Etica
& Politica / Ethics & Politics, 2005, 2
http://www.units.it/etica/2005_2/AGEVALL.htm
Thinking
about configurations: Max Weber and modern social science
Ola
Agevall (1)
School of Social Sciences
Abstract
The article
addresses Max Weber’s relevance for modern social science. The first part is
a quantitative assessment of Weber’s fate in mainstream sociology; the second
part presents an argument that Weber’s work contains elements of
combinatorial thinking which makes it suitable for analysis in terms of the
methodological apparatus developed by Charles Ragin. On the one hand, it is
shown that Ragin’s notation and concepts are useful in bringing out some
important features in Weber’s methodology and substantive writings. On the
other hand, it is suggested that Weber’s use of configurations extend into
areas that have received little attention in mainstream sociology. |
1. Introduction
The ambition of this special issue of Etica & politica / Ethics
& Politics is to assess Max Weber’s relevance for contemporary social science
at the beginning of the third millennium. That is a warranted question, but it
is too vast in scope to be addressed in one piece. It needs to be sliced up
into more manageable chunks. And the question is how. For my part, I would like
to begin by taking my cue from Weber’s Science as a Vocation, and
indulge in that pedantic custom, peculiar to the empirical scientist, of
“always beginning with the external conditions”. (2) That is to say, we start out by treating the question as an empirical
matter, and ask to what extent elements of weberian thought are present in
contemporary social science. This exposition is followed by a discussion where
we take leave of the “is” and enter the “ought”.
From this point of view it seems appropriate to use JSTOR as a data
source. It allows us to perform Boolean searches in a full-text archive that
covers the major Anglophone sociological journals from 1895 onwards. (3) Let us see, first,
how often Max Weber is mentioned in sociological journals from 1895 to 2000,
and how the number of articles mentioning varies from one period to the next. (4) One complication
with this procedure is that the number of journals increases over time. But
this problem is circumvented by reporting ratios, i.e. the number of articles
quoting Max Weber over the total number of articles, instead of absolute
numbers. The corresponding figures for Durkheim and Simmel serve as points of
comparison:
The early peak should be interpreted with care, as should all figures from
the early years. The number of items is low up until the 1920s, varying between
23 and 48 articles annually, which means that percentage levels are sensitive
to very small variations even as we compound the figures into five-year
periods. We are on safer ground from 1925 onwards, for the average number of
articles per annum takes a leap to 180 and continues to rise. (6) But the early years
are interesting from another perspective. It is not until 1930 that “ideal
type” becomes associated with Max Weber. Until then, the terms have no relation
to the German economist, at least not in the sociological articles archived
with JSTOR. As used then, the concept “ideal type” is used in another sense,
and is grafted upon other sources. (7) The next chart portrays the number of articles that mention “ideal
type”, and the corresponding number of articles belonging to the sub-class of
articles quoting both “ideal type” and “Max Weber”.
The figures show that the take-off for the concept coincides with the
upsurge for Weber’s popularity. It came in the second half of the 1930s. From
this point onwards, the concept becomes increasingly entrenched in sociology.
The rise is broken in the 1980s, and quotation frequencies oscillate within the
interval 128 and 148 articles per five-year period (between 2.4 and 3.4 per
cent of the total number of sociological articles). (8)
2. Beyond
the empirical pattern
These few empirical facts about Max Weber’s reception in the
sociological community do not exhaust the discussion of Weber’s relevance for
contemporary social science. By this I do not mean that a larger set of weberian
concepts should be made objects empirical analysis in the same vein as we have
done with “ideal type” and “protestant ethic”. To be sure, a host of candidate
concepts queue up were we to go down that road. What are the fates of, for
example, “charisma”, “value-freedom”, “Zweckrational”, “Wertrational”,
“bureaucracy”, “status”, or “legitimacy”? Each of these concepts are amenable
to empirical inquiry, even if some of them pose technical difficulties.
“Zweckrational” has been translated differently by different authors, as has
“Wertfreiheit”. And apart from parallel renderings of the same
concept, we also have to grapple with the fact that Weber has no monopoly on
everyday words such as “bureaucracy”, “legitimacy”, and “status”, so that
measuring the “weberness” of their uses would require extensive reading of the
texts.
These are formidable problems, but they are not
insurmountable and they are not the reason we stop short of further empirical
analysis. The reason is that we want to pose a question of a different order.
