium studies of non-virtual economic phenomena, inas-
much as it encouraged the development of econometric research of that
nature, was the formation of the Econometric Society in the United
States in 1930. At that time, the businessman Alfred Cowles decided
to try to improve the dismal state of business forecasting, and saw in
2
A bibliography of recent writings that reect the inuence of Walrass legacies, along
with the references for Chapter 11, is placed in a separate section after the references
to writings mentioned before this chapter.
P1: PIG
Concluding comments 315
the Econometric Society a means of achieving his end. He began to
subsidize it and Econometrica, in which econometric research on gen-
eral equilibrium has been published. Likewise, primarily for the pur-
pose of developing general equilibrium economic theory in relation to
statistical empirical research, Cowles obtained a charter for the Cowles
Commission for Research in Economics. After the 1930s, There was
clear recognition of the centrality of general equilibrium analysis in the
development of economic theory sufciently rich to provide a basis for
empirical work (Weintraub 1983, p. 19). A notable example is Wassily
Leontiefs input-output model dealing, for practical purposes, with the
interrelations of industries (Leontief 1941).
The continuedempirical usefulness of Walrasianmodels was foreseen
by Herbert Scarf, writing about a renewed interest in policy questions
at a microeconomic level:
The Walrasian model of competition, even though it is sufciently exible to
incorporate a number of formal modications, is far frombeing the exclusive ana-
lytical framework for the study of microeconomic problems. But it is an important
method of analysis and one whose usefulness, we hope, will be enhanced by the
ability to obtain specic numerical solutions (Scarf 1973, p. 17).
Precisely that has happened, as, for example, in the applications of
general equilibrium models to the analysis of real problems with the
aid of modern computers (see, for example, Scarf and Shoven 1984,
Piggott and Whalley 1985, Chang 1997, Gottinger 1998). One such
study deserves special mention in a survey of Walrass legacy because
of its name as well as its content. It is the multi-country applied econo-
metric general equilibrium model, appropriately called WALRAS, used
to quantify and evaluate the effects of agricultural policies (Delorm e and
Van der Mensbrugghe 1989, Lienert 1989, Martin 19891990, Burniaux
1990). The empirical applications of modern general equilibrium show
the continuing vitality of Walrass ideas in yet another domain.
3
Walrass legacy in special respects and in general has therefore been
to contribute many important strands of the mainstream of modern eco-
nomic research and knowledge. Indeed, his legacy is inseparable from
that mainstream.
3
One writer has referredtothe paucityof the interpretative results producedbyWalrass
equilibrium theory and contended that it is at best a formal synthesis which is not
adequate for stimulating empirical research (M enard 1990, p. 119). Adherents of
those views also describe his theory as a thought experiment (ibid., p. 118), a label
chosen to indicate the belief that the proposition that the economy is an equilibrating
systemcannot be tested and that general equilibriumtheory is useless for applications.
Evidently those views conict with the history of empirical research in the general
equilibrium eld.
P1: PIG
316
P1: PIG
0521858550rfa CB1007/Walker 0 521 85855 0 November 24, 2005 0:30
References
Allais, Maurice (1968), Vilfredo Pareto, Contributions to Economics, in
International Encyclopedia of the Social Sciences, New York: Macmillan
Co. and the Free Press, 2, 399411.
Amoroso, Luigi (1909), La teoria dellequilibrio economico secondo il
Prof. Vilfredo Pareto, Giornale degli Economisti, series 2, 39, October,
35367.
Amoroso, Luigi (1921), Lezioni di economia mathematica, Bologna: N.
Zanichelli.
Antonelli, Etienne (1914), Principes d economie pure, Paris: Librairie des sci-
ences politiques et sociales.
Antonelli, Etienne (1939), LEconomie pure du capitalisme, Paris: Guillaumin
et C
ie
.
Antonucci, Antonio (ed.) (1938), Alcune lettere di Vilfredo Pareto pubblicate
e commentate da A. Antonucci, Rome: Prof. P. Maglione-Editore; Socc. di
Loescher & C.; reprinted in Busino 1973 without Antonuccis comments.
Arena, Richard and Ludovic Ragni (1994), Libre concurrence et m ethodologie
walrasienne: une tentative de mise en relation, Economies et Soci et es, S erie
CEconomia, Histoire de la Pens ee economique, P. E. no. 2021, 1011,
16182.
Arrow, Kenneth (1959), Towards a Theory of Price Adjustment, in Moses
Abramovitz et al. (eds.), The Allocation of Economic Resources: Essays in
Honor of Bernard Francis Haley, Stanford: Stanford University Press, 4151.
Arrow, Kenneth (1968), Economic Equilibrium, in David L. Sills (ed.), Inter-
national Encyclopedia of the Social Sciences, 4, no place: Macmillan Com-
pany and the Free Press, 37686.
Arrow, Kenneth J. and G erard Debreu (1954), Existence of an Equilibriumfor
a Competitive Economy, Econometrica, 22, July, 26590.
Arrow, Kenneth J. and G erard Debreu (eds.) (2001), Landmark Papers in
General Equilibrium Theory, Social Choice and Welfare, in Foundations of
Twentieth Century Economics, 3, Cheltenham, U.K. and Northampton, MA.:
Elgar Publishing Limited.
Arrow, Kenneth J. and F. H. Hahn (1971), General Competitive Analysis, San
Francisco: Holden-Day; Edinburgh: Oliver and Boyd.
Arrow, Kenneth J. and Leonid Hurwicz (1958), On the Stability of Competitive
Equilibrium I, Econometrica, 26, 52252.
317
P1: PIG
318 References
Aupetit, Albert (1901), Essai sur la th eorie g en erale de la monnaie, Paris:
Guillaumin et C
ie
.
Aupetit, Albert (1914), Principes d economie pure, Paris: Guillaumin et C
ie
.
Barone, Enrico (1895), Sopra un recente libro del Wicksteed, unpublished
lost mss.; translated into French by L eon Walras as Sur un livre r ecent de
Wicksteed, in 239, 1895, 2, 64448; translated into English by Donald A.
Walker, in Jaff e 1983, 18286.
Barone, Enrico (1896), Studi sulla distribuzione, Giornale degli Economisti,
series 2, 12, February, 107152.
Barone, Enrico (1908), Il Ministro della produzione nello stato collettivista,
Giornale degli Economisti, series 2, 37, September, 26793; October, 391
414.
Baumol, WilliamJ. and Stephen M. Goldfeld (eds.) (1968), Precursors in Math-
ematical Economics, LSE Series of reprints of scarce works on political
economy, no. 19, London: London School of Economics.
Belloc, Bernard and Michel Moreaux (1987), Allais, Maurice, in Eatwell
et al. 1987, 1, 7880.
Bidard, Christian and Reiner Franke (1987), On Walras Model of General
Equilibrium: A Simpler Way to Demonstrate Existence, Zeitschrift f ur
National okonomie, 47 (3), 31519.
Bini, Piero and Luigino Bruni (1998), Intervista a G erard Debreu, Storia del
pensiero economico, 35, 329.
Blaug, Mark (ed.) (1992a), Leon Walras (18341910), an Elgar Reference
Collection, in Pioneers in Economics, 25, Aldershot, UK and Brookeld,
VT: Edward Elgar Publishing.
Blaug, Mark (1992b), The Methodology of Economics or How Economists
Explain, 2nd ed., Cambridge: Cambridge University Press.
Blaug, Mark (1997), Competition as an End-State and Competition as a Pro-
cess, in B. Curtis Eaton and Richard G. Harris (eds.), Trade, Technology
and Economics: Essays in Honour of Richard G. Lipsey, Cheltenham, UK
and Brookeld, VT: Edward Elgar Publishing, 24162; and in Not Only
an Economist, Recent Essays, Cheltenham, UK: Edward Elgar Publishing,
6686; and in Walker 2000a, 1, 27293.
Bliss, Christopher (1987), Hicks, John Richard (born 1904), in Eatwell et al.,
2, 64146.
Bompaire, Fran cois (1931), Du principe de libert e economique dans loeuvre de
Cournot et dans celle de l ecole de Lausanne (Walras, Pareto), Paris: Recueil
Sirey.
Bousquet, Georges-Henri (1927), Essai sur l evolution de la pens ee econo-
mique, Paris: M. Giard.
Bousquet, Georges-Henri (1928a), Cours d economie pure, Paris: Librairie des
sciences politiques et sociales.
Bousquet, Georges-Henri (1928b), Vilfredo Pareto. Sa vie et son oeuvre, Paris:
Payot.
Bousquet, Georges-Henri (1960), Pareto (18481923). Le savant et lhomme,
Lausanne: Payot.
P1: PIG
References 319
Bousquet, Georges-Henri (1964), Introduction to Pareto 1896/1897, newedi-
tion, Geneva: Librarie Droz, IXXVII.
Bousquet, Georges-Henri (1971), Quelques remarques sur Pareto et certains
d efauts de son oeuvre, Revue Europ eenne des Sciences Sociales et Cahiers
Vilfredo Pareto, 9 (25), 1127.
Bouvier, Emile (1901), L economie politique math ematique, quelques mots
` a propos du trait e de M. L eon Walras, Revue Critique de L egislation et de
Jurisprudence, 30 (12), December, 62329.
Boven, Pierre (1923), Vilfredo Pareto II. Loeuvre et les th eories, Journal de
Gen` eve, 94th year, no. 229, August 22, 12.
Bowley, Arthur L. (1924), The Mathematical Groundwork of Economics: An
Introductory Treatise, Oxford: The Clarendon Press.
Brems, Hans (1986), Pioneering Economic Theory, 16301980. A Mathemat-
ical Restatement, Baltimore and London: The Johns Hopkins University
Press.
Brems, Hans (1987), Zeuthen, FrederikLudvigBang(18881959), inEatwell
et al. 1987, 4, pp. 94445.
Bridel, Pascal (1990), Equilibre, statique compar ee et analyse dynamique chez
Vilfredo Pareto: remarques sur la contribution de Siro Lombardini, Revue
europ eenne des sciences sociales, 27 (88), 183191.
Bridel, Pascal (1997), Money and General Equilibrium Theory; From Walras
to Pareto (18701923), Cheltenham, UK and Lyme, CT: Edward Elgar Pub-
lishing Limited.
Bridel, Pascal (1998), Review of Walrass market models. By Donald A.
Walker. Cambridge; NewYork and Melbourne: Cambridge University Press,
1996, Journal of Economic Literature, 36 (1), 23133.
Bridel, Pascal and Elisabeth Huck (2002), Yet another look at Walrass theory
of t atonnement, European Journal of the History of Economic Thought, 9,
Winter, 51340.
Bruni, Luigino (2002), Vilfredo Pareto and the birth of modern microeconomics,
Cheltenham, U.K. and Northampton, MA: Elgar Publishing Limited.
Buckle, Henry Thomas (1872), The miscellaneous and posthumous works of
Henry Thomas Buckle, edited by Helen Taylor, London: Longmans, Green
and Co.
Burgenmeier, B. (1994), The Misperception of Walras, American Economic
Review, 84 (1), March, 34252.
Busetto, Francesca (1995), Whythe Non-T atonnement Line of ReasearchDied
Out, Economic Notes, 24 (1), 89113.
Busino, Giovanni (ed.) (1973), Vilfredo Pareto: Epistolario, 18901923, 2 vols.,
Rome: Accademia Nazionale dei Lincei.
Busino, Giovanni (1987), Pareto, Vilfredo, in Eatwell et. al. 1987, 3, 799
804.
Cassel, Karl Gustav (1899), Grundriss einer elementaren Preislehre,
Zeitschrift f ur die gesamte Staatswissenschaf, 55 (3), 395458.
Cassel, Karl Gustav (1900), Grunds atze f ur die Bildung der Personentarife
auf den Eisenbahnen, Archiv f ur Eisenbahnwesen, 23, 11646; translated
P1: PIG
320 References
as The Principles of Railway Rates for Passengers, International Economic
Papers, 6, 1956.
Cassel, Karl Gustav (1903), The Nature and Necessity of Interest, London and
New York: Macmillan.
Cassel, Karl Gustav (1918), Theoretische Sozial okonomie, Leipzig: C. F.
Winter; 4th ed., revised, Leipzig: A. Deichertsche Verlasbuchhandlung
Dr. Werner Scholl, 1927.
Cassel, Karl Gustav (1921), The Worlds Monetary Problems, Two memoranda
presented to the International Financial Conference of the League of Nations
in Brussels in 1920 and to the Financial Committee of the League of Nations
in September 1921, London: Constable.
Cassel, Karl Gustav(1923), The Theory of Social Economy, translatedbyJoseph
McCabe from the enlarged German 2nd ed., London: T. F. Unwin.
Cassel, Karl Gustav (1928), The Rate of Interest, the Bank Rate, and the
Stabilization of Prices, Quarterly Journal of Economics, 42 (August),
51129.
Cassel, Karl Gustav (1932), The Theory of Social Economy, translated by
Samuel L. Barron from the 5th German edition, London: E. Benn; reprint,
New York: Augustus M. Kelley, 1967.
Cassel, Karl Gustav (1935), On Quantitative Thinking in Economics, Oxford:
Clarendon Press.
Cassel, Karl Gustav (1937), Keynes General Theory, International Labour
Review, 36, 43745.
Clarke, Desmond M. (1991), Descartess Use of Demonstration and Deduc-
tion, inGeorges J. D. Moyal, (ed.), (1991), Ren e Descartes: Critical Assess-
ments, 4 vols., London: Routledge.
Clower, Robert W. (1965), The Keynesian Counter-Revolution: A Theoretical
Appraisal, in F. H. Hahn and F. P. R. Brechling, (eds.), The Theory of Interest
Rates, London: Macmillan, 10325.
Collins Robert Unabridged French-English, English-French Dictionary (2002),
6th ed., Glasgow: HarperCollins Publishers.
Cournot, Antoine-Auguste (1838), Recherches sur les principes math ematiques
de la th eorie des richesses, Paris: Hachette.
Cournot, Antoine-Auguste (1863), Principes de la th eorie des richesses, Paris:
Hachette.
Creedy, John (1999), The Rise and Fall of Walrass Demand and Supply
Curves, Manchester School, 67 (2), March, 192202.
Currie, Martin and Ian Steedman (1990), Wrestling with Time: Problems in
Economic Theory, Manchester: Manchester University Press.
DAgata, Antonio (2001), Yet Another Proof of the Existence of a Com-
petitive Equilibrium, Revista de Analisis Economico, 16 (2), December,
15558.
Davis, John B., D. Wade Hands, and Uskali M aki (1998), The Handbook of
Economic Methodology, Cheltenham, UK and Northampton, MA: Edward
Elgar Publishing Limited.
P1: PIG
References 321
Debreu, G erard (1952), A Social Equilibrium Existence Theorem, Proceed-
ings of the National Academy of Sciences, 38, 88693.
Debreu, G erard (1959), Theory of Value; An Axiomatic Analysis of Economic
Equilibrium, New York: John Wiley and Sons.
Debreu, G erard (ed.) (1996), General Equilibrium Theory, 2 vols., an Elgar
Reference Collection, in The International Library of Critical Writings in
Economics, general editor Mark Blaug, Cheltenham, UK and Brookeld,
VT: Edward Elgar Publishing Limited.
Debs, Alexandre (2004), ToBe or Ought ToBe: The Questions of Empirical
Content and Normative Bias in L eon Walrass Methodology, Journal of the
History of Economic Thought, 26, December, 47992.
De Vroey, Michel (1987), La possibilit e dune economie d ecentralis ee, Revue
Economique, 38 (4), 773805.
De Vroey, Michel (1999), Transforming Walras into a Marshallian Economist:
A Critical Review of Donald Walkers Walrass Market Models, Journal of
the History of Economic Thought, 21 (4), December, 41335.
Diemer, Arnaud (2002), Economie Pure, Economie Appliqu ee, Economie
Sociale, Un point de vue critique sur loeuvre de L eon Walras?, III
e
Colloque
de lAssociation Internationale Walras, Lyon, September 20, in Les Cahiers
duCERAS, hors s erie no. 3, May, 2004, 23553, withthe title Economie Pure
et Economie Appliqu ee. Retour sur les origines de loeuvre de L eon Walras.
Diemer, Arnaud and J er ome Lallement (2004), De Auguste ` a L eon Walras:
retour sur les origines du march e et de la concurrence walrassiennes, pre-
liminary version, IV
e
Colloque de lAssociation Internationale Walras, Nice,
September 2324.
Divisia, Fran cois (1928), Economique rationelle, Paris: G. Doin.
Dock` es, Pierre (1994), La soci et e nest pas un pique-nique: le socialisme
appliqu e de L eon Walras, Economies et Soci et es, S erie CEconomia, Histoire
de la Pens ee economique, P. E. no. 2021, 1011, 270325.
Dock` es, Pierre (1996), La soci et e nest pas un pique-nique: L eon Walras et
l economie sociale, Paris: Economica.
Dock` es, Pierre (1999), Ce qui est, ce qui devrait etre, ce qui sera: Walrass
economics as he saw it, Revue Europ eenne des Sciences Sociales, 37 (116),
1336.
Dock` es, Pierre, Ludovic Frobert, G erard Klotz, Jean-Pierre Potier, and Andr e
Tiran, (eds.), (2000), Les traditions economiques francaises 18481939,
Paris: CNRS

