The Method of Concept-formation in Critical Theory (1941)
Adorno and Horkheimer's "Notes on Institute Activities" + "Scope and Method"
Editor’s note: from “Scope and Method” to “Notes on Institute Activities.”
Perhaps the single most important reflection on the method of critical theory written in the 1940s is a three-page text published in the ISR’s Studies in Philosophy and Social Science (SPSS) in 1941: “Notes on Institute Activities,” by Max Horkheimer.1 Below, I’ve partially revised and supplemented “Notes on Institute Activities” by reconstructing the earliest drafts, originally titled “Scope and Method” and written to serve as a methodological introduction,2 to the ISR’s second failed “Germany-project” research proposal in a row: Cultural Aspects of National Socialism [CANS] (1941). “Scope and Method” was free to be repurposed into “Notes on Institute Activities” because the very idea of providing a theoretical introduction to what seemed more like a broad social and historical study of German society between ~1900-1933 was met with suspicion and hostility. Even by the more sympathetic American academics who were advising the ISR on how to secure research funding in the US, like the historian Eugene N. Anderson, who recommended the elimination of “Scope and Method” entirely, calling it “too philosophical” and “too tough reading,” and subsequently admonished the social theorists of the ISR: “Do not be too theoretical.”3
In the near future, I plan to post my own transcription of the final CANS memorandum in full, as well as my transcription and translations tracking the changes made to a single thematic section—specifically, Horkheimer’s “Anti-Christianity”: a proposal for a study of the history of the cultural conditions of Nietzsche’s reception in Germany and the cultural impact his works had in the 20th century, ending in a critical account of fascist Nietzsche-reception. In that post, I will take a closer took at the process of composition and revision of CANS, as well as the reasons for the rejection of the final memorandum. To anticipate my argument: the censorship imposed on the ISR in the experience of trying and failing to get funding for any of their ambitious research proposals between 1938 and 1941 ends in the internalization of the demand to censor by the remaining members of the ISR. The ISR core’s ‘esoteric’ technique—namely, to maintain scientific autonomy under increasingly heteronomous conditions of scientific production—will accordingly take on an entirely different form in the 1940s than it had in the 1930s. The lesson the early critical theorists learn from the failure of CANS is that the only way to present critical theory to American social scientists is through adopting a conscious strategy of ‘translation’ between scientific traditions, one that amounts to a ‘de-conceptualization’ of the dialectical (and ultimately aporetic) concepts of the critical theorists, as for them the concept is a “contradictory figure,” one “in which everything is always that which it is only because it becomes that which it is not.”4 The four “prospects” below can thus be read in a dual valence: as positive rules for concept-formation or conceptual construction or as negative measures of the proto-conceptuality of the language of American social science. ‘De-conceptualization’ proceeds by way of taking the four “prospects” as prohibitions: the complex and dynamic unity of the concept is dissolved into isolated, and therefore abstract (even if ‘operationalizable’) definitions and functions. You could read the four prospects in reverse as instructions for arriving at ‘scientific’ concepts for American social science: (IV) delimit the scope of the concept in advance (a) by the assumption of strong disciplinary boundaries, (b) by reference to a single (kind of) empirical referent, and (c) by reserving it for a single purpose or function in social theory (or in the practice of sociologically-informed social administration); (III) return to the traditional model of induction that constructs concepts as generalizations from a set of individual cases collected by arbitrary heuristic; (II) divorce “fact-based” social theories from “value-oriented” research or “normative” social judgments; (I) exclude from the definition of a concept (a) the problem of relating the concept to its object, (b) the conflict between ordinary and specialist uses of the term, or between different special-scientific uses of it, (c) the historical genesis of the concept in scientific theory and practice.
However, “Notes on Institute Activities” (+ “Scope and Method”) also provides the means for the retranslation of the ‘concepts’ of American social science into critical-theoretical ones, commending the critical theorist to take the four “prospects” as positive rules for conceptual construction, which (as in their examples below: the category of the ‘Masses,’ the concepts of ‘Protestantism’ and ‘Reason’) consists in the recollection of abstract elements (one-sided definitions, partial functions, pure descriptions, normative principles, etc) into which concrete concepts are dissolved by their ‘translation’ into the prevailing cliches of American social-scientific language. Read forward, the “prospects” are a roadmap for elevating these scattered elements of possible concept-formation out of their effective reification in the idiom of American social science. The task of interpreting the ISR’s 1940s social research as the kind of theoretical project the critical theorists conceived it to be is largely a matter of reconstructing this double gesture of the dissolution (or decomposition) of concepts into abstract elements and the recollection (or recomposition) of concepts from out of the one-sided definitions and limited functions of American social science. For the early critical theorists, social research was a continuation of social theory by other means.