Granted that at least part of Weber’s work has in fact trickled into mainstream
sociology, we are still entitled to ask whether his continued influence is
justified. Should sociologists devote so much attention to him? Should they use
his conceptual apparatus, reiterate and adapt his arguments, follow his
methodological advice? As long as we remain on the level of general references
to Max Weber, there is only one way of answering these questions. Yes,
sociologists should still read Weber, provided that doing so helps them
formulate, clarify, or solve the problems at hand. This attitude is truly
weberian. It is nevertheless far from clear that sociologists have always
heeded to this rule. I am not convinced, for example, that the analysis of
social action has benefitted from repeatedly returning to Max Weber’s typology
of action. In the end, the value of weberian notions must be assessed by those
who do work in the respective research fields.
Now consider a
complication. Some of Weber’s concepts and ideas have been torn from the vast
edifice and are disposed of freely among modern social scientists. As Peter
Baehr has shown,(9) the “iron cage” – surely
the emblem of Max Weber’s thought as represented in textbooks – is Talcott
Parsons’ invention. Weber’s original metaphor, “stahlhartes Gehäuse”, evokes a
quite different image. But as Baehr himself recognises, the vivid parsonian
metaphor has become canonical in the social sciences.(10) This is exactly the point. Even if
Baehr’s own translation returns to the original metaphor, it is far from likely
that the “iron cage” will go away. The example is admittedly somewhat
inconsequential, but it clarifies a general principle: as the specialist
literature revises old conceptions, the new Max Weber will still be in the
company of some old Webers. Which of these Doppelgänger should we opt
for? In view of what was said above, there is a sound argument for preferring
the fertile-and-incorrect interpretation over the barren-and-correct one. This
being said, however, I would also argue that the Weber of the modern specialist
literature has more interesting things to say than the blander duplicates who
inhabit mainstream sociology. It is perhaps significant that economic
sociology, a field where some leading scholars are also intimately familiar
with the specialist literature on Weber, is one area where Weber’s thought is
put to creative use.
Cross-fertilisation between these literatures would thus appear to be a
means of extracting interesting research problems and avoiding analytical
cul-de-sac. In fact, the entire argument so far has been a prelude to a
concrete suggestion in that direction. To introduce it, we return to the
concepts analysed above. Considering that the concept “ideal type” is used in
roughly three per cent of the articles in sociological journals, this
particular innovation must clearly be perceived to be a valuable instrument for
the researcher. The concept has also been subject to extensive treatment in the
Weber literature. If the frequent use in mainstream sociology signal a
self-contained and self-explanatory concept, the specialist literature is hard
pressed to determine its meaning, uses, and theoretical functions. Specialists
are concerned with how ideal types relate to adequate cause theory. They worry
whether the model for Weber’s ideal type concept is found in Georg Simmel’s,
Heinrich Rickert’s, or someone else’s writings. They follow the different
formulations of the concept, and confront it with its uses in Weber’s
substantive writings. In short, they attempt to specify the concept by aligning
it to proper contexts.
What shall we make of this picture? Mainstream sociologists use “ideal
types” so frequently that it must be perceived to be an analytical
prêt-à-porter, but when Weber specialists get their hands on it, the clarity of
the procedure seems to evaporate. The uncertainty is easy to explain. Max
Weber’s famous formulations basically describe the outer contours of a
procedure, coupled with a list of specifications of what the ideal type is not.
To give a flavour of how he delimits the concept, it is useful to quote from
the entry in Richard Swedberg’s Max Weber dictionary:
Ideal type
(Idealtypus) This is one of Weber’s most celebrated concepts, and it can in all
brevity be described as an attempt to capture what is essential about a social
phenomenon through an analytical exaggeration of some of its aspects. What is
at the heart of an ideal type, according to Weber’s famous formulation, is an
‘analytical accentuation (Steigerung) of certain elements of reality’.
[…] An ideal type is a conceptual tool with which to approach reality; and in
this sense it is a ‘conceptual construct’ (Gedankenbild; MSS: 93). When
confronted with an empirical situation, it is often helpful to introduce a
series of ideal types. In doing so, Weber argues, it is more important to
capture what is essential about a phenomenon than to merely reproduce the often
confusing empirical situation: ‘sharp differentiation in concrete fact is often
impossible, but this makes clarity in the analytical distinctions all the more
important’ (ES: 214). In this sense, the introduction of an ideal type serves
as a first step in analysis. The ideal type ‘serves as a harbor until one has
learned to navigate safely in the vast sea of empirical facts’ (MSS: 104). Once
some order has been brought into an analysis through the introduction of ideal
types, however, it may be important for the sociologist to decide why and how
empirical reality deviates from these. (11)
Thus, unless Weber’s formulations are inserted into a theoretical
context, we have very little to go on. We have known for a long time what the
ideal type is not. But it is a long step from there to a conception of the
operations involved that is sufficiently clear to allow the empirical
scientists to proceed with their business. From the practicing scientists’
point of view, the key issue is obviously how this analytical
accentuation should be done. They presumably need to know how to go about
selecting what is “essential” about their phenomenon, and how to distinguish a
good ideal type from a bad one. So there is good reason for mainstream
sociologists to consult the specialist literature on Weber. But this still
leaves us with the question why the “ideal type” is one of Weber’s most
celebrated concepts. I have a few suggestions, and they will point the way for
the argument in the remainder of the article.