EDITIONS.
Dock` es, Pierre and Jean-Pierre Potier (2001), La Vie et loeuvre economique de
L eon Walras, Paris: Economica; an extract from 249.V, IXLXXVIII.
Dorfman, Robert, Paul A. Samuelson, and Robert M. Solow(1958), Linear Pro-
gramming and Economic Analysis, New York, Toronto, London: McGraw-
Hill Book Company.
Eatwell, John, Murray Milgate, and Peter Newman (eds.), The New Palgrave:
A Dictionary of Economics, 4 vols., London: Macmillan Press; New York:
Stockton Press; Tokyo: Maruzen Company.
P1: PIG
322 References
Edgeworth, Francis Y. (1881), Mathematical Psychics: An Essay on the Appli-
cation of Mathematics to the Moral Sciences, London: C. Kegan Paul; reprint,
Edgeworth, Francis Y. (1889), The Mathematical Theory of Political Econ-
omy. El ements d economie politique pure. By L eon Walras, Nature, 40,
September, 43436.
Einaudi, Manon Michels (1935), Pareto as I Knew Him, Atlantic Monthly,
156, September, 33646.
Fisher, Irving (1892), Mathematical Investigations in the Theory of Value and
Prices, in Transactions of the Connecticut Academy, 9, July, 3126; reprint,
Fisher, Irving (1896), Appreciation and Interest, publications of the American
Economic Association, 3rd series, 11 (4), August, New York: Macmillan
Company.
Fisher, Irving (1892), Mathematical Investigations in the Theory of Value and
Prices, in Transactions of the Connecticut Academy, 9, July, 3126; reprint,
Fontaine, Phillippe (1998), Menger, Jevons, and Walras Un-homogenized, De-
homogenized, and Homogenized: A Comment on Peart, American Journal
of Economics and Sociology, 57 (3), July, 33340.
Gewirtz, Alan (1941), Experience and the Non-Mathematical in the Cartesian
Method, Journal of the History of Ideas, 2 (2), April, 183210.
Giddings, Franklin H. (1896), The Principles of Sociology, NewYork: Macmil-
lan Company.
Gijsel, Peter de (1989), On the Role of General Equilibrium Theory in Wal-
rass Theory of a Just Society, in Donald A. Walker (ed.), Classical and
Neoclassical Economic Thought: Selected Papers from the History of Eco-
nomics Society Conference 1987, Perspectives on the History of Economic
Thought, 1, Aldershot, UK: Edward Elgar for the History of Economics
Society, 13344.
Gustafsson, Bo (1987), Cassel, Gustav, in Eatwell et al. 1987, 1, 37577.
Hahn, Frank (1981), review: M. Beenstock, A Neoclassical Analysis of
Macroeconomic Policy, Cambridge Universtiy Press, 1980, Economic Jour-
nal, 91 (364), December, 103639.
Hands, D. Wade (1998), Positivism, in Davis et al. 1998, 37478.
Hauenschild, Nils and Peter Stahlecker (2002), A Simple Proof for the Exis-
tence of Walrasian Equilibrium under Monotone Preferences, Economics
Bulletin, 4 (4), 18.
Hayek, Friedrich August von (1934), Carl Menger, Economica, n.s., 1,
November, 393420.
Herland, Michel (1996), Three French Socialist Economists: Leroux, Proud-
hon, Walras, Journal of the History of Economic Thought, 18 (1), Spring,
13353.
Herschel, John F. W. (1831 [1830]), A Preliminary Discourse on the Study of
Natural Philosophy, London: Longman, Rees, Orme, Browne and Greene.
P1: PIG
References 323
Hicks, J. R. (1934), L eon Walras, Econometrica, 2, October, 33848.
Hicks, J. R. (1939), Value and Capital; 2nd ed., 1946, Oxford: Clarendon Press.
Hicks, J. R. (1965), Capital and Growth, New York and Oxford: Oxford Uni-
versity Press.
Hildenbrand, W. and A. P. Kirman (1988), Equilibrium analysis: Variations on
Themes by Edgeworth and Walras, in Advanced Textbooks in Economics, 28,
Amsterdam, New York, Oxford, Tokyo: North-Holland.
Hilton, H. C. (1995), L eon Walras on Money and Banking, History of Eco-
nomics Review, 24, Summer, 7278.
Hollander, Samuel and Sandra Peart (1999), John Stuart Mills Method in
Principle and Practice: A Review of the Evidence, Journal of the History of
Economic Thought, 21 (4), December, 36997.
Hotelling, Harold (1929), Stability in Competition, Economic Journal, 39,
March, 4157.
Howitt, P. W. (1973), Walras and Monetary Theory, Western Economic Jour-
nal, 11 (4), December, 48799.
Huck, Elizabeth(2001), La neutralit e dut atonnement auregardde la r epartition
des richesses dans la th eorie de la production de L eon Walras. Le cas du
t atonnement sans bons des trois premi` eres editions des El ements d economie
politique pure, Revue economique, 52 (3), May, 72938.
Hume, David (1777), Enquiries concerning Human Understanding and con-
cerning the Principles of Morals, 3rd edition; reprint, edited by L. A. Selby
Bigge, Oxford: Clarendon Press, 1975.
Hutchison, T. W. (1953), A Review of Economic Doctrines 18701929,
London: Oxford University Press.
Iimura, Takuya (2003), ADiscrete Fixed Point Theoremand Its Applications,
Journal of Mathematical Economics, 39 (7), September, 72542.
Ingrao, Bruna and Giorgio Israel (1990), The Invisible Hand: Economic Equi-
librium in the History of Science, translated by Ian McGilvray, Cambridge,
MA: MIT Press.
Jaff e, William (1935), Unpublished Papers and Letters of L eon Walras,
Journal of Political Economy, 43, April, 187207. All articles of Jaff es that
are listed in these references are reprinted in Jaff e 1983, except Jaff e 1978
and 1984.
Jaff e, William (1967), Walrass Theory of T atonnement: A Critique of Recent
Interpretations, Journal of Political Economy, 75, February, 119.
Jaff e, William(1969), A. N. Isnard, Progenitor of the Walrasian General Equi-
librium Model, History of Political Economy, 1, Spring, 1943.
Jaff e, William (1971), Reections on the Importance of L eon Walras, in
Schaarste en Welvaart, P. Hennipman Festschrift, edited by Arnold Heertje
et al., Amsterdam: Stenfert Kroese, 87107.
Jaff e, William (1976), Menger, Jevons and Walras De-homogenized, Eco-
nomic Inquiry, 14, December, 51124.
Jaff e, William (1977), The Normative Bias of the Walrasian Model: Walras
versus Gossen, Quarterly Journal of Economics, 91, August, 37187.
P1: PIG
324 References
Jaff e, William (1978), Review of Walras Economics: A Pure Theory of Cap-
ital and Money, by Michio Morishima, Economic Journal, 88, September,
574617
Jaff e, William (1980), Walrass Economics as Others See It, Journal of Eco-
nomic Literature, 18, June, 52849.
Jaff e, William (1981), Another Look at L eon Walrass Theory of
T atonnement, History of Political Economy, 13, Summer, 31336.
Jaff e, William (1983), William Jaff es Essays on Walras, edited by Donald A.
Walker, Cambridge, New York, Melbourne: Cambridge University Press.
Jaff e, William(1984), The Antecedents and Early Life of L eon Walras, edited
by Donald A. Walker, History of Political Economy, 16 (1), Spring, 157.
Jolink, Albert (1993), Procrustean Beds and All That: The Irrelevance of
Walras for a Mirowski Thesis, History of Political Economy, 25, Supple-
ment, 15974.
Keynes, John Maynard (1925/1930), F. P. Ramsey 19031930, in Essays and
Sketches in Biography, New York: Median Books, 1956. 11424.
Keynes, John Maynard (1936), WilliamStanley Jevons 18351882, in Essays
and Sketches in Biography, New York: Median Books, 1956, 12560.
Keynes, John Maynard (1936), The General Theory of Employment, Interest
and Money, New York: Harcourt, Brace and Company.
Kirman, Alan P. (1987), Pareto as an Economist, in Eatwell et al. 1987, 3,
8049.
Kirman, Alan P. (1999), The Future of Economic Theory, in Alan P. Kirman
and L.-A. G erard-Varet (eds.), Economics Beyond the Millenium, Oxford:
Oxford University Press, 822.
Klotz, G erard (1994), Achylle Nicolas Isnard, pr ecurseur de L eon Walras ?,
Economies et Soci et es, S erie CEconomia, Histoire de la Pens ee economique,
P.E. nos. 2021, 28 (1011), 2952.
Koopmans, Tjalling (1964), Economic Growth at a Maximal Rate, Quarterly
Journal of Economics, 78, August, 35594.
Koppl, Roger (1992), Price Theory as Physics: The Cartesian Inuence in
Walras, Methodus, 4 (2), December, 1728; and in Walker 2001, 1, 199
210.
Koppl, Roger (1995), The Walras Paradox, Eastern Economic Journal, 21
(1), 4355; and in Walker 2001a, 2, 26880.
Kowalik, Tadeusz (1987), Lange, Oskar Ryszard (19041965), in Eatwell
et al. 1987, 3, 12329.
Kuenne, Robert E. (1961), The Walrasian Theory of Money: An Interpretation
and a Reconstruction, Metroeconomica, 13, August, 94105.
Kurz, Heinz and Neri Salvadori (2000), Walras and Ricardo, in Dock` es et al.
2000, 96785.
Kurz, Heinz and Neri Salvadori (2004), Von Neumann, the Classical Econo-
mists and Arrow-Debreu: Some Notes, Acta CEconomica, 54 (1), 3962.
La Volpe, Giulio (1936), Studi Sulla Teoria Dell Equilibrio Economico Dinam-
ico Generale, Naples: Jovene; translated by Helen Ampt as Studies on
P1: PIG
References 325
the Theory of General Dynamic Economic Equilibrium, Basingstoke and
London: Macmillan, 1993.
Lange, Oskar R. (1936), On the Economic Theory of Socialism, Part I, Review
of Economic Studies, 4 (1), October, 5371.
Lange, Oskar R. (1937), On the Economic Theory of Socialism, Part II,
Review of Economic Studies, 4 (2), February, 12342.
Lange, Oskar R. (1944), Price Flexibility and Employment, Bloomington: Prin-
cipia Press.
Laurent, Hermann (1902), Petit trait e d economie politique math ematique,
r edig e conform ement aux pr eceptes de l ecole de Lausanne, Paris: C.
Schmid.
Lendjel, Emeric (1997), Le biais empiriste dans linterpr etation de Walker
du t atonnement walrassien, Economies et Soci et es, CEconomia, Histoire de
la pens ee economique, S erie P.E., no. 26, 31 (10), 4784.
Lendjel, Emeric (1999), T atonnement walrassien et marchandage par etien:
Une approche comparative, Revue europ eenne des sciences sociales, 37
(116), 295314.
Lengwiler, Yvan (1998), Endogenous Endowments and Equilibrium Starva-
tion in a Walrasian Economy, Journal of Mathematical Economics, 30 (1),
August, 3758.
Lexis, William (1881), Zur mathematisch- okonomischen Literatur, Jahrb u-
cher f ur National okonomie und Statistik, new series, 3, 42734.
Locke, John (1954), Essays on the Law of Nature, edited by W. von Leyden,
Oxford: Oxford University Press.
Mantel, R. (1974), On the Characterisation of Aggregate Excess Demand,
Journal of Economic Theory, 7, 34853.
Marchionatti, Roberto and Enrico Gambino (1997), Pareto and Political Econ-
omy as a Science: Methodological Revolution and Analytical Advances in
Economic Theory in the 1890s, Journal of Political Economy, 105 (6),
December, 132248.
Marchionatti, Roberto and Fiorenzo Mornati (2003), Pareto et l economie
Math ematique au D ebut des Ann ees 1890. Quelques R eexions ` a Propos des
Considerazioni sui Principii Fondamentali Delleconomia Politica Pura,
in Histoire et Th eorie des Sciences Sociales. M elanges en lhonneur de Gio-
vanni Busino, Geneva: Droz, 3, 5184.
Marget, Arthur W. (1931), L eon Walras and the Cash-Balance Approach to
the Problem of the Value of Money, Journal of Political Economy, 39 (5),
October, 569600.
Marget, Arthur W. (1935), The Monetary Aspects of the Walrasian System,
Journal of Political Economy, 43, April, 14586.
Marinitsch, O. (1892), La Bourse, th eorique et pratique, Paris: Paul Ollendorff.
Marx, Karl (1858), Introduction to Contribution to the Critique of Political
Economy, reprinted in David Horowitz, Marx and Modern Economics, New
York: Monthly Review Press, 1968.
Marx, Karl (1867), Capital, 1, Harmondsworth: Penguins Books, 1976.
P1: PIG
326 References
McKenzie, Lionel W. (1954), On Equilibrium in Grahams Model of World
Trade and Other Competitive Systems, Econometrica, 22, April, 14761.
M enard, Claude (1990), The Lausanne Tradition: Walras and Pareto, and
Commentary by Donald A. Walker, in Klaus Hennings and Warren J.
Samuels (eds), Neoclassical Economic Theory: 1870 to 1930, Dordrecht:
Kluwer Academici Publishers, 95136, 13750.
Menger, Karl (1973), Austrian Marginalism and Mathematical Economics,
in Carl Menger and the Austrian school of economics, edited by John R.
Hicks and W. Weber, Oxford: Oxford University Press, 3860.
Mill, John Stuart (1836), On the denition of Political Economy; and on
the Method of Philosophical Investigation Proper to It, London and West-
minster Review, October; in Essays on Some Unsettled Questions of Polit-
ical Economy, London: John W. Parker, 1844, 12064; and in Collected
Works of John Stuart Mill, 22, Toronto: University of Toronto Press, 1967,
30939.
Mill, John Stuart (19811991), Collected Works of John Stuart Mill, 33 vols.,
Toronto: Toronto University Press.
Mirowski, Philip (1989), More heat than light. Economics as social
physics: Physics as natures economics, Cambridge: Cambridge University
Press.
Moore, Henry Ludwell (1929), Synthetic Economics, New York: Macmillan.
Moret, Jacques (1915), LEmploi des math ematiques en economie poltitque,
Paris: M. Giard et E. Bri` ere.
Morishima, Michio (1977), Walras Economics; A Pure Theory of Capital and
Money, Cambridge: Cambridge University Press.
Morishima, Michio (1996), Morishima on Ricardo: Two Replies, Cambridge
Journal of Economics, 20 (1), January, 91109; in Walker 2001, 1, 27896.
Mosak, Jacob L. (1987), Schultz, Henry (18931938), in Eatwell et al. 1987,
4, 26162.
Mouchot, Claude (2000), De quelques probl` emes relatifs ` a lentrepreneur
walrassien, Les Cahiers du CERAS, Actes du Colloque de lAssociation
Internationale Walras, hors s erie n
o
1, juin, 4957.
Murray, Roberto A. (1920), Lecons d economie politique suivant la doctrine
de l ecole de Lausanne, Paris: Payot et C
ie
.
Myrdal, Gunnar (1945), Gustav Cassel in Memoriam, Ekonomisk revy, 2,
313.
Nash, John F. (1950), The Bargaining Problem, Econometrica, 18, 15562.
Nash, John F. (1951), Non-cooperative Games, Annals of Mathematics, 54,
28695.
Negishi, Takashi (1989), History of Economic Theory, Amsterdam: North-
Holland.
Neisser, Hans Philipp (1932), Lohnh ohe und Besch aftigungsgrad im Markt-
gleichgewicht, Weltwirtschaftliches Archiv, 36 (2), October, 41555.
Neumann, John von (1937),
Uber ein