(In future posts, I hope to show how Adorno and Horkheimer “operationalized” their own critical-theoretical concepts into heuristics—or “frames of reference”—for empirical social research and even formed certain critical-theoretical concepts from heuristics first developed in the course of their empirical work—such as the “ticket-thinking” principle, which Adorno and Horkheimer developed through their collaboration with Paul Massing on the ISR’s 1944 “Labor Study.”)5
Despite their frustration with the false modesty in the proud obtuseness of American social scientists towards everything they accuse of being ‘metaphysics,’ none of the early critical theorists considered American social scientists (or anyone at all for that matter) fundamentally incapable of critical theorizing, and considered the task of the critical theorist to be one of articulating back to their peers in American social science the process of concept-formation that was already taking place, though without being recognized as such (“instinctively felt by every good reporter…”). In the end, the critical theorists considered themselves responsible for their own success or failure in showing their American peers that doing justice to the concerns of the pragmatists, positivists, and empiricists of any kind was already, and by necessity, a recollection of dissolution only possible through the “prospects” of the concept.
The dialectical two-step of concept-formation in critical theory requires, to use Marx’s formulation from the methodological ‘Introduction’ to the Grundrisse: in the first gesture, rising from the abstract to the concrete in thought, from abstract categories to comprehensive concepts, and in the second gesture re-embedding the concrete-in-thought in the history of social practice of which it is only ever a part. The three/four “prospects” of the concept elaborated below aim at the same: to get beyond the concept by means of the concept itself. Adorno and Horkheimer even present their approach to CANS by reference to the Marxian critique of political economy, arguing that what Marx did for “the correction or differentiation of [19th century economic] theory” they aim to do for 20th century social science (“[t]he social sciences increasingly require that their problems and studies be oriented to a unified theory of society…”). Their continued reliance on Marx suggests that something much more far-reaching than just another self-criticism of liberal democracies is at work when Adorno and Horkheimer interpret the crisis of the sciences torn between pluralism and holism as a refractory expression of the crisis of society torn between ‘democracy’ and ‘totalitarianism.’ The assumption of CANS is the global, though differential, production of the elements of fascism in liberal democracies. Put in reverse: the bad elements of liberalism flourish under fascism. This much is already clear from the first few sentences of “Scope and Method”: the development of Nazism in Germany deserves the attention (and resources!) of the American academy because it has a “direct bearing” on the present social situation in the United States.
CANS began with a sketch in mid-1940 that was already supposed to be a salvage operation for their previously shot down ‘Germany’-project, “The Collapse of Germany Democracy and the Rise of National Socialism (9/15/1940),”6 and entered a new phase later that October, when Anderson was officially brought on board as an advisor. This phase would last until CANS was definitively rejected by the Rockefeller Foundation in April, 1941. This second phase of the project was devoted to revisions. The revision process was so arduous that the Horkheimer Nachlass contains more than 25 unique drafts of certain thematic sections of the project. Thankfully, the archive only contains four of “Scope and Method”—one in English, listed under the letter “a)” and nearly identical in many parts with “Notes on Institute Activities,” and three in German, or “b)” and “c)” and “d),” the last of which is a one-sheet, two paragraph manuscript written in Adorno’s hand that would in subsequent drafts be typed and extensively revised by Horkheimer. While I have largely retained the text printed in “Notes on Institute Activities” for the non-indented paragraphs below, particularly in the sections with the Roman numerals I-IV, I have revised several sentences of the text in an attempt to integrate “Notes on Institute Activities” with what I considered clearer formulations from draft “a)” (these moments are rare considering the rushed quality of the English draft). Otherwise, the indented sections below (the sub-headings are marked with “[Scope and Method]”) are sourced primarily from the alternative paragraph and sentence constructions to the “prospects” from the latest of the German drafts, “b),” with a handful of excerpts from “c)” and “d)” that seemed unique enough too warrant interpolation. (Note: there are only 3 ‘prospects’ in drafts “d)”-”b),” as the ‘integrative’ prospect of critical-theoretical concept-formation has not yet been distinguished from the others.) The translations from drafts “d)”-”b)” are my own.
