MAX WEBER’S INTERPRETIVE SOCIOLOGY

AND RATIONAL CHOICE APPROACH

Zenonas Norkus

ABSTRACT

This article aims at substantiating two theses: (1) Weber’s programmatic
metatheoretical texts contain a description of the method of socio-
scientific explanation, which anticipate a specific version of the
Rational Choice Approach (RCA) in contemporary sociology, and (2) it
is possible to distinguish two versions of this description; the first, how-
ever, being closer to the RCA than the second. The late Weberian outline
of sociological theory of action is reconstructed out of his famous typol-
ogy of action.

KEY WORDS

•

interpretive sociology

•

Max Weber

•

rational choice

approach

•

sociological explanation

•

theory of action

1. Introduction

Theoretical discussion in the field of sociology over the past two dec-
ades has been largely influenced by the growing popularity of the meth-
odological line of thought known as the Rational Choice Approach
(RCA). Starting out as ‘economic imperialism’, this view is becoming
increasingly popular in sociology. Some advocates of the RCA tend to
justify the new approach in part by referring to specific programmatic
scientific claims by Max Weber, which are interpreted as an anticipation
of RCA. In one of the most recent exchanges between defenders and
critics of RCA, Edgar Kiser and Michael Hechter even designated their
version of RCA as analytical Weberianism (Kiser and Hechter 1998:
798).

The aim of this article is to systematically examine the points of con-

vergence and divergence between conceptions of sociological expla-
nation in Weber’s Verstehende Soziologie and in RCA. In section 1, I
outline the differences and affinities between the two approaches. These

Rationality and Society Copyright

©

2000 Sage Publications (London, Thousand Oaks,

CA and New Delhi), Vol. 12(3): 259–282. [1043–4631(200008)12(3); 259–282; 013478]

260

RATIONALITY AND SOCIETY 12(3)

observations are then specified and defined with consideration of the
internal distinctions of RCA (in section 2) and the modifications of
Weber’s concept of interpretive sociology (in section 3). These modifi-
cations are assessed as thoroughgoing enough to distinguish two ver-
sions of Weber’s interpretive sociology, the earlier one being closer to
RCA than the later, which founds the tradition of specifically ‘sociologi-
cal’ action theory.

2. Interpretive Sociology as a Meta-Theory of

Sociological Explanation

Weber famously defined sociology as ‘a science concerning itself with
the interpretive understanding of social action and thereby with a causal
explanation of its course and consequences’ (Weber 1922/1968: 4). If
we read this definition retrospectively from the perspective of the RCA,
two things stand out: (1) Sociology is not defined here by a specific sub-
ject area, which would distinguish it from other social sciences, above
all economic science, but by the description of an explanatory method.
(2) As I will show, this definition merely contains the description –
already familiar to sociological circles – of a schema of sociological
explanation as a macro–micro–macro transition (Figure 1), which can
be found in all introductory accounts of the leading principles of RCA.

1

It may also be of historical interest and relevance that this schema was
first published in The Achieving Society, by McClelland, where it is used
to illustrate the Weberian methods of argument in his work ‘The Protes-
tant Ethic and the Spirit of Capitalism’ (McClelland 1961: 47). This
schema was adopted by James S. Coleman

2

and has, with slight modifi-

cations, become a type of trademark of RCA in sociology.

In this scheme, the correlation of two (or more) macrosociological

variables (represented in Figure 1 by the dotted arrow) is explained
through the micromechanism of their relationship, by deducing that the
explanandum is a collective consequence of individual actions (right
arrow). These individual actions are explained as the consequences of
acts of rational choice (bottom arrow), which are completed in the
macrosocially determined situations (left arrow).

In the simplest case, a sociological explanation consists of three steps

that link the variables of the macro-level to those of the micro-level: (1)
from the logic of the situation, in which the relationships between the
state of the social system – as the field of activity of individual agents –
and their expectations and wants are analysed; (2) from the logic of

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

261

selection, in which the choice of actions is explained by the expectations
and aims of the agents; (3) from the logic of transformation, which,
through the deduction of the collective consequences of individual
actions, returns to the macro-level of the sociological facts to be
explained.

Those three steps of sociological explanation are found in the Weber-

ian definition of sociology presented earlier. The interpretive under-
standing of social action can be identified with the reconstruction of the
subjective definition of situation. This reconstruction results in a
macro–micro transition starting from the accepted rules, institutional-
ized world views, and uniformities that an agent observes in the pursuit
of ideal and material interests. It is this step that creates the premises for
a causal explanation of the course of social action within the context of
the logic of selection at the micro-level. Weber specifically claims that
the causal explanation of social action has two, not one, explananda.
Both the course and the consequences of social action have to be
explained. In Weber’s view, therefore, the explanation of the course of
social action is not an end in itself but a means to explain its effects, i.e.
the collective or social consequences. After all, interpretive sociology
is, by definition, concerned with the explanation of social action. These
collective effects could even be unintentional or contrary to the inten-
tions of the agent. ‘The final result of political action often, no, even
regularly stands in completely inadequate and often even paradoxical
relation to its original meaning. This is fundamental to all history, a
point not to be proved in detail here’ (Weber 1919/1948: 117; see also
Weber 1922/1968: 585–6).

The third principal idea in Weber’s concept of interpretive sociology,

the one that confirms its status as the anticipation of RCA, is his persist-
ent claim that instrumentally rational (zweckrational) action in the inter-
pretive understanding of social action always has methodical priority or

Macrosociological

Macrosociological

Explanans

Explanandum

Logic

of Logic

of

Situation

Transformation

Logic of Selection

Agent in the Situation

Action

Figure 1. ‘Deep’ Sociological Explanation

262

RATIONALITY AND SOCIETY 12(3)

primacy. An interpretive understanding of social action must always
begin by asking how an instrumentally rational agent, whose expecta-
tions are objectively rational (richtigkeitsrational) would act in a given
situation. These expectations are based on comprehensive and accurate
information on the situation. The answer to this question constitutes a
sort of zero hypothesis, which should be the starting point for the expla-
nation. Similar to present-day advocates of the RCA, Weber recom-
mends this initial hypothesis both for economic science as a way of
explaining the behaviour of market participants, and for an interpretive
understanding of other forms of action.

In working out a concrete causal explanation of individual events, the procedure of
the historian is essentially the same. Thus in attempting to explain the campaign of
1866, it is indispensable both in the case of Moltke and of Benedek to attempt to
construct imaginatively how each, given fully adequate knowledge both of his own
situation and of that of his opponent, would have acted. Then it is possible to com-
pare with this the actual course of action and to arrive at a causal explanation of the
observed deviations, which will be attributed to such factors as misinformation,
strategical errors, logical fallacies, personal temperament, or considerations out-
side the realm of strategy. Here, too, an ideal-typical construction of rational action
is actually employed even though it is not made explicit. (Weber 1922/1968: 21)

However, there are at least three distinctions between the Weberian con-
cept of interpretive sociology and the methodological views of RCA.
Jon Elster pointed out the first of these distinctions:

The two main theorists of rationality in the social sciences are, I believe, Max
Weber and John Neumann. One of Weber’s main legacies to the sociological tradi-
tion is the notion of instrumental rationality, or ends–means rationality. Rationality
is to choose the best means for a chosen end, given the initial conditions. The
assumption of given conditions I refer to as the assumption of a parametric envi-
ronment, and the corresponding notion of rationality as parametric rationality.
(Elster 1979: 68)

Elster is therefore of the opinion that Weber does not have a concept of a
strategic or game-theoretic rationality. However, this concept is essen-
tial, for example, in order to make explicit the ‘ideal-typical construc-
tion of rational action’ of Benedek and Moltke in the war of 1866. In the
above citation, Weber clearly shows no sense of the game-theoretical
problems that emerge if one attempts to demonstrate what ‘fully ade-
quate knowledge both of his own situation and of that of his opponent’
actually means in the case of Moltke and Benedek. Weber’s concept of
instrumentally rational social actions is that of parametric rationality.
The agents in his interpretive sociology are ‘oriented to the past,
present, or expected future behavior of others’ (Weber 1922/1968: 22)
but do not take into consideration how they determine by their expecta-

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

263

tions the expectations of their counterparts, whose object is their own
expectations.

