[2. STATISTICS]
It stands to reason that for economics, statistics, that is, the statistical figure or series of
figures must be of vital importance. In practice this has been recognized at least since the
sixteenth and seventeenth centuries when a large part of the work of the Spanish
políticos, for example, consisted in the collection and interpretation of statistical
figures—not to mention the English econometricians, who were called political
arithmeticians, and their fellow workers in France, Germany, and Italy.
1
We need
statistics not only for explaining things but also in order to know precisely what there is
to explain. But a comment has to be added that is analogous to the comment made in the
preceding paragraph on the subject of history. It is impossible to understand statistical
figures without understanding how they have been compiled. It is equally impossible to
extract information from them or to understand the information that specialists extract for
the rest of us without understanding the methods by which this is done—and the
epistemological backgrounds of these methods. Thus, an adequate command of modern
statistical methods is a necessary (but not a sufficient) condition for preventing the
modern economist from producing nonsense, though very much more so in some fields
than in others: our stake in these methods is too great for us to leave judgment on the
virtues or shortcomings, say, of the variate-difference method to specialists, even if they
were unanimous about it. Again, we shall not be able to live up to the program that
follows from this. But again, we shall recognize, in principle at least: statistical methods
are part of the tools of economic analysis even when not specially devised to meet its
particular needs; and Jacques Bernoulli’s Ars conjectandi or Laplace’s Théorie
analytique stand in the history of many sciences but they have their places also in the
history of our own.
2

[3. ‘THEORY’]
The third fundamental field is ‘theory.’ This term carries many meanings but only two of
them are relevant so far as our own usage in this book is concerned. The first and less

1
and the things (propositions) that we think we have established by admissible
procedure are called theorems. Of course a proposition may figure in one argument as a
postulate and in another as a theorem. Now, hypotheses of this kind are also suggested by
facts—they are framed with an eye to observations made—but in strict logic they are
arbitrary creations of the analyst.
2
They differ from the hypotheses of the first kind in that
they do not embody final results of research that are supposed to be interesting for their
own sake, but are mere instruments or tools framed for the purpose of establishing
interesting results. Moreover, framing them is no more all the economic theorist does
than framing statistical hypotheses is all that the statistical theorist or in fact any theorist
does. Just as important is the devising of the other gadgets by which results may be
extracted from the hypotheses—all the concepts (such as ‘marginal rate of substitution,’
‘marginal productivity,’ ‘multiplier,’ ‘accelerator’), relations between concepts, and
methods of handling these relations, all of which have nothing hypothetical about them.
3

And it is the sum total of such gadgets—inclusive of strategically useful assumptions—
which constitutes economic theory. In Mrs. Robinson’s unsurpassably felicitous phrase,
economic theory is a box of tools.
The rationale of this conception of economic theory is very simple and the same as in
all other departments of science. Experience teaches us that the phenomena of a given
class—economic, biological, mechanical, electrical, and what not—are indeed individual
occurrences each of which, as it occurs, re-
1
By ‘principle’ we shall mean in this book any statement that we (or the authors under discussion)
do not propose to challenge. But it may be a proposition that we (they) have established as well as a
proposition that we (they) have postulated or assumed. The same holds for the objectionable term
‘law,’ the emergence and use or misuse of which will have to be carefully considered: we speak of

work is the first in which
awareness of this last truth is clearly discernible, though economists took over a century
to realize all its possibilities—Léon Walras was in fact the first to do so (see below Part
IV, ch. 6, sec. 5b).
Although it is neither possible nor desirable for us to embark upon an
epistemology of economics and although some of the topics pertaining to
that field will receive attention both in the subsequent chapters of this Part
and in all the subsequent Parts, it will be helpful to insert here a few
additional remarks in the hope that they will do something to scale down
possible barriers between myself and my readers.
First, then, a qualification should be added to the preceding argument
about the nature and functions of economic theory. This argument ran in
terms which are applicable, substantially at least, to all sciences that have
any all-purpose apparatus of analysis. But there are limits to this
parallelism and the most important of them are represented by the two
following facts. Economics lacks the benefits that physics derives from
laboratory experiments—when economists talk about experimenting they
mean something quite different from experimenting under laboratory
conditions—but enjoys instead a source of information that is denied to
physics, namely, man’s extensive knowledge of the meanings of
economic actions. This source of information is also a source of
controversies that will bother us repeatedly on our journey. But its
existence can hardly be denied. Now, when we speak, for example, of
motives that are supposed to actuate individuals or groups, our source of
information may be roughly identified with knowledge of psychic
processes, conscious or
4
The statement above is a brief rendering of E.Mach’s doctrine that every
(theoretical) science is a device for effecting economy of effort (Denkökonomie).
5