NOTES ON INSTITUTE ACTIVITIES.
The research project summarized below formulates certain problems which the Institute of Social Research intended to investigate about a year ago. General world conditions, however, brought to the fore other social problems more urgently connected with American interests and compelled us to postpone our original intention. The Institute plans, nevertheless, to return to this project in due time. As published here, the project contains not only research problems but theoretical conceptions which were in part arrived at through previous research and which would in some measure have to be probed through further investigations. It goes without saying that none of these theses will be treated as dogmas once the actual research is carried through. The publication of the project in the present issue may help further to clarify the conception of critical social research. The prevailing methodological viewpoints of this approach may briefly be characterized as follows.
SCOPE AND METHOD.
The studies we suggest deal with problems of the social development in Germany in the last few decades insofar as it has direct bearing on developments in the American scene. Each one selects a complex of facts that is not only symptomatic of the history of Germany democracy and the rise of German totalitarianism, but is due to the nature of the threat against democracy all over the world. Three viewpoints decided the selection: 1) The importance of an understanding of the breakdown of German democracy; 2) The material available in this country, and especially the material available to the Institute for Social Research; 3) The training of the various scholars responsible for the project and at the disposal of the institute. The entire project was planned in discussions with various specialists. The draft we submit herewith represents a set of abstracts from more extended and detailed projects which have been discussed with the American Committee for International Studies and other experts. We intend to consult regularly with other scholars with a view to obtaining their suggestions. We endeavor here to integrate the results of various branches of science and the experiences of American and European sociologists. The individual collaborators will remain in continual contact during the course of the study and will thus guarantee that the several special studies are adequately adjusted to one another. A unified terminology will be attempted. Those of our members who received their training in Europe have by now acquainted themselves with the methods of the American social sciences, and an interaction of American and European methods will thus become possible. Needless to say, these methods are to be used so far as they can help to clarify and deal with the problems with as much exactness as possible. The social sciences increasingly require that their problems and studies be oriented to a unified theory of society, just as the economic investigations of the 19th century were based on the theory of classical economics of a free market economy. By contributing to the correction or differentiation of this theory, the various economic investigations gain an internal connection. They have reached that high level of development at which they feel a definite lack of connectivity, and the collaboration of European and American scholars is expected to fructify empirical studies through the medium of an integrated social theory. Such a requirement is also associated with the increasing uneasiness about the relativism that has dominated not only the social sciences but the entire cultural pattern of the last decades. One is aware of the fact that democracy has no conclusive theory of society and history such as it could oppose to the ideological onslaught of totalitarianism. Those fears might have some legitimate basis, but we think that at present a conclusive theory cannot be elaborated out of which the problems and ideas of social research may be unequivocally derived. Moreover, there is a certain danger that the recent situation may give birth to a new dogmatism which may prove detrimental to the progress of thought. Dilthey’s criticism that the sociology and philosophy of history of his time indulge in the construction of universal systems without having any appropriate basis for such is still valid for us (cf. SPSS, New York 1940. pp. 431-32). Pluralism as well as monism have become equally questionable. Such a theory cannot be imposed upon the problems from without. The danger exists that the situation may give birth to a new dogmatism which may prove even more deleterious than the relativism it sets out to oppose. We shall attempt in the following four prospects to indicate some characteristics of the method we suggest applying. This method is intended to overcome the all-too-prevalent dissociation of empirical from theoretical studies and of “factual” from “normative” social theory. Some methodological elements through which those conducting this research hope to overcome empiricist pluralism without falling into new dogmatism are the following.
