The Review of Austrian Economics, 14:2/3, 119–143, 2001.


c 2001 Kluwer Academic Publishers. Manufactured in The Netherlands.

On Rationality, Ideal Types and Economics:

Alfred Schütz and the Austrian School
PETER KURRILD-KLITGAARD∗
Department of Political Science and Public Management, University of Southern Denmark (Odense) Denmark;
and Institute for Social and Economic Research and Policy, Columbia University

Abstract. A comparison is made of the views on economic theory and method of the Austrian philosopher and
sociologist Alfred Schütz (1899–1959) and those of his mentor, the Austrian economist Ludwig von Mises (1881–
1973). Schütz basically agreed with the fundamental parts of the Austrian program, but he also had disagreements
with Mises on the epistemological character of the core assumptions, on the formulation and status of the rationality
principle, and on the use of ideal types in economic analysis. In several of these aspects Schütz had important
points of value not only for the use of ideal types in economic modeling, but also within political science and
sociology. In the end, however, there is more which unites than separates Schütz and Mises.

Action is, by definition, always rational.

Ludwig von Mises ([1933] 1981:35)
Action is behavior based on an antecedent
project. Since every project has an “in-order-to”
or “for-the-sake-of-which” structure, it follows
that every action is rational.
Alfred Schütz ([1932] 1967:239)

Introductory Remarks

The Austrian-born philosopher Alfred Schütz (1899–1959), the founder of phenomenologi-

cal sociology, is often seen by sociologists as the anti-thesis of everything economistic. In-
terpretive sociologists, ethnomethodologists, anthropologists, hermeneuticist philosophers

∗ An earlier version of the present paper was presented at the History of Economics Society Session on Schütz

4 January 1999, Annual Meetings of the American Economic Association, New York and at the Colloquium an
Austrian Economics, Dept. of Economics, New York University, 6 March 2000. I am particularly grateful to Peter
Boettke, Roger Koppl, and one anonymous referee for very helpful comments and suggestions, but the paper is
the outgrowth of a larger project, for which I have received help, suggestions and encouragement from a large
number of people, including Richard Ebeling, Walter Grinder, Israel Kirzner, Leonard Liggio, Roderick Long,
Mario Rizzo, Jeremy Shearmur, and Barry Smith. I owe special thanks to Lester Embree, Bettina Bien Greaves,
J. Herbert Furth, Gottfried Haberler, Evelyn Schütz Laing, Kurt Leube and Ilja Srubar, who at various occasions
provided me with invaluable help and information. Finally, I am particularly grateful to the staff of the Institute
for Humane Studies, who in 1993–94 encouraged and supported the research.
120 KURRILD-KLITGAARD

and even strands within ‘post- modernism’ have claimed his sociology of the everyday-life.
In contrast, the Austrian economist Ludwig von Mises (1881–1973) is often viewed as being
the most radical of all economic ‘imperialists’. Mises always championed that all human
action should be analyzed in terms of such economic categories as purpose, choice, utility,
costs, economizing, etc., and, on top of it, argued that all action necessarily is rational.1
In recent years this contrast has been seriously challenged in a number of ways. It has been
shown that Schütz, through his participation in Mises’ Privatseminar was much closer af-
filiated with one particular branch of economic thought than usually acknowledged, namely
the very Austrian School of Economics to which Mises belonged (Prendergast 1986, Helling
1988, Kurrild- Klitgaard 1997a). Furthermore, it has also been shown that important aspects
of Schütz’s insights and those of the Austrian economists are compatible or may be inte-
grated (Rothbard [1973] 1979, 1976, Boettke, Horwitz, and Prychitko 1986, Ebeling 1987a,
1987b, 1987c, Eberle 1988, Koppl 1997, 1998, O’Driscoll and Rizzo 1985, O’Sullivan 1987,
Prendergast 1993).2 Indeed, it has been demonstrated how the most important aspects of
Schütz’s approach can be reconstructed as being basically identical to the rational choice
approach broadly conceived (Esser 1993a, 1993b).
The present study wants to add further to this emerging view. This will be attempted by,
first, briefly introducing Schütz’s writings on economics, several of which have not been
generally accessible until recently (Section 1). Second, a discussion will be undertaken of
some of the main points of potential agreement and disagreement between Schütz and the
Austrians (Section 2). In particular special attention will be given to the different ways in
which Schütz and Mises viewed the delimitation of the field of economic theory and the
usefulness of ideal types. It will be shown that a number of differences may be seen to
exist between the two, but also that Schütz on some points saw disagreement where none
necessarily exists, and that he on some points possibly misunderstood his mentor, Mises.
Indeed, Schütz’s criticism of Mises may be due to a partial confusion of Mises’ concept of
homo agens with the more standard neo-classical homo oeconomicus.
It will also be indicated how these points may have wider importance than merely for the
history of the Austrian School, i.e. for economic theory in general and for the application
of rational choice-like analysis outside the field of economics, e.g., to political science and
sociology.
In this course, it will, however, be necessary to take certain knowledge of the approach
of the Austrian School and the basic concepts of Schütz’s phenomenological sociology for
granted. This is, of course, less than satisfactory, but given the circumstances unfortunately
necessary.

1. Schütz’s Writings on Economics

The first thing to note when aiming at a comparative analysis of the phenomenological soci-
ology of Alfred Schütz and the approach and theories of the Austrian School of economics
is that the two are in fundamental correspondence to each other. Schütz’s scholarly pursuits
may indeed be seen as an integration of Weber’s interpretive sociology with Husserl’s phe-
nomenology in order to support and further develop the economics of the Austrian School
(Prendergast 1986).
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 121

Whenever Schütz wanted to point towards, what he considered to be an ideal case of

how social scientific theories should be formulated, and what errors to avoid, he repeatedly
pointed towards the Austrians as authoritative.3 When he spoke of “pure economics” as
being the “perfect example of an objective meaning-complex about subjective meaning-
complexes” ([1932] 1967:245), he was clearly referring not just to economics in general
but explicitly and specifically to Misesian Austrian economics. Schütz’s view of economics
could indeed be summarized in his own words to his economist friend, Adolph Loewe:
“[My] scanty knowledge of economics is based on what I learned in Vienna some 25 years
ago as economic theory and this was based on the particular brand of marginal theory
developed by the Austrian school.”4
There is, it must be admitted, very little evidence of a direct and obvious influence
specifically from the Austrian School in Schütz’s work, but this is no doubt first and foremost
due to the fact that Schütz wrote so little on economics as such. Schütz was, as he himself
stressed on various occasions, not really an economist by education, and he never considered
himself to be so per se. Rather Schütz’s interest in economics was, on one hand, through his
professional life as a banker, and on the other hand, due to his general interest in the social
sciences, with economics being, in Schütz’s opinion, perhaps the most fully developed of
these. Economics thus, as his intellectual biographer has put it, next to sociology “ranked
second in his private scale of the relative relevance of the various disciplines for his work”
(Wagner 1983:164).
But Schütz never published any essays on specific economic issues, and even his essays
on methodology were, at the most, indirectly on economic theory. Of course if economics,
as Mises ([1933] 1981:xvi) originally saw it, is simply a branch of general sociology, then
one could say that all of Schütz’s writings are related to economics, in so far as they almost
all deal with social scientific methodology in relatively broad terms.
Some of Schütz’s writings have, however, dealt more explicitly with economic themes
and issues than others. This is in particular the case with his lecture for Mises’ Privatsemi-
nar, “Verstehen und Handeln” (“Understanding and Acting”) ([1930] 1996), long passages
in Der sinnhafte Aufbau der sozialen Welt ([1932] 1967), and his very favorable review of
Mises’ Grundprobleme der Nationalökonomie ([1933] 1981). It is even more so the case
with his two unfinished attempts at writing an article on his views on economic methodol-
ogy, “Nationalökonomie: Verhalten des Menschen im sozialen Leben” (“Political Economy:
Human Conduct in Social Life”) and “Untersuchungen über Grundbegriffe und Methoden
der Sozialwissenschaften” (“Investigations into the Fundamental Concepts and Methods of
the Social Sciences”). Most important are Schütz’s written reflections and response to the
draft of his close friend and fellow Miseskreis-member, F.A. Hayek’s essay “Economics
and Knowledge” ([1937] 1949), “Über Wissen und Wirtschaft” (“On Knowledge and Eco-
nomics”) ([1936] 1996). Later came published articles, which dealt in great detail with the
questions of choice and rationality in the social world, often with reference to economic
questions. These include, e.g., “The Problem of Rationality in the Social World” ([1943]
1964),5 “Choosing among Projects of Action” ([1951] 1962), “Common-Sense and Scien-
tific Interpretation of Human Action” ([1953] 1962) and the partly posthumous “Choice
and the Social Sciences” (1972). Finally, in his correspondence with friends and colleagues
Schütz often discussed issues and concepts fundamental to economic theorizing, e.g., with
122 KURRILD-KLITGAARD

F.A. Hayek, Fritz Machlup and Gottfried Haberler.6 This was in particular the case with
a series of interesting letters exchanged in the mid-1950s with his colleague at the New
School for Social Research, the institutionalist economist Adolph Loewe, with whom he
taught a seminar on methodology in 1955.7
These works are together the works which must be considered, if one wants a true
understanding of what Schütz saw as his own views on economic theory.