Weber’s particular interpretation of instrumental rationality also

influenced his stance on the principle of methodological individualism,
which he passionately defended in his foundational theoretical writings.
Weber’s belief in methodological individualism is shared by the modern
advocates of RCA: If sociological explanation is defined as a macro–
micro–macro transition, it results in the application of the principle of
methodological individualism. Therefore, I want not to treat this as a
another one common feature shared by RCA and Weber’s interpretive
sociology but, on the contrary, to emphasize the different ways that this
principle is interpreted in each case. The principle of methodological
individualism can be expressed as follows: All collective facts can and
should be explained as consequences of individual action. This principle
is the standard for assessing socio-scientific explanations and indicates
the direction in which they could and should be improved.

However, two readings of methodological individualism exist: the

strict reading and the weak one.

3

The individualistic ‘deep explanation’

is weak if the logic of a situation is described to already include the facts
of the institutional rules. On the other hand, social theorists can be clas-
sified as advocates of strict methodological individualism, provided
they attempt to explain the existence of the social system on the basis of
a hypothetical presocial state. In this case, the initial conditions of the
explanation of individual action must not include the description of
sociological institutional rules. The issue is how, as a public good ena-
bling agents to cooperate, these rules are created and continue to sur-
vive. This problem can be traced back to Thomas Hobbes. It is the
concept of strategic rationality in modern game theory which allows this
strict, individualistic problem to be raised once more and elaborated in
the context of the RCA. Weber was a passionate advocate of methodo-
logical individualism, but did not hold such a radical individualistic
position. He was solely interested in the typology of uniformities of the
social order and their transformations. It is perhaps worth citing a perti-
nent remark by one of Weber’s interpreters at this point: ‘no action
begins at a social zero state, but is always interwoven with macro-
conditions. One can therefore change the sequence in the “Basic Con-
cepts in Sociology” [Economy and Society, Weber (1922/1968: Ch. 1)]
and begin with the paragraph on social structures and regularities’
(Schwinn 1993: 235).

The third point of divergence between the RCA and Weber’s interpre-

tive sociology relates to the following question: What should one do if

264

RATIONALITY AND SOCIETY 12(3)

the zero hypothesis of the objectively rational action, which always
enjoys the advantage of the greatest clarity and certainty (Evidenz) or
‘adequacy on the level of meaning’ (Weber 1922/1968: 9, 11) does not
prove causally adequate, i.e. if it is contradicted by the actual course of
action? What should replace the discarded zero hypothesis? At this
point, Weber introduces the concept of subjective instrumentally
rational action. These are actions that, although optimizing and maxi-
mizing by intention, contain misinformation, strategical errors, maybe
even logical fallacies in their actual determinants (Bestimmungsgründe)
(Weber 1922/1968: 312). Thus, the observed deviations of the action in
its actual course from its objectively rational course are explained by the
hypothesis that the action was in fact instrumentally rational, but only in
relation to the subjectively, not objectively, defined situation.

Weber’s concept of subjective instrumental rationality is both com-

plex and fascinating. It is fascinating because it can be linked to the cur-
rent discussion on substantive and procedural, perfect and bounded
rationality. Because of the limitations of space, I confine myself to the
minimum of detail needed for a systematic classification of Weber’s
views.

4

The key question regarding this set of problems is the following:

how strictly must one formulate the conditions for attributing (subjec-
tive) rationality? If we formulate these conditions very strictly, compar-
atively few examples of human behaviour can be included in the
concept. A good example of this type of formulation of attributive con-
ditions of instrumental rationality is the theory of subjective expected
utility. The theory was developed by Frank P. Ramsey and Leonard Sav-
age and explicates the concept of rational decision under risk. The
rational action is defined here as an action that maximizes the agent’s
expected value. According to the theory, an agent can act rationally, that
is in a maximizing or optimizing way, only if his intentional system –
i.e. all his wants and beliefs – fulfill specific formal, extremely strict
conditions. The wants of the agent should be consistent and manifested
in an order of preference that is complete, transitive, continuous, etc.
The theory is about the ‘subjective’ rationality, because it allows to
attribute the subjective rationality to an agent whose decision is prem-
ised on false beliefs. It is not the truth but the degree of belief that mat-
ters. This degree of belief is defined as subjective probability. As we all
know, even false statements can have a high degree of subjective prob-
ability and be an object of firm conviction. Even an action influenced by
such beliefs can be classified as instrumentally rational as long as these
beliefs are synchronically and diachronically consistent, which they are
if the subjective probabilities that quantitatively express the degree of

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

265

belief of an agent at a specific time fulfill the axioms of the probability
calculus. The beliefs are diachronically consistent if an agent learns
from new experiences updating the degrees of his beliefs in the manner
described by the Bayesian Rule.

5

Agreement still does not exist as to how often the intentional systems

of agents fulfill all of the consistency conditions. Experiments in cogni-
tive psychology have produced a long list of systematically recurring
discrepancies between the observable decision-making and the consist-
ency conditions, formulated by the theory of subjective expected utility.
These deviations are called anomalies of rational choice. Most social
scientists who subscribe to the RCA remain cool to these findings,
pointing instead to Sir Karl Raimund Popper, who argues for the auton-
omy of the social sciences from psychology: he recommends the
method of situational analysis to the social sciences.

6

Popper’s remarks

on the matter are interesting regarding Weber’s concept of subjective
instrumental rationality on two grounds: (1) His views on the method of
situational analysis were probably directly influenced by Weber,

7

and

(2) he formulates the conditions under which social action can be seen
as instrumentally rational as loosely as possible. In this respect, his con-
ception and the theory of subjective expected utility form to a certain
degree the two polarities between which Weber’s theory is situated.

Similar to Weber, Popper recommends that the socio-scientific expla-

nation begins with the zero hypothesis of the objectively rational action,
and where deviations occur to revert to subjective instrumental ration-
ality. Contrary to Weber, however, he does not allow this assumption to
be dismissed under certain conditions. All conceivable deviations in the
course of action from that which was presupposed as instrumentally
rational are seen not as evidence that the agents did not act rationally,
but as evidence that the explanation was based on erroneously specified
initial conditions. In other words, ‘deviations’ merely show that a
researcher has not adequately reconstructed the logic of the situation.
On the one hand, Popper supports his argument by formulating the
rationality principle as vaguely and noncommittally as possible: ‘agents
always act in a manner appropriate to the situation in which they find
themselves’ (Popper 1976/1994: 172). On the other hand, he under-
stands that the situation is not the objective situation but the situation as
subjectively defined. In his view, the beliefs and wants of the agent con-
stitute part of the situation. ‘We must remember, of course, that the situ-
ation, as I use this term, already contains all the relevant aims and
available relevant knowledge, especially of the various possible means
to achieve these aims’ (Popper 1976/1994: 169).