I. Concepts Are Historically Formed.
The categories we intend to use are not generalizations to be attained by a process of abstraction that yields the genus as the common name for various individuals and species, nor are they axiomatic definitions and postulates as in the trends of mathematics and physical science. The process of forming these categories must take account of the historical character of the subject matter to which they pertain, and in such a way that the categories are made to include the actual genesis of that subject matter. This unique character of the relation of the concept to its “material” does not allow of such abstract concepts as “social change,” “association,” “collective behavior,” “masses,” unless these are used as mere formalistic classifications of phenomena common to all forms of society. The proper meaning of “masses,” for example, cannot be derived through an essentially quantitative analysis or from certain isolated types of “collective behavior,” even though such analysis may be an integral part of any attempt at a theoretical interpretation of the term. Proper methodological usage must recognize that the masses are basically different at the different stages of the socio-historical process and that their function in society is essentially determined by that of other social strata as well as by the peculiar social and economic mechanisms that produce and perpetuate the masses. The category is thus led, by the very nature of its concrete content, to take in other, different sectors of the given social configuration and to follow out the genesis and import of its content within the social totality. The general concept is thus not dissolved into a multitude of empirical facts but is concretized in a theoretical analysis of a given social configuration and related to the whole of the historical process of which it is an indissoluble part. Such analysis is essentially critical in character.
Concept-formation is historical [Scope and Method].
The most important categories are obtained neither through mere abstraction, as one might define the zoological genus “lion” as a common attribute of a variety of individuals and kinds, nor through arbitrary definition, as one might in following the reigning axiomatic tendencies in mathematics and natural science. Both procedures have a role to play in social-scientific methodology: on the one hand, concepts must encompass the relevant social units, and on the other hand, concepts must be delimited according to the subject in order to ensure the fruitfulness of research. Furthermore, there is the methodological requirement that the genesis of the phenomena in question is incorporated into the formulation of the concept itself. For example: if Protestantism is mentioned in connection with German conditions, then the quality of its opposition to feudalism, Roman Catholicism, and France—which only arose because of the uneasy coexistence of Protestantism and Roman Catholicism as religious traditions in the geographical proximity between Germany and France—is implicit in the concept as well. Neither a mere definition nor a mere review of early Protestantism is sufficient for concept-formation, for the early Protestants are not necessarily and expressly aware of their opposition, despite the fact that their religious sentiment, and indeed their entire relationship to their environment, is essentially colored by this tension, something that is quite different for Americans. Likewise with the concept of the rational. Were one to adopt this concept from the literature of today or define it in Pareto’s sense, they would never understand the import this concept has for the whole of modern culture. This is only possible to the degree one remains conscious of the fact that its meaning is only constituted by the various historical phases that ‘reason’ has passed through in the Western world. Mention should also be made of the allegedly positive relationship between revelation and reason in Thomism, the separation of these two terms in the enlightenment, the principle of reason through which human beings have a kind of identity with one another and create a culture in common, the definition of reason as the epitome of the universal human claim to freedom and justice, etc. Historical concept-formation is dynamic because in the process, every single concept must point beyond its own scope and draw in entire chapters of history in order to determine its own meaning. Whereas the danger of historical concept-formation is possible vagueness or inconclusivity, the danger of unhistorical concept-formation is much greater, for it pays a heavy price for its precision: the impoverishment of all of its motifs.
II. Concepts Are Critically Formed.
The critical nature of societal concepts may best be elucidated through the problem of value judgments that animates current discussion among social scientists. The latter is much more than a methodological problem today. The totalitarian states are imposing the political values of imperialist power politics upon all scientific, cultural, and economic activities. This engenders all too much readiness in democratic countries to interpret freedom of science (which is held to include freedom from value judgments) as a drawback of the democratic forms of life. Hence derives a positivist and even skeptical attitude. The attempt has been made to overcome this by a return to old metaphysics, such as neo-Thomism. But this proposed return to the supposedly absolute values of past theological and metaphysical systems may facilitate the destruction of individual liberties to an even greater degree than would the conscious and honest skepticism of the positivists. Social theory may be able to circumvent a skeptical spurning of value judgments without succumbing to normative dogmatism. This may be accomplished by relating social institutions and activities to the values they themselves set forth as their standards and ideals. Thus, the activities of a political party may be investigated in the light of the avowed aims and ends of the party without accepting these as valid or evident. If subjected to such an analysis, the social agencies most representative of the present pattern of society will disclose a pervasive discrepancy between what they actually are and the values they accept. To take an example, the media of public communication, radio, press, and film, constantly profess their adherence to the individual's ultimate value and his inalienable freedom, but they operate in such a way that they tend to forswear such values by fettering the individual to prescribed attitudes, thoughts, and buying habits. The ambivalent relation between prevailing values and the social context forces the categories of social theory to become critical and thus to reflect the actual rift between the social reality and the values it posits.
