Staiti, Andrea - Warren, Nicolas de - New Approaches To Neo-Kantianism-Cambridge University Press (2015) PDF

NEW APPROACHES TO NEO-KANTIANISM
After the demise of German Idealism, Neo-Kantianism flourished as

the defining philosophical movement of Continental Europe from
the 1860s until the Weimar Republic. This collection of new essays by
distinguished scholars offers a fresh examination of the many and
enduring contributions that Neo-Kantianism has made to a diverse
range of philosophical subjects. The essays discuss classical figures and
themes, including the Marburg and Southwestern Schools, Cohen,
Cassirer, Rickert, and Natorp’s psychology. In addition, they examine
lesser-known topics, including the Neo-Kantian influence on theory
of law, Husserlian phenomenology, Simmel’s study of Rembrandt,
Cassirer’s philosophy of science, Cohen’s philosophy of religion in
relation to Rawls and Habermas, and Rickert’s theory of number.
This rich exploration of a major philosophical movement will interest
scholars and upper-level students of Kant, twentieth-century
philosophy, continental philosophy, sociology, and psychology.
nicolas de warren is Research Professor of Philosophy at the

Husserl Archives and the Center for Phenomenology and
Continental Philosophy at the Institute of Philosophy, KU Leuven.
He is the author of Husserl and the Promise of Time: Subjectivity in
Transcendental Phenomenology (Cambridge, 2009) and numerous
articles on phenomenology, philosophy of art, hermeneutics, and
political philosophy.
andrea staiti is Associate Professor of Philosophy at Boston

College. He is the author of Husserl’s Transcendental Phenomenology:
Nature, Spirit, and Life (Cambridge, 2014), and has published many
articles in journals, including Husserl Studies, Continental Philosophy
Review, Philosophy and Social Criticism, and Research in Phenomenology.
NEW APPROACHES TO
NEO-KANTIANISM
edited by
NICOLAS DE W ARREN
ANDREA STAITI
University Printing House, Cambridge cb 2 8bs , United Kingdom
Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781107032576
© Cambridge University Press 2015
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2015
A catalogue record for this publication is available from the British Library
Library of Congress Cataloging in Publication Data
New approaches to Neo-Kantianism / edited by Nicolas de Warren, Andrea Staiti.
pages cm
Includes bibliographical references and index.
i s b n 978-1-107-03257-6
1. Neo-Kantianism. I. Warren, Nicolas de, 1969– editor.
b3192.n49 2015
142′3–dc23
2014048728
i s b n 978-1-107-03257-6 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
In memory of Jonathan Trejo-Mathys (1979–2014)
esteemed colleague and beloved friend
Contents
List of contributors page ix

Acknowledgments xii
Introduction: towards a reconsideration of Neo-Kantianism 1

Nicolas de Warren and Andrea Staiti
part i neo-kantianism and philosophy 17

1. The Neo-Kantians on the meaning and status of philosophy 19
Andrea Staiti
2. Neo-Kantian ideas of history 39
Alan Kim
3. Neo-Kantianism and analytic philosophy 59
Hans-Johann Glock
4. Reise um die Welt: Cassirer’s cosmological phenomenology 82
Nicolas de Warren
part ii ethics and culture 109

5. Philosophy as philosophy of culture? 111
Christian Krijnen
6. The validity of norms in Neo-Kantian ethics 127
Beatrice Centi
7. Neo-Kantianism in the philosophy of law: its value
and actuality 147
Jonathan Trejo-Mathys
vii
viii Contents
8. Neo-Kantianism and the social sciences: from Rickert
to Weber 171
Gerhard Wagner and Claudius Härpfer
9. Simmel’s Rembrandt and The View of Life 186
Karen Lang
10. The binding of Isaac and the boundaries of reason: religion
since Kant 203
Peter E. Gordon
part iii the theory of knowledge 219

11. The philosophy of the Marburg School: from the critique
of scientific cognition to the philosophy of culture 221
Sebastian Luft
12. Natorp’s psychology 240
Daniel O. Dahlstrom
13. Cassirer and the philosophy of science 261
Massimo Ferrari
14. Kant and the Neo-Kantians on mathematics 285
Luca Oliva
Primary sources 307

Index 318
Contributors
beatrice centi is Professor of History of Philosophy at the University

of Parma. She is the author and editor of numerous volumes in both
Italian and English, including L’armonia impossibile. Alle origini del
concetto di valore in R. H. Lotze (1993), Coscienza, etica e architettonica
in Kant. Uno studio attraverso le Critiche (2002), and Values and
Ontology: Problems and Perspectives (2009).
d a n i e l o . d a h l s t r o m is Professor of Philosophy at Boston
University. He has published and edited several books, including
Heidegger’s Concept of Truth (Cambridge, 2001) and Philosophical
Legacies: Essays on the Thought of Kant, Hegel, and their Contemporaries
(2008).
n i co l a s de w a r r e n is Research Professor of Philosophy at the Husserl
Archives and the Center for Phenomenology and Continental
Philosophy at the Institute of Philosophy, KU Leuven. He is the author
of Husserl and the Promise of Time (Cambridge, 2009).
m a s s i m o f e r ra r i is Professor of History of Philosophy at the
University of Turin. He has published articles and books in German,
French, English, and Italian. Recent monographs include Retours à Kant
(2001), Ernst Cassirer: Stationen einer philosophischen Biographie (2003),
Categorie e a priori (2003), and Non solo Idealismo (2006).
hans-johann g lock is Professor of Philosophy at the University of
Zurich, and Visiting Professor at the University of Reading. He is the
author of A Wittgenstein Dictionary (1996), Quine and Davidson
(Cambridge, 2003), La mente de los animals (2009), and What Is
Analytic Philosophy? (Cambridge, 2008).
p e te r e . g o r d on is amabel b. james Professor of History and
Harvard College Professor at Harvard University, and a Faculty Affiliate
ix
x List of contributors
both in the Department of Germanic Languages and Literatures and in the
Department of Philosophy at Harvard University. He is the author of
several books, including Continental Divide: Heidegger – Cassirer – Davos
(2010).
claudius här p f e r is Research Affiliate in Sociology at Goethe-
University Frankfurt. He is the author of Georg Simmel und die
Entstehung der Soziologie in Deutschland (2014) and has published arti-
cles in journals including Società Mutamento Politica.
a la n k i m is Assistant Professor at Stony Brook University. He is the
author of Plato in Germany: Kant – Natorp – Heidegger (2010). He has
also published numerous papers in journals, including Phronesis,
Idealistic Studies, Internationale Zeitschrift für Philosophie, and Oxford
Studies in Ancient Philosophy.
c h r i s t i a n kr i j ne n is Professor of Philosophy at the University of
Amsterdam. He is the author and editor of numerous volumes on Neo-
Kantianism, including Nachmetaphysischer Sinn: Eine problemgeschich-
tliche und systematische Studie zu den Prinzipien der Wertphilosophie
Heinrich Rickerts (2001).
k ar e n l a n g teaches in the Department of History of Art at the
University of Warwick. She is the author of Chaos and Cosmos: On the
Image in Aesthetics and Art History (2006) and served as editor-in-chief of
The Art Bulletin from 2010 to 2013.
sebastian l uf t is Professor of Philosophy at Marquette University. His
publications include Subjectivity and Lifeworld in Transcendental
Phenomenology (2011) and The Space of Culture: Towards a Neo-
Kantian Philosophy of Culture (Cohen, Natorp, and Cassirer) (2015). He
has also edited The Neo-Kantian Reader (2015).
luca oliva is Assistant Professor in the Liberal Studies Program at the
University of Houston. His recent publications include La validità come
funzione dell’oggetto: uno studio sul NeoKantismo di Heinrich Rickert (2006).
a n d re a s t a i t i is Associate Professor of Philosophy at Boston College. He
has published numerous articles on Husserl, the phenomenological tradi-
tion, and Neo-Kantianism. His most recent publication is Husserl’s
Transcendental Phenomenology: Nature, Spirit, and Life (Cambridge, 2014).
j o n a t h a n t r e j o - m a t h y s was Assistant Professor of Philosophy
at Boston College. He has published articles in journals including
List of contributors xi
Ratio Juris, Philosophy and Social Criticism, and Contemporary
Pragmatism.
gerhard wagner is Professor of Sociology at Goethe-University
Frankfurt. He has published numerous articles in journals, including
Società Mutamento Politica and Revue internationale de philosophie.
Acknowledgments
We would like to thank Andrea Cimino, Leonardo Manfrini, Kevin

Marren, Blakely Vogel, and Patrick Eldridge for assistance during the
preparation of this volume. Our gratitude as well to Rosemary Crawley
and Hilary Gaskin at Cambridge University Press for their support
throughout this project.
This book is dedicated to the memory of Jonathan Trejo-Mathys, who
passed away prematurely at the age of thirty-five after a courageous struggle
with cancer. Jonathan was an admirable scholar, a vibrant thinker, an
inspiring teacher, a loving husband, father, and friend. As his superb
chapter in this collection demonstrates, he was a rare Anglophone
philosopher with full command of the German Neo-Kantian tradition in
the philosophy of law. In fact, his chapter was meant to be a précis for a
book that will never be written. Jonathan’s death is a great loss for the
profession, but his legacy continues in the lives of friends and peers who
had the privilege to share with him part of their intellectual and human
journey.
xii
Introduction: towards a reconsideration
of Neo-Kantianism
Nicolas de Warren and Andrea Staiti
In the summer of 1914, T. S. Eliot arrived in Marburg from Harvard

University to attend a summer course in philosophy before taking up
residency at Merton College, Oxford, for a year of study with Harold
Joachim, F. H. Bradley’s successor. At the University of Marburg, Eliot
met Paul Natorp, who assisted him in finding affordable accommodation
and lectured in his course on philosophy. The outbreak of the First World
War would cut short Eliot’s stay in Marburg, but not before he had the
chance to sketch a portrait of the venerable Neo-Kantian Professor. Natorp
strikes a professorial pose, one arm tucked behind his back, the other slung
across his waist. With elven ears and bald cranium, the philosopher appears
endearing in his otherworldliness. Natorp’s face is hidden behind oval
glasses, so large that they seem to constitute a hindrance rather than an aid
to seeing reality.
Eliot’s sketch can be seen as a visual epitome for how Neo-Kantianism
appeared to a younger generation of intellectuals and philosophers who
would come of age in the aftermath of a Europe laid waste through the
cataclysm of the Great War. Eliot’s amusing sketch is an apt illustration for
what Hans-Georg Gadamer, who wrote his PhD dissertation on Plato
under Natorp in 1922, characterized as the Neo-Kantian “calm and
confident aloofness” engrossed in “complacent system-building.”1 With
slightly more bite, Hannah Arendt charged Neo-Kantianism with drown-
ing philosophy “in a sea of boredom,” thereby offering a softer version of
the same hostility that spirited Martin Heidegger’s confrontation with
Ernst Cassirer at Davos in 1929.2 The perception of Neo-Kantianism at
Eliot’s alma mater was similarly unflattering. William James lampooned
1
Hans-Georg Gadamer, Philosophical Hermeneutics, trans. David E. Linge (Berkeley: University of
California Press, 1977), 230, 214.
2
Hannah Arendt, “Martin Heidegger at Eighty,” in M. Murray (ed.), Heidegger and Modern
Philosophy (New Haven: Yale University Press, 1978), 294.
1
2 nicolas de warren and andrea staiti
Heinrich Rickert’s work as an “unspeakable triviality” and a “fanciful flight
of ideas,” against which Rickert could only protest the famous American
psychologist’s “personal unkindness.”3
Time and historical perspective have not rendered more moderate or
judicious this perception of Neo-Kantianism. Commenting on Peter
Gordon’s balanced account of the Davos disputation between Heidegger
and Cassirer, Hans Sluga remarks:
Cassirer was no doubt an accomplished philosopher, an influential teacher,
and above all a thoroughly decent and admirable human being, but he does
not get close in stature to the much more problematic Heidegger, and he
certainly also lacks the philosophical radicalism of a Wittgenstein, Foucault,
or Derrida and the incisive scientific acumen of a Russell, Quine, or Rawls.
Attempts to revive his fortunes are, I am afraid, doomed to failure.4
Even the noted Kant biographer Manfred Kuehn proclaims that, “one of
the reasons why the Neo-Kantians’ popularity has diminished is that their
problems are no longer our problems,” whereas “Kant, on the contrary,
seems to have continued relevance.”5
New Approaches to Neo-Kantianism proposes to challenge these com-
monplace assumptions by offering a fresh examination of the diverse and
enduring contributions that the movement of Neo-Kantianism – in
truth: a fractious constellation of different movements – has provided
and still provides to various fields of philosophical inquiry. Contrary to
Kuehn’s representative judgment, many of the problems that defined
Neo-Kantianism are still very much our own, and in large measure due to
their Neo-Kantian heritage. Rather than accept a monolithic image of
Neo-Kantianism or reduce its identity to trite generalizations, the four-
teen essays in this volume explore anew its rich landscape by means of
critical engagements with central themes of Neo-Kantian philosophy.
Although the chapters in this volume range over the Neo-Kantian move-
ment broadly, New Approaches to Neo-Kantianism does not claim to offer
a comprehensive presentation or exhaustive history of Neo-Kantianism.
Instead, this volume offers a constellation of chapters with numerous
overlaps, divergent interpretations, and a variety of methodological
approaches. Exemplary themes, debates, and figures of Neo-Kantianism
are investigated for the purpose of establishing a more nuanced and
3
Heinrich Rickert, “Zwei Wege der Erkenntnistheorie,” Kant Studien 14 (1909), 169–228; here 173.
4
https://ndpr.nd.edu/news/24626-continental-divide-heidegger-cassirer-davos
5
Manfred Kuehn, “On Interpreting Kant Correctly: On the Kant of the Neo-Kantians,” in S. Luft and
R. Makreel (eds.), Neo-Kantianism in Contemporary Philosophy (Bloomington: Indiana University
Press, 2009), 113–131; here 128.
vibrant understanding of Neo-Kantianism, thus hopefully stimulating
as well as directing future explorations of its complex and contested
philosophical legacy.
*
Although the movement of Neo-Kantianism does not have a clearly
defined historical beginning, emerging instead from a loose chorus of
philosophical reactions to German Idealism during the 1820s and 1830s,
it did have a clearly defined ending. Cassirer’s exile to Sweden in 1935
after the Nazi ascension to power in 1933 marks the definitive end of a
Neo-Kantian presence within German academic institutions. Cassirer’s
exile is equally symbolic of the kind of ending that befell Neo-
Kantianism as a whole. Unlike Adorno, who returned to Europe after
the Second World War, or Heidegger, who, despite his ambiguous
internal exile during the war, found philosophical redemption in
France, or Husserl, whose legacy was promoted through the establish-
ment of the Husserl Archives in Leuven, Neo-Kantianism never
returned to the European scene of philosophy. This total eclipse
of Neo-Kantianism – the reasons for which are to be sure complex, as
yet not thoroughly studied, and reaching back to the First World
War – represents a defining event of twentieth-century philosophy,
the significance of which still reverberates across the full spectrum of
contemporary philosophy.
The purpose of this Introduction is not to provide a survey of Neo-
Kantianism from its multiple points of origin in the early nineteenth
century to its eclipse during the 1920s and 1930s (many of the chapters in
this volume deal in detail with various episodes from this grand adven-
ture), to hazard a general definition of Neo-Kantianism, to make an
inventory of its different movements with philosophical etiquettes, or
to decide who does or does not belong, or partly belong, to Neo-
Kantianism “proper.”6 Rather than promote an antiquarian or monu-
mental approach to Neo-Kantianism, the fourteen contributors to this
volume have been encouraged to adopt a critical approach based on their
own methodological preferences, historical understanding, and argu-
mentative position. For the purpose of situating the contributions to
this volume, it is instructive to begin with two influential verdicts against
Neo-Kantianism that substantially defined the philosophical–cultural
6
An early inventory claims to identify seven different types of Neo-Kantianism, see
Friedrich Ueberweg, Grundriss der Geschichte der Philosophie (Berlin: Mittler, 1923).
memory of its ending.7 The first stems from Lukács, the second from
Heidegger.
In The Destruction of Reason, Lukács, whose early thought was shaped by
Emil Lask, writes: “Granted, we know that even after Hegel academic
systems were created (Wundt, Cohen, Rickert, etc.), but we know also that
they were totally insignificant for the evolution of philosophy.”8 Even if we
reject Lukács’ dialectical materialist view of the history of philosophy, his
underlying condemnation of Neo-Kantianism as historically “insignifi-
cant” is not untypical. Such an appraisal of Neo-Kantianism’s historical
“insignificance” assumes a certain reading of the history of modern
philosophy as a Grand Narrative (whether of progress or decline) that
passes through three points of inflection: “from Kant to Hegel” through
the “hermeneutics of suspicion” of Marx, Nietzsche, and Freud to the
“parting of the ways” of twentieth-century philosophy into the movements
of Analytic (or Anglo-Saxon) Philosophy, Continental Philosophy, and
American Pragmatism. The teaching of the history of philosophy in
European and North American philosophy departments bears out and
reinforces the historical “insignificance” of Neo-Kantianism with the
standard curricular division of courses into Early Modern, Kant and
German Idealism, Marx, Kierkegaard, Nietzsche, Phenomenology, and
Early Analytic Philosophy. On this image of the history of philosophy,
largely constructed against Neo-Kantian historiography of philosophy
during the latter half of the twentieth century, it is as if Neo-Kantianism
effectively never occurred for the institutionalized memory of contempor-
ary thought.
Lukács’ judgment reflects a further commonplace association of Neo-
Kantianism with a sterile form of academic philosophy. Lukács derides
Neo-Kantianism as a Professorensysteme that remained isolated from the
“concrete” historical and cultural forces of “world history.” Such a char-
acterization of Neo-Kantianism fails in part to recognize the essential
contribution of Neo-Kantianism to the establishment of philosophy
within the modern university. This institutionalization of philosophy,
undoubtedly one of the more significant sociological transformations of
7
Important presentations of the development of Neo-Kantianism can be found in Klaus
Christian Köhnke, The Rise of Neo-Kantianism, trans. R. J. Hollingdale (Cambridge University
Press, 1991), Thomas Wiley, Back to Kant (Detroit: Wayne State University Press, 1978),
Manfred Pascher, Einführung in den Neukantianismus (Munich: Fink Verlag, 1977), and Hans-
Ludwig Ollig, Der Neukantianismus (Stuttgart: Metzler, 1979). The best one-volume account
remains Massimo Ferrari, Introduzione al neocriticismo (Rome: Laterza, 1997).
8
Georg Lukács, The Destruction of Reason, trans. P. Palmer (London: Merlin Press, 1980), 322.
philosophy since the early modern revolt (Descartes, Locke, Spinoza,
Hume, etc.) against university-philosophy, or Scholasticism, enjoyed as its
counterpart the formation of a European public space of philosophy with
the founding of philosophical journals (for example: Kant-Studien), the
organization of conferences, and an increased mobility and exchange of
ideas across different European universities. If Neo-Kantianism represents
sociologically the closure of any space for philosophy outside the academic
university, then the institutional rootedness of professional philosophy
today, its organization into different fields of expertise, competing schools,
and professional organizations, can be seen as one of the more enduring
legacies of Neo-Kantianism.
The charge that Neo-Kantianism remained an “aloof” (Gadamer) system
of professorial/professional philosophy fails on another score to recognize the
productive tension in its underlying conception of philosophy as both an
autonomous discipline, demarcated from other sciences in terms of method
and system, and a philosophical discipline necessarily situated vis-à-vis other
scientific fields of knowledge. With the two main movements of the
Marburg School and the Southwestern or Baden School, the philosophical
reach of Neo-Kantianism represented at the time an historically unique
interdisciplinary praxis, the likes of which has since never fully been matched
in breadth or ambition. As significantly, the pervasive influence of Neo-
Kantian thought on other disciplines, especially those new academic
disciplines of the late nineteenth century for which the framework of
Neo-Kantianism played a crucial role in their methodological formation
(sociology, art history, modern jurisprudence, cognitive psychology),
constitutes an important chapter in the development of modern forms of
knowledge. This spectrum of different fields of knowledge that intersect
within Neo-Kantianism is stressed in New Approaches to Neo-Kantianism
with a number of chapters in this volume providing case studies of the
specific contribution of Neo-Kantianism to art history, philosophy of
mathematics, philosophy of science, sociology, philosophy of religion, and
theory of law. If one understands the universality of philosophy in terms of
its ability to speak productively to disciplines other than itself, then Neo-
Kantianism in this regard represents one of the last universal frameworks of
philosophy, not because of its rigid system-building, but on the contrary
due to its “pluralization” of rationality, as paradigmatically expressed in
Ernst Cassirer’s Philosophy of Symbolic Forms.
If Neo-Kantianism has suffered from a perception of “historical insig-
nificance,” it has equally suffered from a widespread identification of its
philosophical center of gravity with an exclusive concern with the problem
of knowledge. In the 1929 Davos disputation between Heidegger and
Cassirer, while Cassirer stressed a “functional” and “historical” under-
standing of Neo-Kantianism, Heidegger promoted a “substantialist”
definition of Neo-Kantianism as essentially a theory of knowledge
(Erkenntnistheorie).9 As Heidegger argues:
We can only understand what is common to Neo-Kantianism on the basis
of its origin. The genesis of [Neo-Kantianism] lies in the predicament of
philosophy concerning the question of what properly remains of it in the
whole of knowledge. Since about 1850 it has been the case that both the
human and the natural sciences have taken possession of the totality of what
is knowable, so that the question arises: what still remains of Philosophy if
the totality of beings has been divided up under the sciences? It remains just
a knowledge of science, not of beings.10
Heidegger’s prejudice has since largely prevailed. Neo-Kantianism
remains mostly known for its promotion of an “epistemological” inter-
pretation of Kant (which still dominates current readings of Kant); it
remains mostly associated with an overly “intellectual” account of know-
ledge; it remains wedded to a perceived uncritical dependency on the
natural sciences, especially mathematical physics as the paradigmatic
form of cognition, and an outmoded conception of the sciences and
mathematics.
Yet, Heidegger’s view of Neo-Kantianism and its historical genesis fails
to acknowledge a more complex dynamic to the Neo-Kantian project of
Erkenntnistheorie. The German term Erkenntnistheorie can be traced back
to Friedrich Eduard Beneke’s Kant und die philosophische Aufgabe unserer
Zeit (Kant and the Philosophical Task for our Times), although it was not
until Eduard Zeller’s 1862 inaugural lecture in Heidelberg that the term
Erkenntnistheorie began to enjoy widespread circulation (and hence diverse
meanings) and function as an organizing signifier for the self-proclaimed
project of Neo-Kantianism.11 Published in Berlin the year of Hegel’s death
in 1831, the title Kant and the Philosophical Task for our Times already signals
9
For studies of this encounter, see Peter Gordon, Continental Divide: Heidegger, Cassirer, Davos
(Cambridge, MA: Harvard University Press, 2010), Michael Friedman, A Parting of the Ways:
Carnap, Cassirer, and Heidegger (Chicago: Open Court, 2000), Cassirer – Heidegger. 70 Jahre
Davoser Disputation (Hamburg: Felix Meiner, 2002), and Pierre Aubenque, “Le débat de 1929
entre Cassirer et Heidegger,” in Ernst Cassirer: De Marbourg à New York, l’itinéraire philosophique
(Paris: Les Éditions du Cerf, 1990), 81–96.
10
Martin Heidegger, Kant and the Problem of Metaphysics, trans. R. Taft (Bloomington: Indiana
University Press, 1997), 171–172.
11
Friedrich Eduard Beneke, Kant und die philosophische Aufgabe unserer Zeit (Berlin: Ernst Siegfried
Mittler, 1832).
how the Kantian task of an Erkenntnistheorie (as Beneke conceives it) is
meant to provide a philosophical alternative to the dominance of German
Idealism. The concept of Erkenntnistheorie must in fact be seen as an
operative concept for three distinct questions: the question of philosophy;
the question of knowledge; the question of culture and politics. From the
beginning, the signifier “Erkenntnistheorie” operated within three domains
of discourse: philosophy, knowledge, culture and politics.
As Julius Ebbinghaus, appointed to the chair of philosophy in Marburg
in 1945, perceptively noted, the disqualification of Neo-Kantianism during
the 1920s and 1930s entailed a rejection of a certain idea of philosophy.12
This ending of Neo-Kantianism as the ending of a certain conception of
philosophy reflects the centrality of the question “what is philosophy?” for
the historical development and self-conception of Neo-Kantianism. The
question “what is Neo-Kantianism?” cannot be approached without rais-
ing at the same time the question “what is philosophy?” As a number of
chapters in this volume explore, different movements within Neo-
Kantianism, as well as different stages within these movements, can be
understood as different orientations within this metaphilosophical
concern.
Beneke’s early use of the term Erkenntnistheorie reveals a second opera-
tive significance, which would become extended, contested, and trans-
formed with further developments of Neo-Kantianism. Through his
contact with John Herschel and William Whewell in England, Beneke
sought to impart a new direction and renewed impetus to German thought
away from German Idealism. Under the heading of Erkenntnistheorie,
Beneke understood a science of experience that would continue Kant’s
critical concern with the conditions of possibility for knowledge, albeit
through the application of psychology and in closer contact with the
empirical sciences.13 Philosophy in this manner aspires to the status of
the “science of the sciences” (die Wissenschaft der Wissenschaften), yet
requires knowledge of the natural sciences, which, during the 1820s and
1830s, were beginning their rapid ascent within German universities, thus
gradually displacing the Naturphilosophie of German Idealism. For later
Neo-Kantians, this foundational relation between the science of phil-
osophy and other sciences would remain central. Contrary to
Heidegger’s narrow assessment, this relationship (in methodological and
12
Julius Ebbinghaus, “Hermann Cohen als Philosoph und Publizist,” Archiv für Philosophie 6 (1956),
109–122.
13
Friedrich Eduard Beneke, Lehrbuch der pragmatischen Psychologie (Berlin: Mittler & Sohn, 1853) and
Skizzen der Naturlehre der Gefuehle (Göttingen: Vandenhoeck & Ruprecht, 1825).
systematic terms) would undergo significant transformation in ways that
reflected the evolution of the natural and human sciences in the late
nineteenth century. In both the Marburg and Baden Schools, the scope
of philosophy’s interest in the sciences kept pace with the emergence of
new sciences (ethnography, linguistics, etc.) as well as revolutions within
established sciences (non-Euclidean geometry, number theory, logic of
relations, etc.).
In advocating for an Erkenntnistheorie as the task of philosophy, Beneke
repeatedly bemoans the “degeneration of German thought” and the
“intellectual barbarism” of German Idealism. In addition to its meta-
philosophical significance, the signifier “Erkenntnistheorie” was forged
with cultural purpose. Already in its early expression with Beneke, the
task of Erkenntnistheorie is implicitly to foster a renewal of (German)
culture. It is not, however, until the 1848 March Revolution that the
cultural and political significance of a new orientation towards Kant in
terms of an Erkenntnistheorie as the future of German thought becomes
more emphatically and widely advocated. In Christian Weisse’s 1847
inaugural lecture In welchem Sinn die deutschen Philosophie jetzt wieder an
Kant sich zu orientieren?, we hear a pronounced amplification of the more
implicit connection between Kant’s critical philosophy and the critical
imperative for a reformation of German culture in Beneke, even if Weisse’s
own speculative theism departed from the thrust of Beneke’s appeal to
Kant’s thinking.14
During the 1870s and the first decade of German unification, the
inseparable connection between political and cultural liberalism and
the Neo-Kantian project of Erkenntnistheorie became exemplified in the
towering figure of Hermann Cohen, founder of the Marburg School, and
whose decisive critique of psychological conceptions of Kant’s critical
philosophy (as with Beneke) ushered, as Heidegger recognizes, the sec-
ond epoch in the development of Neo-Kantianism.15 Cohen’s appoint-
ment at the University of Marburg in 1873 represented the first
appointment of an unconverted Jewish professor in Germany. His two
promoters, Eduard Zeller, whose famous 1862 inaugural lecture in
Heidelberg, Über Bedeutung und Aufgabe der Erkenntnistheorie, is the
veritable launching point for the slogan Zurück zu Kant!, and Friedrich
Albrecht Lange, whose innovative reading of Kant would influence
14
Christian H. Weisse, In welchem Sinn die deutschen Philosophie jetzt wieder an Kant sich zu
orientieren? (Leipzig, 1847).
15
Martin Heidegger, Zur Bestimmung der Philosophie, GA 56–57 (Frankfurt am Main: Klostermann,
1987), 140.
Cohen (even as Cohen contested its psychological leanings) as well as
Vaihinger’s Philosophie des Als Ob, were both cultural radicals. Zeller was
a confessed atheist, who had studied with Bruno Bauer and David Strauss
in Tübingen;16 Lange was a militant socialist who provided a critical
impetus for the development of Kantian Socialism through his political
engagement and political writings.17 An atheist, a social militant, and the
first Jewish Privantdozent of philosophy in Germany launched Marburg
Neo-Kantianism, the target of Heidegger’s critical identification of Neo-
Kantianism with Erkenntnistheorie in 1929.
*
Philosophical slogans have always provided a useful shorthand for the self-
positioning of philosophical movements, but inevitably at the expense of a
genuine understanding of the movement itself. The slogan “Back to Kant!”
has fared no better. Launched during the 1860s, the combination of “back”
and “Kant” unduly came to fix an image of Neo-Kantianism that in many
regards obscured its actual development. Rather than a movement back,
Neo-Kantianism imagined itself as a movement forwards, as most vividly
expressed with the Kantian Socialism of Karl Vorländer’s call Vorwärts mit
Kant as an alternative to Marxism.18 Even the centrality of “Kant” is
historically misleading, since Neo-Kantianism increasingly loosened its
orientation towards Kant (and the crucial question: which Kant?, since
Kant’s three critiques found varied interpretations through-out Neo-
Kantianism) as it developed and diversified historically. Cohen, whose
conception of transcendental thought as dependent on the Faktum of
the natural sciences, more than any other figure framed the historical
perception of Neo-Kantianism as Erkenntnistheorie, himself progressively
moved away from his own critical philosophy in the first decade of the
twentieth century. In a 1905 review of Cohen’s Logik der reinen Erkenntnis,
Leonard Nelson (who influenced Karl Popper) chides Cohen for falling
16
Regrettably, Zeller’s theological writings have been entirely forgotten. A significant exception:
Johannes Zachhuber, Theology as Science in Nineteenth-Century Germany: From F. C. Baur to
Ernst Troeltsch (Oxford University Press, 2013).
17
As Henri Dussort remarks: “Mais la motivation la plus puissante de Lange est d’ordre éthique et
politique. Sa vocation et sa figure sont avant tout celles d’un militant,” L’École de Marbourg (Paris:
PUF, 1963), 53. An excellent treatment of Lange’s political thinking can be found in Thomas Wiley’s
Back to Kant: The Revival of Kantianism in German Social and Historical Thought 1860–1914, chapter
4. See F. A. Lange, Die Arbeiterfrage. Ihrer Bedeutung für die Gegenwart und Zukunft (Duisburg:
Bleuler-Haushier, 1865).
18
Karl Vorländer, Kant und der Socialismus under besonderer Berücksichtigung der neuesten theoretischen
Bewegung innerhalb des Marxismus (Berlin: Reuther & Reichard, 1900).
into “a dogmatic ontology” or, in other words, into a “Hegelian
ontology.”19 In a lecture at Heidelberg in 1910, Windelband positively
addresses a renewal of Hegelianism among contemporary German
thought.20 This shift in the relationship of Neo-Kantianism towards
Hegel spurred Rickert to claim that he considered the philosophy of
Kant and Hegel as representing a fundamental unity.21 In the years prior
to the outbreak of the First World War, who actually remained “Kantian”
among the classic figures of Neo-Kantianism?
If T. S. Eliot had returned to Marburg after the war, he would have
discovered a very different Natorp. The First World War was an intel-
lectual and cultural catastrophe for Neo-Kantianism.22 In an ironic twist,
the demise of Neo-Kantianism precipitated by the war reflected in reverse
the original historical situation of its beginnings: in a 1920 article
published in Kant-Studien, “Der Neukantianismus und die Forderung
der Gegenwart” – a marvelous echo of Beneke’s 1831 title – Karl Sternberg
calls for a renewal of Neo-Kantian thought through a critical reflection
on its historical development in the aftermath of the war.23 After
contributing to the “war-philosophy” undertaken by many German
(and European) professors, Natorp abandoned his orientation towards
the Marburg Neo-Kantianism which had so thoroughly defined his pre-
war thinking – Cohen had even once considered Natorp’s writings as too
infused with Neo-Kantian convictions.24 In his 1922–1923 lecture course
Philosophische Systematik, Natorp began to re-create himself as another
kind of philosopher.25 The primacy of Erkenntnistheorie (so critically
rejected by Heidegger in 1929) became displaced by a new philosophy
of origins centered on the primacy of ontology.26 The problem of know-
ledge now becomes anchored in an original foundation of being; that
there is being (es gibt) is not to be conflated with what is a being – the true
19
Leonard Nelson, “Rezension von Hermann Cohen: System der Philosophie. 1. Teil. Logik der
reinen Erkenntnis,” in Gesammelte Schriften (Hamburg: Meiner, 1973), ii: 1–28.
20
Wilhem Windelband, “Die Erneuerung des Hegelianismus,” in Präludien I (Tübingen: J. C. B. Mohr,
1907), 273–289.
21
Heinrich Rickert, System der Philosophie (Tübingen: J. C. B. Mohr, 1921), xi.
22
For the reaction of German (and British) philosophers to the war, Peter Hoeres, Krieg der
Philosophen (Paderborn: Schöningh, 2004). The impact of the war on Neo-Kantianism as a whole
has yet to be studied.
23
Karl Sternberg, “Der Neukantianismus und die Forderung der Gegenwart,” Kant-Studien 25 (1920),
396–410.
24
Heinz Holzhey, Cohen und Natorp (Basel: Schwabe, 1986), i: 6.
25
Posthumously published: Paul Natorp, Philosophische Systematik, ed. H. Natorp (Hamburg: Felix
Meiner, 2004).
26
See Éric Dufour, Paul Natorp. De la psychologie générale à la systématique philosophique (Paris: Vrin,
2010), especially section ii.
object of knowledge.27 As Natorp writes to the Austrian writer and
literary scholar Hans Prager shortly before his death in 1924, perhaps
with his own earlier Neo-Kantian intellectualism and system-building in
mind, “wir kranken alle an Abstraktion” (we are all sick with abstraction).
What a different portrait T. S. Eliot might have indeed sketched, had he
returned.
*
New Approaches to Neo-Kantianism is organized into three parts. The
presentation of these three parts is meant to redraw the map of our
understanding of Neo-Kantianism in emphasizing the complex legacy of
Neo-Kantianism for contemporary thought (Part i), the centrality of ethics
and culture (Part ii), and the ways in which the concern with the problem
of knowledge and science (Part iii) is situated within a broader vision of
philosophy and human culture.
Part i, “New-Kantianism and philosophy,” examines Neo-Kantian
conceptions of philosophy and the history of philosophy as well as its
relationship to Analytic Philosophy and Husserlian Phenomenology.
In Chapter 1, “The Neo-Kantians on the meaning and status of
philosophy,” Andrea Staiti discusses the metaphilosophical question
“what is philosophy?” within certain strands of Neo-Kantianism, begin-
ning with the writings of Beneke. Staiti emphasizes how Neo-Kantianism
is characterized by an effort to distinguish between the specialized sciences
and the science of philosophy in terms of a distinction between “parts of
the world” and the “world as whole.” After a discussion of Beneke’s works,
Staiti turns to Rickert as a representative figure for the systematic elucida-
tion of philosophy as a theory of the world as a whole.
In Chapter 2, “Neo-Kantian ideas of history,” Alan Kim explores the
salient differences between the Southwestern and Marburger Schools’
conceptions of the history. While the Southwestern School can be seen
as primarily interested in history as a discipline in its own right and
attempted to understand the specificity of historical understanding, the
Marburg School sought to integrate history into their own understanding
of philosophy. This contrast between these schools is best exemplified in
how canonical figures in the history of philosophy, especially Plato, were
historically interpreted.
27
For the importance of Natorp’s ontological thinking on the formation of Heidegger’s thought, see
Massimo Ferrari, Ernst Cassirer: Dalla scuola di Marburgo alla filosofia della cultura (Florence: Leo
Olschki, 1996).
In Chapter 3, “Neo-Kantianism and analytic philosophy,” Hanjo Glock
challenges a standard view that considers Neo-Kantianism and analytic
philosophy as standing in tension to each other. While the first is often
perceived as philosophically “conservative” and “reactionary,” the latter is
celebrated as “revolutionary” and “innovative.” Although Glock identifies
a clear parting of the ways between both movements in terms of doctrine,
method, and style, he explores a number of important points of agreement
and influence. Glock takes issue with different conceptions of the relation
between Analytic Philosophy and Neo-Kantianism, and concludes his
reflections with a reassessment of Frege’s debt to Lotze while rejecting
the claim of any influence from the Southwest School of Neo-Kantianism
(Windelband).
In Chapter 4, “Reise um die Welt: Cassirer’s cosmological phenomenol-
ogy,” Nicolas de Warren contrasts the phenomenological projects
of Edmund Husserl and Ernst Cassirer. De Warren locates the origins of
Cassirer’s phenomenology of knowledge in the transcendental project of
Cohen and Natorp, but whereas each conceived of a system of reason based
on functions of cognition, volition, and value, Cassirer’s conception of
rationality “pluralizes” the transcendental forms of understanding. In this
manner, de Warren examines Cassirer’s critique of Husserlian phenomen-
ology and his argument for a cosmological phenomenology of symbolic
forms that undermines any methodological distinction between “transcen-
dental (constituting) subjectivity” and the constituted forms of possible
experience.
In keeping with the emphasis on the cultural and political orientation
of Neo-Kantianism, Part ii, “Ethics and culture,” offers an array of
studies on various aspects of Neo-Kantian conceptions of practical
philosophy, culture, aesthetics, philosophy of religion, and the
philosophy of law.
In Chapter 5, “Philosophy as philosophy of culture?” Christian Krijnen
argues that Neo-Kantianism, especially in the Southwestern School, can
arguably be seen as the first and perhaps only systematic philosophy of
culture in the history of philosophy. As a philosophy of values, the tradi-
tional project of a critical philosophy, as inherited from Kant, becomes
duly transformed into a critical philosophy of culture. Krijnen further
examines a tension within a Neo-Kantian philosophy of culture between
a creative interpretation of Kant’s philosophy as a philosophy of culture
and the intricate problems posed by the Hegelian demand for a systematic
philosophy that does justice to philosophy as critical self-knowledge, a
project to which Neo-Kantians are equally committed.
In Chapter 6, “The validity of norms in Neo-Kantian ethics,” Beatrice
Centi examines how Neo-Kantian ethics challenged the formalism of
Kant’s ethics. As Centi discusses, rather than begin with a formal account
of moral motivation and action, Neo-Kantian thinkers, beginning with
Cohen’s influential Kants Begründung der Ethik (1877) and Ethik des reinen
Willens (1904), addressed the material content of moral imperatives and
duties. In this manner, Cohen’s ethical thinking – the main focus of her
chapter – advanced a deontological conception of moral life along two
parallel lines: on the one hand, the relationship between ethics and psy-
chology could be profitably explored in terms of the relationship between
feeling (content) and reason (form); on the other hand, the relationship
between ethical norms and the rational structure of values could be
recognized, with the consequence of bringing into sharper focus the ques-
tion of universal values and the problem of culture.
In Chapter 7, “Neo-Kantianism in the philosophy of law,” Jonathan
Trejo-Mathys offers a groundbreaking analysis of the influence of Neo-
Kantianism on twentieth-century developments in the philosophy of law.
After a review of the more notable contributions of the Marburg School
(Cohen, Stammler) and the Heidelberg School (Lask) to questions of
legal theory, Trejo-Mathys examines how Neo-Kantian philosophy of
law, while maintaining allegiance to what it took to be Kantian princi-
ples, innovated both conceptually and doctrinally in formulating
responses to intellectual and socio-political problems that are still worthy
of consideration and surprisingly current. Trejo-Mathys further reveals
the Neo-Kantian lineage of the contemporary discourse theory of law
developed by Hans Kelsen, Robert Alexy, Klaus Günther, and Jürgen
Habermas.
In Chapter 8, “Neo-Kantianism and the social sciences: from Rickert to
Weber,” Gerhard Wagner and Claudius Härpfer focus on the
Southwestern School of Neo-Kantianism and on Max Weber’s
Interpretive Sociology. While Weber at first affiliated himself with
Rickert’s The Limits of Concept Formation in the Natural Sciences and
his project of grounding the historical and cultural sciences through an
“ideographic” approach, Wagner and Härpfer demonstrate how Weber
eventually parted ways with Rickert’s epistemological framework in devel-
oping a sociological method based on a “nomothetic” approach. In light of
this development from Rickert to Weber, Wagner and Härpfer argue that
justice cannot be done to either Neo-Kantianism or the social sciences by
classifying scientific disciplines on the basis of an antinomy: nomothetic
and idiographic (generalizing and individualizing sciences).
In Chapter 9, “Simmel’s Rembrandt and The View of Life,” Karen Lang
explores Simmel’s study of Rembrandt (recently translated into English) in its
cultural and philosophical context. Whereas the importance of Neo-Kantian
thinking for the art historical movement launched from the Warburg
Institute has received scholarly attention, less known is Georg Simmel’s
study Rembrandt: An Essay in the Philosophy of Art. Central themes in
Simmel’s Lebensphilosophie – Inner Life, individuality and universality, reli-
gious art – are expressed in his investigation of Rembrandt’s artistic produc-
tion. Lang devotes special emphasis to Simmel’s conception of portraiture
and the expression of life in art-works, and establishes how the singularity of
the individual and the freedom of life are given expression in artistic creation,
thus anticipating many of the central intuitions in his own philosophy of life.
In Chapter 10, “The binding of Isaac and the boundaries of reason:
religion since Kant,” Peter Gordon tracks the legacy of Kant’s reflections
on the question of religion and the relation between faith and reason. Two
basic interpretations of Kant’s philosophy of religion have largely shaped
its legacy: an “internal” relation between reason and revelation as opposed
to an “external” relation. In this light, Gordon examines Cohen’s religion
of reason as developed in his final writings on religion and Judaism. In an
intriguing turn, Gordon demonstrates the importance of Cohen’s philo-
sophy of religion by contrasting the respective philosophical reflections on
religion in John Rawls and Jürgen Habermas.
New Approaches to Neo-Kantianism concludes with Part iii, “The theory
of knowledge,” on the problem of knowledge, or Erkenntnistheorie.
In Chapter 11, “The philosophy of the Marburg School,” Sebastian Luft
assesses the epistemology of the Marburg School, covering its three leading
figures, Cohen, Natorp, and Cassirer. Luft demonstrates how Natorp’s
critical psychology and Cassirer’s philosophical system of symbolic for-
mation can be seen as innovative applications of the transcendental method,
as first conceived by Cohen. Luft concludes with a useful contrast of these
respective epistemological projects with contemporary forms of transcen-
dental philosophy, such as with the tradition of transcendental argumen-
tation (Strawson) and transcendental pragmatism (Apel).
In Chapter 12, “Natorp’s psychology,” Daniel Dahlstrom offers a
penetrating reading of Paul Natorp’s transcendental psychology and its
complex intersections with Husserl’s phenomenological approach to
subjectivity. As Dahlstrom explores, although Husserl first severely
rejected central elements in Natorp’s psychology in the Logical
Investigations (most significantly, Natorp’s notion of a pure ego), Husserl
came to revise his phenomenological conception of subjectivity and the
self (the ego) in response to Natorp’s criticisms. As Dahlstrom further
investigates, Heidegger devoted a large part of his early Freiburg lecture,
Phänomenologie der Anschauung und des Ausdrucks, to a critical examina-
tion of Natorp’s “reconstructive psychology,” not least in order to address
the question of how the world of the self (Selbswelt) becomes constituted.
In Chapter 13, “Cassirer and the philosophy of science,” Massimo
Ferrari develops a detailed assessment of Cassirer’s philosophy of science
against the backdrop of his classic works such as Substance and Function.
Ferrari demonstrates how Cassirer modified and eventually changed his
earlier understanding of the structure of knowledge through his philoso-
phical confrontation with the developments of modern mathematics,
Einstein’s Theory of Relativity, and quantum mechanics. As Ferrari argues,
Cassirer treads a fine line between keeping a critical distance from certain
Kantian assumptions concerning causality, physical laws, etc., and their
complete abandonment due to the revolution of modern physics.
In Chapter 14, “Kant and the Neo-Kantians on mathematics,” Luca
Oliva situates Heinrich Rickert’s philosophy of mathematics against the
background of debates surrounding Kant’s mathematical thinking. The
form of mathematical knowledge within the framework of the First
Critique as well as the relation to intuition were topics of vigorous debate
among leading figures of Neo-Kantianism. Oliva presents in detail the
neglected account of number in Rickert’s writings and his “transcendental
empiricism” for mathematics. As Oliva argues, Rickert moves towards a
form of mathematical structuralism that brings his conception of number
closer to a Kantian mereological approach, and thus away from the “rela-
tion thought” of Cassirer and Natorp.
part i
Neo-Kantianism and philosophy
chapter 1
The Neo-Kantians on the meaning