The first reason is that, even in this muddled state, the concept names
a common practice. After all, Max Weber’s intention with coining it was not to
create a new practice but to codify and explicate an existing one. Consciously,
or (more often) unconsciously, social scientists deal in ideal typical
constructs: they engage in abstraction, accentuate parts of empirical reality,
and form heuristic concepts from the elements so extracted. Surely there is
some comfort in having a name for what you are doing.
Second, the concept may have been appealing precisely because its
construction principles are not written in stone. Its very indeterminacy made
it adaptable and possible to accommodate within diverse research traditions.
Nascent scientistic sociology could conceive of it in terms of property spaces,
case study researchers could incorporate it into their work, as could their
comparativist colleagues. Hence, even when the rift between different emerging
research strands widened, the ideal type could continue to have a habitat in
several traditions. It remains an open question, of course, just how weberian these
various uses are.
The third reason provides a link between the first two. There is one
aspect of Max Weber’s “ideal type” that makes it savoury for empirical
scientists who are not content to limit
their research repertoire to regression analysis: it is combinatorial by
design. It encourages the researcher to think in terms of configurations,
rather than in terms of a bundle of separate variables. Charles Ragin has
argued, and I believe correctly, that combinatorial reasoning in this sense is
both central to the understanding of social phenomena and a constitutive part
of the language of sociological theorising. (12)
The crux, in Ragin’s view, is that sociological methods have evolved in
ways which tend to obstruct the dialogue between ideas and evidence. A
methodological divide has developed, where one part of the scholarly community
favours variable-oriented studies and another part involves in case-oriented
research. Both strands of research has its merits. Variable-oriented studies
are powerful vehicles of generalisation, case-oriented studies are good at
capturing complexity and diversity. The former uses individual cases to work
out general explanations, the latter use the theoretical apparatus to interpret
the case(s) at hand. On the other hand, both has its limitations. The
homogenising assumptions of variable-oriented techniques, and their tendency to
dissolve cases into single variables, blocks their ability to test the kind of
configurational ideas which are part and parcel of sociological theory. And
they give little guidance for the interpretation of specific cases, considered
as wholes. Case-oriented studies, suitable for interpreting cases as wholes,
are ill-equipped to assess the generality and scope conditions of their
conclusions. Ragin has made an elaborate attempt to combine the virtues of both
strands of research, i.e. to preserve the focus on complexity, diversity, and
interpretation, while at the same time allowing the researcher to make
generalisations in a structured and orderly fashion. But Charles Ragin’s
qualitative comparative analysis and fuzzy-set social science are recent
innovations, and they are not embraced by everyone. The ideal type has long
served as a justification for the case-oriented researcher. Vagueness apart, it
had the promising feature of being a methodological concept which
encouraged configurational thinking and had an air of generality.
This leads up to the “ought” I promised at the outset. Case-oriented
researchers who sensed that there was something useful – for their purposes –
about the concept “ideal type” were not barking up the wrong tree. It is
counter-productive, however, to restrict vision to this one concept. We should
rather take a broader sweep over Max Weber’s methodology and see how his
deliberations are shaped and channelled by a configurational thinking that is
very much akin to that which has been developed by Charles Ragin. By doing so,
we get to see Max Weber as a precursor to recent developments in sociological methods.
Conversely, Charles Ragin’s elaboration of the logic of configurational
analysis provides clues to the interpretation of some features of Max Weber’s
methodology. There is an elective affinity between Max Weber’s Problemlage
and that of modern sociology, and it warrants renewed attention to Weber’s
methodology, from mainstream sociologists as well as from Weber specialists.
3. Concepts, combinations, configurations
How do we preserve a configurational view of cases without sacrificing
generality, and vice versa? How does the task of explaining unique events
square with that of developing new theories? These questions are at the heart
of Charles Ragin’s qualitative comparative analysis. They were no less
important for Max Weber. In the introduction to Economy and Society,
Günther Roth remarks that “…sociologists live, and suffer, from their dual
task: to develop generalizations and to explain particular cases.” (13) The formulation
captures a key concern in Max Weber’s work, but it could equally well serve as
a motto for one of Charles Ragin’s books.