Concept-formation is critical [Scope and Method].
The need in contemporary social science for theoretical viewpoints expresses not only the desire for theoretical unity but also for grounded theoretical development. Given the strength of the prevailing trend that all scientific, cultural, and economic endeavors have towards the pursuit of political power in totalitarian states, in democracies, the separation of knowledge and the positing of values, something hard won over the last few centuries, is often considered a weakness. But a simple return to the old metaphysics is foreclosed to us, just as is theoretical system-formation. Every attempt to present judgments about so-called ‘highest values’ as self-evident insights is bound to fail. Such attempts are even more dangerous than the positivist denial of value judgments entirely. It shows utter disregard for the truth, and leads to frivolous answers for all the most decisive questions. Value-absolutism is just as impossible as value-skepticism, the latter is just more honest. But if science, in its present form, is incapable of setting the goals for society, it can nevertheless go beyond mere skepticism in its critical aspect. All social processes take place within the framework of a culture in which certain values prevail. In reality, there are different value systems with different historical implications at any given time. Despite all the differences, they have agreed at least in certain formal features throughout history. Social science can conduct its investigations in a way that renders the tension between values and actuality, between prevailing values and the process of society, visible in every instance. For example, presenting a political movement from the perspective of its own goal-setting. Its practice can be continually examined from these value-orientations without the researcher having to take them for granted in the least. It is also possible to examine factors of public opinion (radio, cinema, press, etc.) in connection with the value of individuality, a value these factors claim to hold in high regard. There is no need to speak of any one value in every case. But when the researcher orients their investigation towards this aspect, the investigation automatically acquires a critical significance. For with the emergence of features that conflict with the value in question, the value itself becomes negatively—and, therefore, concretely—defined. The social sciences cannot, as some would like (cf. Lynd, Knowledge for what), establish a new religion or a better concept of community. But they can illustrate the lack of religiosity and the isolation of people in the present. By insisting that investigations have their basis in such value-orientation, investigations are given a direction and may be coordinated with one another to a greater extent than would otherwise be the case.
III. Societal Concepts Are “Inductively” Formed.
Social concepts derive their critical coloring from the fact that the rift between value and reality is typical of the totality of modern culture. This leads to the hypothesis that society is a “system” in the material sense that every single social field or relation contains and reflects, in various ways, the whole itself. Consequently, an intensive analysis of a single relation or institution that is particularly representative of the prevailing pattern of reality may be far better able to develop and grasp the nature of the pattern than would an extensive compilation and description of assorted facts. The “pervasive” character of our society, the fact that it makes its peculiar relations felt in every nook and cranny of the social whole, calls for a methodological conception that will take account of this fact. Categories have to be formed through a process of induction that is the reverse of the traditional inductive method which verified its hypotheses by collecting individual experiences until they attained the weight of universal laws. Induction in social theory, per contra, should seek the universal within the particular, not above or beyond it, and, instead of moving from one particular to another and then to the heights of abstraction, should delve deeper and deeper into the particular and discover the universal law therein.
Concept-formation is inductive in a singular sense [Scope and Method].