266

RATIONALITY AND SOCIETY 12(3)

Whether an action is rational or irrational is decided solely by its

appropriateness or inappropriateness in relation to the subjectively
defined situation. It does not matter how the subjective definition of the
situation was produced or how it was justified. Popper is only consistent
when he claims that even the actions of a madman are rational in this
sense: ‘and understanding his actions means seeing their adequacy
according to his view – his madly mistaken view – of the problem situ-
ation’ (Popper 1976/1994: 179).

At this point, the intuitions of Popper and Weber diverge, because

Weber repeatedly defines psychopathic behaviour as irrational and
maintains that it can only be explained psychologically, as in his exam-
ple of the behaviour of the late Friedrich Wilhelm IV (Weber 1903–6/
1975: 137). Weber’s definition of (subjective) instrumentally rational
action is, however, broad enough to subsume a certain action inspired
by a belief in magic or sorcery:

Rubbing will elicit sparks from pieces of wood, and in like fashion the mimetical
actions of a magician will evoke rain from the heavens. The sparks resulting from
twirling the wooden sticks are as much a ‘magical’ effect as the rain evoked by the
manipulations of the rainmaker. Thus, religious or magical behavior or thinking
must not be set apart from the range of everyday purposive conduct, particularly
since even the ends of the religious and magical actions are predominantly econ-
omic. Only we, judging from the standpoint of our modern views of nature, can
distinguish objectively in such behavior those attributions of causality which are
‘correct’ from those which are ‘fallacious’, and then designate the fallacious attri-
butions of causality as irrational, and the corresponding acts as ‘magic’. (Weber
1922/1968: 400)

The conditions under which an action can be considered subjectively
instrumentally rational are not as strict in Weber as in the theory of
subjective expected utility, but neither are they as loose as in the Pop-
perian concept of situative rationality. In recent years, the well-known
French sociologist Raymond Boudon has repeatedly attempted to con-
nect Weber’s concept of subjective instrumental rationality with his
own cognitive model of rationality (Boudon 1987, 1991, 1994: 1–55,
1995, 1996; see also Boudon 1986/1988). In his view, the subjective
rationality is only ostensively definable: ‘behaviour is rational when it
can be explained by a sentence beginning ‘X had good reasons for
doing Y, because …’., without risking objection, and without oneself
having the feeling of having said something incongruous’ (Boudon
1994: 255). The word because should be immediately followed by the
description of the beliefs of the agent that motivated his behaviour.
Boudon illustrates his proposed solution to the problem with exam-
ples:

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

267

Thus, one cannot say: ‘the mother had good reasons to slap the children’s face, for
she was angry’. Such an expression gives immediately a feeling of absurdity. One
would rather say: ‘the mother had no reason to slap the children’s face, but she was
angry’ … By contrast, one would say without meeting resistance: ‘the Puritans had
good reasons to invest their profits, since they were convinced that one should
refrain from superfluous consumption, and that prosperity in business is a sign of
election’. (Boudon 1991: 38)

Whether Boudon can solve the problem of subjective instrumental
rationality is more than doubtful. One cannot assume that common
sense is a guarantee of consensual evaluations in judges. How, for
example, can we judge the reasons of an agent if certain judges have the
‘feeling’ that his reasons are good, and others consider them bad? In sit-
uations where such conflicts of intuition arise, explicitly formulated cri-
teria are required to classify an action as instrumentally rational.
Nonetheless, Boudon’s proposal contains a grain of truth, which can be
predicatively formulated. Behaviour can be seen as subjectively instru-
mentally rational if the motives can be intersubjectively approved. Or,
in Weber’s terminology, the context of meaning (Sinnzusammenhang)
(Weber 1922/1968: 5) to which the subjective meaning of the action
belongs, should be objectifiable, i.e. communicable and intersubjec-
tively shared. So-called madness is always a dropout of social relation-
ships (soziale Beziehung), which, according to Weber, are constituted by
common meaning, into total isolation. Shared madness is no longer
madness but a new (micro) Lebenswelt, or subculture. In other words,
subjective instrumental rationality can be attributed only to actions that
are appropriate with respect to shared definitions of the situation, irre-
spective of how bizarre and grotesque it may appear to others who don’t
accept it. To formulate it in Weber’s own terms: The action is at least
subjectively instrumentally rational if its subjective meaning content
(subjektiv gemeinte Sinn) can also function as the meaning content of a
social relationship.

Weber’s interpretation of subjective instrumental rationality differs

from Popper’s interpretation of situative rationality, however, not just in
the claim that an action must be appropriate to the nonidiosyncratically,
socially defined situation. Weber attributes subjective instrumental
rationality only to those actions performed by an agent in the clear and
distinctly articulated consciousness of its subjective meaning, i.e. those
that are the outcome ‘of rational consideration of alternative means to
the end, of the relations of the end to the secondary consequences, and
finally of the relative importance of different possible means’ (Weber
1922/1968: 26). This condition of conscious deliberation in the choice
of action led Weber to the conviction that interpretive understanding

268

RATIONALITY AND SOCIETY 12(3)

through the hypothesis that the agents (subjectively) acted instrumen-
tally rationally, can rarely be causally adequate:

In the great majority of cases actual action goes on in a state of inarticulate half-
consciousness or actual unconsciousness of its subjective meaning. The actor is
more likely to ‘be aware’ of it in a vague sense than he is to ‘know’ what he is
doing or be explicitly conscious about it. In most cases his action is governed by
impulse or habit. (Weber 1922/1968: 21)

In such cases, Weber sees even subjective instrumental rationality as
having a purely heuristic function: It serves as a passage to other
hypotheses that interpret the action of an agent as value rational, tradi-
tional or affective. We are now at a point where the methodological
views of Weber differ from those of the majority of advocates of the
RCA. If the actual course of the action deviates from that predicted
under the assumption of (subjective) instrumental rationality, the major-
ity tend toward Popper rather than Weber. They do not want to abandon
the assumption of the agent’s instrumental rationality, but to rescue the
explanation based on this assumption by reviewing the descriptions of
the primary conditions. They thus specify shadow incentives and
shadow prices that were not originally considered. In their opinion, the
question whether the behaviour was motivated by conscious delibera-
tion or not is irrelevant for the attribution of instrumental rationality.

3. Weber as a Theorist of the RCA

Despite all of these divergences, Weber’s methodological instructions
on how to approach interpretive understanding and causal explanation
qualify as an anticipation of the RCA. To qualify, however, two assump-
tions implicit in my comparison of the two worlds of ideas must be
explained and revised. I have viewed both the RCA and Weber’s meth-
odology each as monoliths, disregarding the fact that various interpret-
ations of the RCA exist and, moreover, that Weber’s methodological
views have also changed, with the result that two versions of the con-
cept ‘interpretative sociology’ in his work can be distinguished.