6
to denote the uncritical
copying of the methods of mathematical physics in the equally uncritical
belief that these methods are of universal application and the peerless
example for all scientific activity to follow. This history as a whole will
answer the question whether there actually has been such uncritical
copying of methods that have meaning only within the particular patterns
of the sciences that developed them—apart of course from programmatic
utterances that have been numerous enough ever since the awe-inspiring
successes of the physical sciences in the seventeenth century but mean
next to nothing. As regards the question of principle, there cannot be the
slightest doubt that Hayek is right—and so were all who in the nineteenth
century preceded him in uttering protests similar to his—in holding that
the borrowing by economists of any method on the sole ground that it has
been successful somewhere else is inadmissible,
6
F.A.von Hayek, ‘Scientism and the Study of Society,’ Economica, August 1942,
February 1943, and February 1944. This treatise—these articles are nothing less
than a treatise—is strongly recommended both because of the profound
scholarship of which it is the product and because it presents an excellent example
of how near to each other, in discussion of this kind, dwell truth and error.
Interlude I 15
and that the rare and unimportant cases in which this has actually been
done deserve what they get at his hands. Unfortunately this is not the real
question. We have to ask what constitutes ‘borrowing’ before we can
proceed to ask what constitutes illegitimate borrowing. And here we must
beware of an optical illusion similar to the one that makes Marxists so
reluctant to use such terms as price or cost or money or value of the
services of land or even interest when speaking of a future socialist order:
these terms denote concepts of general economic logic and seem to

comes from that has followed it ever since it attracted any attention at all
(which was roughly since the time of the physiocrats) to this day. I shall
simply list the main headings for an answer which our story will amply
verify:
(1) At all times, including the present, in judging from the standpoint
of the requirements of each period (not judging the state of the theory
7
Hayek’s teachers, the Austrian utility theorists, by operating the concept of
marginal utility, actually discovered the calculus. It cannot be a crime to formulate
their reasoning correctly.
History of economic analysis 16
as it was at any time by standards of a later time) the performance of
economic theory has been below reasonable expectation and open to valid
criticism.
(2) Unsatisfactory performance has always been and still is
accompanied by unjustified claims, and especially by irresponsible
applications to practical problems that were and are beyond the powers of
the contemporaneous analytic apparatus.
(3) But while the performance of economic theory was never up to the
mark, that is, never what it might have been, it was at the same time
beyond the grasp of the majority of interested people who failed to
understand it and resented any attempt at analytic refinement. Let us
distinguish carefully the two different elements that enter into this
resentment. On the one hand, there were always many economists who
deplored the loss of all those masses of facts that actually are lost in any
process that involves abstraction. So far as application is concerned,
resentment of this type is very frequently quite justified. On the other
hand, however, there are untheoretical minds who are unable to see any
use in anything that does not directly bear upon practical problems. Or, to
put it less inoffensively, who lack the scientific culture which is required

interesting to note that a propensity
8
of this kind is by no means confined
to our field. Isaac Newton was a theorist if he was anything. Nevertheless,
he displayed a marked hostility toward theory and especially toward
framing of causal hypotheses. What he really meant was not theory or
hypothesis of our second kind but just inadequately substantiated
speculation. Perhaps there was also something else in this hostility,
namely the aversion of the truly scientific mind to the use of the word
‘cause’ that carries a metaphysical flavor. Newton’s example may also be
appealed to in order to illustrate the truth that dislike of the use of
metaphysical concepts in the realm of empirical science does not at all
imply any dislike of metaphysics itself. [J.A.S. intended to have these nine
paragraphs of indented material set in small type so that it would be easy
for the average reader to skip them.]
[4. ECONOMIC SOCIOLOGY]
The reader will have observed that our three fundamental fields, economic history,
statistics and statistical method, and economic theory, while essentially complementing
each other, do not do so perfectly. In writing economic history, there are indeed
statements that should not be added at all unless properly substantiated by pieces of
reasoning that belong to economic theory: such a statement is, for instance, the one that
links England’s great economic development from the 1840’s to the end of the nineteenth
century to the repeal of the Corn Laws and of practically every other kind of protection.
The schemata of economic theory derive the institutional frameworks within which they
are supposed to function from economic history, which alone can tell us what sort of
society it was, or is, to which the theoretical schemata are to apply. Yet, it is not only
economic history that renders this service to economic theory. It is easy to see that when
we introduce the institution of private property or of free contracting or else a greater or
smaller amount of government regulation, we are introducing social facts that are not
simply economic history but are a sort of generalized or typified or stylized economic