Induction is the process which arrives at universal propositions on the basis of individual experiences. According to the traditional doctrine of induction, hypotheses are made increasingly concrete on the basis of as many individual observations as possible until they acquire the weight of laws. Social science cannot, of course, do without such a process. However, there is another type of induction that has always been practiced but has rarely been brought to clear awareness: the attempt to recognize the uniqueness in kind of the whole to which determinate cultural phenomena belong. This induction proceeds, in a manner of speaking, in the reverse direction of traditional induction, in that it does not seek the universal beyond the individual but within the individual itself. This kind of induction does not move from one individual to the next and, finally, to the heights of abstraction, but delves deeper into the individual to discover in it the law of the whole. This is based on the conviction that the society in which we live is a system that is reflected in its individual features in different refractions. In history up to now, society and its culture could by no means be regarded as an organic whole, such that the totality was reflected in every part. This was the view that the Romantics had of the Middle Ages, and it is no longer compatible with modern social research. But neither has society—in any of its historical forms—been a purely mechanical structure, such that every individual part would have had only a completely external relation to the whole. By virtue of a certain power of organization [Prägnanz] in its structure, one way of thinking can acquire a preeminent physiognomic value for a culture. This fact, which is instinctively felt by every good reporter, is also of great importance for the selection and treatment of subjects in the social sciences. It is not only a matter of a heuristic for the selection and treatment of content, but also of one’s choice of theme. We strove to address such singular points, particularly when formulating the research themes in the individual sections. However, selection of theme is only the beginning of this “reverse” induction. What is more, such approach has to prove itself in every single step of the investigation by never losing sight of the intention of the whole.
IV. Social Concepts Are Integrative.
The peculiar kind of induction we have just outlined makes the formation of social concepts an empirical process and yet distinguishes this from the empirical method employed in the specialized sciences. The method is empirical, but by virtue of the historical and social totality immanent in the categories, it is not identical with any empirical concept yielded by any department of science. For example, the concept “youth,” denoting a particular entity in present-day society, is not a biological, psychological, or sociological concept, for it takes in the entire social and historical process that influences the mentality and orientation of youth and that constantly transforms these. Consequently, our concept will assume different functions pari passu with the changing composition, function, and attitudes of youth within the shifting social pattern. And owing to the fact that the concept is to be formed under the aspect of the historical totality to which it pertains, sociology should be able to develop this changing pattern from the very content of the concept instead of adding specific contents from without.
[Conclusion] [Scope and Method].
It is clear to us that it is only through the actual implementation of our research that we can show the significance of the methodological elements presented here. Nevertheless, we hope that the research proposed here can not only form a substantive contribution to the tasks of American democracy, but also to the methodological question of the social sciences.
In: SPSS Vol. 9, No. 1 (1941), 121-123. Horkheimer’s footnote to the title: “Under this heading we shall publish from time to time reports on programmatic and other activities undertaken by the Institute of Social Research.”
Quoted from Roderick Stackelberg, “‘CULTURAL ASPECTS OF NATIONAL SOCIALISM’: An Unfinished Project of the Frankfurt School.” Dialectical Anthropology 12, no. 2 (1987), 254. http://www.jstor.org/stable/29790236.
Stackelberg’s article has the distinction of being both groundbreaking scholarship for its time—since it is one of the few studies in any language that reconstructs the composition, revision, and rejection of CANS—and outdated by the time it was published, since Stackelberg assumes a Habermasian narrative of the “long farewell” that Stackelberg’s own research makes a significant contribution to undermining. If Stackelberg had continued to read the ISR’s social research proposals and reports of the 1940s beyond April 1941 with the care he had CANS, Anglophone critical theory reception might have been spared another quarter century of the same error as the quarter century prior.
Adorno and Horkheimer (1947): “The concept, usually defined as the unity of the features of what it subsumes, was rather, from the first, a product of dialectical thinking, in which each thing is what it is only by becoming what it is not. This was the primal form of the objectifying definition, in which concept and thing became separate, the same definition which was already far advanced in the Homeric epic and trips over its own excesses in modern positive science. But this dialectic remains powerless as long as it emerges from the cry of terror, which is the doubling, the mere tautology of terror itself.” In: Dialectic of Enlightenment: Philosophical Fragments. [1947] Edited by Gunzelin Schmid Noerr, Translated by Edmund Jephcott (SUP, 2002 [1987]), 11.
For more on the construction of the concept of ‘enlightenment’ as a contradictory figure in the first chapter of Dialectic of Enlightenment, see James Schmidt, “Language, Mythology, and Enlightenment: Historical Notes on Horkheimer and Adorno's Dialectic of Enlightenment.” Social Research 65(4) (1998), 807-838. Link: https://open.bu.edu/handle/2144/3898
See: Eva-Maria Ziege, Antisemitismus und Gesellschaftstheorie. Die Frankfurter Schule im amerikanischen Exil. (Suhrkamp, 2009), 224-225.