Not all social scientists who subscribe to the RCA interpret it as the

so-called pure universalism of Donald P. Green and Ian Shapiro (1994:
192–4). This interpretation, advocated by, for example, Gary S. Becker,
amounts to the conviction that all phenomena of socio-scientific interest
can be explained in the framework of the RCA. However, attempts to
do so encounter certain problems of explanation or anomalies, which
cause many supporters of the RCA to modify their core assumptions or

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

269

confine the sphere of application. One example is the legendary para-
dox of election participation, which also illustrates the heuristic value
of the zero hypothesis of strict instrumental rationality already empha-
sized by Weber. The election paradox runs as follows: If we hold the
voters to be strictly instrumentally rational agents, we must expect an
election participation of 0%, because the larger the electorate, the
slighter are the chances for an individual to determine the outcome of
the election. Therefore, the costs of election participation are always
greater than the expected value of election participation, irrespective of
how strongly the voter feels about a particular candidate. Nevertheless,
the reality deviates very strongly from the hypothesized participation
rate of 0%, and this deviation must be causally explained. If we follow
the methodological instructions of Weber outlined earlier, we must first
try to explain the deviation by the hypothesis that, in this case, the
purely subjectively instrumentally rational action diverges from the
strictly instrumentally rational action, and then try to locate the system-
atic error that voters make in estimating the importance of their vote.
However, this hypothesis is not sufficient: Very few people who cast
ballots nurture the illusion that their vote can decide the outcome of the
election.

According to Weber, one should abandon the assumption of instru-

mental rationality here and explain election participation as a value-
rational, affective, or traditional action instead. Supporters of the strictly
universalistic interpretation of the RCA prefer, however, to salvage
instrumental rationality by searching for the ‘shadow incentives’ of
election participation that more than balance the costs of election partic-
ipation. However, some social scientists who consider the RCA a
resource of the socio-scientific market of ideas nevertheless do not take
the view that the anomalies of the RCA can always be surmounted by
searching for shadow costs and shadow incentives. They prefer to apply
the RCA only if specific boundary conditions have been fulfilled. Green
and Shapiro describe this as segmented universalism. In cases that fulfill
these conditions, the RCA can completely explain human behaviour.

This view is advanced by Erich Weede and Michael Taylor, for exam-

ple, who exclude so-called low-cost decisions from the sphere of appli-
cation of the RCA (Weede 1996: 7–8; Taylor 1989: 148–52). In these
situations, an agent hardly risks losing anything if she misjudges them,
or the effects of her actions depend less on her choice of action than on
circumstances outside her control. Consequently, the RCA can explain
the behaviour of politicians, military leaders and, of course, entrepre-
neurs better than that of voters.

270

RATIONALITY AND SOCIETY 12(3)

Another moderately imperialistic interpretation, as it were, of the

RCA is partial universalism (Ferejohn 1991: 282–6; Bates et al. 1998).
Advocates of this interpretation assume that although the RCA is appli-
cable to every behaviour, in many cases it can only partially explain it.
This applies for both low-cost and high-cost decisions. In matters of life
and death, strictly instrumentally rational agents are often embroiled in
strategic interactions, which have the structure of games with more than
one equilibrium. In this case, the game-theoretical model can at best
partially explain or predict how these interactions will develop (certain
outcomes are excluded). Nevertheless, it cannot explain why, of a
number of equilibria, this particular one was realized. Since game-
theoretical models cannot unequivocally predict the results of strategic
interactions, they merely present incomplete (partial) explanations.
Therefore, some people advocate viewing the RCA and the culturalistic
or interpretive social science as mutally complementary, because only
the so-called Third World of the shared linguistic and cultural meanings
places the participants of a strategic interaction, which, for example, is
structured like a game of coordination, in a position to identify a spe-
cific equilibrium as a focal point.

8

This internal differentiation of the RCA allows one to specify the

extent to which Weber’s interpretive sociology can be seen as an antici-
pation of the RCA – i.e. as an anticipation of the segmented universalis-
tic RCA. After all, Weber did not say that ideally typical constructions
arising from social action, which for their part presuppose a strict instru-
mental rationality, had a purely heuristic meaning as generators of
research problems. Even the zero hypotheses of the theory of marginal
utility describing objectively rational economic behaviour can be caus-
ally adequate. Weber mentions a constantly increasing convergence
(Annäherung) of reality with the theoretical propositions of neoclassical
economics, which he calls abstract economic theory, ‘affecting the fate
of continually growing sections of mankind’ (Weber 1908/1968: 305).
‘It is, for example, no coincidence, that a particularly striking measure
of convergence with the theoretical propositions of price fixing, devel-
oped by v. Böhm-Bawerk with reference to Menger, was demonstrated
by the Berlin Stock Exchange fixation under the system of the so-called
uniform price: it could be seen as a perfect example of this’ (Weber
1908/1968: 395–6).

According to Weber, the ability to act in an instrumentally rational

way is an anthropological constant. Another matter, however, is the real-
ization of this ability (as potentiality) at full capacity. Historical epochs
and cultures differ from one another according to how widespread the

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

271

sphere of instrumentally rational and especially strictly or objectively
rational action. When Weber speaks of the rationalization of social
action, he means the existence of institutional and cultural conditions
that force people to mostly act in an instrumentally rational way, at least
subjectively, and which simultaneously enable such action in the first
place. Not only are the zero hypotheses of the RCA first heuristically
valuable under these conditions, but also causally adequate. Weber
attributes the quality of formal rationality to institutional conditions
under which the RCA not only gains heuristic meaning, but can also
explain social action in causally adequate way. It is a matter of insti-
tutions, which render social action calculable and thus decrease the
transaction and information costs for the agents. The conditions for
expanding areas in which the zero hypotheses of instrumentally rational
action are causally adequate include the following:

1. The accumulation of objectively rational knowledge by modern

science, which decreases the information costs of instrumentally
rational action.

2. The invention of semiotic techniques that expand the natural

boundaries of the cognitive capacity of the human psychophysical
apparatus, just as the spade, the bulldozer and the hammer did for
physical capacities. Weber considered one of these inventions –
double-entry bookkeeping – so significant that he included it in
his definition of modern capitalism (Weber 1923/1961: 208, 211–
12). These are inventions that sink information costs and conse-
quently make maximizing action both increasingly possible and
necessary for boundedly rational human beings.

4. Max Weber as Sociological Action Theorist

I have specified my thesis that Weber’s concept of interpretive soci-
ology is the anticipation of the RCA. I also wish to take into account
that this concept was transformed during Weber’s lifetime. Two texts
exist in which this is clearly stated. The first is the article ‘Some Catego-
ries of Interpretive Sociology’, published in 1913 in the journal Logos
(Weber 1913/1968) but possibly written two years previously
(Schluchter 1998: 336–9); the second is the famous Chapter 1 in Econ-
omy and Society (Weber 1922/1986), which he wrote in 1919–20. The
first text can be seen as the conceptual head of the original text in Econ-
omy and Society; whereas the second was intended to perform the same

272

RATIONALITY AND SOCIETY 12(3)

function for a revision of the book, which Weber did not finish before he
died.