The sum total of the historical, statistical, and theoretical techniques that have been
characterized above, together with the results they help to produce, we call (scientific)
economics. This term is of relatively recent growth. A. Marshall’s great treatise was the
first to establish its use, from 1890 on, at least in England and the United States.
1
In the
nineteenth century, the term commonly in use was Political Economy, though in some
countries other terms competed with it in the first decades of that century. This
unimportant matter will be attended to, as we go along, in the subsequent Parts. But it is
just as well to note two points at once. First, political economy meant different things to
different writers, and in some cases it meant what is now known as economic theory or
‘pure’ economics. A warning must therefore be issued right now that in order to interpret
correctly what any given writer said about the scope and method of political economy,
we must always make sure of the meaning he attached to this term—some propositions
about those subjects that have outraged critics become perfectly harmless if this rule be
borne in mind. Second, ever since our science or agglomeration of sciences was baptized
political economy by a not very significant writer of the seventeenth century whose work
owes an undeserved immortality to this fact, there has been the implicit or explicit
suggestion that the exclusive concern of our science was with the economy of the state—
though of course not only of the polis, the city-state of Greece—or, what is almost the
same thing, with public policies of an economic nature. This suggestion, which was still
more emphasized by the German term frequently used as a synonym of political
economy, Staats- wissenschaft, implied of course an altogether too narrow conception of
the scope of economics. Incidentally, it over-emphasized the largely meaningless
distinction between
1
I believe that this phrase is due to Mr. Gerhard Colm.
1
Later on, a parallel usage was introduced, though less firmly established, in Germany. The word
was Social Economics, Sozialökonomie, and the man who did more than any other to assure some

systems of political economy and to any history of the thoughts on economic subjects that
float in the public mind.
[6. APPLIED FIELDS]
Division of labor, in research as well as in teaching, has produced, in economics as
elsewhere, an indefinite number of specialties that are usually described as ‘applied
fields.’ In order to obtain a list of them (which does not claim to be complete) let us draw
inspiration from the courses offered by the larger institutions of higher learning in the
United States that teach economics.
In addition to general survey courses and courses in economic history, statistics,
economic theory, and economic sociology,
1
we find, first, offerings in a group of fields

1
Owing to the discredit that, for good reasons and bad, attaches to the word ‘theory’ in many
minds, this word is occasionally replaced by the word ‘analysis’ which then carries a meaning that
is more restricted than the one attached to it in this book. The field of economic sociology does not,
so far as I am aware, appear under this title or separately, but topics belonging in it are treated in
History of economic analysis 20
courses on history, theory, ‘comparative economic systems,’ the more institutionally oriented
courses on labor, and in a number of others.

which everyone considers part and parcel of ‘general economics’ and which receive
separate treatment only in order to facilitate more intensive treatment of their subjects.
Such are money and banking, business fluctuations (or cycles), foreign trade
(international economic relations), and, occasionally, location. Second, we find a group
of fields, such as accounting, actuarial science, and insurance, that, historically, have
preserved altogether too much independence from general economics (which is slowly
being surrendered in the case of accounting) but are useful or even indispensable for all
or some economists, because they offer both instruments of economic analysis or


2
The field of agriculture offers an interesting example of a department of economics that it is
hardly possible to treat without a considerable command of agricultural technology. In principle,
though mostly to a lesser extent, this is true also in other departments and, so far as this is so, there
is no point in drawing any sharp line between, say, the economics of banking, marketing, or
manufacturing industry and the corresponding ‘technologies.’
Interlude I 21