and status of philosophy
Andrea Staiti
Introduction
There seems to be a recurring pattern in the history of modern philosophy:
an evolution through a series of turns separated by more or less extended
(and more or less turbulent) phases of “normality.” Each turn comes onto
the scene with a claim to revolutionize the whole discipline and finally to set
it on “the secure path of a science” (Kant 1998: 107 [B x]). While shifts of
focus and reappraisals of priorities are as old as the discipline itself, turns,
carried out with the aim of redefining from scratch philosophy’s tasks and
methods, arguably start with Kant.1 His Copernican turn can be considered
paradigmatic for all subsequent turns in the history of philosophy. With it,
Kant created a new language, articulated new problems, and specified the
kinds of questions that philosophers for the next few generations felt
legitimately entitled to ask. He prepared the terrain for both commentators
such as Reinhold and Beck, whose work is primarily in the elucidation and
systematizing exposition of Kant’s thought, and original figures such as the
idealistic triad Fichte, Schelling, and Hegel, as well as the philosophical–
psychological school of Herbart and Fries. A few decades later in the
nineteenth century, a new such turn redefined the discipline in accordance
with the spirit of French positivism. The dawn of the twentieth century
witnessed Husserl’s turn to phenomenology (which, incidentally, can only
be understood in the aftermath of both the Kantian and the positivistic
turns), and the second half of the twentieth century gave way to what Rorty
famously christened “the linguistic turn.”
1
One could argue that, almost two centuries before Kant, Descartes already advocated for some kind of
turn in philosophy. However, I consider it helpful to identify two components as definitive of a “turn”
in philosophy. A turn must involve both methods and tasks. While Descartes did indeed set up a new
method for philosophical thinking, his project revolves around a set of tasks that he inherited from
traditional Scholasticism, such as the proofs of God’s existence and the immortality of the soul.
Therefore, only half of the historical pattern I am trying to describe here is predicated in Descartes.
19
20 andrea staiti
Philosophical turns, of course, do not occur in a linear and irreversible
way. They variously overlap and commingle, so that not infrequently
philosophers working under the aegis of one turn may have hard times
recognizing as philosophical the work of colleagues who haven’t yet turned
or are simply unwilling to do so. Inter- and intradepartmental grudges
between self-branded analytic and self-branded continental philosophers
in our time are a good example of this complex and non-linear develop-
ment of the discipline through overlapping turns.
Whether the list of philosophical turns just outlined is exhaustive is a
question I can leave aside here. If the ones I mentioned are at least some of
the turns characterizing the past two centuries, however, they will suffice to
exemplify the characteristic trend that I would like the reader to observe at
the beginning of this chapter. Unlike in, say, the natural sciences, the phase
of philosophical normal science established in the wake of a turn does not
start to wane because of some groundbreaking discovery, or the like. There
is arguably no equivalent to such discoveries in philosophy. Rather, it
seems that most philosophical turns eventually lose momentum when
practitioners in the field start to register a certain uneasiness about the
excessive technicalization and the restriction of focus that every such turn
inevitably implies. The initial enthusiasm ensuing from the ability to set
boundaries to philosophical questioning and give precise definitions to
philosophical terms tends eventually to morph into boredom and a sense of
insipidity. This uneasiness generally coincides with a desire to recover a
more holistic perspective in philosophy and to address the perennial
questions about our human condition and its meaning in a world rendered
by the advancement of empirical research increasingly foreign to our
innermost aspirations.
Symptoms of “post-turn uneasiness” abound today. The aforemen-
tioned linguistic turn lost momentum even among thinkers who were
trained in accordance with its most rigorous principles. In some cases
accomplished philosophers working in the analytic tradition, such as
Philip Kitcher and David McNaughton, voice disappointment with the
most recent developments of the discipline, developments characterized by
a stifling specialization and an incapacity to address the truly significant
questions that philosophers traditionally ask. In a recent paper tellingly
entitled “Why Is So Much Philosophy So Tedious?”McNaughton writes:
If philosophy is no more than a pyrotechnic display of ingenious argumen-
tation, then Callicles was surely right when he chided Socrates and told him
that philosophy was a fine activity for a youth, but a disgraceful one for
The Neo-Kantians on the meaning and status of philosophy 21
someone of mature years. A life devoted to solving crossword puzzles has
little to commend it – and certainly does not deserve public subsidy. We
should reject this conception of our discipline. Philosophy can and should
deal with important issues. It should enable us both to understand our place
in the world and to live in it.2
Philip Kitcher echoes the same feeling with slightly different words in his
paper “Philosophy inside out,” where, after criticizing the current state of
philosophy in its specialized and technicized form, he points out:
“Philosophy grows out of an impulse toward understanding nature and
the human place in it.”3
Note that both Kitcher and McNaughton counterpose narrow specia-
lization to philosophy’s original commitment to shed light on the world
as a whole and our place in it. Neither of them, however, explains just
what they mean by “the world as a whole”and “our place in it,” probably
taking the meaning of these expressions to be somehow intuitive. And
while there is, indeed, an intuitive meaning to these phrases, the sophis-
ticated system of mereological (whole/parts) and spatial (place) meta-
phors undergirding their intelligibility is worthy of philosophical
scrutiny. The largely forgotten work of the Neo-Kantians offers precious
resources for this endeavor.
This chapter examines some attempts in Neo-Kantianism to clarify the
meaning and status of philosophy. All such attempts revolve around the
notion that philosophy inquires about the world as a whole, as opposed to
the specialized sciences that focus exclusively on isolated parts of the world.
The protagonists of this essay are Alois Riehl, Friedrich Eduard Beneke, and
Heinrich Rickert. The positions of these thinkers diverge significantly in
regard to (1) how we access the world as a whole; (2) how to grapple with
the philosophical problems posed by the world as a whole; and (3) the
relationship between philosophy and the empirical sciences, and, correla-
tively, between the world as a whole and the world’s parts. In spite of their
differences, however, all these thinkers are conversant with Kant; thus, their
disagreement is especially instructive concerning the far-reaching ramifica-
tions of the transcendental tradition at the dawn of the contemporary era,
i.e. before the split between analytic and continental philosophy. My hope is
that this retrieval of the Neo-Kantian work on the nature and status of
philosophy is both illuminating and invigorating for those seeking to rethink
2
D. McNaughton, “Why Is So Much Philosophy So Tedious?” Florida Philosophical Review 9/2
(2009), 1–13; here 5.
3
P. Kitcher, “Philosophy inside out,” Metaphilosophy 42/3 (2011), 248–260; here 252.
22 andrea staiti
the meaning of the discipline beyond the strictures of the so-called linguistic
turn.
The world-whole and the world’s parts