Okonomisches Gleichungssystem und
eine Verallgemeinerung des Brouwerschen Fixpunktsatzes, Ergebnisse
P1: PIG
References 327
eines Mathematischen Kolloquiums, Heft 8, 19351936 [sic]; edited by
Karl Menger, Leipzig and Wien: Franz Deuticke, 1938, 7383; translated
by George Morgenstern as A Model of General Economic Equilibrium,
Review of Economic Studies, 13, 19451946, 19.
Neumann, John von and Oskar Morgenstern (1944), Theory of Games and
Economic Behavior, Princeton: Princeton University Press.
Niiniluoto, Ilkka (1998), Induction, in Davis et al. 1998, 24649.
Osorio, Antonio (1913), Th eorie math ematique de l echange, Paris: M. Giard
et E. Bri` ere.
Pantaleoni, Maffeo (1889), Principii di economia pura, Florence: Barbera.
Pareto, Vilfredo (1893a), Lettre dItalie, Le Monde Economique, July 8, 35
36.
Pareto, Vilfredo (1893b), Considerazioni sui principii fondamentali dell-
economia politica pura, Giornale degli Economisti, series 2, 7, October,
279321.
Pareto, Vilfredo (1896/1897), Cours d economie politique, 2 vols., Lausanne:
F. Rouge; reprint, edited by G.-H. Bousquet and G. Busino, Geneva: Librairie
Droz, 1964.
Pareto, Vilfredo (1897), The NewTheories of Economics, Journal of Political
Economy, 5 (4), 485502.
Pareto, Vilfredo (1898a), Protezionismo italiano, Critica Sociale, Rivista
Quindicinale de Socialismo, 7, February 16, 4950.
Pareto, Vilfredo(1898b), De l economie math ematique, Zeitschrift f ur Sozial-
wissenschaft, 1, 32021.
Pareto, Vilfredo (1899), Liberali e Socialisti, Critica Sociale, Rivista quindic-
inale de Socialismo, 8, September 1, 21516.
Pareto, Vilfredo (19021903), Les syst` emes socialistes, 2 vols., Paris: V. Giard
& E. Bri` ere.
Pareto, Vilfredo (1909), Manuel d economie politique; translated by Alfred
Bonnet from the Italian Manuale deconomia politica, Milan: Societ` a
Editrice Libraria, 1906; translation reviewed and corrected by the author,
Paris: V. Giard & E. Bri` ere. The Italian version translated by Ann S.
Schwier as Manual of Political Economy, New York: Augustus M. Kelley,
1971.
Pareto, Vilfredo (1911b), Economie Math ematique, in Encyclop edie des sci-
ences math ematiques, 1, Paris: Teubner, Gauthier, Villars.
Pareto, Vilfredo (1912a), Economia Dimessa, La Libert ` a Economica,
Rassegna settimanale, 10 (1718), 28590.
Pareto, Vilfredo (1912b), Review of La Teoria della Rendita. By Prof. Guido
Sensini. (Rome: Loescher. 1912.), Economic Journal, 22 (3), September,
46769.
Pareto, Vilfredo (1916), Trattato di sociologia generale, 4 vols., Florence:
Barbera; translated by Andrew Bongiorno and Arthur Livingston as The
Mind and Society; A Treatise on General Sociology, New York: Dover
Publications, 1963.
P1: PIG
328 References
Pareto, Vilfredo (1948), Corrispondenza di V. Pareto, edited by Guido Sensini,
Padua: Cedam (Casa editrice Dr. Antonio Milani).
Pareto, Vilfredo (1960), Lettere a Maffeo Pantaleoni, 18901923, edited by
Guido Sensini, 3 vols., Rome: Banca Nazionale del Lavoro.
Pareto, Vilfredo (19632001), CEuvres compl` etes de Vilfredo Pareto, 31 vols.,
prepared under the direction of Giovanni Busino, Geneva: Droz.
Patinkin, Don (1956), Money, Interest, and Prices: An Integration of Monetary
and Value Theory, Evanston, IL: Row, Peterson; 2nd ed., New York: Harper
and Row, 1965.
Patinkin, Don (1987), Walrass Law, in Eatwell et al. 1987, 3, 86368.
Petri, Fabio (2004), General Equilibrium, Capital and Macroeconomics; A
Key to Recent Controversies in Equilibrium Theory, Cheltenham, UK, and
Northhampton, MA: Edward Elgar Publishing.
Pietri-Tonelli, Alfonso de (1924), Le equazioni generali dellequilibrio eco-
nomico di Vilfredo Pareto, Giornale degli Economisti, series 4, 65,
5561.
Pietri-Tonelli, Alfonso de (1927), Trait e d economie rationnelle, 3rd ed, Paris:
M. Giard.
Pirou, Ga etan (1925), Les doctrines economiques en France depuis 1870,
Paris: A. Colin.
Pirou, Ga etan (1934), Les th eories de l equilibre economique: L. Walras et V.
Pareto, Paris: Les Editions Domat-Montchrestien.
Poinsoit, Louis (1803), Elements de statique; 8th edition, Paris: Bachelier, 1842.
Pokorny, D. (1978), Smith and Walras: Two theories of science, Canadian
Journal of Economics and Political Science, 11 (3), 387403.
Potier, Jean-Pierre (1994), Classication des sciences et divisions de
l economie politique et sociale dans loeuvre de L eon Walras: une ten-
tative de reconstruction, Economies et Soci et es, S erie CEconomia, Histoire
de la Pens ee economique, P.E. no. 2021, 28 (1011), 22377.
Potier, Jean-Pierre (1998), L eon Walras and Applied Science: The Signicance
of the Free Competition Principle, in Gilbert Faccarello (ed.), Studies in the
History of French Political Economy: From Bodin to Walras, London and
New York: Routledge, 369403.
Potier, Jean-Pierre and Donald Walker (eds.) (2004), La correspondance entre
Aline Walras et William Jaff e et autres documents, Paris: Economica.
Rebeyrol, Antoine (1998), The Development of Walras Monetary Theory, in
Gilbert Faccarello (ed.), Studies in the History of French Political Economy:
From Bodin to Walras, London and New York: Routledge, 31968.
Rebeyrol, Antoine (1999), La pens ee economique de Walras, Paris: Dunod.
Rebeyrol, Antoine (2002), Yet Another Look? AComment, European Jour-
nal of the History of Economic Thought, 9 (4), winter, 54149.
Redman, Deborah (1997), The Rise of Political Economy as a Science; Method-
ology and the Classical Economists, Cambridge, MA, and London: MIT
Press.
P1: PIG
References 329
Ricci, Umberto (1924), Pareto e leconomia pura, Giornale degli Economisti,
series 4, 65, 2744; translated as Pareto and Pure Economics, Review of
Economic Studies, 1 (13), 19331934, 321.
Rizvi, S. Abu Turab (2003), Postwar Neoclassical Microeconomics, in A
Companiontothe History of Economic Thought, editedbyWarrenJ. Samuels,
Jeff E. Biddle, and John B. Davis, Malden, MA, Oxford, UK, Melbourne,
Berlin: Blackwell Publishing Ltd., 37794.
Robert, Paul et al. (eds.) (2002), Exp erimental, Le nouveau petit Robert,
Paris: Dictionnaires Le Robert, 1001.
Samuelson, Paul A. (1941a), The Stabilityof Equilibrium: Comparative Statics
and Dynamics, Econometrica, 9 (2), April, 97120.
Samuelson, Paul A. (1941b), The Stability of Equilibrium: Linear and Non-
linear Systems, Econometrica, 9, April, 97120.
Samuelson, Paul A. (1947), Foundations of Economic Analysis, Cambridge:
Harvard University Press.
Samuelson, Paul A. (1993), Gustav Cassels Scientic Innovations: Claims
and Realities, History of Political Economy, 25 (3), 51527.
Samuelson, Paul A. (1998), Samuelson, Paul Anthony, as an Interpreter of
the Classical Economists, in Heinz D. Kurz and Neri Salvadori, eds., The
Elgar Companion to Classical Economics, 2, Cheltenham, U.K.: Edward
Elgar Publishing Limited, 329333.
Schlesinger, Karl (1914), Theorie der Geld-und Kreditwirtschaft, Munich/
Leipzig: Dunker und Humblot.
Schlesinger, Karl (19331934),
Uber die produktionsgleichungen der

okonomischen Wertlehre, Ergebnisse eines mathematischen Kolloquiums,
Heft 6, edited by Karl Menger, Leipzig and Wien: Franz Deuticke, 1935,
1011; translated by William J. Baumol as On the Production Equations of
Economic Value Theory, in Baumol and Goldfeld 1968, 27880.
Schultz, Henry (1929), Marginal Productivity and the General Pricing Pro-
cess, Journal of Political Economy, 37, October, 50551.
Schultz, Henry (1932), Marginal Productivity and the Lausanne School, Eco-
nomica, 12, August, 28596.
Schultz, Henry (1933), Interrelations of Demand, Journal of Political Econ-
omy, 41, August, 468512.
Schultz, Henry (1935), Interrelations of Demand, Price, and Income, Journal
of Political Economy, 43, August, 43381.
Schultz, Henry (1938), The Theory and Measurement of Demand, Chicago:
University of Chicago Press.
Schumpeter, Joseph A. (1910), Marie Esprit L eon Walras, Zeitschrift f ur
Volksvirtschaft, 19, 397402; English translation in Schumpeter 1951, 74
79.
Schumpeter, Joseph A. (1911/1926), The Theory of Economic Development;
An Inquiry into Prots, Capital, Credit, Interest, and the Business Cycle;
translated by Redvers Opie, New York: Oxford University Press, 1961.
P1: PIG
330 References
Schumpeter, Joseph A. (1925), Schumpeter on B ohm-Bawerk, in Henry W.
Spiegel (ed.), The Development of Economic Thought; Great Economists in
Perspective, abridged edition, New York, Science Editions, John Wiley &
Sons, 1964, 37080.
Schumpeter, Joseph A. (1949), Vilfredo Pareto, Quarterly Journal of Eco-
nomics, 63, May, 34559; and in Schumpeter 1951.
Schumpeter, Joseph A. (1951), Ten Great Economists from Marx to Keynes,
New York: Oxford University Press.
Schumpeter, Joseph A. (1954), History of Economic Analysis, New York:
Oxford University Press.
Schwartz, P. (1972), The New Political Economy of J. S. Mill, London: London
School of Economics.
Sensini, Guido (1929), Vilfredo Pareto e la teoria della moneta, Il Giornale
Economico, SeptemberOctober; reprinted in Guido Sensini, Studi di scienze
sociali, 1, Rome: Casa Libraria Editrice Italiana, 1932.
Sonnenschein, Hugo. F. (1972), Market Excess Demand Functions, Econo-
metrica, 40, 54963.
Sonnenschein, Hugo. F. (1973), Utility Hypothesis and Market Demand
Theory, Western Economic Journal, 11, 40410.
Slutsky, Eugen (1915), Sulla teoria del bilancio del consomatore, Giornale
degli Economisti e Rivista di Statistica, 51, July, 126.
Stackelberg, Heinrich von (1933), Zwei Kritische Bemerkungen zur Preisthe-
orie Gustav Cassels, Zeitschrift f ur National okonomie, 4, June, 45672.
Staley, Charles E. (1989), A History of Economic Thought. From Aristotle to
Arrow, Cambridge, MA: Basil Blackwell.
Steiner, Philippe (1999), Lentrepreneur par etien et la th eorie de laction,
Revue europ eenne des sciences sociales, 37 (116), 103118.
Stigler, George (1941), Production and Distribution Theories; The Formative
Period, New York: Macmillan Company.
Syll, Lars Plsson (1993), Wicksell on Harmony Economics: The Lau-
sanne School vs. Wicksell, Scandinavian Economic History Review, 41 (2),
Winter, 17288.
Tatti, Elena (2000),
Etre et devoir etre chez L eon Walras, in Dock` es et al.