Max Weber was above all concerned with interpreting
and explaining particular events in the social world. This follows directly
from the concept of value-relation. This concept, which Weber borrows from
Heinrich Rickert, names the governing point of view which sets the cultural
sciences apart from the natural sciences. To illustrate the logic peculiar to
the latter category of sciences, we do best to consider how they go about
transforming a certain sense datum into a scientific datum. Assume that we try
to test a hypothesis, or work out a theory, by way of experiment. It is clear,
then, that each observation, every single datum, can be substituted for any
other datum in the relevant class. What is important is not that it is this
drop of chemical solution that reacts in this way. Any other drop from the
sample could do. Indeed, if a drop of chemical solution does not behave the
same way as other drops of the same kind, our first guess would be that it has
been contaminated. To be sure, an alternative possibility is that we are on the
track of a scientific discovery, but the drop is nevertheless useless in this
particular experiment. In other words: the natural sciences homogenise data,
and only take an interest in those features which all instances have in common.
All this does of course make perfectly good sense.
The problem is that the cultural sciences operate differently. We are not
interested in the unification of
It has been pointed out by several Weber scholars
that Max Weber failed to provide definite criteria for when a phenomenon is
value-relevant. But perhaps this was not quite the point. As it stands, the
notion that the cultural sciences have value-relations as their governing point
of view ascribes a peculiar dynamics to these sciences. The natural sciences
are piecing together a giant jigsaw-puzzle. Their findings can be valuable
either because they provide new applications of established theoretical
knowledge, or because they contribute to the overall theoretical picture. In
both cases, the dynamics of scientific evolution is intimately linked to the
stock of established knowledge: a problem is taken up because it has a
meaningful relation to theory. This is not necessarily so in the cultural
sciences. We expect the cultural sciences to take an interest in the fall of
the
This, at any rate, was Max Weber’s position. He
placed heavy emphasis on the understanding and explanation of unique events.
But this is not to say that he neglected, or could avoid, the issue of
generalisation. I am not thinking here of the fact that Weber must establish
and make reference to “empirical regularities”. This goes without saying. He
needs such regularities both to create an object of inquiry – as he does, for
example, in the first section of The Protestant Ethic and the Spirit of
Capitalism – and to anchor his causal claims. His adaptation of adequate
cause theory to the cultural sciences, and his frequent allusions to
probabilistic formulations in Economy and Society, make it clear that
empirical regularities play an important role in his work. Yet there is another
reason why issues of generalisation are unavoidable for Weber. Generalisation
is in fact a constitutive part of his endeavour to explain unique events. To
see why this is so, it is instructive to use The Protestant Ethic and the
Spirit of Capitalism as an example, and read it through the lenses provided
by Charles Ragin. In Weber’s famous essay, configurational thinking enters the
argument at two junctures. The first is in the conceptualisation of the
explanandum, i.e. the “spirit” of capitalism: «If any object can be found for
which the use of this term can have any meaning, then it can only be a “historical
individual,” that is, a complex of configurations [Zusammenhänge] in
historical reality which we group together conceptually from the point of view
of their cultural significance to form a single whole.
A historical concept
like this, however, as it relates to a phenomenon which is significant in terms
of its individual characteristics, cannot be defined or demarcated
according to the schema: “genus proximum, differentia specifica”. It must be
composed from its individual elements, taken from historical reality. It will
not be possible to arrive at the ultimate definition of the concept at
the outset but only at the conclusion of the investigation». (14)
The passage is filled with termini technici –
e.g. “historical individual”, “cultural significance”, and the allusion to the
impossibility of demarcating such objects of inquiry along the lines of
aristotelian definition – derived from Heinrich Rickert’s work. They have no
doubt contributed to shaking off a few readers, in our day as well as in his
own time. Be that as it may: Weber wants to identify and explain what he has
chosen to term the “spirit” of capitalism. But he wants to extract it from the
empirical material rather than postulate it. The object of explanation is thus
to be pieced together as we go along, from those individual elements which
actually turn up in the data. By the end of the inquiry, he will thus have
amassed an array of individual elements (a, b, c, …, n) which, taken as an
ensemble, constitute the concept of this “spirit”.
Note, however, that when a series of elements are
added together in this fashion to form a configuration, the result – indeed the
aim – is a highly individualising concept. Think of each “element” as a
dichotomous variable, which can be either present or absent. With two requisite
elements, the concept in question represents one of four possible combinations.