2. Schütz on Economics

What is obvious from these writings is that Schütz shared all the basic points of the eco-
nomics of the Austrian School of the inter-war period. This was so, even if he more often
than not used a terminology quite different from that of, e.g., Mises and Hayek. In the fol-
lowing this will be demonstrated as relating to three of the central elements of the Austrian
approach:8 The questions of (1) methodological subjectivism and methodological individ-
ualism as the proper starting points of the social sciences, (2) the focus on human action
and the principle of marginal utility, and (3) the axiomatic–deductive character of general
economic theory.

2.1. Methodological Subjectivism and the Universality of Economic Theory

Schütz’s writings clearly demonstrate how he shared both the methodological individu-
alism and methodological subjectivism of his friends and colleagues among the Austrian
economists, and indeed rejected the behavioralist view found among many positivists that
economics should only deal with what could be observed. Although his terminology was
different his position was fundamentally Misesian:

All social phenomena can be traced back to actions of actors in the social world who,
in turn, may be observed by social scientists. Therefore it is possible at any time to
pose the further questions: What possible meanings did the actors connect with these
actions which present themselves to us, the observers, as courses of social phenomena?
Posing this question we will no longer be satisfied with establishing a certain form of
curves of supply and demand. Beyond this we ask, What considerations must have
caused sellers and buyers to behave in the market place so that the resulting curves
of supply and demand acquired this or that shape? We will no longer be satisfied
with the prices of end-products and the statistical establishment of the producers when
setting up the plans of production. . . . This perspective of research may appropriately
be called the subjective direction, or better the question about the subjective meaning
of social phenomena. . . . At no stage of our social-scientific investigations can we be
prohibited from referring back to the subjects of the social world. We can object to
this information only if, for reason of one or another problem, this turn of attention
will yield little information of interest, perhaps none at all. Conversely it can be said
that the same turn of attention is unavoidable when we aim at the exact recognition of
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 123

phenomena which are not even viable on the level of objective meaning and become
thematic only at a deeper level of inquiry. (Schütz [1936] 1996:94–95; italics added)

So while–as we shall see–there are specific points, and important ones, where Schütz dis-
agreed with Mises, he was in fundamental agreement with, or even defending, the Misesian
position on this fundamental level. Schütz’s basic adherence to the Austrian program was
in fact obvious already in the first known work, where he touches on economic subjects, the
Miseskreis-lecture “Verstehen und Handeln” ([1930] 1996) which essentially was a preview
of Der sinnhafte Aufbau der sozialen Welt ([1932] 1967). In the lecture Schütz set out with
the purpose of investigating—and defending—the view that all the social sciences deal with
meaningful (purposive) human behavior in the social world and that one branch, economics,
is capable of producing theoretical knowledge consisting of “universally valid propositions”
(Schütz [1930] 1996:84). Schütz’s conclusion is that it indeed is possible to gain such an
objective knowledge of subjective actions, despite the problem of intersubjectivity, and the
reason is that the abstract propositions of economics deal with aspects of action which are
completely interchangeable from one individual to another:

The Thou becomes an impersonal ‘someone’ [‘Man’]. I or anyone else can replace
this Thou by any real alter ego or an ‘ideal type’ or ‘everyone’ without thereby being
able to change anything in the context of meaning in which we fit the action. But
it is just for this reason that the ‘objective meaning’ of the action, its place in the
context of meaning in which we fit the action, its place in the context of meaning of
our experience, remains invariable in the fact of any context of meaning whatever in
which the action happened to be built up polythetically: who executed the action under
regard and when it was done is in principle irrelevant for this mode of observation; it
is sufficient that such an action does not contradict our experiences.
[The] objective meaning (of an action or sign) is exclusively integrated into a context
of meaning in the consciousness of the ‘observer’ . . . Only a science of objective
meaning is capable of forming ‘laws of universal validity’. Political economy is a
science of objective meaning. It does not deal with action which is built up phase-
by-phase in the course of consciousness pertaining to the Thou; it deals instead with
the anonymous processes of actions by an impersonal ‘someone’. Just this sets off the
subject-matter of political economy from that of understanding sociology (and also
that of history). (Schütz [1930] 1996:86; italics added)

These “anonymous processes of actions by an impersonal ‘someone”’ are simply a quite

different way of using the Misesian concept of homo agens, Acting Man, and the “laws of
universal validity” are, of course, a reference to Mises’ view that economic theory consists
of praxeological laws which are a priori (Helling 1988:59).

2.2. Marginal Utility and the Open-Endedness of Human Motivation

Schütz’s adherence to the Austrian program is also clear from his discussions in the mid-
1950s with the institutionalist economist Adolph Loewe. In their discussions Loewe had
124 KURRILD-KLITGAARD

argued against what Schütz understood as the ‘marginal principle’ by trying to show that it
necessarily implied a semi-hedonistic assumption of maximization of satisfaction. Schütz
rejected this and argued that the principle was purely formal:

[As] I see it, the marginal principle does not only consist in establishing the maximiza-
tion of satisfaction as the basic motive of economic action. You reject this function of
the marginal principle by good reasons as an appropriate basic method for economics
and I feel by no means qualified to oppose your arguments which seem to me highly
convincing. It would be preposterous to tell an economist like you what kind of tools
he would need. . . . [But] I do not see, until otherwise enlightened by you, why in the
marginal principle must needs be identified with maximization of satisfaction. Is it not
possible to interpret this principle as a purely formal one?9

Now, what did Schütz mean by such a “purely formal” principle? According to Schütz
the basic assumption—or, as he also called it, no doubt inspired by Mises, “axiom”—of
economic science would be that every actor at any moment of his life is confronted with
what he elsewhere called “problematic possibilities” (Schütz [1951] 1962, 1972). This is a
choice between a set of alternatives, where these are

(1) arranged in systems of relative relevance, each “with a particular weight in terms of
preferences of this individual,”
(2) are mutually exclusive,
(3) are in competition with each other, and thus
(4) involves the renouncement of all other alternatives standing to choice.

Together these points, according to Schütz, constitute “the core of the marginal principle”
when combined with a “second axiom.” The latter is that “the outcome of all these choosing
processes has to be anticipated as if all the choices were made rationally” (italics added).10
This was indeed elaborated in greater detail by Schütz:

The ideal type of social action must be constructed in such a way that the actor in
the living world would perform the typified act if he had a clear and distinct scientific
knowledge of all the elements relevant to his choice and the constant tendency to
choose the most appropriate means for the realization of the most appropriate end . . .
[Only] by the introduction of the key concept of rationality can all the elements be
provided for constitution of the level called ‘pure theory’ . . . [This implies] that all other
behavior has to be interpreted as derivative from the basic scheme of rational acting.
The reason for this is that only action within the framework of rational categories can
be scientifically discussed. (Schütz [1951] 1962:86–87)

We must, Schütz argued, build our

ideal types as if all actors had oriented their life-plan and, therefore, all their activities
to the chief end of realizing the greatest utility with the minimum of costs; human
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 125

activity which is oriented in such a way (and only this kind of activity) is the subject
matter of your science. (Schütz [1951] 1962:87)

These formal assumptions must in any particular model be filled with specific content so as
to meaningful. The interesting point here is that these points have a significant resemblance
with the standard assumptions of standard rational choice theory, e.g., as in game theory—
which, incidentally, was partly developed by another of Mises’ students and friends of
Schütz, Oskar Morgenstern. They are all standard assumptions of choices made among mu-
tually exclusive alternatives on the basis of preferences which are consistent and complete,
and that the choices are made given scarcity and constraints (e.g., a budget). In addition
it must be assumed in any particular model that preferences are constant for the moment
under consideration. This is, Schütz argued against Loewe, a purely formal principle:

Nothing whatsoever is stated as to the specific objects of choice and nothing even as to
the principle according to which this choice is made, except that this choice according
to the second axiom has to be interpreted as a rational one. The concept of ‘satisfaction’
or ‘maximization of satisfaction’ has so far not entered the analysis.11

The parallel to Mises and the Austrians is again striking. Mises had gone to great lengths
to try to distance his economic thought from those based on any simplistic instinct-based
motivation. And so is the immediate inference drawn by Schütz, namely that such abstract,
formal theoretical consideration is a necessary prerequisite for any empirical studies:

Now, I submit that an analysis of this kind precedes even any ‘descriptive analysis’ in
your terminology. The analysis outlined above will refer also to other rational actions
and our common friend Pareto was, I think, right in so far as he made a distinction
between rational and non-rational actions, in stating that the former constitute the
realm of economics and that part of jurisprudence which deals with contracts, whereas
the latter constitutes the domain of sociology. Of course, he had a highly objectionable
conception of rationality.12

Mises no doubt would ascribe to this too, although there would seem to be at least a termi-
nological difference—as Schütz was aware of and as we shall return to. For Mises explicitly
regarded all action is rational and held that one should distinguish between purposive be-
havior which is rational and non-purposive behavior which is non-rational. But otherwise
the twist was the same: That the former constitutes the domain of economics and similar
social sciences, while the latter is that of sociology, psychology, etc.