In 1903 the Neo-Kantian philosopher Alois Riehl gave the following
retrospective description of the foregoing century:
If somebody had resolved to speak about philosophy in public around the
middle of the past century, he would have certainly suffered a disastrous
failure. He would have found no audience for his talk, not even among his
most cultivated contemporaries. Moreover, he would have exposed himself
to the suspicion of being willing to praise something like alchemy in the age
of the natural sciences. (Riehl 1903: 1)
Riehl knew that atmosphere very well from his student years, and his own
brand of critical philosophy was largely influenced by it. It was the heyday
of positivism in Germany following the demise of Hegel’s speculative
idealism and the impressive record of revolutionary discoveries in the
natural sciences. In the eyes of the general public, philosophy had failed
as a discipline for two interconnected reasons: (1) the lack of a clearly
delimited field of inquiry; (2) the lack of a clearly defined and empirical
research-method. As for the first point, the grandiose systems of German
idealism had a pretension to weave together in one meaningful metaphy-
sical narrative each and every segment of reality to form an exhaustive
Wissenschaft. Hegel’s system, for example, encompasses logic, physics,
geography, political science, psychology, and virtually every other
discipline ever cropping up in the history of humankind. The natural
sciences, thus goes the positivist rebuke, were able to achieve their
tremendous results only to the extent that they divided up their labor
into manageable and clearly demarcated areas of inquiry. Philosophy
failed because it was unable to do so. Second, the kinds of methods
employed by philosophical system-builders are not empirical. Consider
Hegel’s dialectics. It is worked out at a very abstract and formal level in
the introduction to the Phenomenology of Spirit and then used throughout
the book to arrange the most diverse kinds of materials conceivable. This
method is not designed to interrogate facts so that they can confirm or
falsify hypotheses. No matter how intriguing and profound a narrative
about absolute spirit it may result in, this is clearly not an empirical
narrative that later practitioners can verify or amend on the basis of better
observations and hypotheses.
Faced with such formidable lines of criticism, German philosophers in
the positivist age had only a few options available: either abandon the
discipline outright and jump on the bandwagon of specialized empirical
research, or rethink the meaning of philosophy from the ground up, so as
to create a legitimate intellectual space for it beside the empirical sciences.
For those favoring this second option, they first sought refuge in Kant’s
critical philosophy. Kant was the last major German philosopher before
the breakthrough of speculative idealism. Philosophy had to resume the
Kantian work that had been interrupted by the advent of system-building
thinkers like Fichte and Hegel. This approach can be found as early as 1832
in the work of the philosopher, pedagogue, and psychologist Friedrich
Eduard Beneke. In his essay Kant und die philosophische Aufgabe unserer
Zeit, written to commemorate the fiftieth anniversary of publication of
Kant’s First Critique, Beneke sets out to separate the wheat from the chaff
in Kant’s seminal book. Beneke finds that the true merit of Kant’s critical
philosophy lies in its pars destruens, that is, in its forceful emphasis on the
limits of scientific knowledge and the consequent rejection of metaphysics
as a science. However, on Beneke’s account, the pars construens of Kant’s
philosophy, that is, its positive spelling out of synthetic a priori knowledge
as the condition for the possibility of empirical knowledge, is deeply
flawed and even responsible for the bad speculative habits of the next
generation of philosophers. This is because this synthetic a priori
knowledge on which all further knowledge allegedly depends cannot
be legitimately considered knowledge by Kantian standards. Kant’s
philosophy is thus self-contradictory. In Beneke’s own words:
According to Kant’s principles, the simple powers or forms of the human
mind are in no way cognizable. Neither immediately from experience (in fact,
experience cannot dig into the simple, into the in-itself, because it is entirely
limited to products or appearances), nor independently of experience (in fact,
for whatever we construct independently of experience out of mere concepts
we have no guarantee regarding its existence), nor, finally, through some
mediation between both dimensions, which would only unite in itself their
respective imperfections. We then see that, in this respect, the Kantian theory
of knowledge is caught up in an insoluble self-contradiction . . . Kant kicked
out merely conceptual speculation through the front door only to let it in
again through the back door. Instead of objective poetizing (poetizing about
God and the world), on which he rightly issued a judgment of condemnation,
he promoted subjective poetizing. (Beneke 1832: 32–33)
Once the notion of synthetic a priori knowledge is rejected as subjective
poetizing, and the genuine spirit of Kantianism is declared to be its original
24 andrea staiti
proximity to the natural sciences and the awareness of their limits, Beneke’s
task is to redefine what the pars construens of a philosophy inspired by Kant
should be. He undertakes this in his next major philosophical work: Die
Philosophie in ihrem Verhältnisse zur Erfahrung, zur Spekulation und zum
Leben Dargestellt (1833). Here we find articulated an idea that characterizes
the whole Neo-Kantian movement for the rest of the century and carries
over to early twentieth-century phenomenology: “Philosophy should be
the highest science, the science of sciences . . . Philosophy has absolutely no
single object; its object is all or nothing, it is the totality, the whole in its
highest unity” (Beneke 1833: 2–3). Interestingly, Beneke’s main argument
for the necessity of a science of the whole (Ganzheitswissenschaft) is driven
by primarily humanistic concerns. While he admits that the division of
labor is a sine qua non for scientific progress, he warns that “from such
division of labor flow a number of pernicious influences for the formation
(Bildung) of human beings” (Beneke 1833: 3). People educated exclusively
in a specialized science are bound to have an extremely narrow and
fragmented conception of themselves and the world. Beneke continues:
“Therefore, beside and above such sciences dealing with specific objects
there has to be a universal science that surveys the whole from a higher
standpoint and regulates the entire intellectual activity” (Beneke 1833: 4).
Note that the kind of “work on the totality” Beneke envisions for philo-
sophy is fundamentally different from the “work on the totality” envi-
sioned and carried out by, say, Hegel. Philosophy for Beneke does not
work on the totality as the sum total of each and every segment of reality. It
does not set out to systematize and weave together in a metaphysical
narrative all the (partial) truths pertaining to this or that specific discipline.
Each discipline works independently and empirically on its allotted seg-
ment of reality. Philosophy surveys and evaluates the whole as a whole, that
is, it sets out to determine the notion of objectivity in general (prior to its
formulation as biological, physical, psychological, or mathematical objec-
tivity) and it asks about the knowability of the world in general (prior to its
specification as knowledge of living things, physical phenomena, mental
states, numbers, etc.). Moreover, philosophy problematizes and assesses
the partitions of reality carried out in the specialized sciences; it tests the
legitimacy of the disciplinary boundaries drawn in empirical research, and
it evaluates its methods. By way of surveying the world as a whole prior to
its subdivision into separate research fields, philosophy is in a position to
ask questions that empirical research is bound to ignore, and to uphold
Kant’s critical vigilance concerning the intrinsic tendency of specialized
research to take a particular field of inquiry for the whole world.
At this point, readers familiar with Kant’s First Critique may be tempted
to raise an eyebrow. By introducing the notion of “the world as a whole” as
the proper subject matter of philosophy isn’t Beneke reverting precisely to
what Kant dismissed as objective poetizing? Didn’t Kant definitively
establish that “the world as a whole” is not an object of possible experience?
In the Antinomy of Pure Reason, Kant pointed out that the “concept of the
world-whole” rests on the idea of an “absolute totality in the synthesis of
appearances” (Kant 1998: 460 [A 408]) which, however, is never given in
actual experience. We always experience this or that particular appearance,
but we never have the total synthesis of all appearances given within one
single experience, so that all theoretical judgments regarding the world-
whole as the total synthesis of appearances are bound to remain unwar-
ranted. How can a philosopher trained in the Kantian tradition then assign
to philosophy the task of theorizing about the world as a whole?
Beneke, to whose determination of philosophy we shall return in a later
section of this chapter, does not directly address this issue; however, if we
look more closely at the kinds of theoretical tasks that Beneke assigns to a
philosophy of the world-whole, we recognize that his notion of such a
philosophy differs from Kant’s. For Kant the questions asked by rational
cosmology (the metaphysical discipline dealing with the world as a whole)
set out to establish factual truths about the total synthesis of factual
appearances. For example, rational cosmology asks whether the world
has a beginning in time, whether it is made of simple elements, and
whether it admits of causality out of freedom over and above natural
causality. The answers to these questions admittedly exceed the power of
the understanding by Kantian standards. However, there is no reason to
reject the possibility of a different perspective on the world as a whole, one
that does not attempt to determine theoretically the world-whole as a fact,
but rather considers the world-whole in terms of what we could provision-
ally call its ontological constituents. From this perspective we would not
ask, say, whether the world-whole has a beginning in time, but rather, for
example, whether we can legitimately construe the totality of what is as
merely physical or we have to admit of other dimensions of reality. In order
to adjudicate on this question, we would not need to have a total synthesis
of all factual appearances given in intuition. It would be enough to show
that if we construe the world-whole as consisting exclusively of physical
appearances we would leave out of consideration an entire domain of
appearances that are de facto acknowledged and successfully investigated
in some existing scientific discipline. To stay with this example, if we
construe the world as consisting exclusively of physical appearances we
26 andrea staiti
arbitrarily rule out the kinds of phenomena studied in psychology, mathe-
matics, and the social sciences. We make a metaphysical move that
obliterates important elements of our experience and significant results in
scientific research. In this regard, to mention a popular example,
Rutherford’s dictum that “all science is either physics or stamp collecting”
is not the pronouncement of a serious physicist working within the
theoretical limits of his discipline but the arbitrary “objective poetizing”
of a rather unsophisticated metaphysician. One specific disciplinary point
of view is illegitimately extended to the totality of knowledge, and one
delimited field of inquiry is taken to be the world-whole.
Although, as we saw, Beneke can be credited for establishing a new task
for philosophy continuing the Kantian tradition and taking seriously the
positivistic challenge, we have to fast-forward almost one full century to
find an explicit reflection on the conditions and methods of philosophy as
the theoretical investigation of the world-whole. It is especially Heinrich
Rickert who supplements the practice of philosophy in the Neo-Kantian
spirit with a sustained reflection on the notion of world-whole. In his last
book Grundprobleme der Philosophie (1934), Rickert points out that “with
respect to its content, the concept of world-whole must remain undeter-
mined at first.” This is because philosophy cannot rely on a simple
conceptual determination to delimit its subject matter like the specialized
sciences do. Every determination is a negation, as Spinoza taught; there-
fore, to determine the world-whole by a simple conceptual definition
would mean to demarcate it from something that it is not, which would
contradict precisely its status of “totality.” Rickert suggests that this funda-
mental difficulty confronting philosophy in the very definition of its task is
the cause of the disunity that historically characterizes the discipline.
Philosophers tend to overlook the conceptual problems connected to the
definition of a science of the whole. They resolve to focus on this or that
particular sector of reality and treat it as if it were the key to revealing the
essence of the whole. A philosophical definition and investigation of the
world-whole as such, however, is unavoidable. The very fact that we are able
to recognize the fields of inquiry of specialized disciplines as parts or
segments of the world logically requires a definition of the whole of
which they are parts. Parts and whole are logically interdependent
concepts.
Rickert’s way out of this impasse draws on what he terms the hetero-
logical principle, one of the tenets of his epistemology and the key to
systematic thinking in general (as opposed to mere system-building
conceptual speculation). First, we have to recognize that the world-whole
as it is given to us in sheer intuition (the Kantian Anschauung) is an
unmanageable, “unsurveyable multiplicity” (Rickert 1934: 39) or, as he
calls this in earlier writings, a “heterogeneous continuum” (Rickert 1962:
34). In the Heraclitean flux of our sensory experience, there are no set
boundaries (everything is continuous with everything else) and no intrin-
sic similarities (everything is different from everything else). If we are to
cognize the whole as such, we have to introduce order and distinction to
it with the aid of conceptual thought. This already holds true for the
cognition of individual objects and classes of objects studied in the
specialized sciences. We have to single out a discrete historical event
(say, the French Revolution) or a discrete natural phenomenon (say,
linear motion) from the immeasurable multiplicity of reality in order to
be able to produce knowledge in history and in natural science. This can
only happen to the extent that we employ conceptual criteria of selection
to organize the heterogeneous material given by the senses in terms of a
historical individual event (individualizing concept formation) or a
general concept encompassing a multiplicity of individuals (generalizing
concept formation).4
In light of these remarks we can now pose the problem of cognition of
the world-whole. It is clear that we cannot isolate the world-whole as if it
were a large-scale individual event or a lawfully regulated phenomenon
against the backdrop of some further material that we decide to exclude.
To continue with the above example, while we can isolate the French
Revolution against the backdrop of historical reality at large or linear
motion against the backdrop of all further properties and qualities of
moving things, no such operation is possible for the world-whole. No
single concept (“French Revolution,” “linear motion”) can ever isolate
and determine the world-whole. Rickert points out, however, that we can
exploit the logical properties of negation and set out to determine the
world-whole with the aid of a pair of mutually exclusive, complementary
concepts. This procedure is what he terms heterology. In Rickert’s
words:
We only need to ask ourselves whether the part of the world-whole
determined through negation (the part which alone is accessible directly to
cognition) could not perhaps be “completed” [ergänzt], in the literal sense of
the word, through the concept of another part of the world. (Rickert 1934: 41)
4
For a thorough discussion of individualizing and generalizing concept formation, respectively, in the
historical and the natural sciences see The Limits of Concept Formation in Natural Science (Rickert
1986).
28 andrea staiti
If we determine a historical event as, say, the French Revolution we are
implicitly negating the rest of history. We are implicitly saying that it is not
any other such event. However, if we try to think about what is “not the
French Revolution” we are at a loss to determine positively the content of
our thought. The concept “French Revolution” does not identify a part of
the whole whose negation results in an intelligible complementary part
completing the whole at issue. However, this is not the case with every
concept:
Wherever we can construct concepts that, through their determinations
(i.e., their limitations), do not merely negate, but they simultaneously
determine positively the negated component that is missing from the
whole, we can think in terms of alternatives. In so doing we are able to
grasp conceptually, through a kind of detour, a whole that cannot be
grasped through concepts in a direct fashion. This is because we can
construct two concepts, under which every element belonging to the
whole necessarily falls. (Rickert 1934: 41)
Heterological pairs of concepts can be employed to determine conceptually
subordinated wholes of reality, such as, say, “the physical world as a
whole.” If we build the concept “living thing” and proceed to negate it
we get as a result the positive concept “inert matter.” We can then see with
evidence that every conceivable entity in the physical world falls under
either one concept or the other. The two concepts taken together thus
circumscribe and determine the physical world as a whole. The same
heterological strategy can be employed to address “the world as a whole”
as the proper object of philosophy:
What remains to do in philosophy, then, is to identify the pairs of concepts
that represent world-alternatives (Weltalternativen) in the above sense, and
thus make us sure that everything in the world falls either under one or the
other of the two concepts . . . Put briefly, as soon as we know the
alternatives that can be considered for the conceptual articulation of
the world as a whole we thereby acquire an all-encompassing knowledge
of the world. (Rickert 1934: 41–42)
Before we proceed to examine two heterological pairs of concepts that lend
themselves to Rickert’s purpose, however, we have to face a challenge
brought to Rickert from another group of sui generis Neo-Kantians. The
challenge comes from Georg Simmel and Wilhelm Dilthey and it
discredits the very project of philosophizing about the world as a whole
in a theoretically oriented spirit and relying exclusively on the power of
conceptual thought.
Weltanschauung and science
In the foregoing section, I drew on Rickert’s last published work
Grundprobleme der Philosophie (1934) to articulate his understanding of
philosophy as the theoretical discipline dealing with the world as a whole.
However, Rickert had been reflecting on these problems and promoting
this view from the very beginning of his philosophical career in the late
nineteenth century. In earlier texts he characterizes the task of philosophy
as the production of a “comprehensive theory of Weltanschauung” (Rickert
1986: 12). The introduction of this term in late nineteenth-century
philosophy was both decisive and controversial. Rickert’s choice of the
term Weltanschauung to designate the task of philosophy is deeply rooted
in a philosophical dispute that I shall attempt to reconstruct in a few
brushstrokes before we can move on to examine the specifics of his
philosophical proposal.
Etymologically, Weltanschauung means literally worldview; however, the
English translation does not retain the philosophical ring of the term
Anschauung. In Kant’s philosophy Anschauung, or intuition, is opposed to
discursive knowledge. Anschauung does not and cannot grant knowledge. It
merely provides the material for conceptual subsumption. Taken literally,
then, the term Weltanschauung alludes to a pre-discursive dimension of our
experience, in which the world is somehow available to us intuitively but not
cognized. This is the conception of worldview holding sway among the
so-called life-philosophers, such as Wilhelm Dilthey and Georg Simmel.
Both thinkers were predominantly interested in the dynamics of cultural life
as studied by the human sciences and they were filled with the ‘historical
sense’ characteristic of the nineteenth century as a whole. They advocated a
reform of philosophy guided by a consideration of historical life, or what
Dilthey came to designate as a critique of historical reason.
Dilthey is arguably the most vocal and influential proponent of philo-
sophy as worldview. He points out that an unbiased “application of
historical consciousness to philosophy and its history” inevitably brings
to light a tension, or “antinomy,” between the claims to universal truth of
philosophical systems and their historical multiplicity (Dilthey 1960: 7).
Such antinomy can only be “dissolved” (1960: 9) by way of tracing back
philosophical systems to the historical life out of which they emerged.
Their claims to timeless validity must be sacrificed, in order for their true
meanings to become manifest. All appearances notwithstanding, the
source of philosophical thinking is not purely theoretical. The conceptual
constructions characterizing philosophical thinking tend to present
30 andrea staiti
themselves as impersonal theoretical truths; however, on closer inspection
they prove to be ultimately rooted in the personality and historical
situation of the philosophers who produced them.
Georg Simmel presents a similar view in his essay Hauptprobleme der
Philosophie (1910), which was so successful as to be reprinted six times in less
than twenty years after its first appearance (Simmel 1996: 7–151). In one of
the opening sections of the essay, Simmel defines the philosopher as the
individual “who possesses the receptive and reactive organ for the totality of
being,” that is, one who is “somehow touched by this totality and responds
intellectually to it” (Simmel 1996: 16, 17). Such intellectual responses, how-
ever, flow from the inner center of the philosopher’s personality, not from
her intellect alone. In this respect, the “native soil of philosophy” is “the
stratum of typical personality within us,” understood as a middle layer
between the factuality of an individual’s bare psychical existence and the
timeless necessity of logical laws and irreal meaning-configurations
(Sinngebilde) (Simmel 1996: 28). Simmel is fleshing out in more definite
terms a version of Fichte’s old adage that the philosophical position one
chooses depends on the kind of person one is. Accordingly, in a short essay
On the History of Philosophy he suggests that each philosophy should be
regarded as a worldview, as “an expression of the existential relationship
between a mind and the cosmos as a whole” (Simmel 1980: 199).
This position bears significant consequences for the conception of truth
in philosophy and the kind of cultural function that philosophy is taken to
fulfill:
What matters in the assertions of philosophy is not the adequacy to an
object – in whatever sense we understand it – but rather their adequate
expression of the philosopher’s own being, of the human type that lives
within the philosopher. (Simmel 1996: 29)
This, of course, is not a way to say that “anything goes” in philosophy. The
expression of a certain personality can be more or less successful, more or
less consistent, and more or less resonant with the broader historical reality
in which it is articulated. However, the core point is that a scientifically
oriented notion of truth does not do justice to the nature of philosophy.
The driving force behind philosophical thinking is not the quest for
scientifically secured results, as it befits the intellect, but the quest for
meaning of the world as a whole, as it befits the human being as a whole.
These considerations stand in sharp contrast to the notion of
Weltanschauung initially employed by Rickert and the Neo-Kantians of
his school. As I pointed out above, in some of his earlier works Rickert
employed the term Weltanschauung to characterize the terminus ad quem of
philosophical thinking, the final achievement of a successful philosophical
inquiry. In the programmatic essay Vom Begriff der Philosophie (1910), for
instance, he insists that philosophy “must produce a Weltanschauung to
shed light on our position in the world as a whole” (Rickert 1999: 4). Note
that for Rickert a worldview must be produced and not merely expressed by
philosophy. Whatever our historical and personal point of departure may
be, when we turn to philosophy we make an effort to think about ourselves
and the world in a purely theoretical fashion, that is, we deliberately choose
to adopt a perspective that transcends the particularity of our contingent
standpoint. Contra Dilthey and Simmel, for Rickert a philosophical world-
view ought to be a thoroughly theoretical construct.
Rickert rejects the premise that an intellectual perspective in philosophy,
because of its one-sidedness, makes it intrinsically impossible to grasp the
world as a whole:
As long as he does not let theory alone decide, that is, as long as he is guided
by the interests stemming from life as a whole, man is interested exclusively
in the world that we could characterize as his own world. The whole human
being, precisely as “whole” or “existing” human being, will never transcend
his “small” world and this small world inevitably takes on different shapes
for different people. (Rickert 1999: 337–338)
The existential pathos and the pressing questions characteristic of life as a
whole, for Rickert, have a restrictive function, rather than a function of
disclosure. They bind us to significant particulars in the world, and they
impede the broadening of perspective that leads one to reflect upon the world
as a whole. In a way that resonates with the Greek notion of the contempla-
tive life, Rickert is arguing that the intellect is the organ of freedom, that
freedom which alone allows our thought to soar over the daily concerns and
chores of our existence: “Therefore, only the one-sided, theoretically oriented
man is able to reflect upon the world in its totality” (Rickert 1999: 337–338).
Rickert is not saying that we should leave out of philosophy those
existential questions and intuitions that pertain to humans as wholes and
often provide their very motivation to philosophize. He does want to say,
however, that such existentially grounded questions and intuitions should
be taken up as part of philosophy’s subject-matter, as opposed to tacitly or
self-avowedly determining the standpoint from which one philosophizes.
Theory does not stand in opposition to life, rather, it is the only stance
from which life as a whole (and correlatively the world as a whole) becomes
visible in the first place.
32 andrea staiti
The subject/object distinction versus the reality/value distinction
How are we to tackle the philosophical determination of the world as a
whole? Early representatives of the Neo-Kantian movement turned to
consciousness and its investigation. Beneke, for example, after determining
philosophy as the “all-embracing, universal science,” insists that this defi-
nition does not rule out the possibility “that philosophy may have a
particularly proximate object or a central point [Mittelpunkt] from which
to survey the whole” (Beneke 1833: 10). This particularly proximate object
is “our own self-consciousness” (Beneke 1833: 11) and “the fundamental task
of all healthy philosophy is the analysis of human consciousness” (Beneke
1832: 49). Riehl subscribes to the same view when he states: “Philosophy is
the theory of consciousness” (Riehl 1872: 27).
The definition of philosophy as the systematic study or analysis of
human consciousness is certainly not new or distinctive of Neo-
Kantianism. We can find earlier versions of this view in the work of the
British empiricists, who are undoubtedly a major source of inspiration for
figures such as Beneke and Riehl. However, the early Neo-Kantians’
appeal to consciousness as the doorway to philosophical understanding
of the world as a whole poses a problem that did not trouble Locke or
Hume: what is the relationship between philosophy and psychology as a
specialized empirical science investigating a mere part of reality, namely,
the human psyche? In Locke’s and Hume’s times psychology did not yet
claim for itself the status of an independent, empirical discipline.
However, by the middle of the nineteenth century, and especially after
the foundation of Wundt’s laboratory for experimental psychology in
Leipzig (1879), philosophy could no longer articulate itself as a
Bewusstseinslehre without somehow clarifying its relationship with
empirical psychology. Isn’t it redundant to have two separate disciplines
dealing with basically the same subject-matter?
Riehl attempts to shed light on this matter in a way that sounds more
like a temporary compromise than a solution:
The difference between the philosophical disciplines in a narrow sense and
psychology consists in this: the philosophical disciplines tackle the content or
the objects of consciousness independently of their origin and independently
of their characteristic of being thought. Psychology, on the contrary, pursues
precisely the question about those processes out of which the content of
consciousness crops up. The philosophical disciplines deal with psychic
products, whereas psychology is the science of psychic processes. Obviously,
if we possessed a satisfactorily worked out psychology, its position in the
system of the philosophical sciences would equal the position of physics in the
system of the natural sciences. Such psychology would be at the same time the
foundation and a part of the philosophical sciences. (Riehl 1872: 31)
A few lines below, Riehl even suggests that the difference between
philosophy and psychology might be substantive only for us, that is from
our limited perspective, and not pertain to the subject matter itself. After
all, the distinction between the processes and the products of consciousness
does not seem to have a fundamentum in re, but rather in our organization
of the universe of knowledge.
The main problem with Beneke’s and Riehl’s perspective is that it is
unclear whether it is really compatible with the notion of philosophy as
the universal science of the world as a whole. First, while it is hardly
deniable that whatever we take to exist must in some way or other relate
back to consciousness, it is not obvious that consciousness grants a
privileged access to the world as a whole and is therefore the mandatory
object of philosophy. Couldn’t we find our way to raise questions about
the world as a whole from within the investigation of other dimensions of
reality, say, history or physics? Second, even if we accept the primacy of
consciousness and the centrality of psychology, do we have sufficient
reasons to believe that the picture of the world-whole offered by this
framework is accurate?
Similar questions led Rickert to revise significantly the early Neo-
Kantians’ understanding of philosophy and to redefine the status of the
world-whole from a new perspective. In his aforementioned manifesto,
Vom Begriff der Philosophie, he sets out to examine the tenability of the
worldview that a philosophy like Beneke’s and Riehl’s necessarily under-
writes. It is a worldview revolving around the subject/object dichotomy
and counterposing consciousness to the rest of reality. To use the language
of heterology, we could say that such a perspective seeks to grasp the world
as a whole with the aid of the conceptual pair “subject” and “object.”
Accordingly, it considers the two dimensions of psychic and physical being
as exhaustive and complementary. This picture of the world-whole, how-
ever, focuses exclusively on reality (Wirklichkeit) and it leaves completely
out of consideration another domain, that of value (Wert), which alone
provides the resources to account for the intelligibility of things and yields
the insights fundamental to the articulation of the meaning of “the world
and our place in it.” Rickert explains:
Besides realities there are values, whose validity we want to understand.
Only these two realms taken together constitute that which deserves
the name “world.” Thereby it is important to notice that values, which
34 andrea staiti
we thus counterpose to realities, are not to be regarded as being themselves
realities. (Rickert 1999: 13)
The notion of value constitutes the hinge of Rickert’s philosophy and it
deserves closer scrutiny. In order to approach it we must first direct our
attention to a “third kind of beings” (Rickert 1934: 78) that are integral to
our experience as much as psychic and physical realities. Let us think, for
instance, of the meanings of words. While having a conversation, we do
not only perceive the physical body of our interlocutor and the psychic
reality attached to it. We also “receive” in the same, direct, immediate way
the meanings of the words that he or she is uttering. These meanings are
not psychic realities. They do not coincide with the thoughts of the speaker
or with the acoustic stimuli of the listener. The very same meanings could
be conveyed by a completely different sensible support, such as written
words on paper. We can even recognize the same meaning as being
instantiated by different words in different languages, such as the identical
meaning of “red,” “rot,” “rosso,” and “rouge.”
Word-meanings for Rickert belong in a sphere of “being” distinct from
that of psychic and physical occurrences, both of which fall under the
broad category of reality. He labels this sphere of “being”that of verstehbare
Sinngebilde, that is, understandable or intelligible meaning-configurations.
The reference is explicitly to Plato’s distinction between “aisthetón” and
“noetón” (Rickert 1934: 81); however, for Rickert, intelligible being does not
exist in a metaphysical otherworld. It is a fundamental ingredient of
precisely the concrete earthly world that we experience, in which besides
physical and psychic entities we undeniably encounter entities like word-
meanings.
Rickert hastens to emphasize that word-meanings are only a tiny
fraction of the intelligible meaning-configurations that we encounter in
experience. “Besides them, there are also many other meaning-
configurations that we likewise do not perceive but we understand”
(Rickert 1934: 81). We can think here of the aesthetic meaning of a
symphony, the moral or political meaning that shines through a certain
gesture, or simply the meaning-configuration displayed by the arrange-
ment of the furniture in our living room. In all these cases, for Rickert, we
must acknowledge that in the concrete content of our experience there is
more than meets the senses. Physical or psychic realities are the “carriers”
of intelligible meaning-configurations (Rickert 1934: 80).
The fact that meaning-configurations are intelligible does not mean that
they are automatically or necessarily understood. If we stay with the
simplest example of word-meanings, we can see that while single words
have meaning per se (i.e. they can be understood), not all possible combi-
nations of word-meanings result, in turn, in intelligible meaning-
configurations. Moreover, even when a combination of word-meanings
is intelligible, like in a grammatically well-formed sentence, this does not
mean that what the sentence expresses is valid. Consider the difference
between these three sentences: (1) “two plus foremost likes or”; (2) “two
plus two equals five”; (3) “two plus two equals four.” The cluster of singly
intelligible meaning-configurations in (1) does not amount to one
overarching intelligible meaning-configuration. Contrariwise, both the
meaning-configurations in (2) and in (3) add up to one intelligible overall
meaning-configuration. We can understand both (2) and (3). Nonetheless,
there is an undeniable difference between (2) and (3). The meaning-
configuration in (3) is not only intelligible, it is also valid, and more
specifically it falls into the distinctive sphere of validity of theoretical truth.
Values for Rickert are the non-empirical “entities” to which validity
attaches and by reference to which we can orient ourselves in the mundum
intelligibilis of non-sensible meaning-configurations. Values cannot be
further explained by reference to other phenomena of meaning or reality.
They are fundamental ingredients of our experience as much as the
meaning-configurations, whose validity they determine. The “validity” of
values is what we have to somehow presuppose if we are to make sense of
our capacity to discriminate between valid and invalid clusters of meaning-
configurations. Rickert is adamant that values can only be experienced on
the basis of things that we hold as valuable (goods) and within concrete acts
of valuation. Goods and acts of valuation are the “venue” where the
domains of reality and value intersect, in the medium of some intelligible
cluster of meaning-configurations or other. However, considered per se,
values are part of the “third realm” of unreal meaning-configurations, of
which they constitute the underlying principles.
What Rickert has in mind with the notion of Geltung, or validity, as the
definitive characteristic of values and just what it means to recognize a
value as valid is by no means self-evident. Anglophone readers can best
approach the meaning of Geltung in German by reflecting on the experi-
ence associated with the word “compelling.” An argument can be compel-
ling; a painting can be compelling; a great singer’s voice can be compelling;
a moral purpose can be deemed compelling; etc. Of course the way in
which an argument is compelling is different from the way in which
someone’s voice is compelling. But there is something common to those
experiences, as marked by the use of the same word. In all these cases,
36 andrea staiti
Rickert would say, we are exposed to values. There is something that we
experience as transcending our own mental acts, revealing itself through
the object at issue (argument, painting, song, etc.) and exerting a compel-
ling force on us. Needless to say, we can decide to stray away. We can refuse
to acknowledge and conform to these compelling forces. We can more or
less deliberately think illogically, despise aesthetic beauty, or reject moral
intuitions. In Rickert’s narrative, however, in so doing we are straying from
the compelling force that a given value is exerting on our mental acts
through an experienced state-of-affairs.
The possibility of experiencing “compelling-ness” in very different
spheres of experience, ranging from aesthetic beauty to logical strin-
gency, should allow us to broaden the ordinary meaning of the terms
“valid/invalid” and to avoid thinking of validity as an exclusively theo-
retical or logical phenomenon. The validity of a moral value can shine
through a certain kind of behavior, and the validity of an aesthetic value
can pervade a musical melody. At an even more elementary level, the play
of our visual experiences happens in a “medium of validity” that we call
reality. This is why they manifest to us objects that we are compelled to
call “existent.”
For Rickert, then, the fundamental heterological pair for investigating
the world as a whole is “reality/value.” Whatever is, is either real (physical
or psychic) or valid (a meaning-configuration displaying some relation to
a value and carried by a real existent, be it physical or psychic). This, of
course, does not end the task of philosophy. Once this fundamental
heterology is established, a broad set of tasks opens up for philosophy.
First, the acknowledgment of values as a fundamental ingredient of
reality introduces a new standpoint from which to address the traditional
controversies between mechanism and vitalism in the philosophy of
nature, and between associationists and gestaltists in the philosophy of
psychology. Both living organisms and complex mental states (Gestalten)
can be considered as wholes irreducible to the sum total of their parts
because their unity is grounded in a value-related configuration of mean-
ing. Second, a lot of further work would be required to articulate a
“system of all values,” which Rickert deems “the only means toward
systematic knowledge of the universe of meaning-configurations”
(Rickert 1934: 91).5
5
For a presentation of Rickert’s system of values and a comprehensive discussion of his philosophy see
my entry “Heinrich Rickert,” in Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy
(Winter 2013 edition), http://plato.stanford.edu/archives/win2013/entries/heinrich-rickert/.
Conclusion
Is there anything we can learn from the Neo-Kantians regarding the
current predicament of philosophy? Can we at least understand better
the nature of the uneasiness expressed by philosophers like Kitcher and
McNaughton?
First, the reason why the linguistic turn feels so unbearably narrow is
because it arbitrarily excludes from philosophical consideration a large
number of perfectly intelligible meaning-configurations. It restricts
investigation to the sphere of linguistic meaning-configurations and
treats all other meaning-configurations as “metaphysical” and therefore
unintelligible. To be sure, this allowed philosophy to make significant
progress in the understanding of language, but it distorted the way we
look at other spheres of meaning, such as aesthetic, ethical, and political
meaning. There is no reason to impose on philosophy such limitation or
to stipulate that we first have to transpose intelligible meaning-
configurations into linguistic meaning-configurations in order to address
them. For instance, we do not need to reduce the experience of ethical
meaning-configurations to the language in which we express ethical
claims. We can study both the language of ethical claims and the
ethical meaning-configurations that we directly encounter in experience.
Moreover, the fact that we use language to articulate ethical claims
entirely presupposes the givenness of intelligible meaning-configurations
revolving around ethical values. If we ignore these latter, the study of the
former is bound to become uninteresting.
Second, upon what conditions can philosophy “enable us both to
understand our place in the world and to live in it”?6 I believe that
Rickert’s “ontological pluralism” is the sine qua non for this purpose
(Rickert 1934: 54–55). As long as philosophy produces only endless
variations of physicalist reductionism, the chances that it will be able
to enlighten us regarding our place in the world and how to live in it are
scant at best. This is simply because the world in which we live includes
precisely us, that is, psychic subjects experiencing meanings and values
as well as physical things. There is no need to be a Cartesian dualist
to admit of psychic reality in the world, and there is no need to be a
metaphysical Platonist to admit of objective meanings and values in
our experience. But to reduce “the world as a whole” to a play
of sub-particles, consciousness to an epiphenomenon of complex
6
McNaughton, “Why Is So Much Philosophy So Tedious?,” 5.
38 andrea staiti
aggregates of sub-particles, and meanings and values to evolutionary
tricks of nature, means to alienate philosophy precisely from our world,
which pays little attention to reductionism and continues to be an
ontologically pluralistic world, populated by psychic, physical, and
meaningful entities.
chapter 2
Neo-Kantian ideas of history

Alan Kim
Introduction
The two main groupings of German neo-Kantians, the Baden or
Southwest School of Windelband and Rickert, and the Marburg School
of Cohen, Natorp, and Cassirer, see history as a central philosophical
problem. For Windelband and Rickert, it is a “logical” problem of the
unity of human knowledge or science (Wissenschaft). On the face of it, the
natural sciences (Naturwissenschaften) obviously differ in subject matter,
method, and style from the sciences of human history and culture, the
so-called Geisteswissenschaften or sciences of the spirit (Geist). But how do
they differ? Are the natural sciences to be taken as models for the
Geisteswissenschaften? If so, how can the latter accommodate themselves
to the natural-scientific form, since their subject matter clearly cannot be
reduced to nature without at the same time destroying the very notion of
a distinct branch of knowledge concerned with spirit or mind? The
Southwest School sees these unresolved questions regarding an important
and undeniably real epistemic activity – historiography – as indicative of
a theoretical crisis that the sciences themselves lack the resources or
perspective to resolve. Only philosophical critique can hope to elucidate
the so-called logic of the historical or cultural sciences, and thus clearly
define both the respective subject matter and methodology of the Natur-
and Geisteswissenschaften. In this way, a critique of the “fact” of the
historical sciences can expose the rational organization of all branches
of human knowledge, of Wissenschaft as a whole. This “critical”enterprise
is simply a continuation of Kant’s project as we see it stated in the
Prolegomena, where he addresses such questions as “How is pure science
of nature possible?” Rickert and Windelband now ask, in effect: “How is
the science of history possible?”
By contrast, the Marburg School approaches history in a way at once
more oblique and profound than Windelband and Rickert’s. The “direct”
39
40 alan kim
path of unearthing the transcendental conditions of possibility of the fact
of (the science of) history is not open to Cohen and Natorp, because they
do not regard history as a science comparable to the natural sciences. The
exact natural sciences remain, from Cohen through Cassirer, the paradig-
matic forms of knowledge, in which the characteristic activity of thinking
can be most clearly observed. History enters the Marburgers’ work not as a
factum of intellectual or academic reality, but rather as an intrinsic part of
analyzing and bringing to reflective clarity the activity of thinking, for this
activity unfolds over time. Thinking, reasoning – whether scientific or
ethico-political – can never be grasped timelessly, but only in and through
its concrete, empirical, and thus historical manifestations. This is why I call
the Marburg approach deeper: history is not considered a scientific
discipline independent of philosophy, but as immanent to the execution
of philosophy’s task, viz., bringing thought’s thinking to self-reflective
clarity. In what way, then, is their historical philosophizing Kantian, and
not, say, Hegelian? I shall return to this question below, but the answer lies
in the Marburgers’ idiosyncratic interpretation of Kantian idealism itself,
and the centrality of activity, spontaneity, and freedom in that interpreta-
tion. In nuce, history is the account we give ourselves of the spontaneous
actions we human beings have taken over time, in both the ethico-political
and theoretical spheres, which very spheres have their genesis in those
actions.
The Southwest School