The most important difference is that Weber’s famous typology of

social action – in which the pure types of the instrumentally rational, the
value rational, the traditional and affectual action are distinguished – is
found only in the later text. Although Weber had already worked with
the differentiation of value rationality and instrumental rationality in the
earlier text, he related this distinction here to the institutional, i.e. to the
macro-level, where he distinguishes the material rationality of legal
norms from their formal rationality, but not to the micro- or action-level.
Weber’s response to the question of what should be done if the zero
hypothesis of instrumentally rational action proves causally inadequate
differs in the two texts. According to the earlier text, an action that is not
even subjectively instrumentally rational must be explained psychologi-
cally. Weber faces the following problem: On the one hand, he has so
strictly defined the conditions under which behaviour can be seen as
instrumentally rational that instrumentally rational action scarcely
occurs; on the other hand, he insists that interpretive sociology should
be independent of psychology. Both have strongly restricted the sphere
of interpretive sociology. ‘Since meaningful social action is the object
of Weberian sociology, he must, as he says, presumably describe 80% of
all social action, that occurs in the shape of semi-conscious or meaning-
ful amorphous habits (‘traditionally determined action’), as not actually
belonging to his theme’.

9

With respect to this difficulty, Weber had to insist in the earlier text

that rational action only differed gradually from irrational action:

For sociology 1. the more or less approximately realized type of the objective
rationality (Richtigkeitstypus), 2. the (subjectively) instrumentally rationally ori-
ented type, 3. the more or less consciously and clearly and more or less unambigu-
osly instrumentally rationally oriented, 4. no longer instrumentally rational but
meaningfully understandable, 5. the more or less understandable behavior, which
is comotivated by the ununderstandable elements in the more or less disconnected
context, and, finally, 6. completely ununderstandable psychic and physical facts
‘in’ and ‘concerning’ a man are connected by the gradual transitions. (Weber 1913/
1968: 435)

Weber can be understood here both in the sense that only case 6 can be
attributed to psychology, and that it can already play a role in case 3. At
any rate, these explanations cannot provide any unequivocal guidelines
for empirical research.

Weber’s action theoretical innovations in ‘Basic Concepts in Soci-

ology’ can be read as a second attempt to surmount these difficulties. In
this text, Weber attempts to supplement the interpretive sociology as

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

273

explanation method through sociological action theory. This involves an
action theory that should satisfy the following conditions:

1. It should be independent of psychology.

10

2. It should be linked to everyday experience or folk psychology.

11

3. It should be more realistic (or descriptively accurate) and have a

broader explanatory scope and analytical power (Heckathorn
1984) than the theory of instrumental rational action, which is
regarded as a special borderline case.

Weber’s action typology basically already contains all the variables of
this kind of comprehensive sociological action theory. These variables
are the opportunities, or in Weber’s terminology, the chances of action
(O), goals (G), the expectations (E) of the agent, the value commit-
ments (V ), the habits (T), and the affects (A). Therefore, the compre-
hensive sociological theory regards the action (H) as the function H =
h(O, E, G, V, T, A). Provided that the meanings of the variables in the
argument concerning the function independently vary from each other
and can also have the meaning 0

12

, the borderline cases are the pure

types of instrumentally rational action [H = h(O, E, G)], value rational
or conviction-ethical action [H = h(O, V )], traditional action [H =
h(O, T)] and of affectual action [H = h(O, A)]. Besides these pure types,
10 mixed types are also possible from the purely combinatory point of
view. In total, Weber’s action typology includes 15 action types:

1. H = h(O, E, G) instrumentally rational action
2. H = h(O, V) value rational action
3. H = h(O, T ) traditional action
4. H = h(O, A) affectual action
5. H = h(O, E, G, V, T, A)
6. H = h(O, E, G, V, T)
7. H = h(O, E, G, V, A)
8. H = h(O, E, G, V ) responsibility-ethical, or verantwortungse-

thisch, action

9. H = h(O, E, G, T, A)

10. H = h(O, E, G, T)
11. H = h(O, E, G, A)
12. H = h(O, V, T, A)
13. H = h(O, V, T)
14. H = h(O, V, A)
15. H = h(O, T, A)

274

RATIONALITY AND SOCIETY 12(3)

Weber himself explicitly pointed out and discussed some of these mixed
types: ‘choice between alternative and conflicting ends and results may
well be determined in a value rational manner. In that case, action is
instrumentally rational only in respect to the choice of means’ (Weber
1922/1968: 26). This obviously refers to type 8, which he treats under
the name of responsibility-ethical action. In the later work, whenever
Weber wrote about rational action, he subsumed at least three types of
action under this concept: 1, 2, and 8. ‘It is a case of sublimation when
affectually determined action occurs in the form of conscious release of
emotional tension. When this happens it is usually well on the road to
rationalization in one or the other or both of the above senses’ (Weber
1922/1968: 25). In this sentence, Weber refers to the mixed types 14, 11,
and 8. A reconstruction of Weber’s entire action typology shows that he
tended to also define as pure types behavioural episodes that in fact
belonged to the mixed types. This is particularly true of his use of the
expression ‘traditional action’. The expression refers not only to the
action of type 3, which has ‘its place in a systematic classification’
merely as ‘a limiting case’ (Weber 1922/1968: 25), but also to the action
of type 13. When Weber writes that ‘attachment to habitual forms can be
upheld with varying degrees of self-consciousness and in a variety of
senses’ (Weber 1922/1968: 25), he means cases 6 and 10, concerning
which he says, ‘in the great majority of cases actual action goes on in a
state of inarticulate half-consciousness or actual unconsciousness of its
subjective meaning’ (Weber 1922/1968: 21).

The action theory outlined by Weber in his late work can perhaps be

better understood by slightly modifying the filter model of instrumen-
tally rational action proposed by Jon Elster.

13

In this model, the choice is

presented as a filtering out of several alternatives. Elster’s model con-
tains two filters. The first consists of the restrictions, which separate the
‘real’ action options, or the opportunity set, of an agent from options
that are possible only logically. The second filter consists of the wants
and expectations of the agent, which in combination determine which
one real action option is selected. The principle of utility maximization
describes how this second filter functions. The whole is represented in
Figure 2.

The Weberian model was intended to include five filters altogether,

which according to the circumstances can be reactivated and can inde-
pendently or, in conjunction with other filters, determine which action is
chosen. By contrast, the theory of instrumentally rational action attempt
to do without additional filters by attributing value commitments,
affects, and habits to the first or second filter. Value commitments can

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

275

be interpreted as special ethical wants or metawants, which have as a
subject other wants of their subject; the traditional action can be rede-
scribed as risk-averse instrumentally rational action or as instrumentally
rational action restricted by the information costs, etc.

Weber’s later outline of the action theory anticipated later attempts at

a sociological action theory. Perhaps the best known among them is the
voluntaristic theory of action formulated by Talcott Parsons and pre-
sented in his early work, The Structure of the Social Action, which was
directly inspired by the action theory attempt in Weber’s later work. If
the Weberian concept of interpretive sociology does anticipate the RCA,
this applies more for the earlier version of the ‘Some Categories of
Interpretive Sociology’ than for the later version of ‘Basic Concepts in
Sociology’, which distanced itself from the positions of the RCA.