Three constitutive elements yield eight combinations, four yield sixteen, five
yield thirty-two, etc. With a dichotomous variable, the number of possible
combinations are given by the formula 2n. Thus, if Weber identifies
10 elements as constitutive for his concept of the “spirit” of capitalism, it
will represent one out of 1024 logically possible combinations. We can compare
it to Alphonse Bertillon’s attempt to construct a procedure for identifying
criminals: describe eleven measures on the body of the suspect, and you will
get a description that is virtually as individualising as the fingerprint.
What sets “historical individuals” apart from the
individuality of the fingerprint is that the capacity to individualise is a
necessary but not sufficient criterion. The construct formed out of these
elements must also be endowed with cultural significance. In the essay, this is
achieved by an argument that the “spirit” of capitalism is tantamount to a
rational Lebensführung adequate for the forms of organisation peculiar
to modern capitalism. To relieve possible doubts that Max Weber thinks of
“historical individuals” in this combinatorial fashion, as configurations of an
array of elements, it suffices to quote the following passage from another of
his works: «Religions cannot … simply be inserted into a chain of types, each
constituting a new “stage” in relation to the others. Rather, they are all
historical individuals of a very complex kind, and the sum total of existing
religions only exhaust a fraction of those possible combinations, which could
imaginably be formed from the relevant factors. Thus, the following exposition
is in no way conceived as a systematic “typology” over religions». (15)
That is to say: each religion must be perceived as a
historical individual, composed of a series of individual elements. But as
noted above, it suffices that we assemble ten element, or
presence/absence-conditions, to get a sum total of over a thousand logically
possible combinations. Even with this very modest number of elements, there
simply are not enough religions to populate each combination. This is the
reason why Max Weber would never aspire to a systematic typology over the world
religions.
Yet Max Weber’s configurational thinking is not
confined to the construction of the explanandum. It is also at the heart of his
explanation, which is – to use Charles Ragin’s terminology – a specimen of
conjunctural and multiple causation. This feature makes it clear why the
problem of generalisation is already contained in Weber’s quest to understand
and explain unique events. As we will see, there is also reason to believe that
the same feature accounts for some of the difficulties which his
contemporaries, and modern commentators, have had in interpreting his argument.
To bring these aspects out, we must first introduce some of Charles Ragin’s
terminology and notation. “Multiple causation”, in Ragin’s terminology, is when
there are several causal paths to the same outcome. “Conjunctural causation”,
on the other hand, is when a particular condition A is not sufficient to
produce the outcome Y, but will produce it if it appears in conjunction with B.
What Ragin has done is to use Boolean algebra, truth tables, and truth
functions to create a way of representing and investigating multiple and conjunctural
causation.
In Ragin’s notation, the presence of a condition is
written in capital letters, e.g. “A”, while the absence of that condition is
represented as “a”. This way of writing corresponds to the ones and zeros of
Boolean algebra. The logical “AND” is written the same way as multiplication in
standard mathematics, while the logical “OR” is written as addition. Thus, the
equation “Y=A+B” tells us that the outcome Y will result if condition A is
present or if condition B is present. This is multiple causation: there are
more than one causal path to the same outcome. The equation “Y=A*B” – or “Y=AB”
for short – tells us that Y will happen if and only if conditions A and B are
simultaneously present. That is to say, the causal relation is conjunctural.
Charles Ragin’s presence/absence-conditions should
have a familiar ring. They correspond to the “elements” that make up
“historical individuals” in Max Weber’s terminology. The key difference is that
Ragin is concerned here with the explanans, Weber with the explanandum. But the
effect of adding more “conditions” to the explanatory model is the same as that
of adding more “elements” to the historical individual: for each added
condition, the number of logically possible combinations doubles. Ragin’s technical
achievement is that of implementing a set of algorithms that allows us to
summarise and minimise boolean equations in such a way that the researcher can
determine which combination of conditions are responsible for a particular
effect. In this context, however, I would rather like to emphasise how his
contribution adds to the language of theory and methodology. Using his
notation, we get the following interpretation of the much more familiar notions
of “necessary” and “sufficient” conditions.
C is necessary
C is sufficient Yes
No
Scholars in Weber’s day knew very well how to
conceive of causes in terms of conditions that are “necessary and sufficient”,
“sufficient”, or “necessary”. But what about the fourth cell? The condition C
is not necessary, for there is a causal path which does not contain this
condition, and it is not sufficient, for it will not produce the outcome unless
it is combined with the absence of condition A. Yet it is clear from Ragin’s
exposition that the condition C should be seen as causally important for the
outcome Y in this case as well. It may be that the path containing condition C
is particularly important, and that C is of particular importance for this
path. It is an explanation of exactly this type which Max Weber had in mind in The
Protestant Ethic and the “Spirit” of Capitalism. In Charles Ragin’s terms,
it is an explanation which posits a multiple and conjunctural causal bond between
the explanandum and a set of conditions.