Neo-Kantianism arose in reaction to the empiricist spirit of its age. Over
the course of the nineteenth century, the natural sciences continued the
great strides they had been making since the Renaissance, and now the
historical disciplines could also look back upon decades of groundbreaking
work in historiography, documentary analysis, archeology, epigraphy,
textual criticism, and the scientific study of ancient languages, as well as,
at a remove, theology, law, anthropology and the nascent disciplines of
psychology and sociology.1 These latter disciplines came to be grouped
under the heading of Geisteswissenschaften, that is, sciences (Wissenschaften)
of the spirit (or mind: Geist). In part due to the great success of the
empirical natural sciences, but also because of their own advances on an
empirical footing, theoreticians of the Geisteswissenschaften saw, in
1
See C. Bambach, Heidegger, Dilthey, and the Crisis of Historicism (Ithaca and London: Cornell
University Press, 1995), 43.
Neo-Kantian ideas of history 41
positivism’s primary attention to empirically accessible and controllable
facts, the foundation of any geisteswissenschaftlich methodology worthy
of the name “scientific.” Further, it led some to argue that the
Geisteswissenschaften should seek their foundation in psychology, which,
at least according to its name, seemed to be the science of spirit par
excellence; psychology thus promised to play a role analogous to that of
mechanics in the physical sciences.
The Southwest School saw the historical sciences’ positivist orientation
towards the natural sciences as misguided, and the source of great confu-
sion in science (Wissenschaft) as a whole, that is, the edifice of all knowledge
(Wissen, Erkenntnis) both natural and historical. The root of this confusion
Windelband and Rickert identified in the very term Geisteswissenschaft and
its suggestion of a domain – that of spirit – ontologically distinct from
nature, and hence subject to investigation by its own method. Just this
dubious metaphysical presupposition, Windelband and Rickert argue, is
responsible for the mistaken view that psychology can somehow found the
practice of the historical sciences, as I discuss below. Instead, they say, we
should look to the actual aims and practices of historians, to the fact of
history as a branch of knowledge, and from this starting point reconstruct
the logic of history as a science. Only in this way can we hope to avoid, on
the one hand, introducing alien and inappropriate notions from the
natural sciences, or, on the other hand, concluding in despair that the
historical disciplines are not science, that is, any kind of knowledge at all.
No: they accept the factum of historical practice, and recognize in it a
genuine project of knowing that answers to a basic rational interest. And
because it, no less than Naturwissenschaft, is a project of reason, they
assume that critique must ultimately succeed in showing the coherence
of the natural and historical sciences.
Windelband tells us that if we compare the natural and historical
sciences, we will notice two things. First, the natural sciences seek general
laws under which “individual” phenomena in the empirical world may be
subsumed and thus explained. By contrast, historians in their writings
bring out particular, individual phenomena for special attention and
treatment. Their goal is to revivify certain significant bygone events,
individuals, nations, or other cultural “formations” (Gebilde), and thus
allow the reader somehow to experience these formations again (nacherle-
ben). Second, as a matter of practice, historians make very little use of the
results of psychology. These two observations are importantly related, for
psychology is not interested in individual phenomena as such, but like the
natural sciences, wants to explain psychic phenomena by discovering
42 alan kim
general patterns or psychological laws. In fact, Windelband and Rickert
argue, psychology (in particular, experimental psychology) is, in both
matter and method, a natural science. We should therefore not be
surprised that historians make no use of it, or think that it could ever
play the role of a “mechanics” of the Geisteswissenschaften.
Indeed, Windelband argues, the unimportance of psychology to history
demonstrates the inaptness of the very term Geisteswissenschaft. The histor-
ical sciences do not study some non-natural domain of being called psychē
or Geist, but rather deal with the very same ontological realm as the natural
sciences do, to wit, empirical reality. It is rather their perspective, that is,
the purpose with which they approach the world that distinguishes them
from the natural sciences. Windelband famously characterizes this differ-
ence in aim by the nomothetic–idiographic distinction. The natural
sciences, in seeking to discover (or as a Kantian would say, to legislate)
the laws of nature, are nomo-thetic or law-laying sciences; the historical
sciences on the other hand seek to “write the peculiar,” and hence are
idio-graphic.2 The natural scientist ascends beyond the particular to a realm
of timeless generality and validity (Geltung), whereas the historical scientist
aims to bring out a bygone particular Gestalt as it once existed in time.
In Rickert, Windelband’s basic insight finely ramifies.3 Like the
latter, Rickert rejects the division of the sciences into Natur- and
Geisteswissenschaften, and argues for relegating psychology to the natural
sciences, thereby clearing the way for an unambiguous theory of historical-
scientific methodology. But he also rejects Windelband’s nomothetic-
idiographic distinction, and proposes in its stead the division of science
into the Kulturwissenschaften and the Naturwissenschaften, the sciences of
culture and the sciences of nature. One of the problems with Windelband’s
dichotomy is that “nomothetic” and “idiographic” lack a common basis for
comparison. Rickert’s proposal introduces such a common term, namely
“value,” with respect to which the disciplines can be contrasted on a formal
or logical basis: the method of the historical sciences is value-relative,
whereas that of the natural sciences is value-neutral.4
Although Rickert dispenses with the nomothetic–idiographic dichotomy
as a classificatory rubric, his own nature–culture opposition nevertheless
2
“Idiographic” is not a neologism, although the sense Windelband gives the term is new: idiographo
means “written in one’s own hand,” i.e., an “autograph.”
3
Bambach, Heidegger, Dilthey, and the Crisis of Historicism, 91ff.
4
Max Weber, Rickert’s colleague and friend at Heidelberg, adopts the term Kulturwissenschaft.
However, Rickert’s identification of value as the distinguishing characteristic of culture is the source
of considerable friction between the two.
implies it at the level of description. For, as he agrees with Windelband, the
natural sciences aim to explain phenomena as such by bringing them under
general concepts or laws, whereas the historical sciences pursue the vivid
Darstellung or presentation of significant phenomena in their individuality.
It is just because they are value-neutral that the natural sciences can view all
phenomena simply “as such,” none more special than another, and so lay
down the laws that must apply to all without exception. By contrast, culture
is the domain of those phenomena that, as cultural, are interpreted with
respect to values, goods, and purposes pursued by human beings.
Consequently, these function as criteria for selecting particular phenomena
from the infinite manifold of events and things that have ever occurred or
existed, so that in “writing” just these phenomena, the so-called cultural
sciences in effect proceed “idiographically.”
The philosopher seeking to bring order to the sciences now faces the
following challenge: to clarify the historian’s (tacit) aim of representing
bygone events of significance in their unique peculiarity. As a criticist
eschewing metaphysical presuppositions, Rickert looks to the actual praxis
of historians and to their products, histories. These aim to give selective
accounts of the past, focusing especially on significant formations in the
realms of politics, war, religion, the arts, science, law, or philosophy itself.
In other words, the historian concerns himself with phenomena (events,
figures, movements, nations) within the main regions of human culture,
and therefore treats them as cultural, and not with respect to their brute
appearance in the natural order. “Value” makes itself first felt in just these
general areas of human striving and action, for politics, the arts, law, etc.
are all constituted with respect to principles considered valuable by their
practitioners. The arts aim at beauty or originality; war at victory; religion
at right worship of the divine; science at truth; law at justice; and so forth.
The particular interpretation of the guiding principle is not at issue here,
just the fact that whatever the principle may be, it is always a value
demanding a sense of obligation that we all recognize, even if we refuse
to acquiesce in that demand. As Aristotle says at the outset of the
Nicomachean Ethics, every human activity aims at some good, which is
just to say: culture is constituted by value.
Again, within each cultural sphere, value provides the criterion for
selecting and darstellen this rather than that poem, author, general, artist,
battle, or temple; for we are interested in the decisive battles, the best
artists, the most aesthetically complete or representative works of art or
architecture. The writer of history, then, seeks somehow to present the
bygone, uniquely important individual human being, event, artifact in its
44 alan kim
significance (Bedeutung), that is, to bring out just those unique qualities
that make it stand out from the countless unimportant, forgettable things
that have happened and passed away (Rickert 1986: 71). And for this
presentation to signify scientifically and be more than a myth or story,
the historian must find a way of conceptualizing the individual in its
peculiarity. It is this task that presents Rickert with his greatest challenge,
viz., to explain the nature and possibility of the individuating – as opposed
to the generalizing – concept.
The task is daunting in that the very term, individuating concept, seems
self-contradictory. Concepts, Begriffe, it is generally thought, are literally
“graspings-together” collecting individuals under one heading on the basis
of certain common features (Rickert 1926: 97).5 Such groupings then allow
us to understand the individual as an instance of this or that genus. Hence
the very notion of Begriff would seem inescapably to involve abstraction
and generalization, the main modes of natural-scientific concept for-
mation. As Rickert writes, if there is no uniquely historical-cultural,
individuating mode of concept-formation, then there can also be no
historical science distinct from the natural sciences. A Wissenschaftslehre
(doctrine of science) must therefore address the possibility of this latter
type of concept, or the epistemic status of history will remain unclear.6
It must be said that in his clearest and shortest treatment of this topic,
Kulturwissenschaft und Naturwissenschaft (Cultural Science and Natural
Science), which Rickert considered an introduction to his magnum opus,
Die Grenzen der naturwissenschaftlichen Begriffsbildung: eine logische
Einleitung in die historischen Wissenschaften (The Limits of Natural
Scientific Concept Formation: A Logical Introduction to the Historical
Sciences), he struggles mightily to explicate the notion of individuating
concept formation. What follows here is but an attempted interpreta-
tion, one that has to contend with additional obscurities raised by
Rickert’s statement that the individuating concept is not to be crudely
opposed to the generalizing concept. Rather, the individuating process of
historical representation cannot take place without resorting to general
conceptual elements, which latter must be combined to produce “a
concept of historical individuality” (Rickert 1926: 66, 70). Further,
as he stresses, the individuating and generalizing modes of concept
formation are but idealized extremes. Natural science often deals with
5
So Kant: a Begriff “[bezieht sich] mittelbar [auf den Gegenstand] vermittelst eines Merkmals was
mehreren Dingen gemein sein kann” (Kant 1998 [A320/B377]).
6
See Bambach, Heidegger, Dilthey, and the Crisis of Historicism, 100ff. Bambach does not grapple with
the difficulties inherent in the very notion of an “individuating concept”; see below.
individual phenomena – say, the Sun, Moon, the eruption of
Mt. Tambora, the radiometric dating of an australopithecine’s skeleton –
just as historians cannot avoid making reference to general, recurring
phenomena of the natural world. Hence, Rickert seems repeatedly to
undermine his chief purpose, the clear “logical” distinction between the
general concepts of natural science and the individual concepts of
cultural science or history, by emphasizing the ways in which they
“factically” involve each other.
The solution to this conundrum lies, I propose, in an ambiguity in the
German usage of the term, Begriff. When Rickert opposes “individuating”
and “generalizing” concepts, these are not two species of a common genus,
“concept”; rather, “concept” itself means two different things in each case.
As I mentioned above, in the paradigmatic natural scientific case, Begriffe
are formed by abstraction from individuals, retaining common character-
istics; alternatively, natural scientific Begriffe may include the laws of
nature, which, again, are abstract and so can be applied to all individuals
“in common.” The natural scientific Begriff just means that by which all
the individuals can be understood or grasped together; when they are so
grasped, through a rule, we know them in the sense of erkennen (savoir). By
contrast, in the individuating case, Begriff must mean the grasp of this here
individual, holding just it clearly before the mind’s eye, knowing it in the
sense of kennen (connaître). This opposition of erkennen and kennen is
supported by Rickert’s contrast of humanity’s natural scientific interest in
Erkenntnis with its equally valid rational and complementary interest in
kennenlernen something particular, i.e., getting to know it, becoming
acquainted with it (Rickert 1926: 78–79).
Thus, when Rickert speaks of individuating concepts, he is using Begriff
in the way it is used in the phrase, sich einen Begriff von etwas machen –
literally: “to make [for] oneself a concept (of something)” – which does not
mean abstracting from the individual to the general, but just the opposite,
namely grasping it, the individual. Now of course this latter grasping of the
individual is done by means of Begriffe in the former sense, that is, with
words or other, usually pre- or non-scientific, concepts. As Rickert says, we
must use “general elements” in order to present an individual as it is in its
uniqueness (Rickert 1926: 55, 60, passim).This presentation of the indivi-
dual, its Darstellung, that is produced through the use of general terms or
Begriffe in the traditional sense – this presentation now just is the Begriff
in the new, individuating sense: a grasp of, and hence corresponding to,
just one person, event, etc. In short, the biographer aims not, per impossi-
bile, to list every last word and deed of Socrates, Hume, or Florence
46 alan kim
Nightingale, but to capture, as we say, the essence of the person, and this
capture, if successful, is the individual Begriff.
The logical (or, as we might say, the methodological) determination of
the cultural sciences in terms of their distinctive mode of individuating
concept formation allows Rickert, he thinks, to construct a harmonious
ordo scientiarum. Rejecting the Natur–Geist dichotomy, he is committed
to ontological monism: there is but one Wirklichkeit, one nature. But this
single reality is phenomenologically bifurcated by value. That is, a certain
subset of phenomena within reality interests us in a peculiar way that the
rest do not, in that they are “meaningful,” “significant,” or “valuable” – as
in some way making a claim on us as human beings. These are the
phenomena whose generation and corruption, while never violating the
laws of nature, cannot be grasped in naturalistic terms, but only with
reference to values. Although reason has an essential interest in making
nature transparent by discovering the laws governing the concatenation
of phenomena, it cannot remain indifferent to culture. For culture just is
the sum-total (Inbegriff ) of deeds done in the course of (someone)
promoting the good and valuable (as she sees it), which goods or values
are for their part never accessible to the senses but only to reason. In other
words, reason looks upon culture as the product of rationality and so in
principle explicable in and on its own terms (nature, by contrast, is
essentially other and “in itself”). Precisely this explication is the task of
the cultural sciences, which take as their objects the activities, achieve-
ments, persons, or groupings thereof that are scientifically, rationally
significant, i.e., that strike reason as essentially homogeneous with itself
(Rickert 1926: 55).
These objects of course include the familiar political, social, and
religious structures that are the topics of the vast majority of historical
writings, but also natural science itself, for this last is also a particular,
highly formalized cultural activity governed by rigid adherence to a value,
namely Truth. Thus, there is and can be no conflict between the natural
and cultural sciences regarding their objects or material, because although
cultural things considered solely as spatiotemporal are natural and hence
subject to the laws of nature, the cultural scientist does not consider them
as merely spatiotemporal, but rather in light of their significance, which is
just what the natural scientist must – as an investigator of nature as the
realm of Sein, not Geltung – abstain from. Similarly, just as the natural and
cultural scientists do not tread on each other’s toes in their respective
objective regions of inquiry, they also avoid conflict in their methods,
since the natural scientist proceeds by generalizing to bring individuals
unter Begriffe and so zu Begriff, whereas the cultural scientist seeks instead
to capture the individuality of the individual by bringing it auf den Begriff.
Hence, on Rickert’s view, the very question whether cultural science
ought to mimic natural science by adapting or adopting the latter’s
methods is nonsensical and betrays a fundamental misunderstanding of
their respective ends. Natural scientific method can never achieve the aim
of cultural science because it is running in the opposite direction. By the
same token, those who would reduce the cultural sciences to the natural
sciences err in their blindness to the cultural scientists’ ineradicable con-
cern with value. Finally, those who would deny the Kulturwissenschaften
the right to be called sciences precisely because they are concerned with
value, whereas science should proceed in a disinterested, value-neutral
way – those critics fall prey to a misunderstanding regarding the relation
of both cultural and natural sciences to values.
On this last point, Rickert’s distinction between Wertung (evaluation)
and Wert-Beziehung (value-reference) is especially important for under-
standing the relative positions of both the natural and cultural sciences in
the ordo scientiarum. As we have seen, Rickert thinks that historical science
can only pursue its proper goal of vivid individuation by using values as
criteria for raising the culturally significant object out of the infinite
manifold of the past. But as I pointed out above, the values that make
this individuation possible need not be values that the historian himself
endorses: he can ascertain in a perfectly disinterested way the virtues and
values of a society of Caribbean cannibals or British imperialists, in order to
understand their motives, even as he personally finds these values wrong
and the actions done for their sake repellent. In fact, Rickert stresses that in
his historical practice, the historian must conscientiously resist evaluating
(werten) bygone cultures if he is to succeed in the scientific task of referring
those cultures to their values (wert-beziehen). In this respect, the cultural-
historical scientist is just as value-neutral as the natural scientist. On the
other hand, despite their necessary abstention from evaluation as a
methodological principle, neither the cultural nor the natural sciences
stand above the realm of value, since both are themselves cultural activities
and achievements, and therefore only explicable with respect to certain
values like Truth (Rickert 1926: 142). Just because science as such is con-
stituted by reference to the value of Truth, it must in practice proceed in a
value-neutral fashion (Rickert 1926: 18, 140). For the natural sciences, this
means studiously ignoring the very notion of value (including utility); for
the cultural sciences, it means considering cultural values without passing
judgment on them, so as not to distort our grasp of the past.
48 alan kim
The Marburg School
The Marburg School, mainly oriented towards the natural sciences, do not
elaborate a theory or logic of history or historiography comparable to the
Southwest School’s; they seem not to consider the discipline of historical
writing to be a science at all. Nevertheless, they are not indifferent to
history, and indeed, history plays a more intimate, immanent role in their
work than it does for Windelband or Rickert.7 The Marburgers are, on this
point at least, no less the heirs of Hegel than of Kant, for they combine the
latter’s notion of the free spontaneity of reason with a Hegelian account of
the dynamic development of that free reason through time. Natorp, for
one, considers philosophy to be the reflection of reason upon the condi-
tions of possibility of its own activity. In reflecting upon the history of
science, reason looks upon itself at its best, and recognizes in scientific
activity not only its own spontaneous nature, but also its long struggle
to liberate that spontaneity from the constant threat of passive, dogmatic
self-satisfaction. In other words: on the one hand, science8 is reason’s
cognition of the world, its painstaking construction of the laws of
experience; on the other hand, history is reason’s self-reflection on its
greatest success, viz. the development of that very same science. Let us
now turn to details.
The Marburgers rarely use the term “reason” (Vernunft), as it appears in
Kant alongside intuition, the understanding, judgment, and will. Instead,
Natorp for example speaks of “thinking” (denken): thinking is judging, and
intuition is ultimately to be reduced to thinking.9 Hence, from among the
Kantian powers of mind, only the understanding and the will remain as
titles for our theoretical and practical powers. Now, when Natorp seeks a
reflective starting point in the fact of scientific thinking, i.e., the cognition
of objective truth, he remarks the following: science does not find
ready-made objects that it then organizes and classifies in genus-species
7
Bambach, Heidegger, Dilthey, and the Crisis of Historicism, 59. I disagree, however, with Bambach’s
statement that the Southwesterners “rob history of its vital, experiential core, leaving only the
desiccated husks of an abstract theory of method” (ibid.). As we have seen, they restrict themselves
to the methodological underpinning of history, which, for its part, they see as focused precisely on
vital experience.
8
In this section, I will stop qualifying science with the terms natural or historical, something that was
necessary to avoid confusion in the discussion of the Southwest School. Unless otherwise noted, in
this section on the Marburgers, science will mean the exact or the natural, not the historical sciences.
9
A. Kim, Plato in Germany: Kant – Natorp – Heidegger (Sankt Augustin: Academia Verlag, 2010), 85ff.
Cassirer, too, distances himself from the term “reason” as “a very inadequate term with which to
comprehend the forms of man’s cultural life in all their richness and variety”; he prefers “symbol”:
“instead of defining man as an animal rationale, we should define him as an animal symbolicum”
(Cassirer 2006: 31).
schemas; objects are not given (gegeben) to mind. Rather, what is given is
an indeterminate phenomenal manifold that calls for determination, i.e.,
demands the construction of determinate objects in and out of that
manifold. The Marburgers encapsulate this view in the dictum that objects
are not gegeben, but aufgegeben, not given as a gift (Gabe), but “given up” to
us as a task (Aufgabe) (Natorp 1981: 262).
On the Marburg view, the object is a task, not an already completely
determined thing, but something to be determined; so, too, the subject
cannot be the independent, passive recipient of the thing. Rather, the
subject, or better, the “subjective” is the act of objective determination.
Its cognitive structure determined by the objects the subject constructs,
object and subject develop in a double helix of mutual constitution. Thus,
at a stage at which the object is less determined, the subject, too, is
correspondingly more diffuse and vague, for as a “thinking-(of)-the-
object” it is not yet thinking anything in particular. As it begins the work
of thinking the object, this activity progressively determines not only the
object, but also, simultaneously, the activity’s own origin, the subject.
Hence, if the fully determinate object is an ideal goal of thinking, then so
too is the fully determinate subject. What is – right now and forever – is the
activity of objectivation, on the way towards the object.
Thinking, then, is an activity of determination, with the goal of con-
structing an object from the chaos of the given. Cohen and Natorp call this
activity “hypothesis” in its literal sense, viz. a “laying,” “positing,” “setting”
(thesis) “under” (hypo-). Thinking lays or sets down a conceptual infra-
structure by which the phenomenal manifold is structured and becomes
intelligible. Insofar as these concepts are the “hypotheses” set down in
thinking, they can be called in German Gesetze (the “set [down things]”),
or Grundlegungen (the “[things] laid [down] at the foundation [or
ground]”): in short, the groundwork of possible experience. Further,
since as groundwork they make experience possible in the first place, and
so are not given in experience, these hypotheses are a priori. Thus, whatever
these concepts ultimately turn out to be, they play the functional role of
Kantian categories, but with an important difference to be mentioned
below.
As thinking constructs its conceptual framework, things come into focus
for it – they are the proto-objects, the correlates of the active subject
responsible for their generation, and against which they come to stand,
as the clay takes form in the potter’s hands. Except this analogy needs
qualification: the person throwing the clay is, as it were, as yet merely an
aspiring potter, a potter-coming-to-be at the same time as the clay is
50 alan kim
coming-to-be-a-pot. Similarly, the thinking origin is forming the object of
experience, and in so doing, itself coming-to-be-a-subject. And just as the
clay thrower comes to be a potter by making and overcoming errors, by
throwing more and more pots, so too thinking finds itself continuously
ensnared in contradiction and inconsistency, both with other of its own
already laid-down hypotheses, but also, more importantly, with the very
empirical world it has generated by determining the manifold (Natorp
1981: 256). Thinking is therefore not to be imagined as an effortless progress
towards the determination of an object in a cosmos of objects, but as a
struggle against several opponents simultaneously: the manifold’s own
indeterminacy; the constructions of other thinkers; and one’s own mis-
constructions that manifest themselves in self-contradiction.
Now this picture of a rationality determines the Marburg conception of
history as a problem and task. For one thing, they call into question the
positivist, empiricist presupposition, in history as elsewhere, of the basic
epistemic unit, the fact. The Latin factum is the neuter past participle of
the verb facere “to make, to do”; a factum is therefore a “done deed,”
something complete. This means that there is nothing for me, the knower,
to contribute to it. “It is what it is,” and all I can do with a fact is take it – as
given. But, as the Marburgers say, no thing is simply given, not even a fact.
Indeed, the fact as “done deed ” always implies antecedent activity. For the
Marburgers, however, the chief point is this: facts are never merely given to
the knower. Rather, the knower constructs them, at least initially, through
a process of discrimination, of collection, and division. There are no pure
data or givens; these are always generated and sifted, and so already
construed by a thinker in advance. Hence, if, as Natorp says, the scientific
object is only ever an X, a something-to-be-determined (Natorp 1981), then
facts, too, as objects and ultimate elements of cognition (Erkenntnis) must
dissolve from the hard and fast elementary particles of science into infi-
nitely contestable and revisable goals of knowledge, both natural and
historical. The factum, as Natorp says, is better understood as a fieri, that
is, not as a deed, over and done with, but an ongoing becoming, the
happening of thinking. Thus, paradoxically, the bygone events of historical
interest cannot be understood as over-and-done-with “facts” simply to be
gathered up and analyzed by the historian.
Instead, the Marburg vision of rationality as the spontaneous legislator
of experience casts history itself in a new and unfamiliar light. For, as I said,
the theoretical construction of experience is a dialectical struggle among
rival hypotheses, not a straight march to truth – just as the ethical emerges
out of a struggle between reason and inclination, and social justice is never
granted but must be won. It is this struggle that constitutes history, for
individuals, nations, cultures, and humanity as a whole. Thus, when the
thinker considers himself as a thinker, he cannot help but think back upon
the train of dialectical point and counterpoint, of (hypo-)thesis and antith-
esis, by which he has arrived at his present understanding of the world and
of himself, of object and subject. Hence, history as a rational, cognitive
enterprise appears not as science (the rigorous dialectical pursuit of the laws
of nature), but as the equal, if not superior10 rational interest a thinker has
in self-knowledge. For I can only become transparent to myself as a
thinking and willing being, as a free, legislating subject, by tracking back
and forth along my path from ignorance and bondage. “I am” not merely
an autonomous legislator, but the one who has had to achieve this status.
Hence just those present traces of the past are historical that, like
footprints, indicate someone having walked. Although the tracker must
first discover the footprints, he is not interested in them as such, but in the
walker they betray – or better, in her walking. So too the historian must
from his own cultural position strive to resolve it again into the activity that
gave rise to it; and so on, ever back. By this process of unraveling, he
discovers ever-deeper “conditions” of his own cultural stage, the very stage,
that is, that makes possible this self-reflection and investigation. He finds –
whether tracking reason’s theoretical or practical achievements – a repeated
cycle of phases: captivity in an unexamined matrix; first insight into that
state of captivity and crisis; ascent towards the light, along with its count-
less setbacks; and finally: emergence into a new, relatively freer condition,
but one that immediately begins to harden and call forth new effort. Only
these actions are the deeds and hence the facts relevant to my history of my
self, my nation, of humanity as a whole. History now appears not simply as
a chronicle or as annals, an amorphous collection of the “precipitates of
life” (Cassirer 2006: 198).11 Rather, for the Marburgers, it is (or should be)
the freely recounted account by free beings of their liberation.
In this manner, history for the Marburg School in general means the
account of reason’s past free activity, its bygone hypotheses, its former acts
of legislation. But it also means representing these actions in their dialec-
tical nexus, in the particularities of the different struggles conditioned by
constantly varying cultural contexts of fait accomplis (Cassirer 2006: 50).
In his work on Plato, for example, Natorp tries to show, on the one hand,
10
On the superiority of history as the realm of the will over nature as the realm of thinking, as well as a
critique of the materialistic notion of history, see Cohen (1981: 426).
11
Even chronicles or annals are never just “data dumps,” as I argued.
52 alan kim
the birth of “critical idealism,” that is, of reason’s recognition of its task of
clarifying to the (mathematical) sciences the logical ground and autonomy
of their method (viz., hypothesis), as well as showing them their limits
(e.g., their inability to reflect upon their own epistemic principles). On the
other hand, Natorp also takes pains to describe the resistance this critical
idealism immediately had to face in what he calls Aristotle’s empirical
dogmatism, the common-sense view that objects and things are simply
given to us, and that scientific conceptualization is not lawful construction
but rather mere arrangement or analysis (Natorp 1994: 385). In point of
fact, Natorp, here following Cohen, sees in Aristotelianism a centuries-
long retreat of the scientific spirit and the silencing of the philosophical
criticism that alone could reassert reason’s creative freedom.
Above I said that whereas the sciences comprise the cognition of the
objective world, history is the recognition of subjective activity by this
activity itself, i.e., by reason (Cassirer 2006: 11). Put less carefully but more
clearly: science is reason’s knowledge of the world; history is reason’s
knowledge of itself. The two are neither opposed to nor separate from
one another, since reason’s historical self-reflection occurs within the
mirror of its (bygone) scientific thinking, in which it bears witness to its
own free activity. More than that, for the Marburgers, thinking, whether
scientific or historical, whether outwardly or inwardly directed, is always
qua rational the “lawful” structuring of a phenomenal manifold. When
reflecting upon itself, then, it is the bygone activities and their contingent
contexts that now make up the “phenomena” out of which reason must
construct the “X ” that is (was) itself (Cassirer 2000: 142). This is why, just
as reason cannot simply intuit the cosmic order, but must laboriously
construct and reconstruct it, it also cannot intuit itself: as I said, the subject
is as much an infinite project as is the object.
Therefore, while biographical, political, or social facts are not of histor-
ical interest as such, some of these disjecta membra will still be necessary to
furnish the transient phenomenal material in which reason can glimpse its
own eternal form, that is, its own constant hypothesizing, objectivating,
legislating activity. Cassirer’s biographical writings, such as Kants Leben
und Lehre, or Rousseau, Kant and Goethe, exemplify this attitude towards
the history of philosophy no less, and perhaps more self-consciously than
Natorp’s Platos Ideenlehre. Natorp in the foreword to the Ideenlehre gruffly
brushes aside all talk of Plato’s personality as irrelevant, vowing instead to
attend solely to the Sache or matter at hand, namely showing Plato’s theory
of forms as the first appearance of the critical idealism described above
(Natorp 1994: vii). Cassirer, too, is sparing in details from Rousseau’s
tempestuous life, contrasting, for example, his wild passions with Kant’s
famous equanimity, all with the purpose of showing the interpretive blind
alleys previous intellectual historians have entered, captivated by just such
superficial facts. Looking beyond historiographical commonplaces to the
thought and texts themselves, Cassirer uncovers the idea that flows from
Rousseau to Kant: human autonomy. Let us briefly examine these two
examples of Marburg philosophical historiography.
I have dealt with the substance of Natorp’s interpretation of Plato else-
where, so I want here to bring out those aspects that illuminate Marburg
historiography. Plato is of central importance to the Marburgers not because
he was faktisch one of the first philosophers, but because his dialogues, as
philosophical, for the first time self-consciously realize the dynamic structure
of thinking. Each dialogue can be read as both a miniature critique and
history of thinking and knowing. Especially the Socratic dialogues are
critical-historical in the Marburg sense, as vignettes of the putative knower
coming to self-awareness of himself as a thinker, and suddenly realizing
thought’s mobility. In the aporetic dialogues, the interlocutor invariably
finds himself mired in cramped self-contradiction. Socrates sets thought into
motion, releasing the interlocutor into the freedom of dialectic. This newly
achieved perspective-in-motion allows the interlocutor to look back and
reflect upon himself as he had been all along without knowing it, to perceive
both his previous ignorance and his dogmatic blindness to that very
ignorance. It is the latter perception that is, as it were, a backward-looking
advance in self-knowledge.
Socrates’ success depends on the insight that a thought expressed in a
definition of piety or virtue is never more than a hypothesis or starting
point for dialectical examination. It is therefore never a conclusion to be
accepted as dogma, which is how people usually and habitually treat their
own opinions, that is, as unhistorical and immune to change. In other
words, Cohen and Natorp’s interpretation of Plato is historical chiefly in
an immanent rather than an external sense. That is to say: they interpret
the ideai (forms) as dynamic hypotheses, and thus as laws inherent in
dialectic, rather than as its transcendent goals. The ideai are the temporary
stepping-stones of thinking-in-motion, not immutable Dinge (Natorp
1994: 74, passim). The defensibility of this notorious interpretation is not
at issue here. Rather, what I want to bring out is how this reading of the
forms as hypotheses ipso facto reveals both Plato’s theory of forms and
rationality itself as historical. Plato is therefore historically important
because his work for the first time reveals the historicity of reason and
thus represents the birth of a new, self-conscious, critical rationality.
54 alan kim
Hence, what matters for Natorp, and, he thinks, for Plato, too, is not the
external historical dimension of the unrepeatability of the peculiar and
particular, but to the contrary, the necessary dialectical unfolding
of concepts placed under relentless interrogative pressure. And such
unfolding is precisely not unique and unrepeatable, but is what every
comprehending reader of the dialogues can and must repeat in his own
train of thought, even millennia later.12
In Rousseau, Kant and Goethe, Cassirer goes beyond Cohen and Natorp’s
focus on the logical spontaneity of thinking, to bring out the perhaps more
perspicuous manifestation of autonomy, the will. I want to distinguish
several intertwined conceptions and praxes of history and historiography in
the first essay of that book, “Rousseau and Kant.” In the first place, more
than any other Marburger, Cassirer comes closer here to traditional histor-
iography, in seeming to return the political to its place as the focus of
historical concern, for Rousseau made his greatest impression upon Kant
with his account of the social contract, and its roots in individual auton-
omy. However, this contact with traditional political history is tangential,
serving only to accentuate the radical difference between Marburg history
and history as commonly conceived.
As Cassirer points out, Rousseau’s deduction of an originary social
contract in the second Discourse is ambiguous, wavering between what
Rousseau himself calls a hypothetical or conjectural mode of history, on
the one hand, and, on the other, long stretches of an apparently empirical
account of prehistoric anthropology. According to Cassirer, Kant resolves
this methodological confusion by clearly separating the ideal and the
empirical: Kant does not claim empirical historicity for the social contract,
proclaiming it rather to be the constitution implicit in the foundation of
any rationally justifiable political order. Thus, on one level, Cassirer’s
historiography highlights a hidden spiritual kinship between Rousseau
and Kant, by looking beyond biographical or psychological facts to
their unanimity on the question of human autonomy. On a deeper level,
it is just this point – the essential freedom of the human being – that then
casts its light back upon the project of historiography itself, by showing,
e.g., that the notion of a state in which every individual’s autonomy is
equally respected and promoted, is the speculative ideal in light of which
all political history can and must be written.
12
It is important to avoid misunderstanding here. In the logic of hypothesis that is thinking, there is a
“before” and “after,” a “following from” what is “pre-supposed.” Considered purely in the context of
inference and implication, this priority is logical, not temporal. Nor does even the thinking of these
thoughts and relations necessarily follow a logical order in time.
Thus, again, we see the deep bond unifying Marburg historiographical
practice with its object, the mind itself, in its theoretical as well as, now, its
practical activity. This unity, unsurprisingly, reflects a deep affinity with
Kant’s philosophy – indeed, I would argue, a much deeper one than is
evident either in Rickert’s external theory of the historical sciences, which
hesitates to intervene in the fact of their practice, or in Windelband’s
intellectual-historical work, which safely stays within the bounds of con-
ventional interpretation. By contrast, the Marburgers work in the spirit of
Kant’s own historical thought, as we can discern it in the First and Third
Critiques, as well as, of course, in such explicitly historical essays as
the “Idea for a Universal History with a Cosmopolitan Intent” and the
“Speculative Beginning of Human History.” What Kant and the
Marburgers as historians share is an ecstatic orientation towards the future
(Cassirer 2006: 59; 1951: 214–215).
When Kant distinguishes in the First Critique between concepts of the
understanding (the Categories) and concepts of reason (the Ideas), he says
that the former are the a priori concepts in accordance with which all
possible experience is constructed, whereas the latter are the a priori
concepts in accordance with which that experience is formed or guided.
In other words, while the categories are the conditions of possibility of the
coherent nexus of phenomena we call experience, these by themselves are
insufficient for accounting for our sense of what experience is of. Here
reason must posit an Idea (in particular, the cosmological Idea, ‘World’)
conceived as the unconditioned totality of which our experience lets us
only ever glimpse some conditioned spatio-temporal segment (Kant 1998
[A508/B536] ff.). It is this Idea of a cosmos, fully determined as a whole and
down to its most minuscule portion, that guides reason’s scientific
enterprise, the methodically purified construction of the laws of the
phenomenal experience produced through the operation of the categories
upon an intuitional manifold (Kant 1998 [A668/B696]; Kant 1950: §56).
Only because reason presupposes the Idea of a cosmos in its totality –
which in principle can never be an object of experience – is it motivated to
explore and progressively determine this structure in a way that asympto-
tically realizes it as a cosmos, i.e., a whole in which every element stands in
lawful, harmonious relation to every other.
This Idea or ideal alone is what underlies and makes explicable the
history of science as a unified account of a single activity – science – despite
its arguments and paradigm shifts: for all these now come into focus as
moments of the dialectical purification of reason’s grasp of nature’s laws,
gradually hypothesizing (legislating) those same laws in accordance with
56 alan kim
the ideal law of lawfulness, of harmonious, cosmic regularity. The history
of science is the account of these dialectical stages. The only temporal
consideration that the historian therefore needs to take into account is the
sequential order of the dialectic, much as the history of a chess game can be
represented algebraically without concern for the time taken for each
move, much less for the various psychological states of the players or
their personalities. And the past stages are comprehensible as stages in
light of the ideal goal, just as the moves of a chess game are comprehensible
as moves only in light of victory, the key difference being that the absolute
knowledge of the cosmos is, as a rational ideal, unattainable, whereas
victory in chess is really possible. It is this atemporal yet nonetheless
historical idea that allows us to discern progress in our theoretical grasp
of the natural world, and which allows the Marburgers both to write the
history of bygone progress as well as to project that progress into an
indefinite future.
In Kant’s practical philosophy, we find similar moral and political ideals
governing the interpretation of bygone moral and political reality, viz. the
ethico-political concepts of the Kingdom of Ends, or the condition of
perpetual peace among a federation of free states. Here again, the ideal is
placed in an unattainable “future,” yet is intimately and immanently
involved in the moral and political history of humanity. That is, we can
only even identify the bygone phenomena relevant to a history of morality
(or immorality), or to a history of politics, in light of an ideal of Freedom.
As Kant says, morality as such is inconceivable without the presupposition
of the freedom of the will. Likewise, as Cassirer argues, Rousseau showed
Kant that the political is only conceivable on the basis of a notion of
autonomy and equality. Thus it is only in light of certain ideals that the
indefinite manifold of “past phenomena” takes shape (Cassirer 2006: 58)13
and stands before us as historically significant, viz. as a connected sequence
of intellectual and moral decisions, guided by a rule. What the historical
investigation of the most ancient such decisions shows, however, is that
this rule or governing ideal was at first but obscurely present to our
ancestors groping towards freedom. The free decisions by which they did
advance were often hindered or reversed, just as has been the case, on
Natorp’s view, in science.
13
Goethe, Cassirer writes, “wanted [in Dichtung und Wahrheit, his autobiography] to discover and
describe the truth about his life; but this truth could only be found by giving to the isolated and
dispersed facts of his life a poetical, that is, a symbolic, shape.” As in autobiography, so in history,
insofar as, as I have been suggesting, Marburg history is the “autobiography” of reason, or, in
Cassirer’s terms, reason’s reflective “symbol” (ibid.).
It is for this reason, in fine, that philosophy is necessary, and why it
necessarily takes on a historical guise. It is necessary as the critical
delimitation of the conditions of reason’s activity, chief among which,
as we have seen, are the ideals that give any sense to that activity –
practical or theoretical – in the first place. Then again, this critique
must needs be historical, in that these ideals can only be discerned and
made clear by examining the past activity in which they “must have been”
all along tacitly presupposed as telē or ends. It is in this light, then, that we
should understand the Marburgers’ philosophical historiography. In
their interpretations of Plato, Kant, Rousseau, Goethe, the thinkers of
the Renaissance and Enlightenment, but also, in the case of Cohen, of
biblical antiquity, the Marburgers always try to reveal the struggle of
the mind to recognize itself in its truth, and so make us so recognize
ourselves – viz. in each case as free activity and not a thing, a fieri, not a
factum – indeed, as just the kind of being that, having recognized itself
in its true active nature, can now only tell itself this kind of story, i.e. a
critical-idealist history that in form is also true to its subject. And since
historical thinking, like all thinking, is “a constructive act” of “interpre-
tation and reinterpretation,” the historian’s task of cross-examining our
past will never end (Cassirer 2006: 199).
Conclusion
In this chapter, I have shown in what sense both the Southwest and
Marburg Schools represent genuinely Kantian, yet very distinct, con-
ceptions of and approaches to the problem of history. Briefly, where the
Southwesterners offer a critique of the historical sciences, the
Marburgers write history as an essential part of their critique of reason
as dialectical activity. For both schools, historiography is essentially a
practice of self-knowledge, the coherent exposition of the development
of the significant structures we call culture. But our exploration has also
revealed fissures and contradictions. The Southwest School’s historio-
graphical practice, in its focus on the particular, falls back despite itself
into historicism and psychologism at odds with its idealistic, Kantian
inspiration. This is, I would argue, precisely because it takes for
granted, as the factum to be transcendentally grounded, historio-
graphical praxes not yet radically, critically clarified, seeking to fit this
factum into a scientific order. By contrast, the Marburgers offer, as I
called it, an idealistic history of idealism, much more in the spirit of
Kant’s own historical essays, filling out the last section of the First
58 alan kim
Critique: “The History of Pure Reason.”14 Instead of accepting the
alleged science of history as practiced, and then justifying it by a new
kind of concept-formation, they directly challenge that practice by
reconceiving history as the confessional autobiography of reason. I
have brought out the contrast between the two schools; its underlying
reason – their conflicting notions of Begriff – will have to await
treatment elsewhere.
14
“Die Geschichte der reinen Vernunft” (Kant 1998: [A852]).
chapter 3
Neo-Kantianism and analytic philosophy