2.3. The Axioms and Levels of Economic Analysis

Schütz believed that the “principle of marginal utility attains a universal character by mark-
ing out theoretical realms of social life” and explicitly advocated that the principle should
be placed as an axiom at the foundation of all economic analysis and combined with the
use of personal ideal types:
126 KURRILD-KLITGAARD

[It] follows that the principle of marginal utility is a postulate directed at economists
that can be circumscribed as follows: 1) Supply the conceptual models of actors in the
social world as such, which you are to form, with experiences of consciousness (goals
of action and motives) so that the resulting actions appear oriented to the principle of
marginal utility; 2) construct only ideal types which conform to this postulate. (Schütz
[1936] 1996:102; italics added)

Schütz explicitly distinguished between different levels of abstraction in economic analysis,

namely on the one hand what he called “the realm of the logic of pure economics” or “general
praxeology” and on the other hand what he called “the social economy of exchange” (Schütz
[1936] 1996:92, 99). He was here no doubt consciously echoing Mises’ distinction between
praxeology as being the “logic of action” and economic theory as being “catallactics”,
i.e. exchange relations between individuals. But it is also worth noting that Schütz made
this distinction in his comments on the concept of equilibrium in his response to Hayek’s
“Economics and Knowledge” ([1937] 1949). Here Hayek had distinguished between “the
pure logic of choice” and the study of market interaction and argued that the former alone
could not explain the occurrence of equilibrium in the latter.
This was indeed echoed by Schütz in his discussion of knowledge and data in economic
modeling:

[As] long as we remain within the realm in which we started [i.e., ‘the logic of pure
economics’, PKK] we will never encounter the question whether or which data are
given to a social scientist. There are no data within the given sphere. Nothing is ‘given’,
everything must be determined by definitions, axioms, so-called tautologies, or else
it must be derived from them. The question of data emerges first when the original
problem-formation is abandoned (expanded) and when the set of insights gained so far
is now to be applied to other strata of problems. What formerly was non-problematic
now becomes the object of a new theme. Economists13 regularly encounter the question
of data; so, for instance, when they apply insights found in the logic of pure economy
to the concrete economy of exchange, or when they apply propositions of the social
economy of exchange to the economy of money; or when they are to apply a general
proposition belonging to the economy of money to the concrete exchange rate of the
pound sterling in 1936. (Schütz [1936] 1996:97; italics added)

3. Schütz’s disagreement with Mises

But while Schütz to a large extent can be seen as championing core elements of the Misesian
program, there were also difference between the two. On two occasions did Schütz explicitly
state his disagreement with his mentor, Mises, first in his “Über Wissen und Wirtschaft”
([1936] 1996) manuscript to Hayek and, secondly, almost twenty years later, in one of his
letters to Loewe (1955).
Schütz’s critique of Mises’ version of Austrian economics can basically be connected to
four points: (1) The epistemological status of the propositions of economic theory, (2) the
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 127

extent of the applicability of the logic of choice, (3) the process of choice itself, and (4) the
meaning and usefulness of the concept of the ‘ideal type’.

3.1. The Status of the Propositions of Economic Theory

Schütz did not disagree with Mises, that the propositions of economic theory are of a
necessary character; they are, he held, about what necessarily must take place. Schütz
actually on several occasions explicitly agreed with Mises’ criticism of Weber’s ideal type
methodology in the essay “Soziologie und Geschichte”. Weber had failed to recognize that
the fundamental laws of economic theory, as Schütz ([1932] 1967:243, [1936] 1996:103)
quoted Mises, “are valid always and everywhere if the conditions assumed by them are
given.” (Mises [1933] 1981:85–86). But Schütz disagreed with Mises about why this would
seem to be the case and the extent to which propositions about economic phenomena may
be seen as being necessarily true.
Mises, a kind of neo-Kantian, in his early works collected in his anthology ([1933] 1981),
distinguished between two fundamentally different ways of achieving knowledge in the so-
cial sciences, “Begreiffen” (“conception”) and “Verstehen” (“understanding”). In the former
knowledge about the categorical character of phenomena is gained by “discursive reason-
ing,” while the latter seeks the subjective meaning of specific actions through observation
and interpretation: “Conception is reasoning; understanding is beholding.” (Mises [1933]
1981:134). Mises held the propositions derived from conception to be true a priori and
due to the mental thought processes of human beings; it was simply inconceivable that they
should not be true—because it is impossible to deny their fundamental assumptions without
getting involved in performative contradictions. In contrast, the propositions deduced from
the understanding of specific phenomena are only true a posteriori. The former is “theory,”
e.g., praxeology and economic theory, while the latter is “history.”
Schütz seems to have disagreed with Mises on the epistemological status of the proposi-
tions of economic theory, and specifically he seems to have been critical of Mises’ apriorism
and rejected, what Barry Smith (1994) has called the ‘impositionist’ character of Mises’
apriorism.
One problem here, however, is that Mises never fully articulated the epistemological
basis of his apriorism, and in particular did he not tie these explicitly to any particular
philosophical basis, not even that of Kant, despite adopting a number of Kantian concepts.
This was in fact something which Schütz, himself primarily a philosopher, on several oc-
casions expressed his desire for. Already in his review of Mises ([1933] 1981) Schütz
regretted that Mises “analyzes the character and methods of the social sciences in a fine
manner though he abstains from a discussion of the question of a general epistemology
basic to the problems of the social sciences” (Schütz [1934] 1996:92). Schütz indeed later
admitted to not quite understanding exactly what Mises meant by “aprioristic,” in particular
whether it signified something beyond the general validity and universality, which Schütz
himself acknowledged, e.g., as used “in the sense of one of the great systems of philos-
ophy” ([1936] 1996:103).14 If that was not the case then he hoped Mises would explain
it in greater detail, or better, suggested Schütz, drop the term and notion of the a priori
altogether. This so since, in his opinion, it will necessarily lead to confusion, because it
128 KURRILD-KLITGAARD

is used in so many different philosophical senses. Furthermore Schütz found the notion
unnecessary:

There is no reason why the question of the current philosophical meaning of the a priori
should intensely occupy methodologists of the social sciences. . . . We social scientists
deal with mundane phenomena and their realities within the world. We do not ask about
the being of the world as such but are satisfied that our propositions are of general
and universal validity within the pregiven world of mundane phenomena—no matter
whether this world hangs together without contradiction as a world of appearance or
as one of genuine being. (Schütz [1936] 1996:103; italics added)

Schütz did not himself provide any elaborate examination of the epistemological status and
validity of the propositions of economic theory. It would, however, seem that the universal
validity which he ([1932] 1967:243–245, [1936] 1996:103)—with explicit reference to
Mises—ascribed to the propositions of economic theory for him were only so exactly qua
commonsensical and by definition:

[It] becomes clear that universal validity is ascribed to these propositions in the sense
‘that they are not an expression of what may occur as a rule but of what will occur
always and necessarily’. And that means that such is the case provided the theoretical
realm, as delimited by the principle of marginal utility, will not be abandoned. This is
nothing but a consequence of the universality of the principle of marginal utility for the
realm of invariance defined by it. To the laws of political economy accrue universality
and objectivity in the same sense in which the laws of physics are ‘universally valid’—
but also only within their established realms of invariance. Both types are hypothetical
in character. (Schütz [1936] 1996:103; italics added)

Schütz accordingly seems to have based his own views on the status of the propositions
of economic theory in a commonsense conception of what we as acting individuals know
typically to be the case, when we are confronted with a problematic choice. The validity
of any models—including the most abstract laws of economics—would thus seem to be
dependent upon the adequacy of their reference to the subjective meanings of the actors,
i.e., upon their empirical grounding in the meaning structures of the commonsense action
of the life-world (Schütz [1932] 1967:244, Srubar 1994:36).
This would seem to be much closer to the views of Weber than those of Mises, and
Schütz ([1936] 1996:103) was quite clear that such a view was contrary to Mises’. But is it
necessarily in conflict with economic theory a la the Austrian School? Some would seem to
think so, e.g., Srubar (1994:36ff), but I will briefly indicate two closely connected reasons
why I do not think that this is so.
One reason is that the fundamental praxeological laws on which Austrian economists
seek to base economic theory need not necessarily be based in a quasi-Kantian apriorism.
Rothbard ([1973] 1979:35–36, 1976:29) thus approvingly referred to Schütz in his attempt
to distinguish his own Aristotelian grounding of the laws of economics from the Kantianism
of his mentor, Mises, by saying that these are based on knowledge which “is ‘empirical’
in the broadest sense” (Rothbard 1976:29, cf. Rothbard 1957:318). Similar to Rothbard
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 129

several other theorists sympathetic to Austrian economics have recently pointed towards
the possibility of grounding the fundamental laws of economics not in Kantian apriorism
but in Aristotelian realism (e.g., Smith 1986, 1990, 1994).15
Another reason is that the disagreement is of little or no practical importance. This is so
since Schütz seems otherwise to have accepted the basic concepts, axioms and theorems
of economic theory. If the same propositions are accepted, and if the employed logic oth-
erwise is the same, then there is little reason to believe that the derived conclusions should
be different.