This terminology was not available in Max Weber’s
day. No wonder, then, that his contemporary and latter-day critics have been
consternated. What was Weber actually saying? Did he think that puritanism was
a necessary condition for the rise of capitalism? In that case he could easily
be rebutted. One need only find instances of capitalism before, or otherwise
unrelated to, the puritans. Or did he think that the puritan sects were somehow
sufficient conditions for capitalism? That claim could also be shown false,
this time by pointing to cases where we have a lot of puritianism but no
capitalism. And the task of refuting Weber would, of course, be all the more
facile if he was arguing that puritanism was both a necessary and a sufficient
condition for capitalism. There was no category and no name for the kind of
argument Weber was pursuing, and it was consequently difficult for his critics
to assess the causal claims involved. Indeed, want of terminology created difficulties
for Max Weber himself. This is evident from Weber’s frustrated replies to his
critics: «Furthermore, I described as “foolish” the attempt to trace the
derivation not only of the capitalist system but specifically also of
the capitalist “spirit” (in my sense of word, be it noted – a matter to
which I shall return) back to the Reformation alone. Additionally, I
explicitly asserted that it was obvious that those religious and psychological
conditions could only bring about the development of capitalism in conjunction
with numerous other “conditions,” especially those of nature and geography». (16)
The first sentence makes it clear that we are
dealing with multiple causation, the second that we are dealing with conjunctural
causation. Throughout the reply, which is addressed to Felix Rachfahl, Weber
complains that he had pointed out these caveats already in his essay, and
accuses his opponent of deliberately misconstruing his argument. His reaction
is more than a bit high-strung. A more plausible explanation for the lack of
communication is that Rachfahl and other critics tried to make sense out of
Weber’s argument by fitting it into recognisable explanatory categories.
When Max Weber declared that he did not take puritanism
to be a necessary or sufficient condition for the rise of capitalism, and
denied that he saw the first as the sole cause of the latter, his critics
perceived him to be retreating. That insinuation made Weber even more furious,
which is why his replies are filled with quotations from and references to the
original essay. But what should the critics think? From their horizon, it
surely looked like Max Weber was stepping back from a position they had shown
to be untenable, to embrace instead a vague claim that puritanism can have some
sort of undefined link to the rise of capitalism, which was perhaps not as
important as he had first claimed. Unless you have embraced the idea that
multiple and conjunctural causation can also be a specimen of causation,
Weber’s replies will only appear as desperate scolastic manœuvres to avoid the
critic cum bird of prey.
To be sure, Max Weber’s polemic with his critics
contains many other elements – a long
list of invectives, squibbles over terminology, and disputes over facts.
Nevertheless, I do think that the configurational nature of explanandum and
explanans is one of the main reasons for the breakdown in communication between
Max Weber and his critics. The methodological structure of The Protestant
Ethic and the “Spirit” of Capitalism is in fact even more complicated than
I have been able to show in these few pages. Even if the historical individual
Max Weber construes from the “spirit” of capitalism is ultimately the
explanandum, his essay does not deal with it in its entirety:
«The task which I
set myself was to reveal one (particularly important) series of causes
which determined the formation of one (again, particularly important)
constituent component of the spirit of modern capitalism: that is, a variety of
this spirit which differed in specific important ways from that of either the
classical period or of the Middle Ages». (17)
The first part of the quote recapitulates Weber’s
insistance that he had all along recognised several paths to the outcome he is
interested in. In the second part of the quote, he states that he does not not
claim to have covered all the elements of the historical individual. What he
also claims, however, is that the causal path containing the condition he
investigates is particularly important, and that it is a particularly important
part of that configuration. Moreover, he holds that the portion of the
historical individual he is investigating is one of its particularly important
elements. He regards the attitudes he has investigated as a synecdoche for the
complex historical individual of the “spirit” of capitalism, and he does so
precisely because they are ‘adequate’ to those forms of organisation which are
peculiar to modern capitalism and set it apart from its predecessors.
A full assessment of Weber’s argument would have to
relate what we have just said to his use of adequate cause theory. That inquiry
will have to be postponed to some other occasion. Instead, we should return to
the question why Max Weber’s preoccupation with explaining unique events also
entails the task of developing generalisations. Let us, for the sake of
simplicity, say that Weber’s image of the rise of modern capitalism has the
form “Y = aC + AB”, where C stands for the specific Lebensführung
described in The Protestant Ethic and the “Spirit” of Capitalism. The
equation points the way to other inquiries. First, it presents Weber with the
task of investigating the second causal path. Second, it requires him to make
additional inquiries to ascertain that the causal conditions of the first path
does not pertain to a great amount of cases without producing the outcome.