2000, 41728.
Vacherot, Etienne (1858), La m etaphysique et la science ou principes de
m etaphysique positive; 2nd ed., 1863, 3 vols., Paris: Librairie de F. Chamerot.
van Daal, Jan and Donald A. Walker (1990), The Problem of Aggregation in
Walrass General EquilibriumTheory, History of Political Economy, 22 (3),
Fall, 489505; and in Walker 2001a, 2, 34157.
van Daal, Jan and Albert Jolink (1989), L eon Walrass Mathematical Eco-
nomics and the Mechanical Analogies, History of Economics Society
Bulletin, 11 (1), Spring, 2532; and in Walker 2001a, 1, 93100.
van Daal, Jan and Albert Jolink (1993), Walrass General General-
Equilibrium Model, in Robert F. H ebert (ed.), Themes on Economic Dis-
course, Method, Money and Trade: Selected Papers from the History of
P1: PIG
References 331
Economics Conference 1991, Perspectives on the History of Economic
Thought, 9, Aldershot: Edward Elgar for the History of Economics Society,
12136.
Wald, Abraham (19341935),
Uber die Produktionsgleichungen der okono-

mischen Wertlehre (II), Ergebnisse eines mathematischen Kolloquiums,
Heft 7; Leipzig and Wein: Frank Deuticke, 1936, 16; translated by William
J. Baumol as On the Production Equations of Economic Value Theory, Part
2, in Baumol and Goldfeld 1968, 28993.
Wald, Abraham (1936),
Uber einige Gleichungssysteme der mathematischen
Okonomie, Zeitschrift f ur National okonomie, 7 (5), 63770; translated by

Otto Eckstein as On Some Systems of Equations of Mathematical Eco-
nomics, Econometrica, 19, October, 1951, 368403.
Walker, Donald A. (1970), L eon Walras in the Light of His Correspondence
and Related Papers, Journal of Political Economy, 78 (4), July-August,
685701.
Walker, Donald A. (1971a), The Static Analysis of Disequilibrium Trans-
actions Markets, Indian Economic Journal, 18 (3), January-March, 305
327.
Walker, Donald A. (1971b), The Determinateness of Equilibrium in Isolated
Competitive Markets, International Review of Economics and Business, 18
(12), December, 11581179.
Walker, Donald A. (1983), William Jaff e, Ofcier de liaison intellectuel, in
The Craft of the Historian of Economic Thought, vol. 1 of Research in the
History of Economic Thought andMethodology, editedbyWarrenJ. Samuels,
Greenwich, CT: JAI Press, 1939.
Walker, Donald A. (1984), Is Walrass Theory of General Equilibrium a
Normative Scheme? History of Political Economy, 16 (3), Fall, 44569.
Walker, Donald A. (1987a) Walrass Theories of Tatonnement, Journal of
Political Economy, 95(4), 75874.
Walker, Donald A. (1987b), Walras, L eon (19341910), in Eatwell et al., 4,
85263.
Walker, Donald A. (1987c), Bibliography of the Writings of L eon Walras,
1st ed., History of Political Economy, 19 (4), 667702.
Walker, Donald A. (1989), A Primer on Walrasian Theories of Economic
Behavior, History of Economics Society Bulletin, 11, Spring, 123;
reprinted in Blaug 1992, 21436.
Walker, Donald A. (1990), Commentary by Donald A. Walker on The Lau-
sanne Tradition: Walras and Pareto, by Claude M enard, in Neoclassical
Economic Theory, 1870 to 1930, edited by Klaus Hennings and Warren J.
Samuels, Boston, Dordrecht, London: Kluwer Academic Publishers, 13750.
Walker, Donald A. (1991), Review of The Invisible Hand: Economic Equilib-
rium in the History of Science, by Bruna Ingrao and Giorgio Israel, 1990,
History of Political Economy, 23, no. 3, Fall, 56065.
Walker, Donald A. (1996), Walrass Market Models, Cambridge, New York,
and Melbourne: Cambridge University Press.
P1: PIG
332 References
Walker, Donald A. (1997a), The School of Lausanne, in Encyclopedia of
Keynesian Economics, edited by Thomas Cate, with associate editors David
Colander and Geoffrey Harcourt, Cheltenham, UK, and Brookeld, VT:
Edward Elgar Publishing Limited, 36165.
Walker, Donald A. (1997b), Advances in General Equilibrium Theory,
Cheltenham, UK, Lyme, CT: Edward Elgar Publishing Limited; in French,
translated by the author: La th eorie de l equilibre g en eral: De nouveaux
eclairages, Paris: Economica, 1999.
Walker, Donald A. (1998), L eon Walras, in The Handbook of Economic
Methodology, edited by John B. Davis, D. Wade Hands, and Uskali Maki,
Cheltenham, UK, Northampton, MA: Edward Elgar Publishing Limited,
54145.
Walker, Donald A. (ed.) (2000a), Equilibrium, 3 vols., in Kevin D. Hoover and
Mark Blaug (general eds.), Critical Ideas in Economics, no. 2, Cheltenham,
UK, and Northampton, MA: Edward Elgar Publishing Limited.
Walker, Donald A. (2000b), La relation entre la Bourse au XIX
e
si` ecle et
le mod` ele de L eon Walras dun march e organis e, in Dock` es et al. 2000,
44357.
Walker, Donald A. (ed.) (2001a), The Legacy of L eon Walras, 2 vols., in Steven
G. Medema, (series ed.), Intellectual Legacies in Modern Economics, Chel-
tenham, UK, and Northampton, MA: Edward Elgar Publishing Limited.
Walker, Donald A. (2001b), A factual account of the functioning of the
nineteenth-century Paris Bourse, European Journal of the History of Eco-
nomic Thought, 8 (2), 186207.
Walker, Donald A. (2003), Early General Equilibrium Economics: Walras,
Pareto, and Cassel, in A Companion to the History of Economic Thought,
edited by Warren J. Samuels, Jeff E. Biddle, and John B. Davis, Malden, MA,
Oxford, UK, Melbourne, Berlin: Blackwell Publishing Ltd., ch. 18, 27893.
Walras, Auguste (1831), De la nature de la richesse et de lorigine de la valeur,
Paris: Alexandre Johanneau.
Weintraub, E. Roy (1983), On the Existence of a Competitive Equilibrium:
19301954, Journal of Economic Literature, 21 (1), 139.
Wicksell, Knut (1893),

Uber Wert, Kapital und Rente, nach den neueren
national okonomischen Theorien, Jena: G. Fisher.
Wicksell, Knut (1898), Geldzins und G uterpreise bestimmenden Ursachen,
Jena: G. Fischer; translated by R. F. Kahn as Interest and Prices; A Study of
the Causes Regulating the Value of Money, London: Macmillan, 1936.
Wicksell, Knut (1901, 1906), Vorlesungen uber National okonomie; 2nd ed.
1911, translated by E. Classen as Lectures on Political Economy, 2 vols.,
London: Routledge & Kegan Paul, 1934.
Wicksell, Knut (1919), Professor Cassels System of Economics, in Wicksell
1934, appendix 1, 21957.
Wicksell, Knut (1934), Lectures on Political Economy, vol. 1, General Theory,
translated from the Swedish by E. Classen, London: Routledge and Kegan
Paul.
P1: PIG
References 333
Witteloostuijn, A. van and J. A. H. Maks (1988), Walras: A Hicksian avant la
lettre, Economie Appliqu ee, 41 (3), 595608.
Wieser, Friedrich von (1891), The Austrian School and the Theory of Value,
Economic Journal, 1, March, 10821.
Witteloostuijn, A. van and J. A. H. Maks (1990), Walras on Temporary Equi-
librium and Dynamics, History of Political Economy, 22 (2), 22337.
Wood, JohnC. (ed.) (1993), L eonWalras: Critical Assessments, 3vols., London:
Routledge.
Zatarin, Joseph de (1972), La th eorie des t atonnements chez Walras, Revue
dEconomie Politique, 82 (6), 122228.
Zawadzki, Wl adyslaw (1914), Les math ematiques appliqu ees ` a l economie
politique, Paris: M. Rivi` ere.
Zeuthen, Frederik L. B. (1928), Den konomiske Fordeling, Copenhagen: Busck.
Zeuthen, Frederik L. B. (1933), Das Princip der Knappheit, Technische Komi-
natin und