3.2. Rationality, Choosing and the Scope of Economic Theory

It was previously mentioned that Schütz explicitly agreed with Mises that it is not an
assumption of maximization of satisfaction, which constitutes the essence of economic
theory. Rather it is, in Schütz’s formulation, the principle of marginal utility, understood as
a formal principle of actors making problematic choices among alternative and mutually
exclusive courses of action. Schütz did, however, on several occasions express disagreement
with Mises on the latter’s claim, that ‘economizing’ is a characteristic of all action and
consequently that all action may be analyzed in terms of marginal utility. Indeed, while
Schütz almost seemed agnostic concerning the epistemological status of the propositions of
economics, he did nonetheless explicitly disagree with one possible interpretation of Mises’
a priori, and notably one which Mises most likely would have subscribed to. This would be
the one that “all acting is economic acting because it implicates preferences and planning.”
(Schütz [1936] 1996:103; italics added).
Now, it is important to be clear about what it is that Schütz objected to, and what it is
not. For Schütz the marginal principle—and hence rationality—is the essence of economic
theory ([1936] 1996:102); in this respect there is not as such anything contrary to Mises,
so it is not at this point that one should locate the disagreement. Furthermore, Schütz
also, as previously mentioned, believed that the operationalization of this principle entails
modeling all actors as being rational,16 so it is neither here that the disagreement should
be located. The disagreement is rather to be found in the role, which Schütz attributed to
the marginal principle as a practical delimitation of what does and what does not constitute
economic theory. Having explicitly stated his basic connection and affinity with the Austrian
School Schütz stated this in response to Loewe who had criticized the principle of marginal
utility:

But, as I see it, the marginal principle performs a second function and this is the
delimitation of the field of economics. It has, therefore, a determinative character and
carries out what I called in a casual remark in one of our last meetings as “actio
finium regundorum”. It answers the question which kind of human activities have
to be considered as economic ones. Although I do not believe . . . that the marginal
principle involves necessarily the idea of maximization of satisfaction, I should like
to maintain that even if it did so it would tell us something about the kind of human
behavior which is a subject matter of economic science (italics added).17
130 KURRILD-KLITGAARD

It is here that Schütz’s qualms with Mises’ equation of action and economic action is
located. By doing so Mises, in Schütz’s view, made economic theory identical with general
praxeology, and thus empties the disciplinary delimitation of meaning:

If one wants to qualify all acting as economic, one must state in what way this action
differs from other manifest cases of human conduct which are called actions or activities
in psychology and philosophy. However, if one does not deprive the term ‘action’ of its
general connotation, one faces in turn the task of deciding what differentia specificae
are characteristic of economic actions. Whichever alternative I may choose, it seems
purposeful to separate acting turned toward so-called economic goods from other
acting. (Schütz [1936] 1996:103–104)

This point was later elaborated by Schütz in his discussion with Loewe, on a point about
the delimitation of economics vis-a-vis other social sciences. Loewe had criticized Lionel
Robbins’ definition of economic behavior as “the disposal of scarce means for alternative
ends” and instead suggested “the transformation of originally given matter into matter suited
for our purposes”. Schütz in return criticized Loewe’s supposed definition and explicitly
raised the question of what he saw as Mises broadening of economics to all areas of human
action:

But are not human activities called ‘services’ also economic activities? And what is
the criterion for these specific activities? Neither Robbins’ definition nor your own
answers these questions unless you take, as my friend Mises does, the shortness of
our life and the impossibility of performing everything we would like during a life
span as a problem of scarcity; then time, like what you call matter, would be a scarce
means for alternative ends. Mises is trying to develop a general praxeology which
he identifies—erroneously as I think—with the theory of economic action, namely
an action according to the assumed scale of preferences of the actors. For example, I
obviously preferred to put my thoughts on this problem before you in the form of the
present letter to my going to the movies or what not. If this were the case there would
be no human action whatsoever which were not an economic action and if gentlemen
prefer blonds they were economic subjects—(the gentlemen, not the blonds!) (italics
added).18

Here it seems, after all, as if Schütz fairly well captured the essence of Mises’ position,
albeit only in order to reject it. Schütz thought that such a view would be what Weber
had called the “rabbiat gewordene Lokalpatriotismus der Fachwissenschaften,”19 or what
we today would call ‘economic imperialism’. Schütz gives this reason for considering it a
problem:

After all, I could say that all human activities are of legal relevance. Sitting here in my
room and writing to you I can do so only because I have rented the room, purchased the
paper, etc. I could also refer everything I am doing to attempts of my organism to come
to terms with its environment according to the nice scheme of stimulus and response,
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 131

and you know that even this idea found advocates in highly esteemed philosophical
circles. Nevertheless, the problem remains: Which is the criterion in accordance to
which the economist considers certain actions and certain events in the world as falling
within his field, whereas he accepts others as mere meta-economical data which he
has to take into account but which to explain is the business of his colleague in another
department.20

This is a valid discussion—and highly so. But it would seem that Schütz here failed to
completely grasp Mises’ position and indeed was somewhat off mark in his critique. Mises
quite clearly made a distinction between what was to be considered rational and economic in
his theorizing and what was not, albeit a very broad one: It is only purposive behavior, which
is the subject of praxeology and economic theory. And since action necessarily is choice
among alternatives and necessarily includes some resources, it must, according to Mises,
necessarily be economizing in nature. For Mises the claim that all action is economizing
is not a feature imputed to a model of an economic agent (i.e., an individual exhibiting
economizing action) but a necessary feature of all agents, real or imagined. Neither is
the claim—as Schütz believed Mises to hold—the limiting device of economics, i.e., what
Schütz elsewhere called the principle of marginal utility. The statement, according to Mises,
is of a much more general character; economics—at least in Mises’ later formulations and
those of Rothbard ([1962] 1970)—is not identical to praxeology as such but rather is a
limited part of it dealing with ‘catallactics’, i.e., human relations based on exchange.
But Schütz may have had another, somewhat more fundamental point than the delimitation
of economic theory from, e.g., sociology and psychology. In the same letter in which Schütz
declared his basic debt and adherence to the Austrian School he also tried to explain to Loewe
where he saw the main difference, in these matters, between Mises and himself:

It is also clear that my attempt of analysing the process of choosing between problematic
possibilities has nothing to do with Mises’ general praxeology because the decisive
problem involved is taken just for granted by Mises, that is, the problem how it comes
that things stand to choice at all. He overlooks also the difference which seems vital
to me, namely on the one hand choosing between objects equally within my reach and
on the other hand, choosing between projects of actions which have to be carried out
by me.21

In this case Schütz truly and explicitly points towards what he sees as genuine differences
between Mises and himself. But there would seem to be two separate questions here.
Considering for a moment the latter first, it is not perfectly obvious why there would be
a categorical difference between the two and why any difference should be relevant for
economic theory per se. The former—choosing between objects equally within reach—
would seem simple, while the latter—choosing between projects of action to be carried
out—would presuppose more complicated reasoning, including considerations of time,
opportunity costs, etc. But from a praxeological view, it would not seem a categorical
difference.
The first of the problems mentioned is that Schütz criticizes Mises for not dealing with
how individuals make decisions and why. This is correct in so far as Mises explicitly stated
132 KURRILD-KLITGAARD

that praxeology and economic theory does not deal with the origin of preferences but
exclusively with the logical implications that follow from the fact that individuals have
preferences and act to realize these. But this would indeed seem simply to be, more or
less, the way in which Schütz himself distinguished between the because-motive and the
in-order-to-motive of an action: Sociology, psychology, etc., as sciences deal with how and
why individuals come to have particular because-motives, while praxeology deals with how
individuals act to achieve their in-order-to-motives.22 In this respect it would seem—to a
large extent—simply to be a question of exactly the delimitation of the economic. Misesian
praxeology does not deal with the origin of because-motives, because this is outside the
logic of choice.
Nonetheless, there would seem to be a fundamental and important point to Schütz’s
distinction, namely that economic theory in many instances—and economic history in
particular—eventually will need to use insights into “how it comes that things stand to
choice at all.” Why is this? The reason is that it might not be quite transparent exactly
what should be understood by a choice being rational. A great deal of literature—as well
as sophisticated concepts and tools—have been developed since the 1920s and 1930s when
Mises developed his praxeology. In order to develop economic theory it is insufficient
simply to use a definition of behavior as being rational that it is that action which will
convey most utility to the agent (or, as Mises might have said, is believed to alleviate
uneasiness the most).23 For does such a concept imply a simple straightforward maxi-
mization of ‘satisfaction’, or is it something more akin to subjective expected utility? Or,
how about agents who may be satisficing rather maximizing? Quite clearly these are all
relevant ways in which individuals meaningfully may make their choices, but how can
knowledge of the relevant reasoning be derived from pure reflection upon the abstract
nature of action itself? On this issue it would seem insufficient to answer that praxeol-
ogy does not deal with the reasons why individuals have the preferences they do, since
that is an altogether different question. Here Austrian economists would seem to be in
serious need for elaborating on exactly what is supposed to be contained in the propo-
sition that individuals choose that action which they believe will give them the most
utility.
Schütz, however, also had a deeper point about whether acting necessarily could be seen
as being economizing. Now, as indicated, this would as such not be a problem for Mises,
who quite explicitly made such a distinction and who certainly must be seen as making a
distinction between praxeology and economic theory. Schütz, however, pointed towards a
more fundamental problem for starting out with preferences and choices:

Should one—as many philosophers do for pertinent reasons—consider cogitare (think-

ing) a form of acting, then one would arrive at completely absurd results if the intention
is to interpret thinking as preferential or selective action. And what kind of acting shall
this preferring and selecting be if it is considered acting on its part? ... One cannot
be satisfied with the concepts of choosing and preferring without careful analyses. As
mentioned before, choosing and preferring themselves are actions.
When I walk with a friend in a park and turn into the left instead of the right path, can
I meaningfully say that I preferred the left path? And are not what we call choosing and
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 133

preferring complex processes occurring in elapsing time? (Schütz [1936] 1996:102,

note and 104)

3.3. Ideal Types

Probably the most obvious point of disagreement between Misesian Austrians and Schütz is
his view of the ideal type, which seems quite different from that of Mises and possibly that
of Hayek too. As set forth in “Begreiffen und Verstehen” the Misesian position on the ideal
type was that it is an entirely historical concept. Hayek certainly seems to have followed
this, when he wrote to Schütz as a response to the draft of Schütz’s article, “The Problem
of Rationality in the Social World” ([1943] 1964), that his own problem with Schütz’s ideal
type approach is “still the element of historism which, at least in my mind, attaches to the
concept of ideal type, a concept that to me always suggests a fictitious entity rather than an
abstraction”.24 This is, indeed, also the way that many modern Austrians seem to interpret
Mises.25
For Schütz all social scientific conceptualizations of actors are ideal types, and Schütz’s
point was that all such necessarily must be ‘less’ than real, red-blooded human beings, al-
though they may come in various degrees of ‘anonymity’. They are either so-called “formal”
(or “course of action”) ideal types where individuals are seen as functioning in particular
roles, or they are “personal ideal types” where some motives are described or postulated
as being constant. The latter is typically the procedure found in economic theory and, ac-
cording to Schütz, so out of necessity; it is simply impossible to formulate economic theory
without postulating some elements as constants:26

The economic subject with which economists deal are not humans of flesh and blood
like you and me, Peter and Paul and everyone. They exist only by the grace of
economists. The latter created these homunculi (artificially animated man-made ma-
chines) in order to experiment with them. Such ideal types set into the world in this way
cannot know, intend and expect anything else but what has been predestined for them
by those economists <who designed them>. It is impossible for them to display any
other action or conduct than that for which they were designed. In short, like any other
ideal type, the ideal type of economic subject cannot transcend the realm assigned to
it; it cannot act atypically. (Schütz [1936] 1996:104–105)

Even the simple “acting man,” homo agens, on which Mises and the Austrians base their
praxeological analysis, is—in Schützian terms—an ideal type, for it is a conceptual con-
struction, an actor model created by an observer, rather than the actual agent himself, even
if it is the most highly abstract imaginable, homo agens is still an abstraction, something
less than reality. That, of course, is also, and in particular, true of homo oeconomicus and
“his class-mates,” as Machlup would have put it:

[Neither] I nor you nor anyone in the fullness of his existence is identical with those
subjects of economic life of which economists speak. We are related to this construction
as to merely one side of our being human, <the side> that responds to those schemes
134 KURRILD-KLITGAARD

which economists posit as economic subjects. What economists consider subjects of

economic life are not humans in the fullness of their existence, but are rather ideal types.
That is, they are fictive beings imagined to be equipped with conscious experiences
(goals of action, motives, actions, etc.) and they are considered sufficient for acting
out those economic events which the economist considers sufficiently relevant for his
problem. (Schütz [1936] 1996:99)

The point which Schütz wanted to make in relation to economic analysis is that ideal types
generally will only suffice for one “specific layer of depth” (Schütz [1936] 1996:99), namely
the one for which it was formulated. This goes, Schütz explicitly says, no matter whether
the actor model—personal ideal type—employed is that of the homo agens of Misesian
praxeology, the homo oeconomicus of price theory or that of an omniscient economic
dictator:

[Economists] treat these conceptual models, on the one hand, as though they were hu-
mans of flesh and blood like you and I and Peter and Paul and everyone who encounters
them in daily life; on the other hand, they endow them with that preknowledge, with
those experiences, and even with that specific orientation of interest which are charac-
teristic only for a few, namely the theorizing economic scientists themselves. (Schütz
[1936] 1996:99)

To illustrate Schütz’s point about the “relevance” of ideal types we may consider an example
inspired by Machlup’s Schützian contrasting of different propositions, all dealing with
economic phenomena and having the same logical form but embodying ideal types of
different degrees of anonymity (Machlup [1936] 1978:64). One of the examples used by
Machlup is that of a central bank director confronted with the choice of whether or not to
intervene given heavy withdrawals of foreign deposits. Will he do so or not? Obviously
the answer to this question will depend upon the preferences of the central bank director,
including his ideology, etc. Equally obvious his choice of action will depend upon his
constraints in terms of knowledge, expectations about the possible behavior of others,
ranging from those of consumers to those of the cabinet members, and the likelihood of these,
etc. “It will make a great difference,” Machlup said somewhat tongue-in-cheek, “whether
Mr. Keynes or Professor von Hayek is governor or expert adviser of the central bank.”
(Machlup [1936] 1978:68). So to answer the question of whether it is rational for the
central bank director to intervene given his preferences, constraints, method of reasoning,
etc., we will need to introduce the relevant assumptions. But by doing so we are moving
beyond the anonymity of the pure homo agens—the behavior of anyone—and we are thus
no longer dealing solely with the pure logic of choice. Indeed, we are moving beyond the
universal validity which may be ascribed to the abstracts laws of praxeology.
Schütz’s point about relevancy may thus be seen to be that, for example, the completely
anonymous homo agens will be of very little use for the analysis of, say, the decision of a
central bank director faced with the choice of whether or not to intervene. It will simply
be too anonymous to allow us to formulate propositions about what the choice will be. On
the other hand, an ideal type of a Keynesianally inclined central bank director will most
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 135

likely be of little use for the analysis of the behavior of consumers responding to changes
in supply in, e.g., the food market; it will not be sufficiently anonymous.
The importance of these points is that when the limits of the ideal type are acknowledged,
so must the limitations upon its universality. If not, then the analysis will no longer be valid:

Here a fictive world comes into existence alongside the actual world. No danger will
arise from this fact as long as the economists keep the rules of typification constant and
their types correspond sufficiently to empirical realities. We pursue political economy
merely in order to recognize and control this economic reality. Yet economists should
avoid the fate of Pygmalion whose sculpture gained a weird life of its own. Economists
should not transpose their models into the mundane world and treat them like humans
with knowledge, experience, error, and freedom. (Schütz [1936] 1996:99–100)

Schütz’s points thus seem to stand; it is necessary to formulate ideal types, and these
should be formulated in accordance with the principles of being (1) commonsensically
understandable and (2) adequate for the phenomena they seek to explain, and when applied
to the field of economics this includes the rationality principle and the principle of marginal
utility.
When theorems are deduced from ideal typical modeling, the implications hold, as Schütz
liked to quote Mises, “are valid always and everywhere if the conditions assumed by them
are given” (Mises [1933] 1981:85–86). But for Schütz they do not do so due to any alleged
aprioricity; rather they do so given that important last part of the quote: They hold so because
and in so far as the assumed conditions allow them.
But what about Mises’ homo agens? Would he not be an example of something where no
specific preferences are held constant and others omitted? And would it not, simultaneously,
be the case that very useful, albeit very general, statements could be deduced, which could
be compared to empirical reality? I suppose the Schützian answer would be something like
this: Yes, homo agens is an example of a highly anonymous ideal type, and it is possible
to deduce very general statements from it. In this respect it would seem that Hayek—and
some later Austrians—have misunderstood Schütz somewhat: For Schütz even homo agens
is an ideal type, but one completely emptied for any history whatsoever.
Under closer scrutiny it would, however, even for homo agens be necessary to impute
some motives in order to deduce meaningful statements of empirical relevance, e.g., that he
prefers more to less and leisure to work, etc. It is, for example, perfectly possible to imagine
actors not considering leisure a consumer good and companies not always pursuing profits,
just as it is possible to imagine some individuals not exhibiting entrepreneurial alertness.
None of these assumptions can be seen as always and everywhere necessarily being a priori
present in each human action, as Mises explicitly required in order for the deduced laws to
hold with necessity (Mises [1933] 1981:78–79).
This is indeed acknowledged by, e.g., Rothbard, who speaks of the ‘Fundamental Ax-
iom of Action’ being supplemented by “a few subsidiary postulates which are actually
empirical.” (Rothbard 1957:315), including leisure as a consumer good.27 Such “subsidiary
postulates” may, as Rothbard (1957:315–316) has argued, be “empirical postulates” which
are legitimate in purely theoretical analysis because they are few in number and “so broadly
based as to be hardly ‘empirical’ in the empiricist sense of the term.” so ‘broadly empirical’
136 KURRILD-KLITGAARD

as to be uncontroversial. But since it is perfectly possible to imagine someone actually