Third, Weber is obliged to investigate just how individualised is the
explanandum. The equation creates a perfectly general reference class. Thus,
even if the explanandum is designed to be radically individualised, and even if
the explanation is derived from and tailored for one particular case, the
suggested explanation is general in scope and it has to be checked against
comparative data. The Gesammelte Aufsätze zur Religionssoziologie, in
which the essay on the protestant ethic is also inserted, is an attempt to do
just that. This description is, admittedly, artificially simplified, for what
Weber also does throughout these essays is to specify the scope conditions for
his thesis. But this is all the more in line with our general argument.
4. Conclusion
This article is an attempt addressed Max Weber’s
relevance for modern social science. In the empirical part, I have tried to
charter to what extent modern sociology makes use of weberian concepts. The
results show that Weber figures prominently in mainstream sociological
discourse. It is another matter, however, whether this continued influence is
justified. The second part of the article argues that there is one particular
area where closer inspection of Max Weber’s work would benefit both mainstream
sociology and the specialist literature on Weber, and where cross-fertilisation
between these literatures promise to yield interesting results, namely
configurations. Recent developments in sociological methodology has revitalised
a concern for combinatorial thinking that was central to Max Weber. And as a
result of these modern efforts we are in a better position to grasp what Weber
was trying to do, why he sometimes failed to be understood, and what other
explanatory tasks his configurational thinking imposed on him. I have used
Weber’s essay on the protestant ethic and the “spirit” of capitalism as an
illustration, but the list of interesting applications can easily be extended.
The often quoted concept “ideal type” – a construct that is combinatorial by
design – is one obvious example where some Ragin-therapy could contribute to
clarification and novel insights.
Nor do the benefits flow one way only, from
mainstream sociology to the specialist literature. Max Weber’s works gain new
relevance once we recognise him as a precursor to these modern methodological
developments. His works are waiting to be plundered and used as vast granaries
full of neatly packed empirical grain. Moreover, it would be worthwhile to see
if his theoretical vocabulary, which was shaped by a concern for preserving a
configurational view of cases, contains useful elements which the modern literature
has overlooked. Insofar as the construction of the explanandum is concerned,
Weber’s conception of “historical individuals” still stand out as an original
contribution to complexity-oriented social science. Now, constructing
historical individuals is one way of arriving at an ideal type. Other
principles are involved in the construction of genetic ideal types, and they
too should be of interest to those who are engaged in qualitative comparative
analysis.
In conclusion, then, Max Weber should not be quoted
and used because he is a classic. He should be used wherever and whenever his
ideas and concepts aid the formulation and solution of concrete research
problems, otherwise not. On the other hand, important aspects of his work are
still awaiting treatment, and precisely these aspects appear to be important
from the point of view of modern social science.
Baehr, P. (2001) "The 'Iron Cage' and the 'Shell as Hard as Steel':
Parsons, Weber, and the stahlhartes Gehäuse Metaphor in The Protestant Ethic and the Spirit
of Capitalism", History and Theory 40: 153-69.
Baehr, P. (2002) "Note on the translation", in: Max Weber, The
Protestant Ethic and the "Spirit" of Capitalism and Other Writings.
Swedberg, R, with the assistance of O. Agevall (2005) The
Max Weber Dictionary: Key Words and Central Concepts, Stanford, CA:
Stanford University Press.
Weber, M. (1970 [1917]) ‘Politics as a
Vocation’, in: From Max Weber, edited by H.H. Gerth
and C.W. Mills, London: Routledge.
Weber, M. (1989 [1920]) Die Wirtschaftsethik
der Weltreligionen, MWG I/19, Tübingen: J.C.B. Mohr.
Weber, M. (2002 [1905]) The Protestant Ethic
and the "Spirit" of Capitalism and Other Writings, edited and
translated by Peter Baehr and Gordon Wells, New York:
Penguin.
Notes
(1) Ola Agevall is
post-doc fellow at Växjö University. He is the author of A Science of Unique
Events: Max Weber’s Methodology of the Cultural Sciences (1999), and
collaborator in Richard Swedberg’s The Max Weber Dictionary: Key Words and
Central Concepts (2005). The author thanks Katarina Friberg for helpful
comments on early versions of this article. The study is funded by a grant from
the Swedish Research Council.
(2) Max Weber (1970
[1917]) ‘Politics as a Vocation’, in: From Max Weber, p.129.