Okonomische Qualit at, Zeitschrift f ur National okonomie, 4, 124.
Zylberberg, Andr e (1988), L economie math ematique chez les actuaires
fran cais au temps de Walras (18201914), Oeconomia, 9, 3564.
Zylberberg, Andr e (1990), L economie math ematique en France 18701914,
Paris: Economica.
P1: PIG
0521858550ref CB1007/Walker 0 521 85855 0 November 24, 2005 0:39
References and Bibliography
for Chapter 11
I. Non-Virtual Walrasian models
A. Exchange models
Bottazzi, J.-M. (1994), Accessibility of Pareto Optima by Walrasian Exchange
Processes, Journal of Mathematical Economics, 23 (6), November, 585
603.
Bronfman, Corinne (2000), An Experimental Examination of the Walrasian
Tatonnement Mechanism, Bargaining and market behavior: Essays in
experimental economics, Cambridge, NewYork, and Melbourne: Cambridge
University Press, 381406.
Cordoba, Jose M. and Peter J. Hammond (1998), Asymptotically Strategy-
Proof Walrasian Exchange, Mathematical Social Sciences, 36 (3), Decem-
ber, 185212.
Grandmont, Jean-Michel (1992), Transformations of the Commodity Space,
Behavioral Heterogeneity, and the Aggregation Problem, Journal of
Economic Theory, 57 (1), June, 135.
Grodal, Birgit and Karl Vind (2000), Walras Equilibrium with Coordination,
Institute of Economics Working paper, University of Copenhagen.
Guesnerie, Roger (2002), Equilibre g en eral, coordination et multiplicit e
sur les marches au comptant, Revue dEconomie Politique, 112 (5),
SeptemberOctober, 67192.
Herings, Jean-Jacques (1997), Equilibrium Adjustment of Disequilibrium
Prices, Journal of Mathematical Economics, 27 (1), February, 53
77.
Ma, Jinpeng and Fusheng Nie (2003), Walrasian Equilibrium in an Exchange
Economy with Indivisibilities, Mathematical Social Sciences, 46 (2),
October, 15992.
Majumdar, Mukul and Vladimir Rotar (2000), Equilibrium Prices in a
Random Exchange Economy with Dependent Agents, Economic Theory,
15 (3), May, 53150.
Negishi, T. (1961), On the Formation of Prices, International Economic
Review, 2, 12226.
Peters, Michael (1997), On the Equivalence of Walrasian and Non-Walrasian
Equilibria in Contract Markets: The Case of Complete Contracts, Review
of Economic Studies, 64 (2), April, 24164.
334
P1: PIG
Uzawa, Hirofumi (1962), AWalrasian Approach to Bargaining Games, Inter-
national Economic Review, 3, 21832.
B. Some structural characteristics of non-virtual exchange
or production models
De Fraja, Gianni and Jozsef Sakovics (2001), Walras Retrouv e: Decentral-
ized Trading Mechanisms and the Competitive Price, Journal of Political
Economy, 109 (4), August, 84263.
Kandel, Eugene and Avi Simhon (2002), Between Search and Walras, Journal
of Labor Economics, 20 (1), January, 5985.
Koutsougeras, Leonidas C. (2003), Non-Walrasian Equilibria and the Law of
One Price, Journal of Economic Theory, 108 (1), January, 16975.
Sertel, Murat R. and Muhamet Yildiz (1999), Double-Edged Population
Monotonicity of Walrasian Equilibrium A Note on the Nature of Com-
petition, Mathematical Social Sciences, 37 (3), May, 23551.
Starr, Ross M. and Maxwell B. Stinchcombe (1993), Exchange in a Network
of Trading Posts, University of California, San Diego Department of Eco-
nomics Working Paper, 9313, April, 114.
C. Non-virtual models in which there is production
Baigent, G. Glenn (2003), Competitive Markets and Aggregate Information,
Eastern Economic Journal, 29 (4), 593606.
Benetti, Carlo (2002), Le probl` eme de la variation de prix: Les limites de la
th eorie walrassienne, Revue Economique, 53 (5), September, 91731.
Bennardo, Alberto and Pierre Andre Chiappori (2003), Bertrand and Walras
Equilibria under Moral Hazard, Journal of Political Economy, 111 (4),
August, 785817.
Berliant, Marcus and Sami Dakhlia (2002), Sensitivity Analysis for Applied
General Equilibrium Models in the Presence of Multiple Walrasian Equilib-
ria, Economic Theory, 19 (3), April, 45976.
Borissov, Kirill (2004), An Intertemporal General Equilibrium Model with
GivenReal Wage Rates, Structural Change andEconomic Dynamics, 15 (2),
June, 20733.
Brown, Donald J. and Chris Shannon (2000), Uniqueness, Stability, and Com-
parative Statics in Rationalizable Walrasian Markets, Econometrica, 68 (6),
November, 152939.
Dore, Mohammed H. I. (1998), Walrasian General Equilibrium and Nonlin-
ear Dynamics, Nonlinear Dynamics, Psychology, and Life Sciences, 2 (1),
January, 5972.
Fisher, Franklin M. (1983), Disequilibrium Foundations of Equilibrium Eco-
nomics, Cambridge: Cambridge University Press.
Flaschel, Peter (1991), Stability Independent of Economic Structure? A
Prototype Analysis, Structural Change and Economic Dynamics, 2 (1),
935.
P1: PIG
336 References and Bibliography for Chapter 11
Funk, Peter (1995), Bertrand and Walras Equilibria in Large Economies,
Journal of Economic Theory, 67 (2), December, 43666.
Gul, Faruk and Ennio Stacchetti (1999), Walrasian Equilibrium with Gross
Substitutes, Journal of Economic Theory, 87 (1), July, 95124.
Hahn, Frank H. (1962), On the Stability of Pure Exchange Equilibrium,
International Economic Review, 3, 20613.
Hens, Thorsten (1997), Stability of Tatonnement Processes of Short Period
Equilibria with Rational Expectations, Journal of Mathematical Economics,
28 (1), August, 4167.
Herings, Jean-Jacques (1997), A Globally and Universally Stable Price
Adjustment Process, Journal of Mathematical Economics, 27 (2), March,
16393.
Horiba, Yutaka (1998), Gross Substitutability and the Laws of Comparative
Statics: ASimultaneous Demand Shift, Keio Economic Studies, 35 (2), 17.
Jensen, Bjarne S. (2003), Walrasian General Equilibrium Allocations and
Dynamics in Two-Sector Growth Models, German Economic Review, 4 (1),
February, 5387.
Katzner, Donald W. (1999), Methodological Individualism and the Walrasian
Tatonnement, Journal of Economic and Social Research, 1 (1), 533.
Majumdar, Mukul and Tapan Mitra (1997), Complexities of Concrete Wal-
rasian Systems, Issues in economic theory and public policy: Essays in
honour of Professor Tapas Majumdar, 4163, Oxford and NewYork: Oxford
University Press, Delhi.
Podczeck, Konrad (2003), Core and Walrasian Equilibria When Agents Char-
acteristics Are Extremely Dispersed, Economic Theory, 22 (4), November,
699725.
Rao, J. Mohan (1998), A Leibenstein-Hobbes-Walras Model of Sharecrop-
ping, Metroeconomica, 49 (1), February, 6296.
Serrano, Roberto and Oscar Volij (1998), Axiomatizations of Neoclassi-
cal Concepts for Economies, Journal of Mathematical Economics, 30 (1),
August, 87108.
II. Uses and developments of special ideas of Walrass
A. Walrass law
Aiyagari, S. Rao (1992), Walras Law and Nonoptimal Equilibria in Over-
lapping Generations Models, Journal of Mathematical Economics, 21 (4),
34361.
Balasko, Yves (2004), The Equilibrium Manifold Keeps the Memory of Indi-
vidual Demand Functions, Economic Theory, 24 (3), October, 493501.
Bidard, Christian (1989), Equilibrium with a Qualitative Walras Law, Jour-
nal of Economic Theory, 47 (1), February, 20305.
Habibagahi, H. and J. Quirk (1989), Hicksian Stability and Walras Law, in
JohnCunninghamWoodandRonaldN. Woods (eds.), Sir JohnR. Hicks: Crit-
ical Assessments, 2, in Critical Assessments of Contemporary Economists,
London and New York: Routledge, 24961.
P1: PIG
Huang, Weihong (2001), Statistical Dynamics and Walras Law, Journal of
Economic Behavior and Organization, 46 (1), September, 5771.
Lange, Oskar R. (1942), Says Law: A Restatement and Criticism, in Oscar
Lange, et al. (eds.), Studies in Mathematical Economics and Econometrics,
Chicago: University of Chicago Press, 4968.
Rhodes, James R. (1984), Walras Law and Clowers Inequality, Australian
Economic Papers, 23 (42), June, 11222.
Yeager, Leland B. and Alan A. Rabin (1997), Monetary Aspects of Wal-
rass Law and the Stock-Flow Problem, Atlantic Economic Journal, 25 (1),
March, 1836.
B. Walras allocations
Armstrong, Thomas E. and Marcel K. Richter (1984), The Core-Walras Equiv-
alence, Journal of Economic Theory, 33 (1), June, 11651.
Armstrong, Thomas E. and Marcel K. Richter (1986), Existence of Nonatomic
Core-Walras Allocations, Journal of Economic Theory, 38 (1), February,
13759.
Herves-Beloso, Carlos and Emma Moreno-Garcia (2004), El poder de veto
de la gran coalici on, Investigaci ones Econ omicas, 28 (3), September, 489
513.
Manresa, Antonio (2002), Can We Identify Walrasian Allocations?, Review
of Economic Design, 7 (1), June, 5773.
Nagahisa, Ryo-ichi (1991), A Local Independence Condition for Characteri-
zation of Walrasian Allocations Rule, Journal of Economic Theory, 54 (1),
June, 10623.
Nagahisa, Ryo-ichi (1992), Walrasian Social Choice in a Large Economy,
Mathematical Social Sciences, 24 (1), August, 7378.
Nagahisa, Ryo-ichi (1994), A Necessary and Sufcient Condition for Wal-
rasian Social Choice, Journal of Economic Theory, 62 (1), February, 186
208.
Nagahisa, Ryo-ichi and Sang-Chul Suh (1995), A Characterization of the
Walras Rule, Social Choice and Welfare, 12 (4), October, 33552.
Nouweland, A. van den, B. Peleg and S. Tijs (1996), Axiomatic Characteriza-
tions of the Walras Correspondence for Generalized Economies, Journal of
Mathematical Economics, 25 (3), April, 35572.
Serrano, Roberto and Oscar Volij (1997), Abstract Equilibria, Interactive
Choice Sets and Walrasian Allocations, Department of Economics Working
Paper, April, Brown University.
Serrano, Roberto and Oscar Volij (2000), Walrasian Allocations without Price-
Taking Behavior, Journal of Economic Theory, 95 (1), November, 79106.
Tauman, Yair and Yan Chen (2000), Acceptable and Walrasian Allocations,
Journal of Mathematical Economics, 34 (3), November, 41538.
Tian, Guoqiang (2000), Feasible and Continuous Double Implementation of
Constrained Walrasian Allocations, Annals of Economics and Finance,
1 (1), May, 1932.
P1: PIG
Yamazaki, Akira (1984), Walras Degrees and Probability of a Blocking Coali-
tion at Pareto Allocations, Journal of Mathematical Economics, 13 (2),
October, 10521.
C. The Walras correspondence
Balasko, Yves (1975), The Graph of the Walras Correspondence, Economet-
rica, 43 (56), SeptemberNovember, 90712.
DeMichelis, Stefano and Fabrizio Germano (2000), Some Consequences of
the Unknottedness of the Walras Correspondence, Journal of Mathematical
Economics, 34 (4), December, 53745.
Jouini, Elyes (1993), The Graph of the Walras Correspondence: The Pro-
duction Economies Case, Journal of Mathematical Economics, 22 (2),
13947.
Korthues, Bernd (2000), Characterization of an Extended Walrasian Concept
for Open Economies, Journal of Mathematical Economics, 33 (4), May,
44961.
Majumdar, Mukul and Bezalel Peleg (1997), An Axiomatization of the Walras
Correspondence in Innite Dimensional Spaces, International Economic
Review, 38 (4), November, 85364.
Saijo, Tatsuyoshi, Yoshikatsu Tatamitani and Takehiko Yamato (1999), Char-
acterizing Natural Implementability: The Fair and Walrasian Correspon-
dences, Games and Economic Behavior, 28 (2), August, 27193.
III. Imperfectly competitive non-virtual general
equilibrium models
A. Oligopoly, Cournot-Walras, and monopolistic models
Alos-Ferrer, Carlos (2004), Cournot versus Walras in Dynamic Oligopolies
with Memory, International Journal of Industrial Organization, 22 (2),
February, 193217.
Benassy, Jean-Pascal (1976), The Disequilibrium Approach to Monopolistic
Price Setting and General Monopolistic Equilibrium, Review of Economic
Studies, 43, February, 6981.
Bobo, Robert J. Gary (1989), Cournot-Walras and Locally Consistent Equi-
libria, Journal of Economic Theory, 49 (1), October, 1032.
Bobo, Robert J. Gary and Philippe Michel (1991), Informative Advertising and
Competition: ANoncooperative Approach, International Economic Review,
32 (2), May, 32139.
Codognato, Giulio (1994), Cournot-Walras Equilibrium: A Reconsideration,
Economic Notes, 23 (3), 388401.
Codognato, Giulio (1995), Cournot-Walras and Cournot Equilibria in Mixed
Markets: A Comparison, Economic Theory, 5 (2), March, 36170.
Codognato, Giulio and Jean J. Gabszewicz (1993), Cournot-Walras Equilibria
in Markets with a Continuum of Traders, Economic Theory, 3 (3), July,
45364.
P1: PIG
Cordella, Tito and Manjira Datta (2002), Intertemporal Cournot and Walras
Equilibria: An Illustration, International Economic Review, 43 (1), Febru-
ary, 13753.
DAspremont, Claude, Rodolphe Dos Santos Ferreira, andLouis-Andr e G erard-
Varet (1997), General Equilibrium Concepts under Imperfect Competition:
A Cournotian Approach, Journal of Economic Theory, 73 (1), March, 199
230.
Huang, Weihong (2003), A Naive but Optimal Route to Walrasian Behavior
in Oligopolies, Journal of Economic Behavior and Organization, 52 (4),
December, 55371.
Kaas, Leo and Paul Madden (2000), Imperfectly Competitive Cycles with
Keynesian and Walrasian Features, Institute for Advanced Studies, Vienna,
Working paper, Economics Series, no. 83.
Kaas, Leo (2001), Cournot-Walras Equilibrium without Prot Feedback,
Economics Bulletin, 4 (9), 18.
Morishima, Michio and Madhavi Majumdar (1996), The Cournot-Walras
Arbitrage, Resource Consuming Exchange, and Competitive Equilib-
rium, in Dynamic Economic Theory, Cambridge, New York, Melbourne:
Cambridge University Press, 27084.
Novshek, William and Hugo Sonnenschein (1978), Cournot and Walras Equi-
librium, Journal of Economic Theory, 19(2), December, 22366, in Andrew
F. Daughety (ed.) Cournot Oligopoly: Characterization and Applications,
Cambridge; New York, and Melbourne: Cambridge University Press, 271
316.
Robert, Jacques (1994), In the Cournot-Walras General Equilibrium Model,
There May Be More to Gain by Changing the Numeraire Than by Elimi-
nating Imperfections: A Two-Good Economy Example: Comment, in Jean
Mercenier and T. N. Srinivasan (eds.), Applied General Equilibrium and
Economic Development: Present Achievements and Future Trends, Ann
Arbor: University of Michigan Press, 22531.
Schenk-Hoppe, Klaus Reiner (2000), The Evolution of Walrasian Behavior in
Oligopolies, Journal of Mathematical Economics, 33 (1), February, 3555.
Siebe, Wilfrid (1990), General Equilibriumwith a Continuumof Oligopolies,
Universitat Bonn Sonderforschungsbereich 303, working paper, May, 176.
Wooders, John (19971998), Equilibrium in a Market with Intermediation Is
Walrasian, Review of Economic Design, 3 (1), 7589.
Yosha, Oved (1997), Diversication and Competition: Financial Intermedi-
ation in a Large Cournot-Walras Economy, Journal of Economic Theory,
75 (1), July, 6488.
B. Nash-Walras, games and bargaining models
Allen, Beth and Heracles Polemarchakis (1994), From Nash to Walras Equi-
librium, in Jean-Fran cois Mertens and Sylvain Sorin (eds.), Game-theoretic