Hans-Johann Glock
Introduction
Until recently, it used to be a commonplace that contemporary philosophy
has little to gain from Neo-Kantianism. Its early opponents were largely
responsible for introducing the term; the motto “Back to Kant!” was seen as
evidence of intellectual backwardness and barrenness.1 From the nineteenth
century, the received story goes, the only philosophers that have stood the
test of time are figures from outside academic philosophy, such as
Schopenhauer, Marx, Kierkegaard, Nietzsche, and Frege. Even in compar-
ison to German idealism, Neo-Kantianism was neglected.
To be sure, since the 1980s there has been a palpable renaissance of
interest. Ironically, this renaissance has tended to reinforce rather than
diminish the impression that the Neo-Kantians lacked originality and
philosophical clout. It is often accompanied by the qualification that
they are of interest mainly for historical reasons. To make matters worse,
often the history in question is not even the history of philosophy or of
ideas, but political and social history, notably, the way in which the
movement reflected the move from democratic to nationalistic liberalism
in nineteenth-century Germany and the baleful influence of anti-Semitism
in nineteenth-century German universities.2 One rarely reads that Neo-
Kantianism has contributed to substantive philosophical debates of current
concern.
The suggestion that Neo-Kantianism is relevant to analytic philosophy
may appear even more far-fetched. Both Neo-Kantianism and analytic
philosophy are often regarded as clearly circumscribed and paradigmatic
philosophical movements. Moreover, the origins of both are standardly
1
See H. Sluga, Gottlob Frege (London: Routledge, 1980) and H. Holzhey, “Neukantianismus,” in
Historisches Wörterbuch der Philosophie, vol. vi (Darmstadt: WBG, 1984).
2
See Holzhey, “Neukantianismus” and M. Pascher, Einfuhrung in den Neukantianismus: Kontext,
Grundpositionen, praktische Philosophie (Munich: Fink, 1997), 7–12.
59
60 hans-johann glock
located in the second half of the nineteenth century. According to those
interested in the history of philosophy, Neo-Kantianism and analytic
philosophy do not just stand in tension to one another; they seem to
occupy opposite ends of the philosophical spectrum. Neo-Kantianism
has the reputation of being a highly conservative, perhaps even reactionary,
academic mainstream that was replaced by vigorous revolutionary move-
ments like analytic philosophy.
In addition to very real contrasts in doctrine, method and style, there are
also important points of agreement and influence. These concern Frege,
Wittgenstein, and some logical positivists (Schlick, Reichenbach, Carnap).
In particular, there is the crucial Kantian idea that philosophy is not part of
the natural or social sciences, but a second-order discipline that reflects on
the conceptual and methodological framework of science.
My contribution starts out by clarifying propaedeutic questions con-
cerning criteria of philosophical influence and conceptions of analytic
philosophy and Neo-Kantianism. I then explore positive and negative
relations between the two traditions, paying equal attention to historical
and substantive issues. While stressing the common ground, I shall place in
perspective some of the more ambitious claims concerning the impact of
Kantian and Neo-Kantian influences on analytic philosophy by Gabriel,
Hanna, and Friedman. In this context, I widen the debate beyond the
prevailing focus on the Southwest School of Neo-Kantianism. For reasons
of space, important issues connected to my topic will not be discussed, but
I have addressed them elsewhere: the relation between Kant on the one
hand, Frege, Wittgenstein,3 and Strawson4 on the other, and the analytic
credentials of Nelson.5
Propaedeutic questions
Anyone purporting to scrutinize the historical relation between Neo-
Kantianism and analytic philosophy, and in particular the influence of
the former on the latter, needs to take a stance on three questions.
1. What counts as a philosophically relevant influence? Under what
conditions are we justified in maintaining that one thinker has influ-
enced another?
3
H.-J. Glock (ed.), The Rise of Analytic Philosophy (Oxford: Blackwell, 1997).
4
H.-J. Glock (ed.), Strawson and Kant (Oxford University Press, 2003).
5
H.-J. Glock, “Nelson und die Analytische Philosophie,” in A. Berger, G. Raupach-Strey, and
J. Schroth (eds.), Leonard Nelson: ein früher Denker der Analytischen Philosophie? (Münster:
Lit-Verlag, 2011), 39–70.
Neo-Kantianism and analytic philosophy 61
2. What is analytic philosophy?
3. What is Neo-Kantianism? Who counts as members of the movement?
As regards question 1, influence is primarily a causal notion. But philo-
sophical influence is not simply a case of efficient causation. A cannot
philosophically influence B by administering a drug to her which makes
her accept his pet theory T. It is only if A’s theory figures in the way in
which B reasons, or, less stringently, in the way she thinks about the topic
of T, that A will have influenced her.
Under what conditions are we licensed to ascertain this kind of influ-
ence? Clearly, mere parallels between the ideas of A and the ideas of B do
not suffice. It is for roughly the same reason that Dummett’s ‘history of
thoughts’ rather than ‘history of thinkers’6 is a doxographical rather than
genuinely historical exercise. By contrast, Baker and Hacker insist that a
“genuine causal connectedness” can be established only by showing that B
has noted a particular view in A “and consequently reached such and such
conclusions.”7 By these standards, one can diagnose a genuine influence
only in cases in which B explicitly acknowledged it and we have reasons for
accepting the statement as sincere and accurate. In my view, this is too
demanding. Philosophers are no saints, and some of them deliberately
conceal certain influences on their thinking. Other philosophers show little
concern for lines of influence, and fail to state influences without being in
bad faith.
In view of this situation, I propose the following compromise between
the extremes of mere doxography and philosophical-depth biography. We
are entitled to state that A has influenced B positively if there are clear
affinities and convergences between the ideas of B and those of A and if B
was familiar with the latter through reading or conversation. Replace
“affinities and convergences” by “disagreements and divergences,” and
you get a criterion for negative influence. It ought to be uncontroversial,
furthermore, that positive influence counts for more than negative influ-
ence, inspiration for more than provocation.
How to understand analytic philosophy is somewhat more challenging. In
What is Analytic Philosophy? 8 I have argued at length that analytic philosophy
cannot be defined by reference to: (i) a geographical habitat and linguistic
medium, notably as “Anglo-American”; (ii) a neglect of or anachronistic
attitude towards the history of philosophy; (iii) topics like a preoccupation
6
M. Dummett, Origins of Analytic Philosophy (Cambridge, MA: Harvard University Press, 1993),
especially chapter 1.
7
G. Baker and P. Hacker, Frege: Logical Excavations (Oxford University Press, 1983), 7n.
8
H.-J. Glock, What is Analytic Philosophy? (Cambridge University Press, 2008).
with logic and a neglect of ethics, or doctrines such as hostility to metaphy-
sics or a “linguistic turn”; (iv) methods – e.g. analysis or rational argument –
or styles of writing – for instance a striving for clarity. Instead, analytic
philosophy is best understood by combining historical or genetic concep-
tions with family-resemblances à la Wittgenstein. It is a loose and diverse
tradition held together by ties of influence and overlapping similarities
concerning topics, doctrines, methods, style, etc.
In summary, analytic philosophy is the most influential philosophical
tradition of contemporary Western philosophy. It predominates in
Anglophone countries, and has recently become the single most important
movement in continental Europe. Analytic philosophy emerged around the
turn of the nineteenth century in Frege, Moore, and Russell – work that
mainly concerned formal and philosophical aspects of logic, semantics, and
metaphysics. It then took a linguistic turn through the early Wittgenstein
and the logical positivists. Two distinctive branches emerged in the 1930s:
logical constructionism (“ideal language philosophy,” Frege, Russell,
Carnap, Quine) on the one hand and conceptual analysis on the other
(“ordinary language philosophy,” Moore, Wittgenstein, Ryle, Austin,
Strawson). Instead of a well-circumscribed movement neatly split into two
branches, contemporary analytic philosophy is a highly diverse movement
that continues to ramify. Nonetheless, there are some overlapping doctrinal,
methodological, and stylistic features that have been shared by many –
though not all – analytic philosophers from its inception to the present
day. In addition to the concern with logic, semantics and language, these
include various methods of analysis, a sceptical or even hostile attitude
towards speculative metaphysics, the attempt to emulate science (especially
the formal and natural sciences), if not in its topics and aims then at least in
its intellectual values, the attempt to address philosophical problems by way
of rational argument, and the ambition to write clearly.
Neo-Kantianism
The label Neo-Kantianism (and the German original Neukantianismus)
has also been used in a variety of ways, some of them highly perplexing.
Neo-Kantians can be distinguished from Kantianism both in a narrow and
in a wide sense. In a broad sense, Kantianism comprises all philosophical
thinkers or theories that are substantially indebted to Kant. In the narrow
sense, Kantianism is a movement that began around the turn of the eight-
eenth century and follows immediately in Kant’s footsteps. It comprises
both close followers – many of them friends and pupils of the sage of
Königsberg like Herz and Jäsche – and the so-called “semi-Kantians” like
Beck and Reinhold. The latter built on Kant, while also departing from
him in order to remedy shortcomings of his transcendental philosophy,
notably concerning the relation between appearances and things as they are
in themselves.
Some figures (Beneke and Fries) can be assigned to Kantianism narrowly
understood and to Neo-Kantianism. Neo-Kantianism is a movement of
revival; it aspires to rehabilitate Kant following the rise and decline of
German idealism. At the same time, there is no consensus even among
experts as regards the contours and sub-divisions of Neo-Kantianism. In
his influential treatment, T. K. Österreich distinguished no fewer than
seven branches:9
the Marburg School (Cohen, Natorp, Cassirer)
the Southwest German or Baden School (Windelband, Rickert, Lask)
the realist approach of Alois Riehl
the metaphysical variety (Liebmann, Volkelt)
the physiological strand (Helmholtz, Lange)
the relativist transformation of Kant’s transcendental philosophy (Georg
Simmel)
the psychologistic variant going back to Fries (Nelson, Cornelius).
More recent commentators have radically pruned this list. Pascher confines
his survey to the first three currents.10 Ollig even makes do with the first
two.11 At this point, the pendulum has swung too far in the opposite
direction. Part of the reason is the aforementioned tendency, especially
among Germanophone scholars, to reduce Neo-Kantianism to a distinctive
institutional phenomenon of imperial Germany, a topic for the sociology of
knowledge rather than substantive philosophy or even the history of ideas.
Both Helmholtz and Lange have reinterpreted Kant’s a priori preconditions
of knowledge as innate physiological features. They thereby provided
important stimuli – both positive and negative – for the interpretation and
rehabilitation of Kant. Like the psychologistic variant, especially Fries and
Nelson, they also indirectly influenced and/or anticipated naturalistic
approaches within or associated with analytic philosophy, notably evolu-
tionary epistemology and naturalized epistemology. Otto Liebmann for his
part cannot be written out of this tale, if only because his Kant und die
9
F. Überweg and T. Österreich, Grundriß der Geschichte der Philosophie. Vierter Teil: Die Deutsche
Philosophie des 19. Jahrhunderts und der Gegenwart (Berlin: Mittler, 1923), §§36–42.
10
Pascher, Einführung in den Neukantianismus, 50–51.
11
H. L. Ollig, “Neo-Kantianism,” in E. Craig (ed.), The Routledge Encyclopedia of Philosophy (London:
Routledge, 1998).
Epigonen of 1865 furnished the slogan “Zurück zu Kant!,” which first made
the return to Kant explicit and marked the “official” start of Neo-Kantianism
(Liebmann 1912).
Analytic philosophy, Austrian philosophy, and Kantianism

To contrast Neo-Kantianism with analytic philosophy is therefore not to
lapse into the discredited prejudice that analytic philosophy contrasts with
Germanophone philosophy. According to Dummett, for instance, the
“sources of analytical philosophy were the writings of philosophers who
wrote, principally or exclusively, in the German language,”12 notably
Bolzano, Frege, Wittgenstein, and the Vienna Circle. In a similar vein,
Simons detects an “Anglo-Austrian Analytic Axis”13 which includes Britain
and the former Habsburg empire (especially Austria, Czechoslovakia, and
parts of Poland). According to these commentators, Germanophone phi-
losophy divides into two distinct branches. On the one hand, there is a
German tradition, which derives from Kant, and stretches through the
German idealists and Neo-Kantians to Heidegger; it is marked by obscur-
antism and idealism. On the other hand, there is an Austrian tradition,
which starts with Bolzano, continues with the Brentano school, and
includes the Polish school of metaphysics and logic founded by
Twardowski and Kotarbinski; it is level-headed and realist, having been
partly inspired by British empiricism.
The idea of a distinctive “Austrian philosophy” requires thorough
qualification and correction.14 Leading Habsburg philosophers like
Bolzano and Brentano attacked Kant, sometimes vehemently, even on
issues on which they essentially agreed with him, like the definition of
truth or the analytic/synthetic distinction. Furthermore, the style of writers
like Bolzano, Brentano or Kotarbinski contrasts favourably with that of the
German Neo-Kantians, even when the philosophical pursuits of the latter
were in close proximity to British empiricism and contemporary science.
Nor is there any gainsaying the fact that other pioneers of analytic
philosophy were critical of Kant and Neo-Kantianism, in particular its
idealist streak. After flirtations with Kant and Hegel, Moore and Russell
rebelled against idealism and initiated the complementary programmes of
conceptual and logical analysis. Subsequently, the credo of the most
12
Dummett, Origins of Analytic Philosophy, ix.
13
P. Simons, “The Anglo-Austrian Analytical Axis,” in J. C. Nyiri (ed.), From Bolzano to Wittgenstein:
The Tradition of Austrian Philosophy (Vienna: Hölder-Pichler-Tempsky, 1986), 98–107.
14
Glock, What is Analytic Philosophy?, chapters 2 and 3.
influential school of analytic philosophers, the logical positivists, was the
rejection of Kant’s synthetic a priori judgements. Proponents of Oxford
conceptual analysis frowned upon the system-building that characterized
both Kant and Neo-Kantianism, and they replaced it by piecemeal inves-
tigations into the use of philosophically relevant expressions. Finally, in the
wake of Quine, analytic philosophy has increasingly been dominated by
naturalism, and hence by the anti-Kantian idea that philosophy is part of,
or continuous with, empirical science.
The “reflective turn” in Neo-Kantianism and analytic philosophy

These anti-Kantian tendencies are only part of the picture, however, since
one needs to consider Neo-Kantianism and early analytic philosophy in the
context of a metaphilosophical constellation that dominated philosophy in
the nineteenth century, and beyond. From Kant until the end of the
twentieth century, one concern has united otherwise diverse traditions of
German philosophy: whether philosophy can preserve a distinct role in
view of the progress of the empirical sciences – a progress that included not
just the natural but also the historical and social sciences. This question
divided the main movements that successively dominated German philo-
sophy in the nineteenth century:
1800–1831 German idealism, which regarded philosophy as a super-
science;
1831–1865 Naturalism, which repudiated philosophy or reduced it to
empirical science;
1865–1900 Neo-Kantianism, which sought to rehabilitate philosophy as
an autonomous, second-order discipline.
Reverting to seventeenth-century rationalism – denigrated as “pre-
critical” by Kantians – German idealists held that all knowledge is a priori.
In their view, philosophy can deduce even apparently contingent claims
from first principles. The naturalists – Vogt, Moleschott, Büchner,
Czolbe – were physiologists by training, who treated the collapse of
German idealism as a sign of the bankruptcy of all metaphysical specula-
tion and a priori reasoning.15 They regarded all knowledge as a posteriori.
15
The term “naturalism,” which we associate with Quine, was not coined by the American pragmatists
(Santayana, Dewey), as has been suggested in A. Danto, “Naturalism,” in P. Edwards (ed.), The
Encyclopedia of Philosophy (New York: Macmillan and Free Press, 1967), 448–450; instead, it goes
back at least to these German physiological naturalists. Thus Czolbe, the most philosophical of
them, referred to his work as a “system of naturalism,” H. Czolbe, Neue Darstellung des Sensualismus
(Leipzig: Costenoble, 1855), 233–234.
The allegedly a priori disciplines were either reduced to empirical
knowledge, including that of psychology or physiology, or rejected as
illusory – this was their favourite treatment of philosophy (this, following
Mill, was their preferred line on logic and mathematics).16
For all their violent differences, the German idealists and the naturalists
were united in propounding a monolithic conception of knowledge as either
wholly a priori or wholly empirical. By contrast, the central strands of Neo-
Kantianism – especially the Marburg and Southwestern schools – retained
Kant’s dichotomy between a priori and a posteriori knowledge. This was
part and parcel of their resistance to attempts to reduce philosophy either to
the natural sciences – as in nineteenth-century German or twentieth-century
American naturalism – or to a branch of belles lettres unrestrained by
academic standards of truth or rationality – as in much French philosophy
since Bergson. If philosophy is to be a cognitive discipline, yet distinct from
the empirical sciences, this is because, like logic and mathematics, it aspires
to knowledge of a non-empirical kind. However, if logic, mathematics, and
philosophy are indeed a priori, that special status needs to be explained. One
explanation is the kind of Platonism favoured by Frege: logic and mathe-
matics are a priori because they deal with entities (numbers, concepts,
thoughts, truth-values) beyond the physical realm. Platonism is a recurrent
option within Neo-Kantianism as well. But most Neo-Kantians are closer to
an alternative explanation of the a priori that I have called Kant’s reflective
turn.17 Philosophy is immune to confirmation or refutation by empirical
evidence (whether through observation or experiment) not because it deals
with abstract entities beyond the empirical world, but because it is a second-
order discipline which reflects on the non-empirical preconditions of empiri-
cal knowledge and logical, epistemological or methodological presupposi-
tions of empirical knowledge that are prior to contingent facts (truth or
falsity). Both positions agree that philosophy (logic and epistemology) is a
science (Wissenschaft) rather than an art. Yet it is one distinct from the special
sciences because it is a priori; and it is a priori because it deals with issues that
lie beyond the scope of natural and social science.
The received contrast between Kantianism and analytic philosophy is
therefore untenable. There is a distinctive anti-naturalist tradition within
analytic philosophy, which insists that philosophy – especially logic,
epistemology, and semantics – differs from natural science (not just
16
Sluga, Gottlob Frege, ch. 1.
17
H. J. Glock, “Philosophy, Thought and Language,” in J. Preston (ed.), Thought and Language
(Cambridge University Press, 1997), 151–169.
quantitatively but qualitatively). Among its godfathers are not just pro-
claimed adversaries of Kant, like Bolzano and Moore, but also Frege and
Wittgenstein. Both of these thinkers developed Kant’s anti-naturalism,
albeit in strikingly different ways. Frege defended the Neo-Kantian idea of
the a priori and autonomous status of philosophy (in particular of logic and
epistemology) against the encroachments of science, in sharp contrast to
Brentano’s naturalism. There are also Kantian themes in Wittgenstein,
whose work owes much more to Schopenhauer and Frege than it does to
the indigenous Austrian tradition of Bolzano and Brentano.
Kant’s account of metaphysics and a priori knowledge set the agenda even
for those who rejected the synthetic a priori. The linguistic turn of logical
positivism was ostensibly directed against Kant’s suggestion that philosophi-
cal propositions are synthetic a priori. Nevertheless, Reichenbach, Schlick,
and Carnap all had strong roots not just in Poincaré’s conventionalism – itself
influenced by Kant – but also in German Neo-Kantianism. Carnap was a
pupil of the Neo-Kantian Bruno Bauch; Schlick’s conventionalism emerged
out of the Neo-Kantian debates on relativistic physics, and he was heavily
influenced by von Helmholtz. Reichenbach studied with Cassirer and wanted
to do his doctorate with Natorp. In spite of their rejection of the synthetic a
priori and occasionally virulent anti-Kantian rhetoric, many logical positivists
accepted the Kantian idea that philosophy is a second-order discipline. Unlike
science or common sense, philosophy is a priori not because it describes
objects of a peculiar kind (such as the abstract entities or essences postulated
by Platonism and Aristotelianism, respectively), but because it articulates the
conceptual scheme that science and common sense employ in their empirical
descriptions and explanations of reality.
This Kantian undercurrent is no coincidence. The Tractatus, arguably the
most important text in the rise of analytic philosophy, sets philosophy the
Kantian task of drawing “the limit of thought,” of demarcating legitimate and
illegitimate forms of discourse, rather than that of adding to our scientific
knowledge of the world. Hanna even goes as far as characterizing analytic
philosophy by reference to Kant: “The history of analytic philosophy from
Frege to Quine is the history of the rise and fall of the concept of analyticity,
whose origins and parameters both lie in Kant’s first Critique.”18 The idea that
analytic philosophy consists of (predominantly hostile) footnotes to Kant also
emerges in Coffa, despite his antipathy to Kant and Neo-Kantianism.19
18
R. Hanna, Kant and the Foundations of Analytic Philosophy (Oxford University Press, 2001), 121.
19
J. A. Coffa, The Semantic Tradition from Kant to Carnap to the Vienna Station (Cambridge
University Press, 1993).
The topic of analyticity is very important, because it is linked to the
status of logical, mathematical, and philosophical (metaphysical) proposi-
tions, which plays such a dominant role in the self-image of analytic
philosophy. It is no coincidence that Quine’s revival of radical empiricism
and naturalism proceeds through an attempt to undermine the analytic/
synthetic distinction, and the associated distinction of a priori and
a posteriori knowledge. The new Kripkean essentialism tries to undermine
the Kantian dichotomies in yet another way. But in doing so it pays
homage to the archetypal Kantian conundrum of how we might come by
substantive knowledge about reality without the aid of experience.
At the same time, the importance of analyticity and the synthetic a priori
must not be exaggerated. Though Moore and Russell on occasion employed
versions of the analytic/synthetic distinction, their work did not revolve
around it. The same goes for Cambridge analysis between the wars, and
for much Oxford philosophy, especially for Austin. The characterization also
excludes a lot of post-positivist and post-Quinean analytic philosophy in the
metaphysical vein, which has moved beyond these topics, if only by treating
Quine’s attack on the analytic/synthetic distinction as axiomatic. Finally, in
the ever-expanding field of moral and political philosophy, the analytic/
synthetic distinction never played a central role. The more general Kantian
problem, namely, whether, and how, philosophy can be conceived as an
autonomous discipline distinct from the empirical sciences, has loomed large
in the work of most analytic philosophers engaged in metaphilosophical
reflections. But not all analytic philosophers are given to such reflections.
Friedman’s account in A Parting of the Ways argues that Neo-Kantianism
was pivotal to the emergence of analytic philosophy.20 According to him, the
notorious split between analytic and so-called “continental” philosophy was
the fallout of a conference in Davos in 1929 attended by Carnap, Heidegger,
and Cassirer. It resulted, he maintains, from their diverse reactions to
problems presented by the Neo-Kantian tradition, notably the problem of
reconciling the logical and the perceptual preconditions of experience.
Unfortunately, Friedman’s explanation suffers from misplaced concreteness.
It ignores the fact that long before Davos, Moore, Russell, and members of
the Vienna Circle launched withering analytic attacks on German and
British idealism, Bergson, and Lebensphilosophie. Indeed, in some respects
these attacks go back to the contrast between Romanticism, empiricism, and
positivism in the nineteenth century. It is difficult to see how a different
course of events at or after Davos could have papered over the widening
20
M. Friedman, A Parting of the Ways: Carnap, Cassirer, and Heidegger (La Salle: Open Court, 2000).
cracks between the philosophical doctrines, methods, and intellectual
demeanour epitomized by these three protagonists.
The idea that the analytic/continental divide springs from a common
root in Kant and Neo-Kantianism is the opposite extreme of the thesis of
an Anglo-Austrian axis. While there is an important Kantian trend within
analytic philosophy, large swathes of the analytic tradition derive from a
break with Kant and hark back to Leibniz. In fact, Friedman himself moves
from a specific Kantian problem concerning sensibility and understanding
to a different and much broader contrast between rationalism (Carnap),
Lebensphilosophie (Heidegger), and a vaguely defined synthesis of the two
(Cassirer). The fact that Kant furnished central topics for analytic philoso-
phy in no way entails that Neo-Kantianism exerted an influence on the
analytic movement, let alone a positive one. The next section makes a start
in establishing such an influence.
Neo-Kantianism and science

Before Quine, the majority of analytic philosophers were united by the
conviction that their activity was distinct not only from that of the poet,
but also from that of the scientist. This is no reason to exclude Quine from
the analytic tradition, given his intellectual roots in logical positivism and
his substantive overlap with other members of the analytic family, notably
concerning the use of logical analysis. But it does establish a crucial
connection between Kant and mainstream Neo-Kantianism, and the
anti-naturalist (and anti-psychologistic) streak of analytic philosophy
(Frege, Wittgenstein, logical positivists like Schlick, Carnap, and
Waismann, and Oxford conceptual analysis).
Frege did not just deny that logic is a natural science; he insisted that it is
more fundamental than either metaphysics or psychology (Frege 1893: xix).
The Wittgensteinian vision of a division of labour between science and
philosophy was preached by Schlick and Waismann. In a more technical
and science-oriented manner, it is also evident in Carnap. Philosophy is
not a doctrine consisting of propositions but a method, namely, of logical
analysis. Negatively, it reveals metaphysical nonsense. Positively, it turns
into the “logic of science” (Carnap 1937: 279), namely, the linguistic
analysis or explication of scientific propositions, concepts, and methods.
This demarcation of philosophy and science underlies Carnap’s distinction
between analytic and synthetic propositions in The Logical Syntax of
Language, and his distinction between internal and external questions
in “Empiricism, Semantics and Ontology” (Carnap 1950). Scientific
philosophy is not philosophy that meddles in the scientific investigation of
reality. Instead, it is philosophy that reflects on this investigation in the
same rationalistic and collaborative spirit as the one that guides the first-
order explorations of the scientists themselves.
Neo-Kantians are close to paradigmatic analytic philosophers in four
further respects. They took a keen interest in the empirical sciences – both
natural and social. Although they resisted the dissolution of philosophy
into empirical science, they insisted that philosophy must take notice of
science. Some historians of analytic philosophy have suggested that
Kantian philosophy was pursued in isolation from developments in the
special sciences.21 This is sheer prejudice. One group of Neo-Kantians (on
a catholic conception) was constituted by eminent scientists such as
Helmholtz, Hertz, and Boltzmann. Moreover, from Kant to Cassirer,
Kantians tended to know more about the special science of their day
than contemporary analytic philosophers tend to know about present-
day science.
The Neo-Kantian interest in science is no coincidence. The Kantian
reflective turn implies that philosophy provides, among other things, an
account of the non-empirical preconditions of empirical science. This idea
was not only taken up by the philosopher-scientists within Neo-
Kantianism widely conceived, it was the core inspiration behind the
Marburg school of Neo-Kantianism founded by Cohen. For Cohen,
philosophy starts out with “the fact of science”; it is nothing other than
the meta-theory of the natural sciences. And when the natural sciences
were revolutionized by the theory of relativity, the Marburg school fol-
lowed suit through the works of Natorp – Die Logischen Grundlagen der
exakten Wissenschaften of 1910 – long before any analytic treatment of
special relativity, and Cassirer’s Zur Einsteinschen Relativitätstheorie
(1921). Indeed, there is only a single step from the idea that philosophy is
the metatheory of science to Carnap’s slogan that philosophy is the logic of
science, especially since the meta-theory of the Marburg school was expli-
citly conceived as a combination of logic and methodology. That step is the
linguistic turn of the Tractatus according to which logic is to be understood
by reference to language. Neo-Kantianism never took a linguistic turn.
Cassirer’s Philosophie der Symbolischen Formen (1923) officially fits that bill.
However, it postdates the demise of Neo-Kantianism, marks a significant
departure from Cassirer’s Neo-Kantian phase, is remote from the linguistic
21
A. Wedberg, A History of Philosophy (Oxford: Clarendon Press, 1984); Coffa, The Semantic
Tradition, 22.
turn of analytic philosophy, and did not influence the latter. In other
respects, however, Carnap’s logic of science is directly influenced by Neo-
Kantianism. Logische Aufbau der Welt (Carnap 1999: §§3–17) aims, among
other things, to carry out the rational reconstruction of the process of
cognition promised by the Marburg school. Carnap takes exception to
what he calls “transcendental idealism,” the view that the objects of knowl-
edge are created rather than recognized by science. However, this is mainly
because he regards the metaphysical dispute between realism and idealism as
idle. He himself insists that we constitute the objects of science. And like the
Marburg school he portrays the constitutive factors as purely formal or
logical: sense-experience is transformed into a sequence of abstract structures.
This brings us to a third feature, which ties many Neo-Kantians to
hardcore analytic philosophy, Quine once more included, including their
emphasis on logic. Logocentrism has been an important but largely unap-
preciated characteristic of mainstream German philosophy from Leibniz to
the early Husserl and the Marburg school. While Descartes and the empiri-
cists from Bacon to Hume disparaged logic, Leibniz assigned a central role to
it both in his methodology and in his metaphysics. He protested, against
Descartes, that clarity and distinctness is a wholly subjective criterion of truth,
which needs to be replaced by an objective, logical one. This Vorsprung durch
Logik was passed on to later German philosophers through Wolff. Whereas
the French and the British relied on purely subjective phenomena like the
clarity and distinctness of ideas or the vividness of sense-impressions, the
Germans tried to erect philosophy on the basis of objective logical principles.
Logocentrism is intimately connected to a fourth feature, which unites the
mainstream of Neo-Kantianism with that of analytic philosophy, namely,
anti-psychologism. For the German naturalists, psychology rather than logic
or metaphysics is the fundamental science.22 Moreover, following Mill, they
insisted that logic is a branch of psychology, and that the laws of logic are
ultimately empirical generalizations based on induction. Unlike British
empiricists, they conceived of psychology and experience in materialist,
indeed, physiological terms, as movements of the nervous system. This
proto-Quinean picture chimes with the psychologistic branch of Neo-
Kantianism. But it was sharply rejected by other Germanophone thinkers,
Neo-Kantians and analytic philosophers alike. Kant, the Neo-Kantians,
Frege, Wittgenstein, and the Vienna Circle all maintain that philosophy
(logic and epistemology) are autonomous, distinct not just from psychology,
but also from other natural sciences such as physiology. Moreover, as part of
22
Czolbe, Neue Darstellung des Sensualismus, 8.
the dualistic epistemology inherited from Kant, they resist any attempt to
reduce logical propositions to empirical ones. Accordingly, the mainstream
of Neo-Kantianism and major currents within analytic philosophy concur
on five points:
– a dichotomy between a priori and a posteriori knowledge;
– an a prioristic conception of philosophy;
– connecting philosophy closely to science yet without reducing it to the
latter;
– logocentrism: logic plays a central role to philosophy;
– anti-psychologism: logic and epistemology cannot be reduced to
psychology.
I shall now show that the Neo-Kantians actually influenced some of the
Germanophone pioneers of analytic philosophy along these lines. Frege
will occupy pride of place here. He was part of the same institutional and
intellectual milieu as Neo-Kantianism, and demonstrably indebted to
some of its central tenets. Given the profound debt Wittgenstein and
Carnap owed to Frege, and the Vienna Circle (Neurath excepted) to
Wittgenstein, the direct influence of Neo-Kantianism on leading logical
positivists will be supplemented by the indirect, yet equally important,
influence through Frege.
Lotze, Liebmann, and Frege