(3) The sociology section
of JSTOR currently contains the following journals: The Academy of Management
Journal, The Journal of the Academy of Management, The Academy of Management
Review, Administrative Science Quarterly, The American Journal of Sociology,
American Sociological Review, Annual Review of Sociology, The British Journal
of Sociology, Comparative Studies in Society and History, Contemporary
Sociology, European Sociological Review, Family Relations, The Family
Coordinator, The Family Life Coordinator, The Coordinator, Gender and Society,
Industrial and Labor Relations Review, Journal of Black Studies, Journal of
Health and Social Behavior, Journal of Health and Human Behavior, The Journal
of Human Resources, Journal of Marriage and the Family, Marriage and Family
Living, Living, The Journal of Modern African Studies, Journal of Palestine
Studies, Middle East Report, MERIP Middle East Report, MERIP Reports, Political
Behavior, The Public Opinion Quarterly, Social Forces, Journal of Social
Forces, Social Psychology Quarterly, Social Psychology, Sociometry,
Sociological Forum, Sociological Methodology, Sociological Theory, Sociology of
Education, Journal of Educational Sociology, Theory and Society
(4) The search has been
limited to articles; no review essays or other items have been included. Since
Weber is a too common name to allow precise measurement, I have used the search
string [“Max Weber” OR “Weber, Max”]. Considering that this is a more
restrictive function, the figures for Max Weber may well be underestimated, as
compared to the other authors, where only one term was employed in the search.
(5) Included in the
search are such variations as “ideal typic”, “ideal-type”, and “ideal typical”.
The formulation “pure type” has not been included in the search.
(6) The number of
articles in sociological journals archived by JSTOR roughly fits a logistic
curve, doubling every ten years. There is one slight deviation, for the total
number of articles in JSTOR declines somewhat from the 1980s.
(7) In this respect, too,
the figures confirm what we would expect on other grounds. We know, for
example, that Auguste Comte and Émile Durkheim used the term and that “types”
of various orders were in vogue in social scientific parlance in the late 18th
and early 19th centuries. For a lengthier discussion, see Ola
Agevall (1999) A Science of Unique Events: Max Weber’s Methodology of the
Cultural Sciences, pp. 166ff.
(8) These facts should also serve as a corrective to
Uta Gerhardt’s assertions in her book on the ideal type: “Die soziologische
Theorie, die heute Max Weber einen ihrer zwei unbestrittenen Klassiker nennt,
verwendet keine Idealtypen. Nicht einmal interpretierende ‚Weberian Sociology’,
die Webers Denken nachkonstruiert, plädiert für den Primat idealtypischer
Verstehens. Zwar umfasst die Sozialforschung heute idealtypische Analysen. Doch
der Idealtypus zur dateninterpretation ist eher ein Außenseiter im
Methodenwettbewerb empirischer Gesellschaftsanalysen.” Uta Gerhardt (2001) Idealtypus:
Zur methodischen Begründung der modernen Soziologie, p.11. It may well be the case that modern sociologists are guilty of clumsy
and erroneous interpretations of Weber’s concept. Yet in purely empirical
terms, the concept does enjoy a fair and stable popularity in the sociological
community.
(9) Cf. Peter Baehr (2001) “The ‘Iron Cage’ and the
‘Shell as Hard as Steel’: Parsons, Weber, and the stahlhartes Gehäuse Metaphor
in The Protestant Ethic and the Spirit of Capitalism”, History and
theory 40: 153-69.
(10) Peter Baehr (2002)
“Note on the translation”, in: Max Weber (2002) The Protestant Ethic and the
“Spirit” of Capitalism and Other Writings, p.lxx.
(11) Richard Swedberg,
with the assistance of Ola Agevall (2005) The Max Weber Dictionary: Key
Words and Central Concepts, pp.119-20.
(12) Cf. Charles Ragin
(1989) The Comparative Method: Moving Beyond Qualitative and Quantitative
Strategies; Charles Ragin (2000) Fuzzy-Set Social Science.
(13) Günther Roth (1978)
“Introduction”, in: Max Weber, Economy and Society, p.XXXVII.
(14) Max Weber (2002
[1905]) The Protestant Ethic and the “Spirit” of Capitalism and Other
Writings, pp.8-9.
(15) Max Weber (1989 [1920]) Die
Wirtschaftsethik der Weltreligionen, MWG I/19, p.116.
(16) Max Weber (2002
[1905]) The Protestant Ethic and the” Spirit” of Capitalism and Other
Writings, p.285.
(17) Max Weber (2002
[1905]) The Protestant Ethic and the” Spirit” of Capitalism and Other
Writings, p.284.