Methods in General Equilibrium Analysis, NATO Advanced Science Insti-
tutes Series D: Behavioural and Social Sciences, 77, Dordrecht and Boston:
P1: PIG
Kluwer Academic in cooperation with NATO Scientic Affairs Division,
22541.
Alos-Ferrer, Carlos and Ana B. Ania (2001), Local Equilibria in Economic
Games, Economics Letters, 70 (2), February, 16573.
Bevia, Carmen, Luis C. Corchon and Simon Wilkie (2003), Implementation
of the Walrasian Correspondence by Market Games, Review of Economic
Design, 7 (4), February, 42942.
Corominas-Bosch, Margarida (2004), Bargaining in a Network of Buyers and
Sellers, Journal of Economic Theory, 115 (1), March, 3677.
Dagan, Nir, Roberto Serrano and Oscar Volij (1998), Comment on McLennan
and Sonnenschein Sequential Bargaining as a Noncooperative Foundation
for Walrasian Equilibrium, Econometrica, 66 (5), September, 123133.
Dubey, Pradeep and John Geanakoplos (2003), From Nash to Walras via
Shapley-Shubik, Journal of Mathematical Economics, 39 (56), July, 391
400.
Faias, Marta, Emma Moreno-Garcia and Mario Rui Pascoa (2002), Non-
manipulability in Walrasian Cost Games, Reviewof Economic Design, 7 (1),
June, 93104.
Geanakoplos, John (2003), Nash and Walras Equilibrium via Brouwer,
Economic Theory, 21 (23), March, 585603.
Ghosal, S. and H. M. Polemarchakis (1997), Nash-Walras Equilibria, Re-
search in Economics, 51 (1), April, 3140.
Kunimoto, Takashi and Roberto Serrano (2004), Bargaining and Competition
Revisited, Journal of Economic Theory, 115 (1), March, 7888.
Minelli, Enrico and Heracles M. Polemarchakis (1999), Nash-Walras Equilib-
ria of a Large Economy, Universit e Catholique de Louvain CORE Discus-
sion Paper, July.
Ponsati, Clara (2004), Search and Bargaining in Large Markets with Homoge-
neous Traders, Contributions to Theoretical Economics, 4 (1), 1st quarter,
125.
Sabourian, Hamid (1999), Bargaining and Markets: Complexity and the Wal-
rasian Outcome, Yale Cowles Foundation Discussion Paper, December.
Serrano, Roberto (2002), Decentralized Information and the Walrasian
Outcome: A Pairwise Meetings Market with Private Values, Journal of
Mathematical Economics, 38 (12), September, 6589.
Sinha, Deepak K. (1997), Diffusion of Innovations in Globally Stable
Walrasian Equilibrium, Journal of Economics (Zeitschrift f ur National-
okonomie), 66 (1), 122.
Sonn, Sang-Young (2001), Bargaining and Walrasian Equilibrium in Simple
Production Economies, Seoul Journal of Economics, 14 (4), Winter, 379
405.
Szymanski, Stefan (2004), Professional Team Sports Are Only a Game:
The Walrasian Fixed-Supply Conjecture Model, Contest-Nash Equilibrium,
and the Invariance Principle, Journal of Sports Economics, 5 (2), May,
11126.
P1: PIG
Trockel, Walter (1996), A Walrasian Approach to Bargaining Games, Eco-
nomics Letters, 51 (3), 295301.
Trockel, Walter (2000), Implementations of the Nash Solution Based on
Its Walrasian Characterization, Economic Theory, 16 (2), September,
27794.
Wooders, John (1998), Walrasian Equilibrium in Matching Models, Mathe-
matical Social Sciences, 35 (3), May, 24559.
Yildiz, Muhamet (2003), WalrasianBargaining, Games andEconomic Behav-
ior, 45 (2), November, 46587.
C. Fix-price models
Benassy, Jean-Pascal (2000), Rigidit es nominales dans les mod` eles d equilibre
g en eral intertemporel stochastique, LActualit e Economique/Revue DAna-
lyse Economique, 78 (4), December, 42357.
De Vroey, Michel (2004), Th eorie de d es equilibre et chomage involontaire:
Un examen critique, Revue Economique, 55 (4), July, 64768.
Dr` eze, Jacques (1975), Existence of an Exchange Equilibrium under Price
Rigidities, International Economic Review, 16, June, 30120.
Grandmont, Jean-Michel (1982), Temporary General Equilibrium Theory, in
KennethJ. ArrowandMichael D. Intriligator (eds.), Handbook of Mathemati-
cal Economics: Volume II, Amsterdam: North-Holland Publishing Company,
879922.
Herings, P. Jean-Jacques, Gerard van der Laan and Dolf Talman (1998), Price-
Quantity Adjustment in a Keynesian Economy, Tilburg Center for Economic
Research Discussion Paper, November.
Herings, Jean-Jacques, Gerard van der Laan and Richard Venniker (1998), The
Transition from a Dreze Equilibrium to a Walrasian Equilibrium, Journal
of Mathematical Economics, 29 (3), April, 30330.
Kades, Eric (1985), New Classical and New Keynesian Models of Business
Cycles, Economic Review, Federal Reserve Bank of Cleveland, Quarter IV,
2035.
IV. Micro- and macro-economics: blending the ideas
of Walras and J. M. Keynes
Allen, Ralph C. and Jack H. Stone (1998), Walrasian, Non-Walrasian and Post-
Walrasian Perspectives on the ADAS Model: Theoretical and Pedagogical
Implications, Aggregate demand and supply: Acritique of orthodox macroe-
conomic modelling, NewYork: St. Martins Press; London: Macmillan Press,
6781.
Backhouse, Roger E. (1998), The Neo-Walrasian Research Programme in
Macroeconomics, in Explorations in economic methodology: FromLakatos
to empirical philosophy of science, in Frontiers of Political Economy, 17,
London and New York: Routledge, 1338.
P1: PIG
Benassy, Jean-Pascal (2002), The macroeconomics of imperfect competition
and nonclearing markets: A dynamic general equilibrium approach, Cam-
bridge and London: MIT Press.
Citanna, Alessandro (2001), Continua of Underemployment Equilibria
Reecting Coordination Failures, also at Walrasian Prices, Journal of
Mathematical Economics, 36 (3), December, 169200.
Colander, David (1999), A Post-Walrasian Explanation of Wage and Price
Inexibility and a Keynesian Unemployment Equilibrium System, Growth,
employment and ination: Essays in honour of John Cornwall, New York:
St. Martins Press; London: Macmillan Press, 21125.
De Vroey, Michel (1999), Keynes and the Marshall-Walras Divide, Journal
of the History of Economic Thought, 21 (2), June, 11736.
Dr` eze, Jacques (1997), Walras-Keynes Equilibria Coordination and Macroe-
conomics, European Economic Review, 41 (9), December 173562.
Kaas, Leo (1998), Stabilizing Chaos in a Dynamic Macroeconomic Model,
Journal of Economic Behavior and Organization, 33 (34), January,
31332.
Kaas, Leo and Paul Madden (2000), Imperfectly Competitive Cycles with
Keynesian and Walrasian Features, Institute for Advanced Studies, Vienna,
Working paper, Economics Series, no. 83.
Palley, Thomas I. (1998), Walras Law and Keynesian Macroeconomics,
Australian Economic Papers, 37 (3), September, 33040.
Palley, Thomas I. (1999), General DisequilibriumAnalysis with Inside Debt,
Journal of Macroeconomics, 21 (4), Fall, 785803.
V. Applied and computational general equilibrium models
Brown, D. J. and R. L. Matzkin (1996), Testable Restrictions on the Equilib-
rium Manifold, Econometrica, 64 (6), November, 124962.
Burniaux, Jean Marc et al. (1990), Economy-Wide Effects of Agricultural
Policies in OECD Countries: A GE Approach Using the WALRAS Model,
in Ian Goldin and Odin Knudsen (eds.), Agricultural Trade Liberalization:
Implications for Developing Countries, Paris: Organisation for Economic
Co-operation and Development; Washington, D.C.: World Bank, 283
306.
Burniaux, Jean-Marc, Fran cois Delorm e, IanLienert, andJohnP. Martin(1990),
WALRAS A Multi-sector, Multi-country Applied General Equilibrium
Model for Quantifying the Economy-wide Effects of Agricultural Policies,
OECD Economic Studies, 13, Winter, 69102.
Carvajal, Andres (2004), Testable Restrictions on the Equilibrium Manifold
under Random Preferences, Journal of Mathematical Economics, 40 (12),
February, 12143.
Chang, S.-I. (1997), The Effects of Economic Integration between North and
South Korea: A Computable General Equilibrium Analysis, International
Economic Journal, 11 (4), Winter, 116.
P1: PIG
Cheng, John Q. and Michael P. Wellman (1998), The WALRAS Algorithm: A
Convergent Distributed Implementation of General Equilibrium Outcomes,
Computational Economics, 12 (1), August, 124.
Delorm e, Fran cois and Dominique van der Mensbrugghe (1989), Assessing the
Role of Scale Economies and Imperfect Competition in the Context of Agri-
cultural Trade Liberalisation: A Canadian Case Study, OECD Economic
Studies, 13, Winter, 20536.
Dhillon, Upinder S., Dennis J. Lasser, and Taiji Watanabe (1997), Volatil-
ity, Information, and Double versus Walrasian Auction Pricing in US and
Japanese Futures Markets, Journal of Banking and Finance, 21 (7), July,
104561.
Gottinger, H. W. (1998), Greenhouse Gas Economics and Computable General
Equilibrium, Journal of Policy Modeling, 20 (5), October, 53780.
Leontief, Wassily (1941), The Structure of American Economy, 19191929,
Cambridge, MA: Harvard University Press.
Lienert, Ian (1989), Quantifying Agricultural Policies in the WALRAS
Model, OECD Economic Studies, 13, Winter, 10330.
List, John A. (2004), Testing Neoclassical Competitive Theory in Multi
lateral Decentralized Markets, Journal of Political Economy, 112 (5),
October, 113156.
Martin, John P. (19891990), Economy-Wide Effects of Agricultural Policies
in OECD Countries: Simulation Results with WALRAS, OECD Economic
Studies, 13, Winter, 13172.
Noland, Marcus (2000), Avoiding the Apocalypse: The Future of the Two
Koreas, Washington, D.C.: Institute for International Economics.
Mensbrugghe, Dominique van der, John P. Martin, and Jean Marc Burniaux
(1989), How Robust Are WALRAS Results?, OECD Economic Studies,
13, Winter, 173204.
Piggott, John and John Whalley (1985), NewDevelopments in Applied General
Equilibrium Analysis, New York: Cambridge University Press.
Rao, J. Mohan (1998), A Leibenstein-Hobbes-Walras Model of Sharecrop-
ping, Metroeconomica, 49 (1), February, 6296.
Scarf, Herbert E. (1973), The Computation of Economic Equilibria, New
Haven: Yale University Press.
Scarf, Herbert E. and John B. Shoven (eds.) (1984), Applied General Equilib-
rium Analysis, New York: Cambridge University Press.
Yao, Shuntian (2002), Walrasian Equilibrium Computation, Network For-
mation, and the Wen Theorem, Review of Development Economics, 6 (3),
October, 41527.
P1: PIG
344
P1: IWV
0521858550ind CB1007/Walker 0 521 85855 0 November 24, 2005 4:1
Index
abstract sciences, 38, 39, 43
accounting/inventory control, 93, 9697
activity analysis model of production,
301
adjustment process
inequality of average cost/price effect
on, 152
innovation effect on, 284
path dependency and, 11, 163
in virtual model, 11, 306
without parametric changes, 148149.
See also tatonnement, in mature
comprehensive model
agents. See capitalist; entrepreneur
Allais, Maurice, 306
allocation, of resources, 157, 169170,
272273
Amoroso, Luigi, 279
Antonelli, Etienne, 279
Antonelli, Giovanni Battista, 279
applied economics, 86, 98102
applied sciences, Walras on, 4547
applied ethical science, 46
applied natural science, 4546
industry and, 46
mores and, 46
archrationalism thesis, 2829, 40, 45, 71,
8586
Arena, Richard, 179
Arrow, Kenneth, 150, 306308, 309310
asset holding, as real economic behavior,
93
assumptions, in theories, 5657
astronomy, inuence on Walras, 56, 73
atomless exchange economy, 308309
auctioneer
absence in Walras models, 150
ctional, in Arrow/Hahn models, 308
Aupetit, Albert, 187, 279
Auspitz, Rudolph, 1415
average cost of production, 67
change in quantity produced and, 165
consumer demand and, 152, 169
entrepreneur role in, 143144
equation for, 143
equilibrium price and, 154
new durable capital goods and, 157
Pareto and, 272
prots and, 144, 145, 148
supply functions and, 152
written pledges sketch and,
178179
Bakounine, Allesandrina, 260
bank notes, need for mandatory volume
of, 109
Barone, Enrico, 14, 271, 279, 289
barter vs. money, in Walras model, 92
bibliography of Walras works. See
denitive bibliography of
Walras writings
Bidard, Christian, 308
Blaug, Mark, 12, 31, 288
Bliss, Christopher, 311
Bompaire, Fran cois, 7
Boninsegni, Pasquale, 279
bons, Walras use of, 176
Bourse, incorporated into mature
comprehensive model,
141142, 166
Bousquet, Georges-Henri, 200, 264
Bouvier, Emile, 108
Bovert, Pierre, 280
Bowley, Arthur, 279
Bresciani-Turroni, Constantino, 279
Bridel, Pascal, 29, 33
on productivity of Walras, 183, 189,
190
on Walras monetary theory, 183
business cycle theory, 110
capital accumulation
tatonnement and, 154155, 159161.
See also savings
345
P1: IWV
346 Index
capital goods
average cost of production and, 157
entrepreneur role and, 143144,
154155, 271272, 284
maximum utility of new, 274
new, not used during tatonnement,
158159
rate of net income from, 155, 282
reprise in tatonnement and, 156157
disequilibrium production and,
156157
existing/new capital goods, 157
time element in production effect on,
283
in Walrass law, 153. See also
tatonnement
capital-good services, 157
capitalist, entrepreneur as, 146147
capital theory, non-aggregative, 301
cash balances, 167, 168169, 184
Cassel, Karl Gustav, 289, 290292, 294,
301
approach/theoretical foundations of,
292293
background of, 290292
models of, 293295, 309
character/limitations in, 295296
dynamized model, 294295
as inuential, 296
stationary state model, 294
virtual model of pure exchange, 294
scarcity/marginal utility and, 297
Walrasian equation system and, 300
Centre Auguste et L eon Walras
(TRIANGLE), 199
Cheysson, Jean, 106
circulating balances, 168169
circulating capital/money, markets for
absence of written pledges in, 179, 182
incomplete model of in 4th edition, 183
coefcients of production, 93, 96, 142,
144, 273, 293
commercial credit institution, 110
commodities
ctional perpetual net income
commodity, 81
functional relation with utility, 131,
132133
indivisible, 309
net income, 80
Walras evaluation of Mill on, 94
commodity money
determining price of, 7980
unique vs. double standard for,
100101
commodity substitution, 79
competition
denition of, 6567
imperfect, 308
pure, 275, 285, 300, 302. See also free
competition; monopolies
competitive equilibrium
Arrow on, 307
Koopmans on, 301
Wicksell on, 282283
Comte, Auguste, 261
concrete facts, 266
concrete particular facts vs. essences,
3235
Condillac, Etienne Bonnot de, 94
consumable services, written pledges
and, 179
consumer demand
average cost of production and, 152,
169
theory of, 8, 267268
corporations, system/regulation of, 110
costs, xed, 9192
Cournot, Antoine-Augustin, 4, 9, 27, 58,
187
Cours d economie politique (Pareto),
264
Cousin, Victor, 46, 98
Cowles, Alfred, 314315
credit institution, commercial, 110
credit markets, 154155
Creedy, John, 135
Currie, Martin, 149, 167
DAgata, Antonio, 309
Darwin, Charles, 65, 261
Debreu, G erard, 140, 307308, 309310
decade of sacred fertility, 9
deduction
in Bacon-Mill tradition, 43
hypothetico-deductive method, 43
in science, 2627, 30, 31, 32, 37, 38
Walras meaning of, 41, 56. See also
induction/deduction
denitive bibliography of Walras