Lotze influenced leading Neo-Kantians like Liebmann and Windelband.
Nevertheless, he is normally regarded as the leading representative of post-
Hegelian metaphysics. This misleading classification is probably due to two
points: Lotze’s teacher C. H. Weise was Hegel’s student and Lotze appears to
repudiate the primacy of epistemology, which was accepted by many Neo-
Kantians and propagated by Zeller. Instead, he raises his voice in favour of
the claims of metaphysics: “The constant sharpening of knives is tedious,
however, unless one has something to carve” (1879: §ix). This recalls Hegel’s
taunting remark about Kant’s project of ascertaining the possibility of purely
rational knowledge through a prior “critical” examination:
other instruments can be investigated and assessed by other means than using
them for the specific work to which they are dedicated. But the investigation
of cognition (Erkenntnis) cannot proceed in any way other than through
cognition; as regards this so-called instrument, investigating it means nothing
other than to cognize. Yet wanting to cognize, ahead of cognizing, is just as
absurd as the wise undertaking of the scholasticus, namely to learn how to
swim before venturing into the water. (Hegel 1977: §10)
However, two important differences need to be noted. First, the context reveals
that Lotze is mainly intent on rejecting a psychological justification of meta-
physics.23 Second, in contrast to Hegel, Lotze does not dispute the feasibility of a
theory of knowledge. Instead, he merely notes, in accordance with Kant, that it
is supposed to serve as a prolegomena to metaphysics. Indeed, the third volume
of Lotze’s Logik (1989b) provides a theory of knowledge in the spirit of Kant.
Finally, Lotze’s conception of philosophy is much closer to Kant than to
Hegel. As a scientist by training, he repudiated the claim of German
idealism to encompass all knowledge. Metaphysical categories cannot
simply be derived a priori, without reference to the natural sciences.
Only empirical disciplines are capable of grasping reality through observa-
tion and experiment. By contrast, metaphysics clarifies the concepts and
theories of the special sciences and arranges them into a maximally sys-
tematic whole. Metaphysics is concerned not with “being,” i.e. reality or
facts, but with “validity.” Philosophy is a study which for its direct object
has the concepts which, in the special sciences as well as in life, have validity
(gelten) as principles for assessing things and actions (Lotze 1902: 94).
In the aforementioned metaphilosophical battle, Lotze lined up on the
same side as mainstream Neo-Kantianism. He repudiated both the ultra-
rationalist pretensions to deduce the way the world is a priori from first
principles and the naturalistic reduction of philosophy to natural sciences
in the name of a crude materialism.
Sluga maintained that the main influences on Frege “were no doubt
Kant and Leibniz, Lotze and Herbart.”24 Dummett protested that he is
much closer to the Austrian realists and (proto-)phenomenologists
Bolzano, Brentano, Meinong, and Husserl than to the idealist and psy-
chologistic Neo-Kantians. There is some justice in this remark, especially
as regards Bolzano. But it overlooks two important facts.
First, for better or worse, Frege was more exposed to the ideas of Neo-
Kantianism than to those of the Austrian school. The only member of that
tradition that he had contact with was Husserl. And while the received story of
Frege’s influence on Husserl’s renunciation of psychologism may be exagger-
ated, there is no evidence for a substantial influence in the reverse direction.
Second, it is illegitimate to associate Neo-Kantianism as a whole with idealism
and psychologism. Even the German Idealists did not reduce reality to
episodes in the minds of individuals, as Dummett wrongly supposes.
23
G. Gabriel, “Einleitung des Herausgebers. Objektivität: Logik und Erkenntnistheorie bei Lotze und
Frege,” in Lotze 1989b, xiii.
24
Sluga, Gottlob Frege, 40.
Instead, they insisted that reality is intelligible only because it is the
manifestation of a divine spirit or rational principle. This “objective,” i.e.
non-psychologistic, idealism influenced Lotze and Rickert, among others.
As regards psychologism, i.e. the conflation of logical and epistemo-
logical with psychological issues, Dummett is even less on target.
Admittedly, there is a psychologistic side to Kant, which inspired the
psychologistic Neo-Kantians Fries and Nelson. Yet Kant also inaugu-
rated crucial anti-psychologistic modes of thought. He distanced trans-
cendental philosophy from “empirical psychology,” notably Locke’s
“physiology of the human understanding.” It elucidates not the “genesis
of experience,” but rather “what lies within it,” i.e. its content. What
makes a true belief either a priori or a posteriori according to Kant is the
justification we have for it, rather than the way we acquired it. The same
anti-genetic point is made by his distinction between the question of how
we acquire a certain kind of experience or belief (quaestio facti) and the
question of what the logical and epistemological status of that experience
or belief is (quaestio iuris).
Leibniz had already distinguished a priori from innate knowledge. Like
many Neo-Kantians, Lotze was influenced by Leibniz as well, and might
have imbibed the idea from either Leibniz or Kant. In any event, it is one of
his central claims. What makes a knowledge claim a priori is not its genesis
(Genese), but the kind of validity (Geltung) it has (1989b: ch. 3). The
“genetic perspective” is unsuited to the philosophical purposes of logic
and epistemology (1989b: §322). For the same reasons Lotze condemns
psychologism as an “illusion of psychology and a vitiation of logic” (Lotze
1989b: §333; see also 1989a: §§18, 47). Indeed, he can lay claim to be the real
founder of anti-psychologism. For he did not just influence Husserl on this
score, but also the British Idealists (Bosanquet, Bradley), and hence,
indirectly, Moore and Russell.
The repudiation of psychologism and naturalism more generally deci-
sively shaped all strands of Neo-Kantianism indebted to Lotze. While
“being” (Sein) and “genesis” (Genese) are investigated by empirical
science, investigating the validity of knowledge claims is the prerogative
of philosophy. The Marburg school revolves around the idea that trans-
cendental philosophy should be conceived in a logical and analytic vein
rather than a psychological or physiological one. The transcendental
preconditions of knowledge are not mental acts of a transcendental
subject, but a system of propositions presupposed by empirical science
(Cohen 1877).
Recent research leaves no doubt that a significant part of Frege’s
philosophical background was constituted by Lotze and Neo-Kantians
influenced by him.25 Frege
(a) attended Lotze’s lectures on the philosophy of religion;
(b) mentioned Lotze as one of his teachers in the unprinted CV accom-
panying his doctoral dissertation;
(c) was labelled a “pupil of Lotze” by his colleague Bruno Bauch;
(d) was reported by Bauch to have stated that “the impulses he got from
Lotze were of decisive significance”;
(e) was familiar with Lotze’s theory of judgement.
Following my criterion of parallels plus acquaintance, it is probable that
Frege took over some important ideas and phrases from Lotze, such as the
contextualist primacy of propositions as against concepts (1989b: §321;
Frege 1884: §§60–62), characterizing axioms as “neither in need of nor
allowing of a proof” (1879: §1; cp. Frege 1884: §3), and his three-world
ontology (1989b: §§347–348; Frege 1918).
More important, however, is that through Frege Lotze influenced the
whole anti-naturalist strand within analytic philosophy. This holds in
particular for Frege’s anti-genetic stance. Frege’s attendance of Lotze’s
lectures is usually dismissed as insignificant, since the latter concerned
the philosophy of religion. But a prominent contention of those lectures
was precisely that the correctness or incorrectness of an idea is independent
of its origins. Right at the start of his manuscript Lotze writes, “taken by
itself, the history of the origins [Entstehungsgeschichte] of an idea can never
decide its correctness” (1882: 1). In exactly the same way Frege distinguishes
between the rational justification and the causal explanation of knowledge:
Whether what you take for true is false or true, your taking it comes about in
accordance with psychological laws. A derivation from these laws, an
explanation of a mental process that ends in taking something to be true,
can never take the place of proving what is taken to be true (1918: 58–59, see
also 1977: iii; 1884: xvii–xxiii).26
25
See L. Kreiser, Frege: Leben – Werk – Zeit (Hamburg: Meiner, 2001), 99–100; Gabriel, “Einleitung
des Herausgebers”; H. Sluga, “Frege on Meaning,” in Glock, The Rise of Analytic Philosophy. This is
not to belittle the mathematical background highlighted by Baker and Hacker. At the same time,
Kantian themes played a role even among mathematicians in nineteenth-century Germany. Indeed,
Frege’s PhD dissertation starts out from a Kantian question: how can geometry depend on the
nature of our faculty of intuition, even though imaginary figures possess features that run counter to
intuition? (Kreiser, Frege, 93–94, 108).
26
Similarly, Frege granted that “the historical perspective” has a certain justification, while denying
that one can divine the nature of numbers from psychological investigations into the evolution of
numerical thinking (1884: xix).
In the spirit of Kant and Lotze, Frege concedes to empiricism:
without sense-impressions we are as thick as a plank and know nothing of
numbers or of anything else; but this psychological proposition does not
concern us here at all. I emphasize this again because of the constant danger
of confusing two fundamentally different questions. (Frege 1884: §105 n)
The similarities are even more palpable when one compares Frege to
Liebmann, who was his colleague at Jena from 1882 onwards. We know
that Frege read Liebmann’s Zur Analysis der Wirklichkeit at a time when he
wrote Foundations. The demonstrable parallels can therefore once more be
treated as the result of an actual impact.
Neither Liebmann nor Frege employ Lotze’s terminology of “genesis” vs.
“validity.” Both, however, distinguish between the psychological question of
how we acquire our knowledge and the logical-cum-epistemological ques-
tion of its justification. Liebmann sets apart “psychic laws of nature accord-
ing to which representations arise” and “norms, categorical prescriptions, the
compliance with which makes the natural course of thoughts hit upon the
truth, the violation of which leads to missing it.” Someone who understands
logical or arithmetical statements and nonetheless denies them “would for us
be like a madman” (1880: 251–252, 240, see also 77, 79). Frege for his part
distinguished the “psychological laws of holding to be true” from the “logical
laws of being true.” Rejecting the latter would amount to “a hitherto
unknown form of madness” (Frege 1893: xvi).27
Following the same anti-genetic and anti-psychologistic trajectory,
Liebmann and Frege attack the attempts by physiological materialists to
reduce thought to neurophysiological phenomena. Both refer in particular
to Moleschott’s infamous slogan “No thoughts without phosphates.”
Liebmann’s reaction: “And what have proteins, calcium and phosphate in
the brain to do with logic?” (Liebmann 1880: 531). Similarly, Frege observes
how absurd it is to analyze thoughts by reference to the amount of phosphate
in the brain, once one realizes that the laws of thought are the laws of logic.
And he reminds us of the difference between the conditions for holding a
belief and the conditions for its truth, lest we think that the proof of
Pythagoras’ theorem might have to mention the phosphate content of our
brain (Frege 1884: xviii; 1979: 5). Finally, the rejection of psychologism goes
hand in hand with a rejection of empiricism and materialism. It is hardly a
coincidence that Lotze (1989b: §330), Liebmann (1880: 208), and Frege (1884:
27
Liebmann also characterizes the relation between logic, geometry and mathematics essentially in
line with §§14 and 26 of Foundations of Arithmetic.
§3 n, 4 n) are united in repudiating the empiricist credo that all knowledge is
based on induction, and on the same grounds: without some general truths,
at the very least the principle of induction itself, induction from particular
observations could not get off the ground.
Frege and the Southwest German School

This argument also appears in Windelband (Windelband 1921: ii/107), the
founder of the Southwest German School. In comparing Frege to
Windelband and Rickert one must keep two points in mind, however.
First, indisputable affinities notwithstanding, there is no evidence that
Frege was actually familiar with these authors. Second, the style of
Windelband and in particular of Rickert is even more remote from
Frege’s lucid and well-reasoned writing than that of Lotze and Liebmann.
Lotze had transformed the dichotomy between Sein (being) and Sollen
(ought) – which he inherited from Fichte – into one between being and
value (Wert). Being is, values hold or are valid (gelten). But the contrasts are
not identical. X can be valid in one of three ways: being true, being good or
being beautiful; yet only the second case implies that X ought to obtain.28 The
Southwest German School nevertheless associated the whole domain of
validity/value with normativity. Philosophy has the task of establishing the
values of truth, goodness, and beauty as valid, against naturalism, empiricism,
and relativism (Windelband 1921: i/26, 43–44; Rickert 1939: 172). The princi-
ples of transcendental philosophy are reinterpreted as norms, which ought to or
must be acknowledged, if one is to understand how objects can be given to us in
experience. Psychologism is condemned, since psychology can only furnish
inductive hypotheses about the genesis of our claims to validity, without being
able to justify these claims. In line with Frege, Windelband writes:
For philosophy it may be of interest to determine, whether a representation
came about in one way or another: for epistemology the only concern is
whether the representation is valid, i.e. whether it ought to be acknowledged
as true. (1921: i/24; see Rickert 1910: 31)
Like Frege, Windelband compares the normative status of logic to that
of ethics. At the same time, both expand on this idea by maintaining that
the normative character of logical laws for our thinking derives from its
validity, i.e. from being appropriate to its topic.29
28
H. Schnädelbach Philosophie in Deutschland 1831–1933 (Frankfurt am Main: Suhrkamp, 1983), 201.
29
G. Gabriel, “Frege als Neukantianer,” Kantstudien 7 (1986): 84–101; here 94.
It is undeniable that the Southwest German School shared an anti-
naturalist perspective with Frege. This can easily be explained, however, by
reference to their common inheritance from Lotze. Gabriel and Sluga go a
step further in contending that Windelband and Rickert actually influ-
enced Frege’s views on more specific matters. They start out from Frege’s
theory of judgement, in particular his distinction between the act of
judging – what Frege calls a “judgement” – and the content of the
judgement – what Frege calls a “judgeable content” and later on a
“thought.” The distinction lies behind the so-called “Frege-point,” the
insight that assertive force is not associated with the copula, since judgeable
contents with a copula can occur in conditional judgements without being
asserted.
Frege maintained that the distinction between a judgement and its
content was entirely his own. But it was anticipated not just by Bolzano,
but also by Kant and Neo-Kantians, including Lotze, Christoph Sigwart,
Julius Bergmann, Windelband, and Rickert.30 Thus Sigwart – another
pupil of Lotze – distinguished “the living act of thinking” and the ideas to
which the act is directed. Windelband distinguished between the “assess-
ment” (Beurteilung) and the judgement, and also claims that this distinc-
tion goes back to Lotze (1921: i/30, 32n1). Once again, there is no need to
assume that Frege derived his distinction from Windelband.31 This con-
sideration does not apply to another affinity. Rickert tried to motivate the
distinction between the act of judgement and the judgement by propos-
ing an argument that also occurs in Frege. Every assertion of the form
“The sun is shining” can be transformed into a question “Is the sun
shining? Yes,” which is supposed to show that one and the same judge-
ment or judgeable content can be the object of different kinds of acts
(1904: 98–101).
Gabriel and Sluga have made an additional claim, namely that Frege
borrowed the notion of a truth-value from Windelband and Rickert. Both
had a general theory of the contrast between fact and values. Prominent
among values is that of truth:
30
That he might have been aware of this fact is suggested by Über den Zweck der Begriffsschrift, where
he defends himself against the charge of having confused the act of judgement and the judgeable
content, a charge that was no doubt fuelled by his new terminology in which “judgement”
designates what other authors regarded as the content of the judgement.
31
In any event, Frege’s terminology is a novelty, though not necessarily an improvement. He contrasts
the (act of) judgement and the content. This may have invited the charge that he confused act and
content, which, ironically, he was at pains to avoid (see 1977: 101; 1918: 61n). That he unwittingly
invited this accusation suggests, incidentally, that he was not familiar with Windelband’s prior
terminology.
The laws of logic hold for us only on the assumption of the purpose (Zweck) of
truth; it is the general validity, i.e. the value, of seeking recognition by every-
one which disinguishes the logical forms of thought from the other associa-
tions that take place within processes governed by natural law. (1921: ii/73)
Elsewhere, Windelband calls this value a “truth-value” (Wahrheitswert):
All sentences making knowledge claims are combinations of ideas whose
truth-value has been determined by affirmation or negation. (1921: i/32)
In other words, the act of judging is a transition from the judgement
(Frege’s judgeable content or thought) to the truth-value. This idea was
subsequently elaborated by Rickert.32 Unlike Windelband but like Frege,
Rickert admits a second truth-value-falsity. According to Rickert, in every
judgement a value is affirmed, i.e. truth or falsity, which attaches timelessly
to the content of the judgement. Truth and falsity are values in the strictest
sense of the word, which is why we take a positive or negative stand to
judgeable contents that have these values. The objects of knowledge,
therefore, are not things in the world of facts, but truth itself conceived
as a positive value. Rickert also suggested that the truth-value True might
be conceived as the totality of judgements that have a positive value, a
suggestion that has parallels in Frege’s Grundgesetze der Arithmetik.
Finally, Rickert is close to Frege in his Platonist leanings. He accepts
Kant’s account of the sensible world of science. But he insists, against Kant
and with Frege, that there is also an intelligible world of non-sensuous
objects that we know not by perception but by understanding (Verstehen).
The similarities are indeed striking. But before concluding that Frege
engaged in plagiarism of gigantic proportions, the following points should
be borne in mind. As mentioned above, although it is possible that Frege
read some Windelband and Rickert, there is no evidence that he actually
did. Furthermore, it is unlikely that Frege hit on the notion of a truth-value
as a result of reading Windelband or Rickert. His whole logic was based
on a generalization of mathematical function-theory, and the idea of a
function is that of a mapping of arguments onto values.
Gabriel recognizes this point, yet insists that it cannot be the whole
story, since truth-values can also be arguments of other functions, namely
the truth-functions (e.g. the function expressed by Φ ∧ Ψ). But this
response is feeble. The mathematical notion of a function from which
Frege starts out in Funktion und Begriff (1891) allows for precisely the same
thing. Numbers are the values of numerical functions like f(x) = x2, but
32
Sluga, “Frege on Meaning,” 222–223.
they are also their arguments. Moreover, in certain cases argument and
value are identical: for the argument 1, for example, the aforementioned
function has the value 1.
At the same time, Gabriel is right to point out that in one passage from
Sinn und Bedeutung (1892), Frege introduces the notion of a truth-value in
a quasi-teleological fashion reminiscent of Windelband. Frege here
makes out the following case for holding that sentences do not just have
a sense – the thought they express – but a meaning – a truth-value:
The fact that we concern ourselves at all about the meaning of the part of a
sentence indicates that we generally expect and recognize a meaning for the
sentence itself. The thought loses value for us as soon as we recognize that
one of its parts lacks meaning. We are therefore justified in not being
satisfied with the sense of a sentence, and in inquiring also as to its meaning.
But now, why do we want every proper name to have not only a sense, but
also a meaning? Why is the thought not enough for us? Because, and to the
extent that, we are concerned with its truth-value. (Frege 1884: 163)
Gabriel concludes that Frege took truth-values to be the meanings of
sentences because of the following line of thought: it is their truth or
falsehood which determines the value of sentences, and the meaning
(Bedeutung) of a sentence is its significance, that which is important
about it, i.e. its value or purpose. But this interpretation is not compulsory.
From the start Frege insisted that his formal system – his Begriffschrift – be
confined to sentences with a truth-value. According to him, only true
sentences can function as premises of valid arguments and are thus suitable
for the logicist derivation of arithmetic from logic. Furthermore, Frege hit
upon the notion of a truth-value entirely without teleology, namely, by
admitting functions the arguments and values of which are no longer
confined to numbers, but include objects of any kind.
Conclusion
There is no compelling evidence for the hypothesis that Frege adopted
details of his philosophical logic from the Southwest German school. The
demonstrable and paramount Neo-Kantian influence originates in Lotze
and Liebmann and is predominantly general in nature. Negatively, it
consists of the anti-naturalistic repudiation of psychologism, materialism,
and empiricism. Positively, it is the idea that philosophical (logical,
epistemological) problems and theories are sui generis. Frege himself turned
this idea into a Platonist doctrine, like Rickert, but in a less abstruse
manner. Philosophy, by contrast to natural science, is concerned not
with the physical world, but with logical entities in a “third realm” beyond
space and time.
In many respects, Wittgenstein was strongly indebted to Frege. But he
reconceived the autonomous status of philosophy in the more genuinely
Kantian spirit of a reflective turn. Philosophical problems and investiga-
tions concern not the world itself, but the conceptual framework that we
employ (among other things) in our thought and talk about the world.
Wittgenstein also revived a theme going back to Lotze and only mentioned
in passing by Frege, namely, that of normativity. In his wake, that idea
has taken centre-stage in contemporary dissidents from naturalism such
as Kripke, McDowell, and Brandom. But their indebtedness to Neo-
Kantianism is indirect and vague, since Wittgenstein was hardly familiar
with Kant, let alone Lotze or the Southwest German School. For that
reason, the normativism he inspired is a topic for another occasion.
chapter 4
Reise um die Welt: Cassirer’s cosmological

phenomenology
Nicolas de Warren
Il faut être absolument moderne

What has been hazarded as the final philosophy of Europe was launched in
earnest a silent decade after its initial breakthrough in the Logical
Investigations with the publication of Edmund Husserl’s Ideen I in 1913.
As Husserl announces, “pure phenomenology” is “an essentially novel”
enterprise that seeks to establish the “foundational science” of philosophy.
This incarnation of philosophy as a “rigorous science,” a call earlier
professed in a 1911 essay in the Neo-Kantian journal Logos, is motivated
by a host of ambitions: to realize the Western ideal of philosophy as a
science; to clarify the relationship between philosophy and the natural
sciences by way of a foundation for their respective constitutions; to
culminate the modern philosophical revolution set into motion by
Descartes; to lead human existence toward “true humanity” based on a
practical life in truth. In fulfilling the “secret desire of modern philosophy,”
phenomenology represents the advent of a philosophical modernism
through which the European project of modernity finds its completion.
As with many of the orchestrated surprises in Ideen I, it is only after
patiently following its complex progression for some length that the way of
phenomenology towards the science of philosophy dawns as the way of
transcendental idealism. This transcendental alignment was clear to all, not
least of all to Husserl’s own disappointed students, as it saddled Ideen I with a
contentious inheritance: Kant’s project of forging a new science of transcen-
dental philosophy. As Kant writes in the Preface to the Critique of Pure
Reason, his aim is to find “a secure path” for philosophy to become a science –
an aspiration mirrored in the opening sentence of Husserl’s Ideen I:
Pure phenomenology is an essentially new science, one that, by virtue of its
very distinctiveness, lies far afield of natural thinking and consequently
presses now, for the first time, for development. Here we are seeking the
way into this science, we characterize its unique position, relative to all other
82
Cassirer’s cosmological phenomenology 83
sciences, and we aim to demonstrate it to be the fundamental science of
philosophy. (Husserl 2014: 3)
Unlike Kant, Husserl’s transcendental science does not have as its principal
task the discovery of the system of all principles of pure reason independent
of experience brought under the idea of the possibility of a priori synthetic
judgments. Heralding a new science of phenomena, Husserl’s enterprise
remains unrecognizable as a transcendental science from Kant’s original
standpoint; phenomenology must therefore struggle with and against Kant
to achieve its own transcendental promise. But if philosophy as a rigorous
science is thus placed on a secure path, pure phenomenology must in turn
confront the exact sciences since, as Husserl remarks: “Other sciences long
known to us also attend to phenomena” (Husserl 2014: 3). Indeed, every
material science (in contrast to formal sciences) is inherently a science of
phenomena: physics, biology, sociology, psychology, etc. What need is
there for an additional science of phenomena given that every kind of
phenomenon (nature, consciousness, etc.) already enjoys a corresponding
science? Is there some aspect of the world for which a science is still lacking?
The answer is evident from the scientific triumph that is the nineteenth
century with its explosion of new sciences and revolution within estab-
lished sciences that collectively reached every region of possible knowledge.
Most significantly, as David Hume foreshadowed and Auguste Comte
celebrated, the historically elusive final frontier of that phenomenon closest
to us – our own consciousness – finally received the quintessentially
modern science of a psychology without a soul from an empirical stand-
point. The opening paradox of phenomenology is that it is a science
without a phenomenon immediately of its own – a phenomenon not
already pre-given, as with physical phenomena or mental phenomena.
If phenomenology were not transcendentally directed, its pursuit would
necessarily prove either futile or trivial. As Husserl insists, phenomenology
does not discover a new kind of phenomenon, as if some piece of the world
or region of being had somehow miraculously been overlooked by the
history of human curiosity, and now finally brought to the light of scientific
thought. Cultural objects, social institutions, material bodies, empathy,
consciousness, etc. – these are the “phenomena” of phenomenological
inquiry insofar as phenomenology is a universal science of experience with-
out prejudice. As Husserl incorporates methodologically into the epoché and
the reduction, the science of phenomenology “intends” or “approaches”
(geht auf ) phenomena “in a wholly different attitude whereby any sense of
the word ‘phenomenon’ which we find in the all too familiar natural sciences
84 nicolas de warren
becomes modified in a certain way” (Husserl 2014: 3; translation modified).
This change in theoretical attitude is not merely a change of optics or
perspective (even if Husserl at times speaks in this way). It is more
profoundly a transformation of how phenomena are thought, or known,
encountered in thinking.
This transformed sense of how phenomena are problematized and
reflected upon is placed under the general heading of the problem of
transcendental constitution. Setting aside the complexities of Husserl’s
notion of constitution, is phenomenology defined by one central problem
of constitution (the subject, the body, the other, etc.) or by a constellation
of problems without a fixed center? This issue has no straightforward
answer; it animated in equal measure the development of Husserl’s own
thinking as well as its contested legacy. Yet, as evident from its progression,
the organizing problem of Ideen I is the problem of reason. Phenomenology
is essentially a phenomenology of reason; it aims to develop a system of
reason through a far-reaching phenomenological analysis of constitution in
the three domains of rationality: knowledge, value, and action. This
explicit self-understanding is associated with a salient metaphor that
Husserl invokes time and again to convey the thrust of his enterprise. At
critical junctures in Ideen I, Husserl characterizes phenomenology as the
exploration of a “new world” – not a different world, but the world of
experience considered philosophically anew. As Husserl writes: “The
premier task of this first book will be to seek ways in which the enormous
difficulties of penetrating into this new world can be overcome bit by bit,
so to speak” (2014: 5). Even though it would take years to enter into focus
properly with the theme of the Lebenswelt, Husserlian phenomenology
gravitates around the problem of the world, even if the full import of this
organizing problem never became entirely transparent or mastered. The
fundamental “phenomenon” of phenomenology is the world as the totality
of the many senses of phenomena or, in other words, the multiple ways
in which the world manifests itself, as “carried” or “constituted” in trans-
cendental (inter)-subjectivity.
Nous sommes embarqués