However, the action typology in Weber’s later work contains only the

outline, not the formulation of an action theory that would fulfill the
above three conditions for such a theory. To fulfill these conditions,
Weber’s action theory should contain at least one nomological relational
statement, which would specify the functional relationship between all
of its variables. Let us make a comparison: If we know that the attrac-
tion force F stands in functional dependency F = f(r, m

1

, m

2

) from the

distance r and the masses m

1

and m

2

, we still do not have the attraction

law. The shape of this functional dependency must be specified, too (in
this case, F = q(m

1

m

2

)/r

2

). As long as Weber’s action theory does not

include nomological statements, it can be seen only as a scheme of clas-
sification, not as an explanatory theory. The same applies for all later
efforts to construct a sociological action theory, from Parsons to Jürgen
Habermas. They merely contain classification schemata and are not
explanatory theories. Precisely because these efforts failed to produce
an explanatory sociological action theory, social scientists, who do not
want to renounce the autonomy of their discipline from psychology,

Logically

Re-

Opportunity

Beliefs

Chosen

possible

stric-

set

and

actions

tions

wants

action

Figure 2. Elster’s Filter Model

276

RATIONALITY AND SOCIETY 12(3)

hold the theory of rational choice – despite all its anomalies – to be the
best action theoretical offer available.

5. Conclusion

I have restricted myself to an analysis of Weber’s programmatic scien-
tific texts and not investigated the relationship of Weber’s meta-
theoretical concept of interpretive sociology to the logic-in-use of his
substantive work.

14

This topic has been reserved for separate treat-

ment.

Comparing Weber’s concept of the verstehende Soziologie and the

methodology of RCA, I applied the distinction made by Donald P.
Green and Ian Shapiro between the strictly universalist, segmented uni-
versalist, and partially universalist versions of the RCA. I also distin-
guished two versions of Weber’s concept of interpretive sociology, the
earlier one anticipating both the method of situational analysis pro-
pounded by Karl Popper and the segmented universalist version of the
RCA, and the later pointing in the direction of the voluntaristic theory of
social action promulgated by Talcott Parsons. The main reason for the
divergence between Weber’s early interpretive sociology and main-
stream RCA is Weber’s very restrictive concept of instrumental rational
action. In the later version of Verstehende Soziologie, Weber was about
to embed instrumentally rational action in the comprehensive theory of
action. As I argue, the only tangible result of this project was Weber’s
famous typology of action, which in its complete form includes 14 types
of action.

It is obvious that RCA, with heavy infusions of ideas direct from

economics, long ago surpassed Weber when it comes to analysing sys-
tematic properties of instrumental rationality. Has Weber’s quest for the
fuller action theory any remaining present relevance for RCA? The vir-
tues of empirical theory (both discursive and mathematically formu-
lated) are descriptive adequacy, universality of explanatory scope (in
our case, the ability to explain all episodes of human behaviour), and
analytic power, which means ‘the efficiency of a theory in converting an
informational input (e.g. quantitative specifications of initial conditions)
into an informational output (i.e. predictions or explanations). Thus ana-
lytic power can be defined as the ratio of informational output (IO) to
information input (II), i.e. AP = IO/II’ (Heckathorn 1984: 297). The
attraction of the theory of rational action consists in its considerable
analytical power. Given beliefs and preferences, it yields the determi-

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

277

nate predictions of choices (with some important exceptions, e.g. in
games with several undominated equilibria

15

); given choices and

beliefs, the preferences (or expected utility function) are derivable; for
given choices and preferences, the beliefs can be triangulated. The
broadening of the explanatory scope of this theory up to pure universal-
ism doesn’t succeed, however, without sacrificing descriptive accuracy
or analytical power. ‘The obvious interpretation of the statement that an
action is rational is that the agent can give reasons for it – that he can
explain why he does it, or did it. The clearest instances are those where
he has had to make a case for doing this, not that’ (Hicks 1986: 102).
This robust descriptive intuition goes lost in the reformulation of the
theory of instrumentally rational action by Gary S. Becker in terms of
microeconomic theory of production (Becker 1976: 87–149), which
nevertheless achieves greater analytical power. As for Popper’s permis-
sive view of the instrumental rationality as appropriateness with respect
to situation as defined subjectively, it results in tautological expla-
nations, which achieve universality of explanatory scope at the expense
of descriptive adequacy and analytical power.

It was the preference for descriptive accuracy that motivated Weber

and other sociological action theorists to seek the specifically sociologi-
cal action theory. This preference was satisfied, however, at the expense
of analytical power: Weber’s later action theory and its kin yield no test-
able predictions. Nevertheless, the efforts to produce such a theory were
not completely fruitless. They enriched the vocabulary of qualitative
sociological research, providing the valuable tools for ‘thick descrip-
tions’. Weber’s concepts of value rational, traditional, and affectual
action are of remaining value for such work. By now, Weber’s types of
action achieved in the sociological discourse the status of Nelson Good-
man’s well-entrenched predicates (Goodman 1955/1965: 98–106), gov-
erning sociologists’ ‘feeling of the real’. So they can be useful as the test
conditions for the descriptive adequacy of the elaboration of the theory
of instrumentally rational action striving after the greater descriptive
adequacy without refusing pure universalism and sacrificing analytical
power. Such elaboration can count as descriptively adequate if it doesn’t
eliminate or explain away distinctions between instrumentally rational,
value rational, and nonrational behaviour by redescribing or reclassify-
ing the problematic behaviour episodes, but rather, if it provides the
conceptual resources to explain them, preserving their descriptions in
types-of-action terms. But is such a task not self-defeating? How can the
same behaviour episodes be both described as nonrational and explained
as effects of instrumentally rational choices?

278

RATIONALITY AND SOCIETY 12(3)

Valuable suggestions for answering these questions can be found in

the work of German sociologist Hartmut Esser (Esser 1990, 1996, 1999;
see also Lindenberg 1989). He elaborates a pure universalist version of
the theory of instrumentally rational action, distinguishing the first-
order decisions from those of the second order

16

. The objects of the

first-order decisions (or selections) are the alternative courses of the
outer behaviour. The objects of the second-order selections are mental:
the modes of information processing and the frames, named by Esser
alternatively as models of situation and codes. The modes differ with
respect to the heuristics used. There are costly heuristics, which demand
plenty of time, attention, and other scarce mental resources to process
information, and cheap heuristics. They differ in the probability of find-
ing the course of outer behaviour that is optimal with respect to objec-
tive situation. Ceteris paribus, the more costly heuristics provide
objectively rational decisions with higher probability. Frames or codes
are dominant leitmotifs or dominant goals – criteria for evaluation of the
prospective courses of outer behaviour. The framing or coding of the
situation means that the actor considers only one value or supreme goal
relevant for evaluating those courses and suppresses alternative points
of view. In this way, the task of making a decision is simplified.