writings
comparison with rst edition, 205206
compilations of jottings/brief notes in,
201
completeness of entries in, 16,
201202
completeness of list of entries in,
200201
El ements citations, 1718
form of entries, 206
P1: IWV
Index 347
omission/exceptions to omissions of
uvres economiques
compl` etes in, 1617
previous autobibliography by Walras,
199200
previous bibliography by Bousquet,
200
sources for information in, 202
19th century works, 207243
20th century works, 243255
unpublished manuscripts in, 201. See
also Writings acknowledging
Walrass inuence
Del Vecchio, Gustavo, 279
demand
average cost of production and, 152,
169
consumer demand theory, 8, 267268
demand curves
de Zatarin on,
individual, 135, 137
market/competition and, 6567
demand curves, market,
de Molinari, Georges, 263
Descartes, Ren e, 26, 27, 28, 45, 47
de Vroey, Michel, 183, 190
de Zatarin, Joseph, 180181
Diemer, Arnaud, 66
Divisia, Fran cois, 280
division of labor, 124
Dupuit, Jules, inuence on Walras, 58
durable goods, new
average cost of production and, 157
reprise in tatonnement and production
of, 156157
dynamic economics, Hicks on
foundations of, 303306
dynamic general equilibrium system
Moore on, 285
Pareto on, 271
virtual
Cassel and, 294295
von Neumann and, 300302
dynamization, 294
economic growth
Cassel and theory of, 294295
mature comprehensive model and, 161
steady-state growth model, 295
economic justice, 98
economic philosophy/methodology, of
Walras. See mature
comprehensive model, steps in
development of
economics
as experimental science, 70
as natural science, 62
as observational/experimental, 71
Econometric Society, 314315
economics of rm, 8, 142143, 273274.
See also production
economic system, relationships between
parts of, 5
economic theory
as basis for social/economic justice, 98
as theory of social wealth, 66
of Walras as positive, 61, 6263. See
also individual theory
Edgeworth, Francis Y., 267, 268, 271
Einaudi, Luigi, 279
elementary phenomena, 34
El ements d economie politique pure
(Walras)
Bousquet on difculty of, 264
economic reality in, 6061
erroneous interpretation of Walras as
archrationalist, 8586
Jaff e on, 60
English theory of interest, 95
entrepreneur, 142147
costs/output and prots/losses, 157
denitions of, 144, 146
function/role of, 143146
adjust production to achieve
equilibrium, 145
change input proportions to
maximize output, 144
choose technology for production
process, 144
combine inputs to produce output,
144
connect real input/output markets,
143144
direct resource ow to protable
lines of production, 144
direct transformation of services
into products, 144
as managerial worker, 145
order output diminished/increased,
144
redirect resources to alter output
composition, 157
as ideal/real type, 9596
income of, 144146
in equilibrium, 145, 146
losses and, 144, 146147
Say theory on, 5
use of savings in capital-goods
industry, 154155
P1: IWV
348 Index
epistemological idealism, 53
equations, general equilibrium
static, 13, 294
uniqueness of solutions to, 299300
equation system, Walrass
Schlesinger and, 300
Zeuthen and, 300
equilibrium
competitive
Arrow on, 307
Koopmans on, 301
Wicksell on, 282283
existence of
Hicks on, 302303, 309
mathematical treatment in mature
comprehensive model,
162163
Pareto on, 269
Wald on, 298299, 309
presumed convergence to, 148150,
165
stability of (see stability). See also
tatonnement, in mature
comprehensive model
equilibrium of exchange
as ideal, 77, 148
non-virtual behavior and, 175
equilibrium prices, law of the variation
of, 68
exchange economy
atomless, 308309
mathematical model of Isnard, 5
pure, 309
in temporarily isolated freely
competitive market, 8
exp erience/exp erimentales, 4243, 69
experimental science, 43, 56, 7071
facts
concrete, 266
3235
general, 31, 3536, 37, 4245
humanistic, 3435
natural, 3435, 6263, 69, 88
personal, 8788
reality and, 4142
vs. theory, 35
Fanno, Marco, 279
at-money model, 168169
ctional perpetual net income
commodity, 81
ctional written pledges, 81
rm, economics of. See economics of
rm
Fisher, Irving, 1415, 279
xed costs, free competition and, 9192
Fontaine, Philippe, 113114
Franke, Reiner, 308
free competition
analyzing with pure theory, 99100
determining rate of production and,
as fact, 110112
hypothetical regime of, 6365
large-scale enterprises and, 293
as not self-sustaining, 9192
price determination and, 61, 62, 7879,
135137
production/exchange and, 6667
speculators and, 271272
welfare economics and, 170, 283284
free inputs, 297
freely competitive markets
construction of model, 7382
assumption all markets freely
competitive, 7375
ctional vs. real types, 8081
induction/deduction, 7678
realism in/applications of, 82
real referents of model components,
7576
in practice, 7880
exchange model in temporarily
isolated, 8
ubiquity of in real economy, 7273
free will, human nature and, 118124
free will as empirical fact, 119120
free will as notion/not empirical fact,
119
free will/determinism and human
behavior, 121
free will vs. determinism, 120121
Gambino, Enrico, 280
game theory, 307
general equilibrium
initial comprehensive model of, 910.
See also last comprehensive
model; mature comprehensive
model
general equilibrium equations
static equations, 13, 294
uniqueness of solutions to, 299300
general equilibrium theory, inuence of
on non-virtual economic systems,
1415
on virtual competitive models, 15
(see also virtual Walrasian
modeling)
P1: IWV
Index 349
general facts, 31, 3536, 37, 4245
general socio-economic equilibrium, of
Pareto, 276277
general theory of society, 83
geometrical forms, Walras on, 4749, 51,
52
German scholarship, philosophy of
science in, 4647
Giddings, Franklin H., 38
Gijsel, Peter de, 99
Gossen, H.-H., 114, 128129
Grotius, Hugo,
Hahn, Frank, 308, 309310, 311,
312
harmony economics, 283284
Hauenschild, Nils, 309
health, of Walras, 183187, 190191
Herschel, John F. W., 3536
Hicks, J. R., 12, 169, 177, 302305,
306
on marginal utility, 12
on Samuelson, 289
Walras inuence on
all markets purely competitive,
302
counting procedure, 303
existence of equilibrium, 302303,
309
interrelatedness of economic
variables, 302
model in essence virtual, 303305
Hilton, H. C., 167
Hobbes, Thomas, 39
Hollander, Samuel, 3031
homo conomicus, 138139
connection with functioning of
markets, 131134
functional relation between
commodity/utility, 131
maximization of utility through
economy/prudence, 133134
properties of utility, 132133
consequences of characteristics of,
135137
price determination in freely
competitive markets, 135137
supply/demand, 135
Hotelling, Harold, 289
humanistic facts, 3435
human nature, 138139
facets of, 114118, 138
animal/human nature distinction,
114115
essential unity in behavior, 117118
malleability, 115116
multiple facets, 115
free will and, 118124
free will as empirical fact, 119120
free will as notion/not empirical
fact, 119
free will/determinism and human
behavior, 121
free will vs. determinism, 120121
general features applicable to society,
126128
introduction to, 113114
society and individual, 124126
critique of individualism, 124125
human nature as basis of social
phenomena, 125126
use of scientic methods in studying,
137138
in Walras economic models, 128131.
See also homo conomicus
Hume, David, 117
hypothetico-deductive method, 43
idealism thesis, facts in, 6263
idealist, Jaff e on Walras as, 113114
ideal society of Walras, 81
ideal types, abstraction of, 7880
Iimura, Takuya, 309
imaginary, Walras use of, 80
imagination, Walras on, 4749, 5051,
52
imperfect competition, 308
income commodity, net, 80
individual demand curves, 135, 137
indivisible commodities, 309
induction/deduction
in construction of freely competitive
model, 7678
in natural sciences, 40. See also
deduction
induction/inductivism, 43
Ingrao, Bruna, 312
initial comprehensive model of general
equilibrium, 910
innovating entrepreneur, 284
inputs/outputs
combining inputs to maximize prots,
144
combining inputs to produce output,
144
connecting real input/output markets,
143144
free inputs, 297
problem with in written pledge sketch,
179
P1: IWV
350 Index
interest
market/natural rates of, 282283
Walras on English theory of interest, 95
Interest and Prices (Wicksell), 282
intimate reality, in economic theory,
8788
inventory control/accounting, 93, 9697
Isnard, Achylle Nicolas, 5
Israel, Giorgio, 312
Jaff e, William, 12
on El ements, 60
on tatonnement, 149
on Walras as idealist, 52, 113114
on Walras compared with other
economist, 113114
on Walras legacy, 13
on Walras growth model, 161
Jevons, W. S., 113114
business cycle theory of, 110
creative years for, 9, 10, 1415
on human desires, 128129
trading bodies and, 95
Jolink, Albert, 27, 58
Kant, Immanuel, 46
Keynes, J. M., 9, 10, 169
Kirman, Alan, 267, 311, 312
Koopmans, Tjalling, 301
Koppl, Roger, 99, 113114
Kurz, Heinz, 7
laissez-faire/laissez-passer, 98
Lallement, J er ome, 66
land, Walras on state ownership of, 23,
81, 109
land rents, 109
Lange, Oskar, 153, 288289
last comprehensive model, 181183
markets in, 182, 183
monetary theory in, 182183
reasons for poor quality of revisions in,
183187
written pledges in, 181182
Launhardt, C. F. W., 135
Laurent, Hermann, 279
Lausanne school, 279, 289
law of the variation of equilibrium prices,
68
Lendjel, Em eric, 29, 270
Lengwiler, Yvan, 308309
Lerner, Abba, 288
Lexis, William, 297
Lieben, Richard, 1415
Locke, John, 26, 46, 47, 62
losses. See prots/losses
macroeconomic savings function,
154155
Maks, J. A. H., 149150, 155, 167
Malthus, T. R., Walras on population
theory of, 50, 51, 94
Manuale deconomia politica (Pareto),
264
Marchionatti, Roberto, 280
Marget, Arthur W., 167, 279280
marginal productivity theory, 143, 279
marginal utility functions, 107
marginal utility theory, 12, 13
proportionality of prices to marginal
utilties, 88
market demand curves,
individual, 135
market/natural rates of interest, 282283
markets
assumption as purely competitive,
302
credit, 154155
oral pledges, 149
problems with in written pledges
sketch, 178, 179, 182, 183
production/consumption sides of, 5
rates of interest and, 155, 282283
stocks/bonds, 141142. See also freely
competitive markets
Marschak, Jacob, 289
Marshall, Alfred, 114, 141, 303
Marx, Karl, on economic theory scope,
120, 121122
mathematical model of exchange, of
Isnard, 5
mathematics, use in theoretical work, 27
mathematization of economics
dealing with exchange models, 3
inuence of Cournot on, 4
mature comprehensive model, 1011
absence of central price quoter in,
150
assumption of entrepreneur as
sometimes price-taker in, 150
Bourse incorporated into, 141142,
166
consumer sovereignty in, 169170
entrepreneur in (see entrepreneur)
equilibrium prices in, 154
as growth model, 161
legacy of, 170171
P1: IWV
Index 351
mathematical treatment of phenomena
in, 162166
comparative statistical analysis,
166
contradiction between model/
equations, 163164
equation solution need to be
non-negative, 164, 297
neglect of free goods, 163164
path dependency, 163
mathematical treatment of existence,
162163
contradiction between model/
equations, 163164
treating stability with equation
system, 164166
monetary processes in, 167169
availability of money, 167
cash-balance approach, 167
question of presence of no
uncertainty in model, 167168
role of true money, 168169
non-virtual behavior in disequilibrium,
175
pricing behavior/price formation in,
150
production/consumption ows in, 142
real economy as basis of dynamics of,
147149
as realistic/application to real
problems, 82
stability of, 151
stocks/bonds market incorporated into,
141142
tatonnement in (see tatonnement, in
mature comprehensive model)
wages in, 154
welfare economics and, 170. See also
mature comprehensive model,
steps in development of; Walras
evaluations of own model
mature comprehensive model, steps in
development of
rst step, economy as objective reality,
5967
denitions of market/competition,
6567
objective economic reality, 5960
objective reality aspects treated by
economic theory, 6365
positive character of economic
theory, 6063
second step, procedures for studying
economy, 6772
observation/experience/experiments,
6772
third step, ubiquity of freely
competitive markets in real
economy, 7273
fourth step, construction of freely
competitive model, 7382
assumption all markets in model as
freely competitive, 7375
ctional vs. real types, 8081
induction/deduction method, 7678
realism in/applications of, 82
7576
in practice, 7880
fth step, secondary elements of real
economy introduced into
model, 8284
sixth step, criteria for evaluating
theories/models
applications/evaluation of
mathematics in economics,
102108
equations based on economic reality,
105107
justication for using
mathematics in economics,
102104
restrictions on using mathematics,
107108
theoretical uses of mathematics,
104105
applications of economic theory,
97102
rationale of applications, 97
social/economic justice, 98102
conclusions about, 110112
conrmation of
hypotheses/conclusions by
experience, 108110
evaluation criteria, 8597
logical experiments/historical
evidence, 8789
theorizing/reality
aspects of economic models, clearly
dened/precise, 86
realistic theory based on empirical
knowledge, 86
science as deductive process, 86
theory must be applied, thus
realistic, 86
Walras evaluations of own model,
8993
P1: IWV
352 Index
mature comprehensive model (cont.)
general equilibrium equations,
9092
xed costs, 9192
prices/income, 8990
supply/demand, 89
theory of capital, 9293. See also
Walras evaluation of other
economist theory
Walras evaluation of other
economist theory, 9397
Condillac/Say utility theory, 94
English economists, 9496
English theory of interest, 95
Malthus population theory, 94
Mills commodities, 94
Mills tenant farmership, 9596
Ricardos theory of rent, 9394
summary of, 59
maximizing output, 93, 96
M enard, Claude, 150, 289, 294
Menger, Carl, 113114, 298, 300
metallic-money model, 168169
metals, quantity of precious, relations to
utility, 9293
Mill, John Stuart, 37, 60
on homo conomicus, 130
on human behavior, 122
on human nature, 129
inuence on Walras, 26, 27, 2932,
58
on physical sciences as models, 68
on political economy, 76
on science, 45
on science of society, 126
on validity of theories, 31
Mirowski, Philip, 58
modern economics, Walras ideas in,
313315
monetary aspects, of all phases of Walras,
167
monetary theory
commodity money
determining price of, 7980
unique vs. double standard for,
100101
in last comprehensive model, 182183
Schlesinger work on, 293, 296297
verisimilitude of, 93, 99101
Walras vs. Pareto, 279280
of Wicksell, 282283
monopolies, 80, 293, 302
as obstacle to resource allocation,
272273