Husserl was not alone in his ambition to invent a new science. Within the
ambit of philosophy, Husserl’s ambition placed his enterprise in direct
alignment but equally in direct confrontation with the Neo-Kantianism
launched in Marburg under the tutelage of Hermann Cohen. If we
consider Neo-Kantianism broadly as the commitment to the realization
of philosophy as a science in the form of transcendental thought, and,
moreover, as dedicated to what Jürgen Habermas called the unfinished
project of modernity, then Husserlian phenomenology shares in this
aspiration while circumscribing an alternative horizon of philosophical
progress. Husserlian phenomenology and Marburg Neo-Kantianism, the
latter taken in its most expansive form with Ernst Cassirer’s philosophy of
symbolic forms, are two different realizations of Kant’s transcendental
revolution. As I shall explore, this parting of ways equally marks a parting
of ways between two contrasting philosophical modernisms as two differ-
ent realizations of the last philosophy of Europe: phenomenology.
The call “philosophy as rigorous science” in the form of transcendental
thought is essential to the enterprise of Marburg Neo-Kantianism. As
Cohen proclaims: “Kantian philosophy means to me nothing other than
philosophy as a science” whereby “science is the Ideal of System on the
basis of continuous methodological work” (Cohen 2001b: iv). As with any
call, however, one must be able to hear its directive properly, and contrary
to entrenched prejudices defining our view of Neo-Kantianism today, this
call “Back to Kant” was originally a call of philosophical emancipation. As
Cohen writes: “When we call upon Kant’s critical method for our vision,
we thus feel ourselves liberated from dogmatic dependencies” (Cohen 1928:
231). It is not a call back to Kant-Philologie, but a demand for renewed
engagement with the problems themselves through a direct confrontation
with Kant’s unfinished project. Cohen’s return to Kant exhibits a complex
dynamic: the turn back to Kant’s philosophical writings (of the past), first
announced in Kants Theorie der Erfahrung (1871), calls for the “restitution”
(Wiederherstellung) (in the present) of the task of philosophy as a science
(for the future), as neither a speculative adventure nor as a reduction to
positivist natural sciences.
In keeping with Kant’s conception of critical philosophy as the system of
the pure principles of reason, Cohen understands transcendental philoso-
phy as the critical exploration of the three domains of reason: knowledge,
feeling (values), and action. Against the psycho-physiological reduction of
Kant’s thinking in Hermann von Helmholtz and F. A. Lange, Cohen
defines transcendental philosophy as primarily a method for the analysis
of Erzeugung, or “constitution,” of the factum of scientific knowledge.1
1
The extension of this critical enterprise to the factum of culture first becomes formulated in Cohen’s
Kants Begründung der Ästhetik (1889). Cohen’s complete system envisions three domains of know-
ledge (scientific knowledge, morality, aesthetics), each grounded on principles of reason, and each
irreducible to the other, and each of these orientations within the world account for different forms of
cultural objectivity.
That philosophy “depends on” or “refers to” (angewiesen) the Faktum der
Wissenschaft does not represent, as often contended, a slavish fascination
with science or a tacit embrace of philosophy as the “ideology” of
science.2 The philosophy of the future is not a philosophy of science
but the science of philosophy. In the same vein, Cohen’s conception of
transcendental method fashions a critique of metaphysics and the spec-
ulative turn of Kant’s thinking with German Idealism, as well as the
naturalization (or “anthropologization”) of Kant’s thinking prevalent in
J. F. Fries and Jürgen Bona Meyer. As Cohen argued in Kants Theorie der
Erfahrung, Kant’s theory of experience is inseparable from the discovery
of the transcendental function of the a priori in its three meanings: as the
form of objective knowledge irreducible to empirical machinery (sensa-
tions, psychological genesis, etc.); as a cognitive (or “logical”) form not to
be conflated with a psychological faculty or innate idea (or disposition);
as an a priori condition of possibility and therefore, most crucially, as not
devoid of material content, which it “produces” or “constitutes”
(erzeugt).3
Cohen’s pursuit of Kant’s unfinished revolution gave impetus to an
historically distinctive philosophical movement. Unlike Husserl, whose
phenomenological movement arguably gravitated around his thinking,
even if his students and (wayward) protégés were in many cases brilliant
in their own right and not always in accord with the Master, Marburg Neo-
Kantianism formed a research collective (Arbeitsgemeinschaft) of a different
kind that became progressively displaced through its three central figures:
Cohen, Natorp, and Cassirer. This generational fertilization and, in the
case of the relationship between Aby Warburg and Cassirer, cross-
fertilization, reflected within the movement of Marburg Neo-Kantianism
the dynamic of Bildung essential to the historical progress of reason itself.
In this regard, Cassirer understands ThePhilosophy of Symbolic Forms as
realizing the enterprise of Marburg Neo-Kantianism. The irony of
Cassirer’s philosophy of symbolic forms, however, is that it stands as a
“monumental homage” to a Neo-Kantian conception of philosophy as a
philosophy of cultural forms at a moment when leading figures of Neo-
Kantianism (Natorp, Cohen, Rickert) had already parted with this vision
2
W. Marx, Transzendentale Logik als Wissenschaftstheorie (Frankfurt am Main: Klostermann, 1977),
17–18.
3
As Dussort observantly remarks: “c’est la condition formelle, sine qua non de la possibilité de
l’expérience, dans la mesure où il contribue à la constituter, à la ‘construire’ (Terme usuel de la
première édition de KTE, supprimé dans la seconde)” (H. Dussort, L’École de Marbourg, Paris: PUF,
1963, 91).
of philosophy.4 Cassirer’s The Philosophy of Symbolic Forms presents an
enriching transformation of Marburg Neo-Kantianism’s inner motivations
through an openness to its own historical situation, marked by the catas-
trophe of the First World War, and the contingency of a decisive encounter
with the eccentric brilliance of an anti-philosopher, Aby Warburg, whose
self-styled Critique of Pure Unreason (Kritik der reinen Unvernunft) pro-
vided the catalyst for Cassirer’s own Critique of Symbolic Reason.5 These
complex inflections notwithstanding, as Cassirer explains, “the ‘philoso-
phy of symbolic forms’ grows out of [the] critical transcendental question
and builds upon it. It is pure ‘contemplation,’ not of a single form, but of
all – the cosmos of pure forms – and it seeks to trace this cosmos back to the
‘conditions of its possibility’” (1996: 189).
Essential to this transformation of Kant’s transcendental enterprise is a
phenomenology of knowledge (Phänomenologie der Erkenntnis) – the sub-
title of the third volume of The Philosophy of Symbolic Forms. Yet, as
Cassirer explains: “in speaking of a phenomenology of knowledge I am
using the word ‘phenomenology’ not in its modern sense but with its
fundamental meaning as established and systematically grounded by
Hegel.” Indeed, Cassirer understands “modern phenomenology” as “start-
ing much less from Kant than from Brentano’s concept of consciousness”
as intentionality; the promise of which would only become fully achieved
in Husserl’s phenomenology (Cassirer 1985: 196). By distancing his phe-
nomenology of knowledge from Husserl’s “modern phenomenology,” one
should not consider Cassirer’s phenomenology as “pre-modern” in its
ostensible turn back to Hegel, but should instead recognize an alternative
modern phenomenology – and hence an alternative philosophical modern-
ism. In fact, the Hegelian reference invokes the external casing, as it were, of
a concept of phenomenology without adopting or endorsing its Hegelian
content and method.6 The Hegelian precedent provides a focal (“funda-
mental”) meaning for the concept of phenomenology as the “totality of
meaningful forms” (geistigen Formen), as “objectifications” or “works” of
“spirit,” which can only come to know itself through a reflection on the
4
M. Ferrari, Introduzione al neocriticismo (Rome: Laterza, 1997).
5
Tagebuch der Kulturwissenschaftlichen Bibliothek Warburg, ed. K. Michels and C. Schoell-Glass (Berlin:
Akademie, 2001), 144. The First World War was also decisive for Natorp, albeit in an entirely different
manner, as it provoked a transformation in his thinking that would lead to his abandonment of
Marburg Neo-Kantianism. In his 1922–1923 lecture courses (Philosophische Systematik), Natorp devel-
ops a philosophy of being (crucial for the young Heidegger) in opposition to a Neo-Kantian philosophy
of knowledge. Cohen had arguably also abandoned his critical platform with the “ontological turn”
(and for many: Hegelian turn) of his philosophical thinking before the war.
6
A. Philonenko, L’école de Marbourg: Cohen, Natorp, Cassirer (Paris: Vrin, 1989), 150ff.
historical forms of its appearances. With this first invocation, “phenomen-
ology” situates within a transcendental framework the question of the
“world” or “cosmos” as the question of the whole in its different forms of
manifestation. As Cassirer remarks:
The philosophy of symbolic forms does not want any metaphysics of
knowledge but wants instead to be a phenomenology of knowledge. It thus
takes the word “knowledge” [Erkenntnis] in the broadest and most encom-
passing sense. It does not understand with this word only the act of scientific
conception and theoretical explanation, but understands every spiritual activ-
ity through which we construct [aufbauen] a “world” in its characteristic
validity, order and in its “being-so” [So-Sein]. (Cassirer 1956b: 208)
As he further observes: “We start with the concept of the whole: the whole is
the true (Hegel). But the truth of the whole can always only be grasped in a
particular ‘aspect.’ This is ‘knowledge’ in the broadest sense – ‘seeing’ the
whole ‘in’ an aspect, through the medium of this aspect” (Cassirer 1996: 193).
Cassirer’s reference to Hegel is here deliberately misquoted, for Hegel in fact
writes: das Wahre ist das Ganze, not: die Wahrheit ist das ‘Ganze’ (Hegel 1977:
11). This discrepancy between Hegel’s das Wahre and Cassirer’s die Wahrheit
marks a distance between a closed metaphysical totality and an open totality
of the world in perpetual motion.7 Cassirer’s phenomenology, in other
words, is not the first step towards a science of logic, which would in turn
represent the first moment of an encyclopedia. It is instead a “voyage around
the whole world” (eine Reise um die Welt) exploring within a transcendental
framework the different forms of meaning-formation (“symbolic functions”)
and world-disclosure in a cosmological (and cosmopolitan) vision of the
whole.
Die Sachen Selbst

The rallying cry Zurück zu Kant! – formulated in Eduard Zeller’s 1862
lecture “Über die Aufgabe der Philosophie und ihre Stellung zu den
übrigen Wissenschaften,” incorporated as a refrain into Otto Liebmann’s
1865 Kant und die Epigonen, and absorbed into Marburg Neo-Kantianism –
is the recognizable foil for Husserl’s own celebrated call Zurück zu den
Sachen selbst! Husserl, however, understood this call quite differently from
how the term “phenomenology” is bandied about today, as now
7
For differences between Hegel and Cassirer, see D. P. Verene, The Origins of the Philosophy of
Symbolic Forms (Evanston: Northwestern University Press, 2011), 45–48, 78ff., and Philonenko,
L’école de Marbourg, 150–151.
designating “the study of structures of consciousness as experienced from
the first-person point of view” – with which Husserl’s own conception of
phenomenology has in fact little to do. As Husserl stresses time and again,
the meaning of phenomena within phenomenology proper cannot be literal:
phenomena are not there to be discovered by squinting our eyes more
intensely.
Under the motto “back to the things themselves,” Husserl understood in
spirit “back to what matters most for philosophy,” and what matters most
for philosophy is thinking, or reason, for which Husserl articulated an
original conception in Ideen I. This call motivates more specifically a return
to “phenomena,” not as naively experienced in the natural attitude, but as
critically experienced through the method of reduction and framed by the
problem of constitution. The phenomena of phenomenological considera-
tion are irreal in the sense of discovered through an eidetic method of
analysis. The method of Husserl’s transcendental thinking is phenomen-
ological insofar as this eidetic method leads to intuition, or an original
“givenness” of “the things themselves.” Husserl’s “back to the things
themselves” is therefore also a call back to the essence of intuition and
the intuition of essences, forged anew in a manner that still remains not
entirely understood today, and having little to do with the “Platonism” or
“Metaphysics” with which it is all too naively conflated. As Husserl makes
clear:
But obviously here, as elsewhere in phenomenology, this does not mean to
engage straightforwardly in actual experience, i.e., to proceed empirically. . .
The task is rather to examine, in eidetic intuition, the essence of the experienced
in general and as such . . . (Husserl 1989: 97)
In its own manner quietly dramatic, Ideen I culminates with a new
phenomenological conception of reason. As Husserl proposes: “the first
fundamental form of rational-consciousness: originary and giving ‘seeing’
[die erste Grundform des Vernunftsbewusstseins: das originär gebende ‘Sehen’]”
(Husserl 2014: 270). For Marburg Neo-Kantianism, this claim to an
intuitive reason (das originär gebende ‘Sehen’) was perceived as regressing
to a conception of intuition against which Cohen had so mightily
struggled. Husserl’s provocation against the Marburg transcendental pro-
gram is elsewhere pointedly formulated with his declaration: “we are the
genuine positivists” (Husserl 2014: 38). Husserl’s “positivism” is not the
dogma of empiricism but the critical evidence of intuition. With this
declaration, Husserl tacitly reappropriates the meaning of “positivism”
from Paul Natorp’s criticism of the empirical positivism of Ernst Mach’s
“physical phenomenology” and its program of finding a foundation for
knowledge in brute sensations (Natorp 1887). Husserl’s pure phenomen-
ology is not Mach’s, nor comparable, as Husserl always stressed, with
notions of “phenomenological analysis” in Carl Stumpf and Theodor
Lipps. These differences notwithstanding, as spokesperson for Marburg
Neo-Kantianism, Natorp repeatedly criticized precisely this conception of
intuition in his reviews of The Logical Investigations and Ideen I (Natorp
1901, 1917).8 Cassirer follows in turn when he characterizes “modern”
phenomenology as “the way of intuition.” As he writes: “Yet another way
of intuition is found in Husserl. He too is a ‘rationalist,’ but in a broader
sense than that of Cartesian ‘mathesis universalis’” (Cassirer 1996: 171).9
The architect of this founding critique of intuition for Marburg Neo-
Kantianism was Cohen, for whom the receptivity of intuition represents an
unacceptable notion of the given (das Gegebene) as an “irrational
Remainder” (unbegriffener, unvernünftiger Rest) within the function of
knowledge.10 Receptivity is not just “saddled,” as John McDowell would
have us now say, with conceptual function (i.e., conceptual content) and
the spontaneity of the understanding. “The given” is not merely “aufge-
hoben” but “aufgegeben” in a more decisive sense insofar as “the given”
(das Gegebene) is configured dynamically into an Aufgabe, or task.11 Cohen
in this fashion reformulates Kant’s Ding-an-sich into a limiting-concept, or
horizon, for a determinate function of knowledge, and thus rejects the
“two world” interpretation of the distinction between “appearances” and
“thing in itself.” The “thing in itself” is neither the mysterious cause for the
sensible manifold nor an inscrutable thing beyond the appearances that
could be apprehended in an intellectual intuition. Rather than a founda-
tion for knowledge in “the given,” knowledge is a dynamic striving to
8
The debate regarding intuition represents a threefold parting of ways: Husserl, Heidegger, and
Marburg Neo-Kantianism. See Heidegger’s discussion of Natorp’s critique of intuition and the
given in Heidegger (2005: 224ff.). As Husserl complains to Heidegger: “Im Wesentlichen ebenso in
der Allgemeine Psychologie . . . aber zeigt, dass Natorp ganz und gar nicht fähig war den klaren, doch
auf der Hand liegenden Sinn der Phänomenologie, als einer Wesensanalyse des reinen Bewusstseins . . . zu
fassen und überhaupt ein Sehen und sehend Gegebenes gelten zu lassen” (letter of September 10,
1918; my emphasis). In this regard, Heidegger and Levinas join ranks with Natorp and Cassirer in
their respective critiques of Husserl’s conception of intuition.
9
It is a critique that Husserl would learn to direct against himself – at least in the form in which
phenomenology first appeared in The Logical Investigations: “Danach fiel in den Logischen
Untersuchungen phänomenologische Wesenslehre und rationale Psychologie zusammen. Dass dies
in verschiedenem Betracht unrichtig ist . . .” (Husserl 1971: 24).
10
This critique of intuition becomes fully articulated in Cohen’s Das Prinzip der Infinitesmal-Methode
und seine Geschichte (1883) (Cohen 1984).
11
As Natorp writes: “Der Gegenstand ist nicht gegeben, sondern aufgegeben: Wissenschaft wird damit
zur Aufgabe” (2013: 15).
expand and to enrich the scope of determinateness of the “thing” as an
“indeterminate X” (Unbestimmte X), or Idea of reason. The object of
knowledge thus becomes “temporalized,” i.e., discovered historically
through the progression of knowledge. As Cohen argues, the proper
subject of astronomy are not the stars in heaven (i.e., the visible phenom-
ena) but the scientific understanding of their lawful motion, which alone
gives to the visible stars the “dignity” (i.e., validity) of reality (Cohen
2001b: 111). As Cohen elsewhere writes:
[F]or those in the know, it suffices to invoke Lambert, who also starts from
‘the simplest,’ as final and enduring elements. Transcendental method,
however, does not search for such final form-elements; instead, it seeks
the ‘highest basic principles’ as found in printed books and in the history of
the actual becoming of experience. (Cohen 2001b: 35)
This invocation of Lambert is revealing, for it was Lambert who first
introduced the term “phenomenology” into philosophy in his 1764 Neues
Organon – the source for Kant’s own sporadic deployment of the
notion.12 Cohen’s renewal of transcendental idealism rests on the fore-
closure of any independent status for intuition, sensible or intellectual,
and hence, the possibility of phenomenology in its modern forms, as
either “sensuous” (Stumpf, Lipps) or “eidetic” (Husserl). This exclusion
of phenomenology does not imply an exclusion of experience, but rather
an inclusion of experience within the space of reason; experience is only
constituted in its mediated form as the object or “fact” of a scientific
project in its historical development, or Bildung (and as materialized in
printed books). Expressed differently, the “transcendental aesthetic”
becomes absorbed into the “transcendental analytic” such that any pos-
sible entrance of phenomenology into transcendental thought through
the opening of the transcendental aesthetic is effectively closed. Yet, if
Cohen’s launch of Neo-Kantianism rests on the rejection of any possible
“phenomenology,” how can Cassirer in turn claim to fulfill the ambition
of Marburg Neo-Kantianism in the form of a phenomenology, albeit
neither in the way of Husserl or Hegel?
12
Lambert defines the science of phenomenology as the “transcendent optic or perspective” and
“theory of appearance” (Lehre von dem Schein). The contrast between “appearances of visible things”
(Schein der sichtbaren Dinge) and “the true” (der Wahren) is introduced in this discussion by way of
the example of astronomy and the distinction between sphärischen and theoretischen Astronomie.
Johann Heinrich Lambert, Neues Organon (Leipzig: J. Wendler, 1764), ii §4. For Kant, “general
phenomenology” is a “negative science” for the principles of sensibility (cf. his letter to Lambert,
September 2, 1770); in his letter to Marcus Herz (February 1772), Kant speaks of general phenom-
enology as the “science of all principles of a priori sensibility” as the first part of metaphysics.
Kant or the unfinished revolution
In the 1929 Davos disputation, Heidegger baptized a prejudice against
Marburg Neo-Kantianism that has since become entrenched in the histor-
iography and consciousness of philosophy. This construal of a parting of
ways between Neo-Kantianism and phenomenology reflects in part
Natorp’s influence, yet takes for granted the possibility of an alternative
phenomenology as the final realization of the critical project of Marburg
Neo-Kantianism. As Cassirer challenges: “What does Heidegger under-
stand by Neo-Kantianism? . . .What does Heidegger have in mind when he
employs the phenomenological critique in place of the Neo-Kantian one?”
Moreover: “Neo-Kantianism is not a kind of philosophy as a dogmatic
theoretical system; it is a direction of questioning [eine Richtung der
Fragstellung]”(in Heidegger 1991: 274). Cassirer’s reply here echoes the
spirit of Cohen’s thinking as well as his own “counter-history” of the
Enlightenment as a “process of doubting and seeking, and in ‘action,’
the constantly evolving process of thought,” which expresses nothing less
than the “task of shaping life itself,” or, in other words, the autonomy of
human existence (Cassirer 1951: xv). How it is that the progressive
cosmopolitanism of this philosophical movement came to be branded by
above all Heidegger as “conservative” and “reactionary” is one of the
supreme (and telling) ironies of twentieth-century philosophy.13
Kant’s “revolution in the manner of thinking” (Revolution im Denkart)
principally centers on discovering eine Richtung der Fragstellung. This new
direction of philosophical questioning marks the end of Western meta-
physics, for which Kant provided its essential definition as a thinking
centered on the problem of being in contrast to the genuine transcendental
problem of sense (Sinn) – not to be understood, however, as the question of
the sense of being, but as the transcendental question of under what
conditions is sense constituted in its manifold ways of appearing.
Transcendental thought brings to a close the philosophical adventure of
“fundamental ontology.” Metaphysics considers “what is it to be?” through
the concept of substance and a corresponding notion of thinking as the
apprehension of an “absolute object” independent of the form of its being-
known. As significantly, Cassirer identifies metaphysics with spatial think-
ing where “transcendence,” or the transcendent character of the object of
13
Let us recall: Marburg Neo-Kantianism was launched by the unholy trinity of the first non-
converted Jewish professor in Germany (Cohen), an avowed atheist (Zeller), and a socialist militant
(Lange).
knowledge, is tacitly understood as “outside” the “inner” space of thinking
or “inner” theater of mental representation.
Kant’s philosophical revolution discovers knowledge as a function, or
more specifically, as the synthetic unification of multiplicity. The unity of a
multiplicity is not itself a hypostasized entity enjoying independent exis-
tence or an abstraction from a pre-given concrete manifold. An “object” of
knowledge just is a multiplicity unified according to a determinate func-
tion of sense, or concept. On this account, knowledge is not the knowledge
of an object as “absolute” (as independent of being-known) but of the
object under a determinate function of knowledge as a synthetic activity
conditioned by an a priori form. Objects conditioned by a function of
knowledge are appearances. For Kant, the synthetic function of concepts
was interpreted as the unification of a manifold under a rule in the form of
a judgment. Hence, Kant’s central question of how a priori synthetic
judgments are possible.
Kant’s revolution remained, however, unfinished. Kant’s First Critique
(I shall not explore here Cassirer’s assessments of Kant’s First and Third
Critiques) is structured around an intolerable dualism between form and
matter within its own proper concern with das Erkenntnisproblem (a notion
all too loosely rendered in English as “epistemology”) and an “intellectual,”
or “cognitive,” conception of synthesis. Whereas the former problem
(relation between form and matter) is overcome within Cohen’s Neo-
Kantianism, it resurfaces according to Cassirer in Husserl’s transcendental
phenomenology (as it was known to Cassirer through Ideen I ). On the
other hand, even if Marburg Neo-Kantianism dissolves the dualism
between form and matter, Cohen and Natorp are each beholden to the
primacy of scientific cognition that is the hallmark of an intellectual
conception of synthesis. Cassirer’s phenomenology of knowledge speaks
to both issues: it provides a “non-intellectualist” or “non-cognitive” con-
ception of synthesis while overcoming the dualism between form and
matter within the field of knowing, taken broadly as different ways of
“understanding the world” (Weltverstehen). In this manner, Cassirer posi-
tions his alternative phenomenology in contrast to Husserl’s (modern)
phenomenology as well as (Cohen’s and Natorp’s) Marburg Kantianism.
In the First Critique, Kant understands the synthetic unity of know-
ledge as the synthesis of form and matter under a priori conditions in the
form of a determinate judgment. Kant conceives of sensible experience
as “das schlechthin Gegebene” – the brute given for the basis of any knowl-
edge. How can “the given” become synthesized with a priori forms of
sensibility and the understanding; how is the receptivity of intuition and
the spontaneity of the concept brought together into an object of know-
ledge? In articles spanning Substance and Function (1910) and The
Philosophy of Symbolic Forms (1923–1929), Cassirer formulates Kant’s legacy
in terms of an intolerable oscillation between the “myth of the given” and
“conceptual holism” (to speak in a contemporary idiom).14 Each repre-
sents, and for different reasons, unacceptable solutions to the Kantian
problem of how to think together receptivity and spontaneity. Unlike
others since Hegel who struggled with this dualism within Kant’s thinking,
Cassirer contends that Kant was himself aware of this debilitating dualism
within the First Critique; one need not await the Third Critique for Kant’s
critical self-awareness against himself.
Three insights are essential for situating Cassirer’s phenomenology of
symbolic forms within Kant’s original “direction of questioning.” The first
consists in Cassirer’s argument that the distinction between sensibility and
understanding does not mark a real distinction between two “absolute
Seinspotenzen” (i.e., two faculties or powers, receptivity and spontaneity).
This cardinal distinction is neither ontological nor psychological; it does
not demarcate an “outside” from an “inside.” It marks instead a relative
and methodological distinction within the movement of transcendental
constitution, or, in Cassirer’s own framework, a determinate symbolic
function. The genuine meaning of Kant’s Copernican Revolution thus
resides with a novel conception of the relation between form and matter
(concept and intuition): receptivity and understanding are neither
“successive phases” in knowing nor a distinction in being, but a strenges
In-Einander, or rigorously entangled with each other.
The second insight consists in Cassirer’s emphasis on Kant’s distinction
in the Analytic of Pure Understanding between objective and subjective
deductions of the categories. Whereas the objective deduction addresses
the form of objective knowledge, the subjective deduction is directed
towards the conditions and originality of perceptual consciousness. In
perceptual consciousness, the “world of perception is not merely a form-
less matter of sensations but already formed and determined” according to
the threefold synthesis of apprehension, reproduction, and recognition.
The direction of questioning delineated with the objective deduction
reflects the centrality of objective consciousness for Kant’s critical enter-
prise. As Cassirer notes, Kant’s 1772 letter to Marcus Herz already under-
stood the question of how representations are related to their objects as
14
See “Erkenntnis nebst den Grenzfragen der Logik” (1913), “Erkenntnis nebst den Grenzfragen der
Logik und Denkpsychologie” (1927), and “Inhalt und Umfang des Begriffs” (1936) in Cassirer 1993.
“the key to the whole mystery of the metaphysics that has hitherto been
concealed from itself” (Cassirer 1985: 234). Along the direction of ques-
tioning in the subject deduction, Kant merely gestures towards a founda-
tion for “objective consciousness” in perceptual experience. This
distinction between objective and subjective deductions effectively struc-
tured the legacy of Kant’s thinking in the nineteenth century: the natur-
alization of Kant’s thinking in Helmholtz, for example, represents the
psychologization of the subjective deduction at the expense of the objec-
tive deduction; the critique of psychology and the myth of the Given in
Cohen, for example, represents a rejection of any subjective deduction in
favor of the critical development of the objective deduction.
The third insight consists in Kant’s discovery of synthesis as grounded in
the transcendental apperception. The transcendental apperception is the
condition of possibility for all knowledge, including perception. The
sensible manifold unified in perceptual consciousness is conditioned by a
connection to possible self-consciousness: the “I think” must be able to
accompany any consciousness of an object. This transcendental require-
ment reflects Kant’s argument that perceptual consciousness is formed as a
manifold of appearances, and yet that the structural connections within the
sensible manifold cannot be intrinsic to the manifold itself. Rather, the
linkages within perceptual consciousness are provided by the spontaneity
of the understanding and the rules (concepts) that organize and unify the
sensible manifold into a fully fledged object of experience. Perceptual
consciousness is always (already) formed perception, saddled, as some
would now say, with conceptual form and the spontaneity of the under-
standing. Because the unity of perceptual consciousness possesses the form
of a synthetic unity according to a rule (i.e., a concept) of the under-
standing, perceptual consciousness must have a necessary relation to
consciousness and, more specifically, to the ego, since it is the activity of
the ego to synthesize.
The transcendental deduction is thus organized around two capital
insights: perceptual experiences must have a reference to an object (as the
unity of a sensible manifold brought under a concept or rule of the
understanding) and perceptual consciousness must belong to an ego, and
in this sense, must be constituted for an ego through its own power of
reason, or a priori synthesis. This dual insight organizes the problem of the
transcendental deduction: to establish the necessary connection between
experience and its object as well as the reference of any objective experience
to an ego – to demonstrate that consciousness of objectivity and self-
consciousness are distinct, yet necessarily bound together. Both of these
requirements are only possible with the subsumption of appearances to
pure concepts of the understanding; concepts, as Kant’s fundamental
discovery, are rules for the synthesis of the manifold (in the form of
judgment), and these rules constitute perception as both objective and
subjective, as both an experience of consciousness and as an experience
(of something) for consciousness. Expressed in Husserl’s terms, Kant tacitly
discovers the outlines of intentionality: consciousness is consciousness of
something as well as self-awareness of this consciousness as consciousness
of something. Expressed in Cassirer’s terms, Kant tacitly discovers the
problem for the concept of “symbolic pregnance” that forms the core of
his “true phenomenology of perception.” This symbolic relation “can be
reduced neither to merely reproductive processes nor to a mediated intel-
lectual process – it must ultimately be recognized as an independent and
autonomous determination, without which neither an object nor a subject,
neither a unity of the thing nor a unity of the self would be given to us”
(Cassirer 1985: 235).
Husserl or the modern necromancer