Esser proposes rational reconstruction of Weber’s typology of

action,

17

considering Weber’s types of action as frames or codes

selected in the inner, mental behaviour. While the actor frames or codes
a situation in the instrumentally rational way, she makes her first-order
choices in the self-interested and calculative way, using costly and
efficient heuristics to process information. Obviously, this picture corre-
sponds closely to Weber’s restrictive view of the instrumentally rational
action. While the actor uses frames other than Zweckrationalität or
cheap heuristics (e.g. habits), she selects the course of her outer behav-
iour in the manifestly nonrational way. Nevertheless, the behaviour
which is coded or framed value-rationally, traditionally, or affectually is
latently instrumentally rational, because the selection of these frames or
codes and heuristics is itself the effect of the instrumentally rational
inner behaviour, maximizing the actor’s subjective expected utility
under the restrictions imposed by her bounded rationality (the scarcity
of mental and cognitive resources). So the pure universalism of the
theory of instrumentally rational action is preserved at the level of the
second-order decisions, which produce the subjective definition of situ-
ation. At the same time, phenomena are saved, too: the distinctions
described by Weber’s types of action are accounted as real at the level of
the first-order decisions.

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

279

NOTES

I wrote this paper as a 1998–99 Fellow at Wissenschaftskolleg Berlin, where I read the
first version at Dienstagskolloquium. I thank Lynda O’Riordan for improving my English
and two anonymous reviewers for helpful comments.

1. This point is made by Esser (1993/1996: 3–6).
2. For the first time in: Coleman (1986: 1320–3). See also Coleman (1990: 1–23). Cole-

man considers Weber’s argumentation in ‘Protestant ethic thesis’ as specimen of
sociological explanation via macro–micro–macro transition. He criticizes Weber for
missing elaboration of the concluding link (micro–macro transformation). However,
he presents Weber’s case in a not completely correct way, considering as Weber’s
final explanandum ‘capitalist economic system’ (Coleman 1986: 1322). McClelland
describes Weber’s explanandum correctly as ‘spirit of modern capitalism’. For recent
interesting suggestion how to improve the micro-macro link in Weber’s argument see
Swedberg (1996: 319–23).

3. In this distinction, I follow Udehn (1987).
4. The useful systematic overview of the field with relation to the metatheoretical prob-

lems of sociological explanation is provided by Goldthorpe (1998).

5. See, for example, Eells (1982), Howson and Urbach (1989), Morton (1977/1997).
6. See Popper (1944–45/1957), Popper (1961/1980) Popper (1967/1985), Popper (1976/

1994).

7. Popper makes no reference to Weber in his description of the method of situational

analysis. But because of Popper’s good knowledge of Weber’s methodological work
there is no reason to doubt Weber’s influence on Popper on this point. Such influence
is also assumed by Goldthorpe (1998: 181, 189).

8. Regarding the concept of ‘focal point’, see Schelling (1960).
9. Baumgarten (1964: 603–04). This statement by Baumgarten (80%) about Weber’s

views cannot be justified by the direct textual evidence but, substantially, it is cor-
rect.

10. Weber presents his case for the independence of social science from psychology most

strongly in his review essay on Lujo Brentano’s attempt (Brentano 1908) to deduce
the law of diminishing marginal utility from the psychophysical Weber–Fechner law.
See Weber (1908/1968).

11. This condition was called ‘Postulate of Adequacy’ by Alfred Schütz: ‘Each term in a

scientific model of human action must be constructed in such way that a human act
performed within the life-world by an individual actor in the way indicated by the
typical construct would be understandable for the actor himself as well as for his
fellow-men in terms of common sense interpretation of everyday life’ (Schütz 1953/
1962: 44).

12. In this case the variable has no influence on behaviour.
13. See Elster (1986: 22–3); Elster (1989: 13–18).
14. With respect to this problem, see Udehn (1987: 151–7).
15. On the problem of indeterminacy of rational choice, see Elster (1993).
16. I am using the distinction between the first-order and second-order decisions to

present Esser’s theory in a shorter, but somewhat simplified way. He locates the
decisions which I refer to as ‘second-order’ in the context of the ‘logic of situation’
(see Figure 1). The decisions I refer to as the ‘first-order’ belong to ‘logic of selec-
tion’. Esser conceives his theory of second-order decisions as the theory which
explains the phenomena described in the sociological tradition as ‘definition of

280

RATIONALITY AND SOCIETY 12(3)

situation’. On the second-order decisions, see most recently Sunstein and Ullmann-
Margalit (1999).

17. See Esser (1990: 244–5), Esser (1996: 30–1) and especially Esser (1999: 224–30),

where he presents his view in the most detailed way.

REFERENCES

Bates, R.H., Rui J.P. De Fiqueiredo Jr and B.R. Weingast 1998. ‘The Politics of Interpreta-

tion: Rationality, Culture, and Transition.’ Politics and Society 26: 221–56.

Baumgarten, E. 1964. Max Weber: Werk und Person. Tübingen: J.C.B. Mohr (Paul

Siebeck).

Becker, G.S. 1976. The Economic Approach to Human Behavior. Chicago: The University

of Chicago Press.

Boudon, R. 1987. ‘The Individualistic Tradition in Sociology.’ In The Micro–Macro Link,

eds J.C. Alexander, B. Giesen, R. Münch and N.J. Smelser, pp. 45–70. Berkeley: Uni-
versity of California Press.

Boudon, R. 1986/1988. Ideologie. Geschichte und Kritik eines Begriffs. Hamburg: Row-

ohlt.

Boudon, R. 1991. ‘Weber’s Notion of Rationality and the Theory of Rationality in Contem-

porary Social Sciences.’ In Verstehen and Pragmatism. Essays in Interpretative Soci-
ology, ed. H.J. Helle, pp. 33–46. Frankfurt: Peter Lang.

Boudon, R. 1994. The Art of Self-Persuasion. The Social Explanation of False Beliefs.

Oxford: Polity Press.

Boudon, R. 1995. Le juste et le vrai. Etudes sur l’objectivité des valeurs et de la connais-

sance. Paris: Fayard.

Boudon, R. 1996. ‘The “Cognitivist Model”. A Generalized Rational-Choice Model.’

Rationality and Society 8(2): 123–50.

Brentano, L. 1908. ‘Die Entwicklung der Wertlehre.’ Sitzungsberichte der Königlichen

bayerischen Akademie der Wissenschaften. Philosophisch-philologische und histor-
ische Klasse. 3. Abhandlung.

Coleman, J.S. 1986. ‘Social Theory, Social Research, and a Theory of Action.’ American

Journal of Sociology 91: 1309–35.

Coleman, J.S. 1990. Foundations of Social Theory. Cambridge, MA: The Belknap Press of

Harvard University Press.

Eells, E. 1982. Rational Decision and Causality. Cambridge: Cambridge University Press.
Elster, J. 1979. ‘Anomalies of Rationality: Some Unresolved Problems in the Theory of

Rational Behaviour.’ In Sociological Economics, ed. L. Levy-Garboa, pp. 65–85. Lon-
don: Sage.

Elster, J. 1986. Introduction. In Rational Choice, ed. J. Elster, pp. 1–33. Oxford: Basil

Blackwell.

Elster, J. 1989. Nuts and Bolts for the Social Sciences. Cambridge: Cambridge University

Press.

Elster, J. 1993. ‘Some Unresolved Problems in the Theory of Rational Behavior.’ Acta

Sociologica 36: 179–90.

Esser, H. 1990. ‘ “Habits”, “Frames” und “Rational Choice”. Die Reichweite von Theorien

der rationalen Wahl (am Beispiel der Erklärung des Befragtenverhaltens).’ Zeitschrift
für Soziologie 19: 231–47.