preferring work to leisure or less money to more money, they are assumptions which nec-
essarily must be introduced as auxiliary assumptions. That in itself necessarily limits the
analysis—although, as Rothbard pointed out, probably without much difference. 28
This does, however, not invalidate the propositions of economic theory; it only means
that the conclusions only apply when the relevant assumptions are given as being relevant—
but when they do, the conclusions follow with necessity. This is indeed what, e.g., Mises
([1933] 1981) and Rothbard (1957) may be seen as saying themselves.
But how about when we thus move beyond the pure Misesian praxeology—what Schütz
called the “formal analysis of action” and Hayek and James Buchanan the “pure logic of
choice”—and focus on the interaction on two or more individuals, exchanging goods and
services in a money-based market? In that case, Schütz argues, we will very soon need to
introduce further assumptions as “data” into our modeling. Schütz was very conscious that
as soon as we as social scientists adopt “the subjective direction” of research and move
beyond the observable behavior and prices and into that of subjective choice and aim “at
the exact recognition of phenomena”. Here we will need to take into consideration the
expectations of consumers and producers, their knowledge, their perceptions of risk and
possibilities of error, etc. (Schütz [1936] 1996:95).
Mises would, of course, not disagree with the importance—and necessity—of including
such concepts; it was, indeed, one of his major points, as Schütz was very well aware. But
Schütz’s point was deeper. If, for example, we want to develop a theory of how the market
works or study a specific, actual market, then we must necessarily include assumptions
about, e.g., the role of knowledge and expectations. Surely, these concepts may be viewed,
as what Mises would have called praxeological categories, or concepts belonging to the
category of action. Koppl (1997, 1998) has in his comparisons of Mises and Schütz made
these points as regards expectations:

If Mises was right to contrast conception and understanding, there can be no theory of
economic expectations. Expectations are formed through the unfolding of the market
process. They are a matter of entrepreneurial understanding. Thus there is an essential
extra-rational element to them. Given Mises’s epistemology, a theory of expectations
would have to be a matter of discursive reason. But we can have no discursive concep-
tion of acts of understanding except to recognize them as such. We can only ‘behold’
entrepreneurial expectations through acts of understanding. Here conception cannot
penetrate. Thus no ‘theory’ of economic expectations is possible for Mises. (Koppl
1997: 67; italics added)

Similar points may be made about the knowledge of the actors more broadly. Indeed, the
perhaps most important aspect in the modeling of agents, according to Schütz, is the question
of knowledge. For Schütz the methodological subjectivism which he, Mises and the other
Austrian economists shared, represented nothing less than a “Copernican turn”: “Once this
turn is carried out, the question of the problem of knowledge in economics expands itself
to that of grasping the knowledge characteristic of one economic subject or else some or
all of them at a specific time.” (Schütz [1936] 1996:99). But whether the actors of a model
(or reality) have perfect knowledge, some knowledge or no knowledge cannot be deduced
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 137

from the concept of homo agens and the category of action alone. It must necessarily be
introduced through some form of auxiliary assumptions, and what is introduced will have
considerable significance for the resulting theorizing. Schütz may actually very well be seen
as embracing an argument for the relevance of what today might be called ‘the economics
of knowledge’:

An especially good example of the importance of such reflection is the problem of

‘data’ in political economy. It is certainly time to stop using the concept of ‘data’, or
even of the ‘given datum’, as a refuge for ignorance. The definitely legitimate question
is: To whom are such data given? To me, living my daily life, the whole environment
is, as it were, given as an environment of data. It is a datum for me that the sun will
rise tomorrow, but also that the streetcar will work, and that I can buy merchandise in
a store for my money. It is a datum for me, further, that the stockmarket in Paris is in
disorder and that the price of tea will rise because if the poor harvest in India. In this
sense, all my experiences of the external situation on which I have no influence are
data.
According to my given situation of interest, I will pay attention to one or another
factor; I will even analyze sufficiently the phenomena of importance for me to the
extent called for by the situation and allowed by my experience. For instance, it will
suffice for me as passenger to be informed about travel-times and frequencies of the
streetcar lines. However, if I am not a technician and if it is not relevant for me for
other reasons, I am not concerned with the construction and technical layout of the
electric streetcar—even those these too are data for me. Thus, what is a datum for us
in daily life and what we keep in view of our considerations is what is relevant for us
in the given case. We are ready to analyze it and do this either on the basis of a review
of the annual stock of our experiences in the matter or on hand by way of potential
information. (Schütz [1936] 1996:96)

While Austrian economists over the last decades in increasing degrees have paid attention
to the role of knowledge, it is certainly an area which needs to be more fully integrated into
the Austrian corpus.
Does this have any negative consequences for Austrian economic theory? What should
be clear for the present purposes is that Schütz does not deny the possibility of formulating
abstract laws applicable to the field of economics a la those of Austrian economics (cf. Schütz
[1932] 1967:245ff, Srubar 1994:39). The point which Schütz tried to make in his criticism
of Misesian economic theory is that the praxeological theorems constituting the core of
economic theory can maintain the universal validity desired by Mises and the Austrians if
and only if the actors remain completely anonymous, i.e., only as long as the individuals
simply are seen as doing what they do, because that is something which they prefer to
not doing so. But, in contrast, if further assumptions are added then the universality and
necessity must give way to less general validity. They may, however, be very close to being
universally valid, since it probably is true—as Rothbard argues—that such assumptions are
so broadly empirical as to be uncontroversial. Indeed, as Schütz himself said: “The principle
of marginal utility does not contain inner contradictions and agrees with the postulate of
138 KURRILD-KLITGAARD

possibility. And more: a good chance exists that you and I and everyone orients his conduct
to it.” (Schütz [1936] 1996:102; italics added).

Conclusion

The present study has demonstrated several things:

First, Schütz should definitely be seen as being not just sympathetic to economic reasoning
and what today would be called rational choice theory but indeed largely as a contributor
to and even a proponent of that approach. In this way the paper is in line with a number of
the more recent studies, and perhaps even takes the points further.
Second, Schütz may, however, be seen as only admitting a limited role to rational choice
analysis. The use of simple and anonymous rational choice models for the explanation of
phenomena, such as those typically found in economic theory, may be relevant and sufficient
in some cases, but the more specific the situation the more rational choice analysis alone
becomes insufficient.
The Schützian “message” to economists—Austrians as well as others—may thus be
boiled down to the following points:

• The completely tight distinctions between “conception” and “understanding” and be-
tween “theory” and “history” do not hold (Koppl 1998). There is only a matter of degrees
(albeit many so) between, on the one hand, “the logic of pure economics” (Schütz),
the “pure logic of choice” (Hayek), and “praxeology” (Mises), and, on the other, the
formulation of more or less specific models of complex phenomena.
• In the more full and complete analysis of specific situations of choice it may be insufficient
merely to consider what follows given that individuals act on the basis of given preferences
and constraints. The more empirically oriented the analysis, the more the scientist will
need to incorporate assumptions about a number of issues, e.g., the particular form of
reasoning of the agent, his beliefs, knowledge, expectations, etc.

These are points of relevance for the study of economic phenomena, but also for the appli-
cation of rational choice models to such related disciplines as political science (Langlois
1988, Kurrild-Klitgaard 1997b) and sociology (Esser 1993a, 1993b).
Would Schütz’s criticism of Mises indicate that he should not be seen as—even remotely—
a member of the Austrian School? Some contemporary Austrians might think so, but they
would—in my opinion—then also have to discard of, e.g., Hayek, Haberler, Machlup and
several other of the most important inter-war members of the school. Whatever one might
think of Schütz’s points of disagreement with Mises, one must remember two things. First,
Schütz’s intellectual project may to a significant extent be seen as an attempt at supporting
the Austrian program against its increasing number of opponents rather than the opposite
(Prendergast 1986). Second, it would seem that Schütz otherwise was in general agreement
with all the core points of what in the 1920s and 1930s was seen as constituting the Austrian
program (Kurrild-Klitgaard 1997a, Koppl 1998).29
Some contemporary Austrians may find many of Schütz’s points hard to swallow, even
if, e.g., Mises and Rothbard in general endorsed his work. But one certainly does not need
to agree with all Schütz’s points in order to appreciate that the questions he raised were
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 139

original and of a fundamental importance to the development of Austrian economics, or for

that matter for the discipline of economics itself. Without greater attention to these, some
serious lacunas may remain.