monopolistic entrepreneur, 272
monotone/non-monotone preferences,
309
Moore, Henry L., 14, 285, 289
Morgenstern, Oskar, 307
Morishima, Michio, 7, 8, 149150, 161
Mouchot, Claude, 144
Murray, Roberto A., 280
Nash, John, 307
natural facts, 6263, 69, 88
vs. humanistic facts, 3435
natural monopoly, 81
natural sciences, 3940
applied, 4546
induction/deduction method in, 40
Nature and Necessity of Interest (Cassel),
291
Negishi, Takashi, 167
Neisser, Hans, 290, 297
net income commodity, perpetual, 80
Newton, Isaac, 26
non-aggregative capital theory, 301
non-conrmation of hypotheses and
conclusions, 44
non-durable goods, reprise in
tatonnement and production of,
156157
non-virtual model, characteristics of,
313314
numeraire, 90
On Quantitative Thinking in Economics
(Cassel), 291292
oral pledges market, 149
Osorio, Antonio, 279
Ostwald, Wilhelm, 9
Pantaleoni, Maffeo, 262263, 279
Pareto, Raphael, 259
Pareto, Vilfredo, 14, 289
background of, 259262
attitude toward own nobleness, 260
education/studies of, 259260
inuences on, 261
as laissez-faire economic liberalist,
261262
marriage/lack of children, 260
as pacist/humanitarian, 261
parents of, 259
as railway engineer, 260
runs for public ofce, 261
socialism and, 262
conservatism in later years, 277278
P1: IWV
Index 353
contributions to economics, 267278
economics of rm, 273274
entrepreneur, 271273
existence/uniqueness of equilibrium,
269
general socio-economic equilibrium,
276277
Paretos law, 275276
stability of equilibrium, 269271
theory of consumer demand,
267268
welfare economics, 274275
greatness of, 278
real world connection of works of,
265267
contrasts with Walras, 266267
on equations evaluated by reference
to concrete facts, 266
on successive approximations of
theory to real economy, 8384,
266
Walras inuence on, 262265
contact with work of Walras,
263264
meets Walras, 265
teaches at University of Lausanne,
265
theory of general equilibrium
comprehensibly written,
264265
Pareto optimum, 307
Paretos law, 275276
Patinkin, Don, 153
Peart, Sandra, 3031
perfect circle, as ideal type, 77, 120, 122
personal facts, in economic theory,
8788
Petri, Fabio, 160, 183, 311
philosophical/methodological views, of
Walras
antecedents of, 2627
archrationalism thesis, 2829
Descartes, 2728
Mill, 2932
applied sciences, 4547
applied ethical science, 46
applied natural science, 4546
industry and, 46
mores and, 46
3235
general facts, 3536
pure sciences, 3745
abstract sciences, 39
classifying according to subject
matter, 3738
concrete sciences, 39
empirical investigation/discovery
process, 4042
individual facts effect on
assumptions/conclusions,
4245
method in, 3840
natural sciences, 3940
induction/deduction method in,
40
real/ideal, 4755
doctrine/methods syntheses,
5355
idealism/realism, 54
rationalism/empiricism syntheses,
5455
imagination, 4749, 5152
real/ideal types, 4750
Phipps, Cecil G., 273274
physical sciences, 58, 68
as observational/experimental, 7071
Physiocrats, 60, 95
physiology, as
observational/experimental,
7071
physico-mathematical sciences, 56, 60,
109
Piccard, Antoine Paul, 10
Pietri-Tonelli, Alfonso de, 279
planned economy, 175
Platonic idealism, 41
Platonic realism, 25, 41
pledges
oral pledges markets, 149. See also
written pledges sketch
Poincar e, Henri, 132
Poinsot, Louis, 56, 27
population theory, 50, 51, 94
Potier, Jean-Pierre, 37, 179, 192
price determination
by free competition, 61, 62, 7879
historical record and, 8889
by utility, 68
prices
foodstuffs in industrial/agricultural
differential, 93
law of the variation of equilibrium
prices, 68
prices, equilibrium, 154
in freely competitive markets, 74
law of variation of, 68
P1: IWV
354 Index
pricing, Walrasian pricing, 74
privately owned monopoly, 81
private property, 124125
production
activity analysis model of, 301
coefcients of, 93, 96, 142, 144, 273,
293
consequences of above average
production for industry, 9192
period of, 283
rate of production, determination
by free competition,
production function, 142143
productivity, marginal productivity
theory, 143, 279
prots/losses
entrepreneur and, 144, 146147, 157
lack in virtual purely competitive
economy, 145
Proudhon, Pierre-Joseph, 27
public nance theory, 279
pure competition, 275, 285, 300, 302
pure economic theory, 6062
as free of normative considerations,
6162
theory of production in, 6667
Walras denition of, 6365
criticism of, 64
pure exchange economy, 309
purely competitive model, 308
pure moral sciences, 38, 61
pure natural sciences, 3738
pure sciences, 3745
abstract sciences, 39
as basis of applied science/practice, 45,
47
classifying according to subject matter,
3738
concrete sciences, 39
empirical investigation/discovery
process, 4042
general facts of real world in, 31, 37
individual facts and, 4245
method in, 3840
rationalism/empiricism synthesis in,
5455
Quesnay, Fran cois, 5
rate of net income, 155
rate of production, determination by free
competition,
ratios of exchange, in economic theory,
66
real/ideal, Walras on, 4755
doctrine/methods syntheses, 5355
rationalism/empiricism syntheses,
5455
imagination, 4749, 5152
real/ideal types, 4750
reality
drawing components of theory from,
41
facts and, 4142
industry relation with economic theory
and, 93, 9697
intimate/personal/external facts/reality,
in economic theory, 8788
realism
philosophical, 25
Walras on, 25
Rebeyrol, Antoine, 168, 179
Redman, Deborah, 27
Renouvier, Charles, 130
rent theory, 9394
representations of imagination, 119
Ricardo, David, 7, 60
Walras on theory of rent of, 9394
Ricci, Umberto, 279
Robertson, D. H., 169
Rossi, Pelegrino, 58
Rousseau, Jean-Jacques, 95
Ruchonnet, Louis, 3, 191192
Salvadori, Neri, 7
Samuelson, Paul, 289, 293, 306
savings
of capitalists, 79, 154155
macroeconomic savings function,
154155. See also capital
accumulation
savings balances, 168169
savings/investment behavior, 154155
Say, John Baptiste, 5, 58, 60, 94
scarcity, 297
Scarf, Herbert, 315
Schlesinger, Karl, 296, 297298, 300
Schmoller, Gustav, 265
Schultz, Henry, 288, 289
Schumpeter, Joseph A., 14
on creative years of scientist, 9, 10, 11
on Pareto, 280
on Walras difference from other
economists, 113114
on Walras legacy, 13
on Wicksell, 281
P1: IWV
Index 355
science
abstract, 38, 39, 43
applied ethical, 46
as inductive/deductive process, 2627,
30, 31, 32, 37, 38, 86
experimental, 43, 56, 62, 70
Mill on, 45, 68
physical, 58, 68, 7071
a priori vs. experimental, 32, 38
social, as observational/experimental,
7071. See also natural
sciences; pure sciences
science of society, 126
scientic socialism, 99
Scottish Enlightenment, 117, 126
Sensini, Guido, 280
shareholder, entrepreneur as, 147
silver, demonetization of, 107
Smith, Adam, 126
social economics, 98
socialist economy model, 279
social science, as observational/
experimental, 7071
social wealth theory, 66
society, general theory of, 83
socioeconomic interests, differing, as
reconcilable, 98
Sonnenschein, Hugo, 310
speculation, nancial, 110
speculators, social role of, 271272
Spencer, Herbert, 261
stability
Pareto on, 269271
tatonnement and, 147149, 150152,
154155
treating with equation system, 164166
Stahlecker, Peter, 309
state ownership of land, 23, 81, 109,
266
static equations, 13, 294
stationary state model, 294
steady-state growth model, 295
Steedman, Ian, 149, 167
Stigler, George, 12
subjective value theory, 12
supply/demand, 89, 135, 152
tatonnement, in mature comprehensive
model, 74, 147162
contradictions in mature model,
158162
capital accumulation, 160161
capital use/accumulation, 159160
denition of, 67
new capital goods not used during
tatonnement, 158159
parametric constants as not
contradictory, 161162
Currie, Martin on, 149
as dynamic, 161
equilibrium/stability of model,
150154
average cost/price, 152
capital use/accumulation, 154155
equilibrium/stability of real economy,
147149
general equilibration and equilibrium
of model, 149
Jaff e criticism of, 149
Koppl criticism of, 149
Maks, J. A. H. on, 149150, 155
M enard, Claude on, 150
model of, as realistic, 90
Morishima, Michio on, 149150
reprise in, 155157
conditions for, 155156
existing/new capital goods, 157
non-durable/net new durable
capital goods production,
156157
redirection of resources to alter
output composition, 157
Steedman, Ian on, 149
uniqueness, 162
van Witteloostujin, A. on, 149150,
155
taxation, 81, 98
theory of business cycles, 110
Theory of Social Economy, The (Cassel),
291, 292
theory of social wealth, 83
Theory of Value (Debreu), 140
time, capital and, 283
trade in cereals, liberty of, 110
trading bodies, Jevons on, 95
uniqueness of solutions to general
equilibrium equations,
299300
utility
Condillac/Say utility theory, 94
functional relation with commodities,
131, 132133
maximizing through
economy/prudence, 133134
maximum utility of new capital goods,
274
as price determination element, 68
P1: IWV
356 Index
utility (cont.)
properties of, 132133
continuous/discontinuous functions,
133
decreasing intensity, 133
independent of consumption of
other commodity, 132133
measurability, 132
quantity of precious metals relations
to, 9293
utility theory, marginal. See marginal
utility theory
Vacherot, Etienne, 26, 50, 51, 98, 99, 119
value in exchange theory, modern
industry role in, 93, 96
van Daal, Jan, 27, 58, 179
van Witteloostujin, A., 149150, 155, 167
virtual competitive models, 15
virtual model, 140
virtual Walrasian modeling, 293,
296297
inuence on economist, 1930s,
296302
Neisser, Hans Philip, 297
Schlesinger, Karl, 297298
von Neumann, John, 300302
Wald, Abraham, 298299
post-Hicksian inuence on economist,
306309
Allais, Maurice, 306
Arrow, Kenneth, 306308
Bidard, Christian, 308
DAgata, Antonio, 309
Debreu, G erard, 307308
Franke, Reiner, 308
Hahn, Frank H., 308
Hauenschild, Nils, 309
Iimura, Takuya, 309
Lengwiler, Yvan, 308309
Stahlecker, Peter, 309
present state of, 309312. See also
Cassel, Karl Gustav;
Hicks, J. R.
von B ohm-Bawerk, Eugen, 9, 283
von Neumann, John, 294, 300302, 307,
309, 312
von Pufendorf, Samuel,
von Stackelberg, Heinrich, 290
von Wieser, Friedrich, 93
wages, in mature comprehensive model,
154
Wald, Abraham, 296, 297
existence of general equilibrium and,
298299, 309
on uniqueness of solutions to
general equilibrium system,
299300
virtual Walrasian modeling and,
298299
work of von Neumann and, 301
Walker, Francis A., 144
WALRAS econometric model, 315
Walras, Auguste (Walrass father), 2
inuence on Walras, 34, 58
on mathematics in economics, 102
notions about natural law, 98
Walras, L eon
brief biography of, 23
creative period of, 9, 10, 11
fails to acknowledge debts to
predecessors, 6667
inuences on, 36, 2627, 30
Auguste Walras, 34
Cournot, 4
Isnard, 5
Poinsot, 56
Quesnay, 5
Say, 5
scarcity and, 297
teaches at University of Lausanne, 3,
265
Walras, L eon, inuence on other
economists, 1216
contemporaries/immediate successors,
1415
Barone, 14
Moore, 285286
opinion on most important aspects
of, 1213
general equilibrium concept, 12,
13
marginal utility concept, 12, 13
Pareto (see Pareto, Vilfredo)
school of Lausanne, 278280
Schultz, 286287
Schumpeter, 284
innovating entrepreneur of, 284
Wicksell, 281284
contributions of, 281284
contributions of, to theory of capital,
283
indebtedness to Walras, 281.
See also virtual Walrasian
modeling; writings, on
Walras
virtual modelers, 288312
P1: IWV
Index 357
Walras, L eon, legacy of
achieving consensual understanding
of, 69
approach of present book, 78
foundations of consensus, 89
past interpretations of, 67
initial model of general equilibrium,
910
mature comprehensive model of
general economic equilibration
and equilibrium, 1011
Walras on, 1314
written pledges sketch, 11, 175183
Walras, L eon, productivity of, 187192
appendix for, 192196
average number of pages per year
(18581910), 188
decline (18961910), 188
effect of ill health on, 190191
great creativity phase (18721877),
188
intellectual maturity phase
(18781895), 188
journalistic/experimental phase
(18581871), 188
measurement unit, 188
pages newly published/written
(18581910), 188
pages newly published/written
(19011910), 191
Walras, L eon, writings of
professional attention paid to, 12.
See also denitive bibliography
of Walras works; Writings
acknowledging Walrass
inuence
Walrasian pricing rule, 74, 151
Walras-Pareto doctrine, 293, 296297
Walras-Pareto school, 279
Walrass law, 152154
equation for, 153
welfare economics
free competition and, 170, 283284
mature comprehensive model and, 170
Pareto on, 274275
Wicksell, Knut, 14, 185, 289
on utility/scarcity, 293
Schumpeter and, 281
Wood, John C., 12
Worlds Monetary Problems, The
(Cassel), 291
Writings acknowledging Walrass
inuence
applied/computational general
equilibrium models, 342343
imperfectly competitive non-virtual
general equilibrium models,
338341
x-price models, 341
Nash-Walras and games/bargaining
models, 339341
oligopoly/Cournot-
Walras/monopolistic models,
338339
micro/macro-economics,
Walras/Keynes blend,
341342
non-virtual Walrasian models,
334336
exchange models, 334
exchange/production
models/structural
characteristics of, 335
production not present in,
335336
place/use in present book, 1619
uses/development of special ideas,
336338
Walras allocations, 337338
Walras correspondence, 338
Walrass law, 336337
written pledges, ctional, 81
written pledges sketch, 1112. See also
virtual Walrasian modeling;
written pledges sketch,
rationale/characteristics of
written pledges sketch,
rationale/characteristics of
problem/Walrass solution, 175183
de Zatarin on, 180181
production without written pledges,
180
problem overview, 175179
incompleteness, 178179
inputs/outputs, 179
uncertainty, 177
virtual markets, 179
written pledges markets, 178
problem review, 175
Zeuthen, Frederik L. B., 290, 297, 298,
300

Walrasian Economics - Donald A.walker

Uploaded by

Copyright:

Available Formats

Walrasian Economics - Donald A.walker

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Walrasian Economics - Donald A.walker

Uploaded by

Copyright:

Available Formats

What are the main contributions of Leon Walras according to the text?

How did Walras' ideas influence other economists?

P1: JPJ

0521858550pre CB1007/Walker 0 521 85855 0 November 24, 2005 7:31

Etat. Formule g en erale de

ECONOMIE POLITIQUE PURE, and bears

, 22, 11; and in the X, 4959,

, XXIV, 524, 6 plates.

that consumer demand be

that the results of the process of production be accurately predicted.

Uber die produktionsgleichungen der

Etre et devoir etre chez L eon Walras, in Dock` es et al.

Uber die Produktionsgleichungen der okono-

Uber einige Gleichungssysteme der mathematischen

Okonomie, Zeitschrift f ur National okonomie, 7 (5), 63770; translated by

You might also like