Esser, H. 1993/1996. Soziologie. Allgemeine Grundlagen. Frankfurt am Main: Campus.

NORKUS: INTERPRETIVE SOCIOLOGY AND RATIONAL CHOICE

281

Esser, H. 1996. ‘Die Definition der Situation.’ Kölner Zeitschrift für Soziologie und

Sozialpsychologie 48(1): 1–34.

Esser, H. 1999. Soziologie. Spezielle Grundlagen. Bd. 1. Situationslogik und Handeln.

Frankfurt am Main: Campus.

Ferejohn, J. 1991. ‘Rationality and Interpretation: Parliamentary Elections in Early Stuart

England.’ In The Economic Approach to Politics. A Critical Reassessment of the Theory
of Rational Action, ed. K.R. Monroe, pp. 279–305. New York: Harper Collins Publish-
ers.

Goldthorpe, J.H. 1998. ’Rational Action Theory for Sociology.’ The British Journal of

Sociology 49: 167–92.

Goodman, N. (1955) 1965. Fact, Fiction and Forecast. Indianapolis: The Bobbs-Merill

Company.

Green, D.P. and I. Shapiro. 1994. Pathologies of Rational Choice Theory. A Critique of

Applications in Political Science. New Haven: Yale University Press.

Heckathorn, D. 1984. ‘Mathematical Theory Construction in Sociology: Analytical Power,

Scope, and Descriptive Accuracy as Trade-Offs.’ Journal of Mathematical Sociology
10: 295–323.

Hicks, J. 1986. ‘Rational Behavior – Observation or Assumption?’ In Subjectivism, Intel-

ligibility & Economic Understanding, ed. I. Kirzner I, pp. 102–10. New York: New York
University Press.

Howson, C. and P. Urbach. 1989. Scientific Reasoning: The Bayesian Approach. La Salle,

IL: Open Court.

Kiser, E. and M. Hechter. 1998. ‘The Debate on Historical Sociology: Rational Choice and

Its Critics.’ American Journal of Sociology 104: 785–816.

Lindenberg, S. 1989. ‘Choice and Culture: The Behavioral Basis of Cultural Impact on

Transactions.’ In Social Structure and Culture, ed. H. Haferkamp, pp. 175–200. Berlin:
Walter de Gruyter.

McClelland, D.C. 1961. The Achieving Society. Princeton: D. van Nostrand Co.
Morton, A. 1977/1997. A Guide Through the Theory of Knowledge. Oxford: Blackwell.
Popper, K.R. 1944–45/1957. The Poverty of Historicism. London: Routledge.
Popper, K.R. 1961/1980. ‘Die Logik der Sozialwissenschaften.’ In Positivismusstreit in der

deutschen Soziologie, eds Th. W. Adorno et al., pp. 103–24. Darmstadt, Neuwied:
Luchterhand.

Popper, K.R. 1967/1985. ‘The Rationality Principle.’ In Popper Selections, ed. D. Miller,

pp. 357–65. Princeton, NJ: Princeton University Press.

Popper, K.R. 1976/1994. ‘Models, Instruments and Truth. The Status of the Rationality

Principle in the Social Sciences.’ In The Myth of the Framework. In Defence of Science
and Rationality, ed. M.A. Notturno, pp. 154–84. London: Routledge.

Schelling, Th. 1960. The Strategy of Conflict. Cambridge, MA.: Harvard University Press.
Schluchter, W. 1998. ‘Max Webers Beitrag zum “Grundriss der Sozialökonomik”.

Editionsprobleme und Editionsstrategien.’ Kölner Zeitschrift für Soziologie und
Sozialpsychologie 50: 327–43.

Schütz, A. 1953/1962. ‘Common Sense and Scientific Interpretation of Human Action.’ In

Collected Papers, vol. 1. The Problem of Social Reality, ed. A. Schütz, pp. 3–47. The
Hague: Martinus Nijhoff.

Schwinn, T. 1993. ‘Max Webers Konzeption des Mikro–Makro-Problems.’ Kölner

Zeitschrift für Soziologie und Sozialpsychologie 45: 220–37.

Sunstein, C.R. and E. Ullmann-Margalit. 1999. ‘Second-Order Decisions.’ Ethics 110: 5–

31.

282

RATIONALITY AND SOCIETY 12(3)

Swedberg, R. 1996. ‘Analyzing the Economy: On the Contribution of James S. Coleman.’

In James S. Coleman, ed. J. Clark. London: Falmer Press.

Taylor, M. 1989. ‘Structure, Culture and Action in the Explanation of Social Change.’ Pol-

itics and Society 17: 115–62.

Udehn, L. 1987. Methodological Individualism. A Critical Appraisal, PhD thesis. Uppsala,

Uppsala University.

Weber, M. 1903–6/1975. Roscher and Knies: The Logical Problems of Historical Econ-

omics. Translated with an introduction by Guy Oakes. New York: The Free Press.

Weber, M. 1908/1968. ‘Die Grenznutzlehre und das “psychophysische Grundgesetz”.’ In

Gesammelte Gesammelte Aufsätze zur Wissenschaftslehre, ed. J.Winckelmann, pp.
384–99. Tübingen: J.C.B. Mohr (Paul Siebeck).

Weber, M. 1913/1968. ‘Über einige Kategorien der verstehenden Soziologie.’ In Gesam-

melte Aufsätze zur Wissenschaftslehre, ed. J. Winckelmann, pp. 427–74. Tübingen:
J.C.B. Mohr (Paul Siebeck).

Weber, M. 1919/1948. ’Politics as a Vocation.’ In From Max Weber. Essays in Sociology,

eds. H.H. Gerth and C. Wright Mills, pp. 77–128. London: Routledge.

Weber, M. 1922/1968. Economy and Society. An Outline of Interpretive Sociology, vols 1–

3, eds G. Roth and C. Wittich. New York: Bedminster Press.

Weber, M. 1923/1961. General Economic History. Transl. F.H. Knight. New York: Collier.
Weede, E. 1996. Economic Development, Social Order and World Politics. Colorado:

Lynne Rienner Publishers.

ZENONAS NORKUS is an Associate Professor of Philosophy and
Social Theory at the Vilnius University (Lithuania). He graduated from
the University of Saint Petersburg (formerly USSR), where he defended
his PhD thesis on Edmund Husserl’s philosophy of logic. He was Hum-
boldt Research Fellow at the Mannheim University and Fellow at Wis-
senschaftskolleg Berlin. His academic interests are the history of
philosophy, history and philosophy of social science, and metatheory of
historical studies. He has published articles (mainly in Lithuanian, Ger-
man, and Russian) on German and Austrian intellectual history (the
works of Edmund Husserl, Franz Brentano, Alexius Meinong, Johann
G. Droysen, and Max Weber; German historism; Austrian theories of
value), social theory (positivism and postpositivism in social theory, the
concept of power), and theory of action (heuristics of suspicion; the
‘thin’ and the ‘thick’ concepts of rationality). His books include Study
of History (Historik, in Lithuanian, 1996); Max Weber and Rational
Choice (forthcoming, in German).

ADDRESS: Department of Philosophy, Philosophy Faculty, Vilnius
University, Didlaukio 47, Vilnius LT-2057, Lithuania [email: zeno-
[e-mail]].