Notes

1. This point is actually noted indirectly by one of the leading ‘economic imperialists’, public choice school
co-founder, Gordon Tullock. He has on a number of occasions, e.g. Tullock (1994:53), pointed to his early
reading of Mises’ praxeology as being the primary inspiration for his own application of economic reasoning
to political processes. For a criticism of hermeneuticist philosophy from a student of Mises sympathetic to
Schütz’ work, see Rothbard (1989).
2. For a dissenting view, indirectly opposed to such an interpretation and instead favoring a ‘psotmodern’ version
of economics, supposedly inspired by Schütz, see the works of Pietrykowski (1996, 1999). It is, however,
worth noting that Pietrykowski undertakes this interpretation without utilizing any of Schütz’s writings on
economics or economic methodology and only relies on some of the very late essays by Schütz.
3. See, among several other examples, e.g., Schütz ([1932] 1967) pp. 197n, 199n, 212, 227 and—in particular—
pp. 242–246, as well as Schütz ([1934] 1996).
4. Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University).
5. It may be noted that Schütz originally gave this important paper as a lecture in 1940 at the Faculty Seminar
at Harvard, headed by Talcott Parsons and Schütz’s countryman, Joseph Schumpeter, and with Gottfried
Haberler as the middleman (cf. Wagner 1983:74, 82).
6. Gottfried Haberler’s, F.A. Hayek’s, and Fritz Machlup’s archives are located at the Hoover Institution, Stanford
University, and in particular the Machlup archives contain much of the correspondence with Schütz. Schütz’s
part of the correspondence is located at the Beinecke Library, Yale University.
7. This correspondence, which is located in the Schütz archives at Beinecke Library, Yale University, includes
a number of interesting letters from 1955–56, e.g., a letter from Schütz dated 17 October 1955.
8. Cf., e.g., Machlup (1981) who on several occasions identified these issues as being “the essential distinguishing
characteristics of Austrian economics” and “the most typical requirements for a true adherent of the Austrian
school”: (1) Methodological individualism, (2) methodological subjectivism, (3) an emphasis on tastes and
preferences, (4) opportunity costs, (5) the principle of marginal utility, (6) the time structure of production and
consumption, (7) consumer sovereignty, and (8) political individualism (Machlup 1981:21–22). As Machlup
noted some of these tenets have since been become parts of mainstream economics.
9. Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University).
10. Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University). See also
Schütz (1972:584–587).
11. Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University).
12. Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University).
13. The translation erroneously says “Political scientists.”
14. This lack of understanding of the meaning of the a priori might surprise contemporary students of Mises,
who are familiar with Mises’ books ([1949] 1966, [1957] 1984, [1962] 1978). Schütz, however, explicitly
pointed out that his lack of understanding was based on “the works of von Mises published so far” ([1936]
1996:103; italics added). Here it must be kept in mind that at the time Mises’ only writings on economic
methodology were the essays collected in his anthology on the subject ([1933] 1981). Furthermore, there are,
as I hope to explore further at a later point, indications that Mises radicalized his views on the extent of the a
priori character of the laws of economics.
15. Such a modern Aristotelian realism might indeed be the bridge between Mises’ aprioristic but semi-essentialist
analysis of the logic of action and Schütz’ brief remarks about the possibility of formulating ideal types and
derived propositions not necessarily based exclusively in “empirical” knowledge in the sense of “derived from
the senses”. Such knowledge, according to Schütz, is “eidetic” knowledge which is “derived from essential
insight.” (Schütz [1932] 1967:244).
16. Cf., e.g., Schütz ([1932] 1967:239, [1936] 1996:102, [1951] 1962:86–87, 1972:584–587) and Alfred Schütz,
Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University).
140 KURRILD-KLITGAARD

17. Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University).
18. Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University). Cf. also:
“Authors who consider all acting ‘preferential acting’ do not realize that with this thesis they provide a
definition of acting as such and that thereby <unwittingly> have set themselves the task of separating this
acting from other manners of human existence.” (Schütz [1936] 1996:102, note).
19. Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University).
20. Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University).
21. Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University).
22. Cf., e.g., Alfred Schütz, Letter to Adolph Loewe, 7 December 1955 (Beinecke Library, Yale University), and
Eberle (1988:86).
23. Rothbard does not answer this either. His most elaborate statements would seem to be these: “All action is
an attempt to exchange a less satisfactory state of affairs for a more satisfactory one. The actor finds himself
(or expects to find himself) in a nonperfect state, and, by attempting to attain his most urgently desired ends,
expects to be in a state.... Therefore, all action involves exchange—an exchange of one state of affairs, X,
or Y, which the actor anticipates will be a more satisfactory one (and therefore higher on his value scale). If
his expectation turns out to be correct, the value of Y on his preference scale will be higher than the value
of X, and he has made a net gain in his state of satisfaction or utility. If he has been in error, and the value
of the state that he has given up—X—is higher than the value of Y, he has suffered a net loss.” (Rothbard
[1962] 1970:16; italics in original) and “Since man is always acting, he must always be engaged in trying
to attain the greatest height on his value scale, whatever the type of choice under consideration. There must
always be room for improvement in his value scale; otherwise all of man’s wants would be perfectly satisfied,
and action would disappear. Since this cannot be the case, it means that there is always open to each actor
the prospect of improving his lot, of attaining a value higher than he is giving up, i.e., of making a psychic
profit. What he is giving up may be called his costs, i.e., the utilities that he is forgoing in order to attain a
better position. Thus, an actor’s costs are his foregone opportunities to enjoy consumers’ goods. Similarly,
the (greater) utility that he expects to acquire because of the action may be considered his psychic income,
or psychic revenue, which in turn will be equal to the utility of the goods he will consume as a result of the
action. Hence, at the inauguration of any action, the actor will believe that this course of action will, among
the alternatives, maximize his psychic income or psychic revenue, i.e., attain the greatest height on his value
scale.” (Rothbard [1962] 1970:62; italics in original). Judging from his brief and early discussion ([1956]
1970:16–17) Rothbard should probably be seen as also rejecting the subjective expected utility maximization
approach of von Neumann–Morgenstern game theory.
24. F.A. Hayek, Letter to Alfred Schütz, 25 December 1942 (Beinecke Library, Yale University).
25. See, e.g., Selgin ([1988] 1990:20ff), who, however, in his discussion fails to fully understand the only gradual
difference between highly anonymous ideal types and less anonymous ones.
26. On these points in relation to economic theory, see, e.g., Schütz ([1932] 1967:241-249, [1936] 1996:95,
101–102, 104–105).
27. The three other auxiliary assumptions (other than leisure as a consumer good), which are necessary to analyze
barter and a monetary economy, are, according to Rothbard (1957:316): (1) An existence of a variety of natural
and human resources, and, “as limiting subdivisions,” (2) that “indirect exchanges are being made,” and (3)
that “firms always aim at maximization of their money profits.”
28. Schütz would no doubt have found Rothbard’s procedure quite acceptable: “Every [personal ideal] type, so to
speak, carries an index; it points to the purpose for which it was formed. . . . The same goes for the types formed
by social scientists. Their situations of interest are determined by the problems they had posed themselves
and by the layers of depth on which they wished to deal with these problems. This is no empty phrase. As I
have shown, with the selection of a problem and the determination of the layer on which to treat it, what is
thematically relevant is already established as well as what can be treated as a datum—and, accordingly, what
is considered extra-thematic and thus invariant. This is what economists mean by the expression, ‘ceteris
paribus’.” (Schütz [1936] 1996:101).
29. I have here not considered the possibility that Schütz may have changed his views on economics over the
years. An anonymous referee has indeed argued that Schütz “grew increasingly impatient with economics
over the years” and that this was a result of the failure of economists to develop a genuinely “empirical
subjective approach”.
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 141

References

Boettke, P. J., Horwitz, S., and Prychitko, D. L. (1986) “Beyond Equilibrium Economics: Reflections on the
Uniqueness of the Austrian Tradition.” Market Process, 4(2): 6–9.
Bradley, Jr., R. (1979) “Symposium on Theory and Method in the Social Sciences Held in Milwaukee.” Austrian
Economics Newsletter, 2(1): 4.
Ebeling, R. M. (1987a) “Expectations and Expectations Formation in Mises’ Theory of the Market Process.”
Market Process, 5(1): 12–18.
Ebeling, R. M. (1987b) “The Roots of Austrian Economics.” Market Process, 5(2): 20–22.
Ebeling, R. M. (1987c) “Cooperation in Anonymity: Review of ‘Anonymity: A Study in the Philosophy of Alfred
Schütz’, by Maurice Natanson.” Critical Review, 1(4): 50–59.
Eberle, T. S. (1988) “Die deskriptive Analyse der Oekonomie durch Alfred Schütz.” In List, E. and Srubar, II.
Alfred Schütz: Neue Beiträge zur Rezeption seines Werkes, pp. 67–120. Amsterdam: Rodopi.
Esser, H. (1993a) “The Rationality of Everyday Behavior: A Rational Choice Reconstruction of the Theory of
Action by Alfred Schütz.” Rationality and Society, 5(1): 7–31.
Esser, H. (1993b) “How ‘Rational’ Is the Choice of ‘Rational Choice’? A Response to Randall Collins, Christopher
Prendergast, and Ilja Srubar.” Rationality and Society, 5(3): 408–414.
Hayek, F. A. von [1949 (1937)] “Economics and Knowledge.” Economica n.s. 4: 33–54. Revised and reprinted in
F. A. von Hayek: Individualism and Economic Order, pp. 33–56. London: Routledge & Kegan Paul.
Hayek, F. A. von (1994) Hayek on Hayek: An Autobiographical Dialogue. Kresge, S. and Wenar, L. (eds.)
Chicago: University of Chicago Press.
Helling, I. K. (1984) “A. Schütz and F. Kaufmann: Sociology Between Science and Interpretation.” Human Studies,
7(2): 141–161.
Helling, I. K. (1988) “Alfred Schütz, Felix Kaufmann, and the Economists of the Mises Circle: Personal and
Methodological Continuities”. In List, E. and Srubar, I. (eds.) Alfred Schütz: Neue Beiträge zur Rezeption
seines Werkes, pp. 43–68. Amsterdam: Rodopi.
Koppl, R. (1997) “Mises and Schütz on Ideal Types.” Cultural Dynamics, 9(1): 67–76.
Koppl, R. (1998) “Alfred Schütz and F. A. Hayek as Misesian Methodologists,” Paper presented at the Workshop on
“Spontaneous Orders: Austrian Economics, Philosophy, and Aesthetics,” 3 November 1998, Carlsberg Akademi,
Copenhagen.
Kurrild-Klitgaard, P. (1997a) “Alfred Schütz and the Austrian School: The Viennese Connection,” Paper presented
at the Austrian Scholars Conference, 3–4 April 1997, Ludwig von Mises Institute, Auburn University,
Auburn, AL. Forthcoming in Quarterly Journal of Austrian Economics.
Kurrild-Klitgaard, P. (1997b) Rational Choice, Collective Action and the Paradox of Rebellion. Copenhagen:
Institute of Political Science and Copenhagen Political Studies Press.
Langlois, R. N. (1988) “Are Economic Models Applicable to Politics?” Economia della Scelte Pubbliche/Journal
of Public Finance and Public Choice, 6(2): 83–93.
Machlup, F. [1978 (1936)] “Why Bother with Methodology?,” Economica n.s. 3: 39–45. Reprinted in Fritz M.,
Methodology of Economics and Other Social Sciences, pp. 63–70. New York: Academic Press.
Machlup, F. (1978) Methodology of Economics and Other Social Sciences. New York: Academic Press.
Machlup, F. (1981) “Ludwig von Mises: A Scholar Who Would Not Compromise.” In Andrews, Jr. J. L. (ed.)
Homage to Mises: The First Hundred Years: Commemorative Papers for the Ludwig von Mises Centenary, pp.
19–27. Hillsdale: Hillsdale College Press.
Mises, L. von [1981 (1933)] Grundprobleme der Nationalökonomie: Untersuchungen über Verfahren, Aufgaben
und Inhalt der Wirtschafts- und Gesellschaftslehre. Jena: Gustav Fischer. Translated as Epistemological Prob-
lems of Economics. New York: New York University Press and Institute for Humane Studies.
Mises, L. von (1940) Nationalökonomie: Theorie des Handelns und Wirtschaftens. Geneve: Editions Union.
Mises, L. von [1966 (1949)] Human Action: A Treatise on Economics, 3rd rev. ed. Chicago: Henry Regnery.
Mises, L. von [1984 (1957)] Theory and History: An Interpretation of Social and Economic Evolution. Auburn:
Ludwig von Mises Institute.
Mises, L. von [1978 (1962)] The Ultimate Foundation of Economic Science: An Essay on Method. Kansas City:
Sheed Andrews & McMeel.
O’Driscoll, G. P., and Rizzo, M. J. (1985) The Economics of Time and Ignorance. Oxford: Basil Blackwell.
142 KURRILD-KLITGAARD

O’Sullivan, P. J. (1987) Economic Methodology and Freedom to Choose. London: George Allen & Unwin.
Pietrykowski, B. A. (1996) “Alfred Schütz and the Economists.” History of Political Economy, 28(2): 219–
244.
Pietrykowski, B. A. (1999) “Postmodernizing Schütz: Phenomenology and Postmodern Economics,” Paper pre-
sented at the the History of Economics Society Session on Schütz 4 January 1999, Annual Meetings of the
American Economic Association, New York.
Prendergast, C. (1986) “Alfred Schütz and the Austrian School of Economics.” American Journal of Sociology,
92(1): 1–26.
Prendergast, C. (1993) “Rationality, Optimality, and Choice: Esser’s Reconstruction of Alfred Schütz’s Theory of
Action.” Rationality and Society, 5(1): 47–57.
Rothbard, M. N. [1977 (1956)] “Toward A Reconstruction of Utility and Welfare Economics.” In On Freedom and
Free Enterprise, pp. 224–262. Princeton: D. Van Nostrand. Reprinted as Occasional Paper No. 3. New York:
Center for Libertarian Studies.
Rothbard, M. N. (1957) “In Defense of ‘Extreme Apriorism’.” Southern Economic Journal, 23(1): 314–320.
Rothbard, M. N. [1970 (1962)] Man, Economy and State: A Treatise on Economic Principles, 2nd ed. Los Angeles:
Nash Publishing and Institute for Humane Studies.
Rothbard, M. N. [1979 (1973)] “Praxeology as the Method of the Social Sciences.” In Natanson, M. (ed.) Phe-
nomenology and the Social Sciences. Vol. 2, pp. 311–339. Evanston: Northwestern University Press. Reprinted in
Murray N. Rothbard, Individualism and the Philosophy of the Social Sciences, Cato Paper No. 4. San Francisco:
Cato Institute, pp. 29–62.
Rothbard, M. N. (1976) “Praxeology: The Methodology of Austrian Economics.” In Dolan, E. G. (ed.) The
Foundations of Modern Austrian Economics, pp. 19–39. Kansas City: Sheed Andrews.
Rothbard, M. N. (1989) “The Hermeneutical Invasion of Philosophy and Economics.” Review of Austrian Eco-
nomics, 3: 45–60.
Schütz, A. [1996 (1928–29)] “Pragmatismus und Soziologie, notes for four lectures,” Miseskreis December
1928–March 1929. Edited and translated as: “Toward a Viable Sociology.” In Wagner, H., Psathas, G.,
and Kersten, F. (eds.) Alfred Schütz, Collected Papers, Vol. IV, pp. 75–85. Dordrecht: Kluwer Academic
Publishers.
Schütz, A. [1996 (1930)] “Verstehen und Handeln, notes for a lecture,” Miseskreis 27 June 1930. Edited and
translated as: “Understanding and Acting in Political Economy and Other Social Sciences.” In Wagner, H.,
Psathas, G., and Kersten, F. (eds.) Alfred Schütz, Collected Papers, Vol. IV, pp. 84–87. Dordrecht: Kluwer
Academic Publishers.
Schütz, A. [1967 (1932)] Der Sinnhafte Aufbau der sozialen Welt. Wien: Julius Springer. Translated as The
Phenomenology of the Social World. Evanston: Northwestern University Press.
Schütz, A. [1996 (1934)] Staat-Gesellschaft-Recht-Wirtschaft, review of Mises ‘Grundprobleme der Na-
tionalökonomie’. Deutsche Literaturzeitung 1: 36–42. Translated as: Basic Problems of Political Economy. In
Wagner, H., Psathas, G., and Kersten, F. (eds.) Alfred Schütz, Collected Papers, Vol. IV, pp. 88–92. Dordrecht:
Kluwer Academic Publishers.
Schütz, A. [1996 (1936)] [Über Wissen und Wirtschaft], untitled MS, 15 October 1936. Edited and
translated as: “Political Economy: Human Conduct in Social Life. In Wagner, H., Psathas, G., and
Kersten F. (eds.) Alfred Schütz, Collected Papers, Vol. IV, pp. 93–105. Dordrecht: Kluwer Academic
Publishers.
Schütz, A. [1964 (1943)] “The Problem of Rationality in the Social World.” Economica, 10: 130–149. Reprinted
in Brodersen, A. (ed.) Alfred Schütz, Collected Papers, Vol. II: Studies in Social Theory, pp. 64–88. The Hague:
Martinus Nijhoff.
Schütz, A. [1962–1951] “Choosing among Projects of Action.” Philosophy and Phenomenological Research 12
(December): 161–184. Reprinted in Natanson, H. (ed.) Alfred Schütz, Collected Papers, Vol. I: The Problem of
Social Reality, pp. 67–96. The Hague: Martinus Nijhoff.
Schütz, A. [1962 (1953)] “Common-Sense and Scientific Interpretation of Human Action.” Philosophy and Phe-
nomenological Research 14 (September): 1–37. Reprinted in Natanson, H. (ed.) Alfred Schütz, Collected Papers,
Vol. I: The Problem of Social Rationality, pp. 3–47. The Hague: Martinus Nijhoff.
Schütz, A. (1972) “Choice and the Social Sciences.” In Embree, L. E. (ed.) Life-World and Consciousness: Essays
for Aron Gurwitsch, pp. 565–590. Evanston: Northwestern University Press.
ON RATIONALITY, IDEAL TYPES AND ECONOMICS 143

Selgin, G. A. [1990 (1988)] “Praxeology and Understanding: An Analysis of the Controversy in Austrian Eco-
nomics.” Review of Austrian Economics 2: 19–58. Reprinted as Praxeology and Understanding: An Analysis of
the Controversy in Austrian Economics. Auburn: Ludwig von Mises Institute.
Tullock, G. (1994) On the Trail of Homo Economicus: Essays by Gordon Tullock. Brady, G. L. and Tollison, R.
D. (eds.) Fairfax: George Mason University Press.
Wagner, H. R. (1983) Alfred Schütz: An Intellectual Biography. Chicago: University of Chicago Press.