It may be surprising to discover Cassirer’s claim that the “first true
presuppositions for any true phenomenology of perception” were estab-
lished in Kant’s Critique of Pure Reason. In the context of The Philosophy of
Symbolic Forms and its notion of “symbolic pregnance,” Cassirer under-
stands a “true phenomenology of perception” to be centered on the
Kantian insight that “the understanding is the simple transcendental
term for the basic phenomenon that all perception as conscious perception
must always and necessarily be formed perception,” and that the determi-
nate form, or meaning, of perception is not grafted afterwards; “perception
‘is’ only insofar as it takes determinate forms” (Cassirer 1985: 194). The
inherence of meaning within perception reflects the integration of any
perceptual experience, as a particular experience of this chair, this red ball,
etc., within a meaningful whole textured with implications of other pos-
sible appearances of meaning. Yet, despite this delineation of a “true
phenomenology” within the transcendental deduction, Kant’s Critique of
Pure Reason effectively remained unable to exploit its own discovery. This
unthought dimension to Kant’s critical thinking would cast a long shadow
across the nineteenth century, and it is this Kantian unthought that
Cassirer aims to render both perspicuous and productive through his
notion of “symbolic pregnance” and, more generally, his conception of
symbolic forms.
The reason for the unthought dimension to Kant’s own guiding thought
is revealing of how Cassirer conceives of his phenomenology as an alternative
to Husserl’s “modern phenomenology” while also understanding his project
as fulfilling the direction of questioning rejuvenated within Marburg Neo-
Kantianism, even as he moves beyond the frameworks of Cohen and Natorp.
Kant’s own critique of rational psychology notwithstanding, Cassirer argues
that the language of the First Critique suggests that “sensibility” and “under-
standing” are two separate sources of knowledge (as Kant himself at times
speculates), thus further insinuating either a psychological reality or an
obscure metaphysical power of the soul. As Cassirer observes, “the transcen-
dental insight which he [Kant] was striving to establish is expressed in the
concepts of the eighteenth-century faculty psychology” (Cassirer 1985: 194).
The two legacies of Kantianism in the nineteenth century – the naturaliza-
tion of Kant with Helmholtz (and others) and the speculative adventures of
German Idealism – are each inscribed within this error of transcendental
grammar. In both instances, whether through a “psychologization” or a
“metaphysicalization” of the Analytic of the Understanding, a certain con-
ception of subjectivity comes to the fore as what Cassirer dubs a
Formgebungsmanufaktur – as a “manufacturing of forms.” Transcendental
subjectivity becomes either “naturalized” or “idealized” with the same phi-
losophical consequence that transcendental conditions become “derived
from a self-subsistent transcendental subject as their author.” The genuine
functional meaning of concepts thus becomes “re-substantialized,” to wit,
naturalized as psychological realities, or idealized into the speculative self-
positing activity of Spirit. Cassirer expresses this point with unusual flair in
speaking of “a magician or necromancer animating dead sensations, awaken-
ing it to the life of consciousness” (Cassirer 1985: 195).
This unthought dimension to Kant’s thinking – its “true phenomen-
ology” – would seem to present a non-negotiable impasse for the basic
tenets of Marburg Neo-Kantianism. On the one hand, Cohen critically
rejected any psychological interpretation of Kant’s transcendental insight
into the function of understanding; there can be no “true phenomenol-
ogy,” if one here understands a further development of the direction of
questioning broached with the subjective deduction. In the development
of Kant’s legacy in the nineteenth century, Cohen’s Kants Theorie der
Erfahrung marks in this respect a “new epoch” in the history of German
Neo-Kantianism – a second revolution, as Heiddegger recognized
(Heidegger 1987: 140). On the other hand, the problem posed by the
subjective deduction, which misfired with Kant’s recourse to the language
of eighteenth-century rational psychology, nonetheless outlines a “true
phenomenology of perception” – and hence of consciousness – without
which the force of questioning in the transcendental deduction cannot be
fully developed. How can the problem of consciousness find room and
method within transcendental thought without forfeiting transcendental
thought’s self-defining critique of psychology, while equally avoiding the
speculative adventures of Spirit within German Idealism?
As Cassirer acknowledges, “we are led to a similar methodological
question by another school of thought, which not only differs from
Kantian thinking in its historical point of departure but seems to be
diametrically opposed to it.” As he continues: “Modern phenomenology
starts less from Kant than from Brentano’s concepts of consciousness” and
its discovery of the “distinctive factor of consciousness,” intentionality
(Cassirer 1985: 196). The promise of this modern phenomenology resides
with its methodological way of integrating the problem of consciousness
into a transcendental framework without the risk of empiricism or reduc-
tionism (let alone a metaphysical notion of the soul). At the center of
Husserl’s phenomenological thinking stands the emancipation of inten-
tionality from any vestige (as is still the case with Brentano’s intentional
relation as “mental phenomenon”) of the “myth of the mental,” or, in
Cassirer’s own terminology, the “mythology of activities.” Husserl thus
characterizes intentionality as a function of consciousness. As Husserl
writes: “The viewpoint of function is the central viewpoint of phenomen-
ology” (Husserl 2014: 169). This emphasis on intentional acts of conscious-
ness in terms of their functional meaning is but another way to stress the
form-giving, synthetic activity of consciousness. Although Husserl arrives
at his conception of function via Stumpf (while clearly rejecting Stumpf’s
phenomenalism), the primacy of a functional conception of intentionality
signals a convergence with Cassirer’s Kantian-derived conception of
knowledge as synthetic function of sense formation. Moreover, as
Cassirer acknowledges, Husserl’s conception of intentionality formulates
in its own way the decisive transcendental insight that “to have meaning” or
“to mean something” constitutes the “basic character of all consciousness”
(Cassirer 1985: 198). The basic meaning of consciousness is not “to be” but
“to mean.” As Cassirer himself makes known to Husserl in a letter of April
10, 1925: “Since the publication of the first volume of the Logical
Investigations I have always held the conviction that between the tasks
that phenomenology sets for itself and the basic insights of critical
philosophy there lies a deep commonality: to be sure, what is at stake for
both of us is what you call . . . ‘the radically executed science of the
transcendental, to be executed ad infinitum’” (Husserl 1994: 5).
Despite this common cause of a “radically executed science of the
transcendental,” Cassirer nonetheless discerns the persistence of a
dualism between form and matter within Husserl’s phenomenology.
More significantly, Husserl’s conception of transcendental subjectivity,
as essentially defined by the power of “bestowal of meaning”
(Sinngebung), remains beholden to the spell of the necromancer: subjec-
tivity as a magician of experience bringing dead matter to life, or meaning.
For even if Husserl’s conception of intentionality has effected in its own
way a passage from the problem of “substance” to the problem of “func-
tion” (i.e., from the primacy of the question of being to the primacy of the
question of sense) and likewise established a phenomenological conception
of consciousness irreducible to any descriptive psychology, Husserl’s think-
ing still embraces an “absolute opposition” between “matter” and “form”
within the immanence of pure consciousness – the central discovery of
Husserl’s method of transcendental reduction. As Husserl expresses the
matter in Ideen I:
[I]t is also absolutely indubitable that here “unity” and “manifold” belong to
totally different dimensions and, of course, that everything hyletic belongs to
the concrete experience as a really obtaining, integral part of it, while what
“displays” and “profiles” itself in it as a manifold belongs, by contrast, to the
noema. (Husserl 2014: 196)
Cassirer takes issue with this Husserlian insistence on a “radical sharp-
ness” between immanent, non-intentional hyleticcontent and animating
noetic acts of consciousness. Although this “radical distinction” does not
replicate a metaphysical distinction between “mind” and “world,” nor a
Brentanian distinction between “mental phenomenon” and “physical
phenomenon,” this particular dualism within intentionality organizes
Husserl’s phenomenology of knowledge around an internal dualism
between “meaning-bestowing acts” (intentionale morphé) and “dead
matter” (sensuelle hylé) within the sphere of immanence, or pure
consciousness. As Husserl stresses, the relation between “form” and
“matter” must be seen as “intentional functions,” yet he still maintains:
“in any case, this strange doubling and unity of sensual hylé and
intentional morphé plays a dominant role across the entire phenomen-
ological field” (Husserl 2014: 165; translation modified). Albeit without
the same dramatic consequences, Cassirer in his own way strikingly
anticipates Jacques Derrida’s critique (albeit more sharply delineated
on Husserl’s theory of linguistic expression and meaning) of
the phenomenological distinction between “living” and “the dead,”
meaning-bestowing form and dead hylétic matter, within Husserl’s
account of constitution and the noetic-noematic structure of intention-
ality in Ideen I.15
Cassirer’s reservation against this “strange duality” between “two realms
of being,” immanent consciousness and transcendent objectivity, and its
underlying structuring distinction between “matterless form” and “form-
less matter” within pure consciousness, hinges on two objections. The
phenomenological justification for “formless matter,” or non-intentional
content, is directly called into question. Even if Husserl’s notion of hylé is
not to be conflated with a classical notion of sensation (since the sensation
of red is itself an objectified object of consciousness), some variant of the
“myth of the given,” even as inscribed within the immanence of conscious-
ness, returns in Husserl’s conception of hyletic content as the experienced
sensing of red. Yet, as Cassirer notes, “from the standpoint of phenomen-
ological inquiry” (1985: 199; my emphasis), any sharp distinction between
form and matter is not only illicit but also obscures the genuine phenom-
enological constitution of phenomenon as “total experiences.” This
emphasis on phenomenon as “total experiences” prohibits any sharp
distinction between “form” and “matter.” Instead, the distinction between
“form” and “matter” within any kind of experience must be understood as
a functional relation. Cassirer thus takes issue with the one-sided account
of the constitution of meaning; on Husserl’s view, it is the noetic acts of
consciousness that bestow meaning onto a “meaningless hylé,” which is
somehow present (as reelle content), or contained, in consciousness. As
Cassirer objects:
For no content of consciousness is in itself merely present, or in itself merely
representative; rather, every actual experience embraces both factors, and it
is therefore this mutual relationship, and not the form, the noetic factor
alone, that constitutes the foundation of all animation and spiritualization.
(Cassirer 1985: 199)
Cassirer further targets Husserl’s “apprehension-content of apprehension”
scheme of constitution. On this view, as Husserl first presented in The
Logical Investigations and – on Cassirer’s reading – substantially retains in
Ideen I (despite its expanded theory of the noetic-noematic correlation),
non-intentional content is apprehended by different noetic acts in view of an
intentional, transcendent object: the same non-intentional content can be
apprehended differently through different noetic acts, and thus constitute
15
See J. Derrida, “La forme et le vouloir-dire. Note sur la phénoménologie du langage,” in Marges de la
Philosophie (Paris: Les Éditions de Minuit, 1972).
different intentional objects. The non-intentional content at the basis of
such an objectifying act is indifferent, or neutral, to the specific meaning
character of the act as well as the constituted, transcendent object. Yet, as
Husserl would himself revise through his complex inquiry into inner time-
consciousness, and fully exploit in his genetic analysis of passive synthesis,
non-intentional content cannot in fact be understood as an identical sub-
strate underlying different noetic acts in view of their respective transcendent
objects. As with the paradigmatic case of time-consciousness, whereas on the
schema of “apprehension-content of apprehension,” the same non-
intentional content can either be apprehended as “past” or as “present,” on
Husserl’s revised account – unknown to Cassirer – non-intentional content
must already be differentiated, and hence, pre-constituted, through a deeper
and more pervasive form of passive constitution. Otherwise, the same hyletic
content would underpin the apprehension of a sound as both now and as
just-now – in both cases, the consciousness of the now as well as the just-now
would be based on something present (i.e., the immanent presence of the
underlying non-intentional content).
What matters here is less the extent to which the progression of Husserl’s
thinking after Ideen I, especially in its genetic development, transforms the
“static” conception of intentionality and its schema of “apprehension-
content of apprehension.”16 What matters here is rather how Cassirer
positions his own “true phenomenology” with regard to his critique of
Husserl’s “modern phenomenology” and its “static” conception of consti-
tution, its organizing distinction between “noetic” meaning-giving acts
and “dead matter,” and the conception of pure consciousness as a necro-
mancer bringing dead-matter – the given – to the “flesh and blood” of
meaning. Intentionality must give way or become transformed into the
concept of “symbolic pregnance.” Expressed more faithfully to Cassirer’s
Marburg spirit, the problem of intentionality must become rediscovered
through the concept of “symbolic pregnance” within the horizon of a
philosophy of symbolic forms.
Symbolische Prägnanz
A central insight in The Philosophy of Symbolic Forms is the recognition that
the constitution of perceptual experience cannot be understood through
16
This development remained unknown to Cassirer since it was never fully made visible in published
form (the only form in which Husserl’s transcendental phenomenology could have been known
publicly during the 1920s).
any kind of dualism between “form” and “matter,” or, in other words, any
kind of “myth of the mental” and/or “myth of the given.” Perceptual
content is always “saturated” with a sense-giving function of understanding
(Verstehen). Cassirer identifies three basic “symbolic forms,” each as a
determinate manner of “sense-formation.” The three symbolic functions
of myth, language, and pure concept correspond respectively to three
forms of “world-understanding” (Weltverstehen): expression, representa-
tion, and scientific cognition. Each of these symbolic forms structures in an
a priori manner a way of having the world in view (Sichtweisen) and
orientation of subjectivity within the world. With the progression from
expression through representation to thinking, the world becomes pro-
gressively (historically as well as systematically) expanded and diversified
through a multiplication of “sheets or bands of sense” (Sinnverbänden).
Although a symbolic form “saturates” the content of experience, meaning
is thereby not in consciousness but in the world. As Cassirer writes:
“content is not simply in consciousness, filling it by its mere existence –
rather, it speaks to consciousness and tells it something” (Cassirer 1985: 191).
The prose of the world; yet, the world only speaks to consciousness in
disclosing, or making visible, different ways of its meaning.
Central to Cassirer’s alternative to Husserl’s phenomenology is the
concept of “symbolic pregnance.” Introduced through a critique of
Husserl’s account of constitution and the distinction between “noetic
acts of meaning-bestowal” and non-intentional content, “symbolic preg-
nance” anchors Cassirer’s “true phenomenology of perception.” Its guiding
insight is not just the claim that “matter,” or perceptual content, is always
“saturated” by a determinate (symbolic) form of sense, but that the content
of perceptual experience is never fixed to any single determinate form.
Since the relation between “matter” and “form” is relative, not absolute or
sharp, the synthesis of perceptual content and its possible forms of sense is
variable and hence plural. By way of illustration, Cassirer takes the percep-
tion of “a simple line.” Depending on the symbolic function of under-
standing at work, a line (drawn on a chalkboard, etched in wood, etc.) can
“mean” differently: a mood is expressed in a line drawn by Matisse; a
geometrical object is cognitively apprehended in a line drawn in a geometry
text-book; a boundary between spaces is marked in a line drawn in the
sand. In each instance, the visibility of the line, its materiality, is meshed
with its legibility of meaning, with the “meaning” of meaning in each
instance variable: an expression of mood, a visualization of geometrical
form, etc. In each of these instances, the line appears differently since it
means differently (i.e., incarnates different senses). Yet with each of these
instances, and especially with the prosaic line expressing a mood, “we do
not merely read our own inner states subjectively and arbitrarily into the
spatial form [of the line]; rather, the form gives itself to us as an animated
totality, an independent manifestation of life” (Cassirer 1985: 200). Sense
inheres in perceptual sensation; the mood is expressed in the animation of
the line itself.
A “same” physical inscription can thus take on different forms of
visibility and legibility: “its [the line’s] pure visibility is never conceivable
outside and independently of a determinate form of vision” (Cassirer 1985:
200). There is, in other words, no underlying inscription (“primary sensa-
tion”) of a line to which is (then) grafted a particular sense as “mood,”
“geometrical object,” etc. The concept of “symbolic pregnance” entwines
“visibility” and “legibility” (or “speakability,” as it were) into one form of
manifestation. Whereas for Husserl, we come to speak about what we
originally perceive, for language, as Husserl argues in Ideen I, does not
create a new dimension of meaning, and whereas for Heidegger, we come
to see what we originally speak about, “symbolic pregnance” entails both
perception and speaking.17 The structure of perception is embedded within
the logos of an “as structure” of understanding; likewise, the logos of an
“as structure” of understanding is embedded within the structure of
perception. Language and perception are mutually implicating; it is there-
fore a false problem to inquire which comes first, perception or language.
Experience of the world is originally pregnant with both speech and
perception.
“Symbolic pregnance” also operates within a particular sense modality.
Revealing once again in his writings the profound and pervasive influence
of Goethe, Cassirer rejects an analysis of color perception that would claim
to identify an “indifferent substrate” of color underlying its different
appearances. The visibility of color depends on the context and function
in which it is experienced; sensations are saturated so as to become vehicles
for the incarnation of sense. As Cassirer writes (with Husserl in mind):
“Here we are not dealing with bare perceptive data, on which some sort of
apperceptive acts are later grafted, through which they are interpreted,
judged, transformed” (Cassirer 1985: 202). Each particular perception of
the world is both a “differential” – a distinct aspect of the world rendered
salient – as well as an “integral of experience,” by which Cassirer under-
stands, a concentrated apprehension of a totality of experience, as reflected
17
See R. Bernet, “Perception et herméneutique (Husserl, Cassirer et Heidegger),” in La vie du sujet
(Paris: PUF, 1994), 139–161.
and refracted along further horizons, or “lines,” of symbolic relations.
More trenchantly expressed, each perceptual experience of the world is a
“vector” of ever-diversified (and self-diversifying) experiences. Cassirer’s
insight here would prove inspirational and perhaps even foundational for
Merleau-Ponty. As the latter writes in The Visible and the Invisible: “this red
is what it is only by connecting up from its place with other reds about it,
with which it forms a constellation . . . a punctuation in the field of red
things, which includes the tiles of roof tops, the flags of gatekeepers and of
the Revolution . . . [red] is a concretion of visibility, it is not an atom”
(Merleau-Ponty 1968: 132).18 In Cassirer’s own description:
Here we are not dealing with bare perceptive data, on which some sort of
apperceptive acts are later grafted . . . rather, it is the perception itself
which by virtue of its own immanent organization, takes on a kind of
spiritual articulation . . . it is not only subsequently received into this
sphere but is, one might say, born into it. It is this ideal interwovenness,
this relatedness of the single perceptive phenomenon, given here and now,
to a characteristic total meaning that the term “pregnance” is meant to
designate. (Cassirer 1985: 202; my emphasis throughout)
Spirit or the authentic Proteus

Cassirer’s conceptions of symbolic form and “symbolic pregnance” mark
significant breaks with Cohen’s original reframing of Kant’s transcendental
enterprise in moving beyond the bounds of a transcendental logic into the
domain of perceptual consciousness. As Cassirer remarks, the concept of
“consciousness” is philosophically freighted with nearly insurmountable
difficulties, ambiguities, and equivocations (Cassirer 1985: 57); in its myriad
forms and conceptions, consciousness is the “authentic proteus of philo-
sophy.” Especially in light of Cohen’s critique of any psychological intru-
sion into Kant’s critical enterprise, the challenge of how to integrate the
problem of consciousness within a transcendental framework, without
relapsing into psychology, and thus forfeiting the methodological rigor
of Kant’s revolution in the manner of thinking, represented the Holy Grail
of philosophical thought towards the end of the nineteenth century.
18
Despite the strong imprint of The Philosophy of Symbolic Forms on Merleau-Ponty’s thinking, he
espouses a Heideggerian critique of Cassirer’s notion of symbolic form as a variation of “intellec-
tualism” for its apparent lack of reflection on the “ground of being.” As Merleau-Ponty, revealing
his attachment to the primacy of ontology, remarks: “If consciousness [and hence symbolic
functions] is placed outside of being, then consciousness can never be penetrated by being”
(Merleau-Ponty 2012: 126).
Numerous were the attempts to develop a genuinely philosophical approach
to consciousness that would avoid any conflation or confusion with a
descriptive psychology or cognitive science.
The significance of “symbolic pregnance” and, more generally,
Cassirer’s claim to a phenomenology of knowledge centrally situates the
problem of consciousness in The Philosophy of Symbolic Forms and its
transcendental enterprise. A proper inclusion of the problem of conscious-
ness within the transcendental framework of philosophy of symbolic forms
thus requires a fundamental transformation in the meaning of subjectivity.
“Spirit” (Geist) becomes for Cassirer the general marker for a new concep-
tion of consciousness (or subjectivity) that distances itself from the mean-
ing of the term “Spirit” in Hegel, while simultaneously guarding against
either a relapse into a naturalistic conception of consciousness or a
Husserlian notion of “pure consciousness.” “Spirit” marks a conception
of subjectivity as a totality, or whole (Ganzes), of different symbolic
functions of world-understanding (Weltverstehen). The imprint of
Hegel’s thought on Cassirer is nonetheless visible in his vigilance against
any illicit “substantialization” of Spirit into a substrate underlying the
development of history. The Life of Spirit is different ways of world
making. Such ways of world making are not semiotic constructions of the
world, as argued, for example, by Nelson Goodman.19 More robustly,
symbolic forms constitute the veritable appearance of the world, its “phe-
nomenality,” as it were, as an ordered cosmos of diversified and irreducible
sense. A transcendental mapping and exploration of the world – eine Reise
um die Welt – requires a new conception of subjectivity and, indeed, of
human existence, or life, as homo symbolicus.
Despite Cohen’s original stricture against the transcendental intellig-
ibility of the problem of consciousness, Cassirer finds its promising “direction
of questioning” within Kant’s First Critique, yet not in the Transcendental
Deduction (and its distinction between objective and subjective deduc-
tions), but more promisingly with the fundamental problem behind the
Doctrine of Transcendental Schematism. As Cassirer already argues in his
early work Das Erkenntnisproblem (1907), “the genuine theme of the theory
of schematism” is not the (transcendental) logical possibility of pure
concepts – the problem of the Transcendental Deduction – but their
“psychological possibility” (Cassirer 1971: 713). In fact, Kant in the First
Critique distinguishes between two problems: the question of the trans-
cendental deduction concerns the validity of pure concepts (categories of
19
N. Goodman, Ways of Worldmaking (Boston: Hackett, 1977).
the understanding). Yet, insofar as pure concepts of the understanding
must be considered as “general ideas” (i.e., as concepts), the argument of the
transcendental deduction remains insufficient as it presupposes the
question concerning the “applicability” of pure concepts. For this
question, Kant proposed his celebrated, if obscure, doctrine of the
schematism – a section that attracted the interest of Heidegger and
Cassirer in their clashing interpretations of Kant. The hidden source of
the “productive imagination” becomes the cardinal reference point within
Kant’s First Critique for Cassirer’s development of his conception
of symbolic forms as well as for the integration of the problem of
consciousness into his transcendental framework.
Leaving aside its complex genealogy, Cassirer further develops Kant’s
transcendental schema into the conception of symbolic functions as three
basic types of transcendental schemata, which are not to be considered as
images, or metaphors, or abstract generalities. A symbolic function is a
dynamic formation of sense and activity of the productive imagination that
grounds, or constitutes, the “applicability” of concepts. Concepts have a
grip on the world neither in terms of representation (image) nor as abstract
notions for particulars (I cannot here enter into Cassirer’s subtle critique of
abstraction, which exhibits many striking similarities with Husserl’s critique
of abstraction and an image-theory of concept in The Logical Investigations).
Concepts such as “time,” “space,” “number,” “person,” etc., can be sche-
matized in different ways, to wit, in the three symbolic forms of expression
(myth), representation (language), and pure concept (scientific cognition).
Under the concept of “symbolic form,” Cassirer aims to think at once the
(productive) imagination and the “world-understanding” (Verstehen) in a
transcendental frame through its historical pluralization. Symbolic forms
do not stand opposed or in contrast to “the world” but constitute the ways
in which the world manifests as well as the ways in which consciousness, or
spirit, historically manifests itself in the world.
Strictly speaking, the symbolic ways of world constitution are not the
singular achievements of a pure consciousness; the world as constituted
transcendentally is not “carried” in consciousness, but instead “flows
through” (durchflutet) consciousness “like a single stream of life”
(Cassirer 1985: 202). In this manner, symbolic forms do not span an
inner world of consciousness and the outer world of reality. In fact, as
Cassirer stresses, symbolic forms do not “bridge” a finished inner world and
a finished outer world since, on the contrary, symbolic forms originally
constitute the “separation and sorting out of the I and the world”(Cassirer
1996: 60). Symbolic forms are transcendental forms of tension, the original
“in-between” or “spacing,” of mind and world, giving sense and direction
to both. As Cassirer writes, a philosophy of symbolic forms is “concerned
with the dynamics of the giving of meaning (Dynamik der Sinngebung), in
and through which the growth and delimitation of specific spheres of being
and meaning occur in the first place” (Cassirer 1996: 4). In this respect, the
transcendental dynamic of symbolic forms is neither “active” nor “passive.”
Consciousness is not a necromancer that would enliven what is merely
given and “dead.” The original phenomenon of “symbolic pregnance,”
“without which neither an object nor a subject, neither a unity of the thing
nor a unity of the self would be given to us,” is medial in character. The
“mediality,” as it were, of “symbolic pregnance” forecloses the question of
whether a transcendental subject constitutes the world, if we are here to
understand subjectivity as a meaning-bestowal activity that could be iso-
lated (to be sure in reflection, i.e., Husserl’s phenomenology reflection)
from symbolic formation. There is no position exterior to symbolic for-
mation for a possible subject to occupy (and hence: no possibility for a
reduction to pure consciousness). To borrow an insight from Émile
Benveniste’s analysis of the medial voice, the medial form of activity
situates the subject within its process; subjectivity is born from the man-
ifestation of the world that it has constituted for itself.20 For this reason,
Cassirer insists that the discovery of “symbolic pregnance,” as anchoring a
“true phenomenology of perception,” reveals “a new angle on how the
analysis of consciousness can never lead back to absolute elements: it is
precisely the pure relation which governs the constitution of consciousness
and which stands out in it as a genuine a priori, an essentially first factor”
(Cassirer 1985: 203). As Cassirer announces: “the philosophy of symbolic
forms directs its regard not exclusively and not in the first place to pure
scientific, exact Weltbegreifen, but to all directions of Weltverstehens”
(Cassirer 1985: 16). As an alternative to Husserlian phenomenology,
Cassirer’s cosmological regard (Betrachtung) is directed not to the field of
“pure consciousness” in which the world is constituted but undertakes
instead a perpetual “voyage around the world, or the actual whole” (eine
Reise um die Welt, oder das wirkliche Ganze) with the aim of unceasingly
pursuing the forms of an authentic Proteus perpetually in search of itself.
20
Benveniste remarks: “L’actif alors n’est plus seulement l’absence du moyen, c’est bien un actif, une
production d’acte, révélant plus clairement encore la position extérieure du sujet relativement au
procès; et le moyen servira à définir le sujet comme intérieur au procès.” Émile Benveniste, “Actif
et moyen dans le verbe,” in Problèmes de linguistique générale, 1 (Paris: Gallimard, 1966), 168–175;
here 173.
part ii
Ethics and culture
chapter 5
Philosophy as philosophy of culture?

Christian Krijnen
Introduction
One of the basic ideas of Neo-Kantianism, especially of its Southwest School
(Wilhelm Windelband, Heinrich Rickert, Emil Lask, Bruno Bauch, Jonas
Cohn et al.), is that philosophy is philosophy of values, i.e. of determining
factors of human orientation, and as a comprehensive philosophy of values
philosophy turns out to be philosophy of culture. Whereas, on the one hand,
this idea of philosophy stems from a Fichte-inspired creative interpretation
and appropriation of Kant’s philosophy, rejuvenating his Copernican turn
in order to deal with problems the Neo-Kantians were facing in their time,
on the other hand, framing philosophy as philosophy of culture leads
to intricate problems concerning the Hegelian challenge of a system of
philosophy doing justice to the project of philosophy as self-knowledge, a
project the Neo-Kantians also adhere to.
In what follows, I will sketch and critically discuss the Neo-Kantian idea
that philosophy is philosophy of culture. In contemporary philosophy (and
beyond), this idea as such is attractive to many, as living in a postmodern
world seems to be characterized by a kind of plurality, expressed by notions
like, e.g., the multicultural society, that is in need of a philosophy concep-
tualized as a philosophy of culture rather than of a philosophy conceptua-
lized as a philosophy of an overarching “reason” or “idea.” Indeed,
Neo-Kantianism is what has recently been called the “serious beginning”
of the philosophy of culture as an academic discipline.1
The first section, “Neo-Kantianism: philosophy as philosophy of culture,”
develops the Neo-Kantian idea of philosophy as a philosophy of culture,
positioning this idea within a set of other ideas essential to Neo-Kantian
philosophy: ideas about, e.g., Kant and transcendental foundations, validity,
epistemology, worldviews, subjectivity, and the like. Taking Hegelian
1
R. Konersmann, “Einleitung,” in R. Konersmann (ed.), Handbuch Kulturphilosophie (Stuttgart and
Weimar: Metzler, 2012), 1–12; here 7.
111
112 christian krijnen
challenges into account, the second section will articulate three basic problems
that arise within the scope of philosophy as philosophy of culture: the relation
between self-knowledge and self-formation, the problem of practical values
within the whole of values, leading to the problem of a general concept of
actualizing validity, i.e., values, hence freedom.
Neo-Kantianism: philosophy as philosophy of culture

To get to grips with Neo-Kantianism as a philosophical movement, it is
important to see that Neo-Kantianism primarily understands philosophy
as a science of foundations. As such, Neo-Kantianism underscores the basic
intention of metaphysics to address fundamental questions concerning our
understanding of the world and ourselves.
From a historical point of view, Plato is an important sparring partner
for the Neo-Kantians.2 Plato showed that we can only understand the
foundations of both things and our knowledge of them if we assume ideas
that transcend sensible experience. The Neo-Kantians agreed with Plato
that philosophy should be idealism. However, although Plato tried to
understand ideas as principles for all that is, his classical metaphysical
position insufficiently differentiates between being and knowledge,
or ontology and logic. He understands ideas as themselves a type of
being, i.e., general, transcendent, non-sensible, and proper being.
Kant’s project of transcendental philosophy puts an end to such a reification
of ideas. The domain of philosophical foundations, ‘the transcendental’ so to
speak, is discovered to be a domain of principles that are the ground for the
validity of thought and action as such. These principles should not be under-
stood as a type of being, but rather as a whole of principles of validity. That is,
they must be seen as conditions that first enable and direct our thought and
action. Thus, principles are to be conceived of as preceding experience without
losing their intimate relation to experience. Put in more general terms, any
putative ontology presupposes a transcendental logic. For Kant, knowledge has
its ground in the cognitive relation which is defined in terms of the a priori
conditions that first make knowledge and the objects of knowledge possible.
The objective validity of these conditions lies in their function to enable
possible experience, not in their relation to a supersensible world.
2
See H. Holzhey, “Platon im Neukantianismus: Einleitung und Überblick,” in T. Kobusch and
B. Mojsisch (eds.), Platon in der abendländischen Geistesgeschichte (Darmstadt: Wissenschaftliche
Buchgesellschaft, 1997), 227–240; Karl-Heinz Lembeck, Platon in Marburg: Platonrezeption und
Philosophiegeschichtsphilosophie bei Cohen und Natorp (Würzburg: Königshausen und Neumann,
1994).
Philosophy as philosophy of culture? 113
This systematic link to the history of philosophy is only one aspect of Neo-
Kantianism. Another aspect concerns the fact that a philosopher never oper-
ates in a cultural vacuum, but is also always imbued with the spirit of his own
age. Neo-Kantianism also reacts to its cultural context and must be understood
from this perspective. The cultural or spiritual context of Neo-Kantianism is a
complex one, but I will highlight one important line of influence.
This line starts with Hegel’s death as a historical date that has symbolic
meaning for the history of philosophy: German Idealism had lost its
leading spiritual position in Germany. Henceforth, natural science, a
more historical orientation, realism, and the general “loss of illusions”
gradually came to dominate intellectual culture; this provoked a kind of
post-idealistic identity crisis. With Hegel’s death, his philosophy and the
Hegelian conception of the unity of facticity and meaning, of reason and
reality, had also faded. As a result, not only could the influential theme of
“worldview” (Weltanschauung), suggesting a situated perspective on total-
ity, spring up and become popular, but also all kinds of naturalism and
scientific reductionism, evoking loss of meaning, of the richness and depth
of life, sprouted. The ghost of nihilism, of a metaphysical void, dawned.
This spiritual background already points to Neo-Kantianism: Neo-
Kantianism tries to overcome the above-sketched post-idealistic gap between
“is” and “ought.” The situation becomes even more complex, as the empiri-
cal sciences appeared to be emancipated from philosophy and became
wholly independent. Hence, the question arose: why then still philosophy?
By the middle of the nineteenth century, marking out and making sense
of the field of properly philosophical investigations had become proble-
matic. This problem leads us directly to the beginning of Neo-Kantianism.
In reaction to the identity crisis of philosophy, Neo-Kantianism, both in its
early and its mature forms, makes a case for the rehabilitation of philoso-
phy. This rehabilitation starts with a clear commitment to epistemology
(Erkenntnistheorie) as the ultimate foundational discipline of both
philosophy and the other sciences. To be sure, this does not at all imply
a reduction of philosophy to epistemology. Neo-Kantian philosophy is
about culture in the broad sense, not just about knowledge and science in
the narrow sense. With respect to the rehabilitation of philosophy, the
Neo-Kantians, as the epithet suggests, return to Kant.
Of course, many regard Neo-Kantianism as primarily an epistemological
Kantian movement. There are plenty of reasons for doing so. Widespread
topical and methodological uncertainty in the universities led philosophers
such as Eduard Zeller and scientists like Hermann von Helmholtz to attempt
to provide philosophy with its own topic and its own method, while at the
same time discussing the methods and principles of the non-philosophical
sciences, which were developing ever so rapidly in their time. Such attempts led
to what at the end of the 1870s became known as the Marburg and Southwest
Schools of Neo-Kantianism. Fairly soon, these schools came to dominate the
epistemological debates of the nineteenth century. It is therefore not entirely
untrue to see Neo-Kantianism as primarily an epistemological movement.
However, more recent research on Neo-Kantianism suggests that this
view is responsible for much confusion about Neo-Kantianism. Especially
the cultural-philosophical nature of Neo-Kantianism as a reaction to a
crisis has, as a result, been insufficiently acknowledged. Recent research has
emphasized that questions regarding worldview were in fact the driving
force behind the “logical” preoccupations of the Neo-Kantians. Despite
the many differences in Neo-Kantian theories, it is a modern philosophy of
culture that unites the Marburg and Southwest Neo-Kantians, and the
interpretation of Kant’s philosophy is equally part of this goal.
The concept of “worldview” serves as an abbreviation for the problem of
the validity of values and hence points to the dispute about how culture is to
be shaped. Nihilism, the loss of faith in the rationality of the world and the
values assumed to be valid, is not only a major challenge to Neo-Kantianism,
but was also a concern of many other scientists and thinkers towards the end
of the nineteenth century, e.g. Bergson, Sorel, Durkheim, Dilthey, Weber,
Simmel, Michels, Mosca, and Pareto. To understand Neo-Kantianism
properly, however, it is important that the emphasis on the cultural-
philosophical aspect does not lead one to disregard the specific way in
which the Neo-Kantians put culture on the philosophical agenda. Not just
that Neo-Kantianism can be understood as a philosophy of culture and that
it understands itself as such, but also how it is to be seen as a philosophy of
culture is what makes up the peculiar nature, unity, the relation to Kant, and
the argumentative potential of Neo-Kantianism.
The labels “Neo-Kantianism” or “Critical philosophy” (Kritizismus) are
best restricted to the Marburg School – whose main representatives are
Hermann Cohen, Paul Natorp, and Ernst Cassirer – and the Southwest
German School, also called the Baden School or Heidelberg School – whose
protagonists were in particular Windelband, Rickert, Lask, and Bauch. Both
schools were formed at the end of the 1870s. The Marburg and Southwest
Schools represent the mature theories of Neo-Kantian philosophy.3
3
Of course, there also exist broader conceptions of Neo-Kantianism. Some identify as many as seven
sub-schools. Recently a discussion has arisen concerning the question of whether the division of Neo-
Kantianism proper should consist of three schools: the aforementioned two and a type of Neo-
Kantianism called “realistic critical philosophy” (realistischer Kritizismus). The latter includes
The famous dictum “Back to Kant,” originating with Otto Liebmann,4
one of the pathfinders for Neo-Kantianism, encapsulates in a concise
and programmatic way the determined recourse to Kant of the leading
Neo-Kantians. However, Kantian motives can be found not only in Neo-
Kantianism, but in almost every philosophical school of thought in the
nineteenth and twentieth centuries (at least in the continental tradition).
Therefore, an additional feature marking out Neo-Kantianism is needed.
Windelband formulated a dictum no less famous than that of Liebmann:
“To understand Kant rightly means to go beyond him” (Kant verstehen,
heißt über ihn hinausgehen) (Windelband 1921: iv). For the leading figures
of Neo-Kantianism, this dictum means that the return to Kant is not a
mere reproduction of his historical position; to understand Kant means to
further the development of philosophy with the help of Kant.
Not even the tendency to advance philosophy by Kantian means is
specific to Neo-Kantianism. Already German idealists such as Fichte and
Hegel were committed to this goal. Again, an additional feature is needed to
determine the specific nature of Neo-Kantianism. At the center of the efforts
of the Neo-Kantians, as indicated earlier, is the problem of validity (Geltung,
Gültigkeit). Taking the validity of our theoretical and non-theoretical –
practical, aesthetic, religious – endeavors as its theme constitutes the core
of Neo-Kantian philosophy. For Neo-Kantianism, philosophy is the theory
of validity, as insinuated in the above remarks on Plato and Kant.
In developing a theory of validity, the Neo-Kantians not only follow Plato’s
conviction that philosophy can only succeed as idealism, but at the same time
emphasize a fundamental aspect of Kant’s critical philosophy, namely that the
determinacy of human endeavors, being products of reason, is to be estab-
lished by means of a determination of the principles of their validity.
Neo-Kantians, thus, especially appreciate Kant’s insight into the pro-
blem of validity (cf. paradigmatically the quid juris issue in Kant’s
Transcendental Deduction of the categories).5 At the same time, they
find it important to further develop Kant’s concept of philosophy, rather
philosophers like Alois Riehl, Otto Liebmann, Richard Hönigswald, and Bruno Bauch. See, on this
subject, the dispute between Krijnen and Zeidler in C. Krijnen and K. W. Zeidler (eds.),
Wissenschaftsphilosophie im Neukantianismus (Würzburg: Königshausen & Neumann, 2014).
4
In his book Kant und die Epigonen in which he compared the German idealists, Herbart, Fries, and
Schopenhauer with the Critical philosophy of whose ‘“absoluteness” and “certainty” (Liebmann
1912: 13) he was convinced, Liebmann wrote at the end of each chapter: “Hence, we must return to
Kant” (Also muß auf Kant zurückgegangen werden).
5
The terminology used is in this respect only of secondary relevance. Terms like validity (Geltung,
Gültigkeit), value (Wert), meaning (Sinn, Bedeutung, Gehalt), justification (Rechtfertigung), founda-
tion (Grundlage, Grundlegung, Begründung) specify the general problem of validity.
than resort to metaphysical speculation as did, according to their opinion,
the German idealists, or regard the method of philosophy in terms of a
positivistic approach to the nature of validity. Neo-Kantians therefore not
only reactivate Kant’s contribution to philosophy; their aim is to renew it
in the light of a different constellation of philosophical problems from
Kant’s.
Against post-Kantian German Idealism, Neo-Kantians stress that
philosophy should not study things qua their being, but focus on the
validity of thinking things qua their being.6 In some respects one could
see Hegel’s logic as a development of Kant’s transcendental logic. On this
reading, Hegel’s analysis of the determinations of thought leads to a
fundamental set of a priori conditions. For the Neo-Kantians, however,
Hegel’s logical system is not just a whole of logical conditions for the
validity of thoughts; they reproach him for having conceptualized thought
as itself a metaphysical reality of spirit. In their view, Hegel contaminates
the radical foundations of modernity with classical metaphysics, hence
departing from the framework of Kantian transcendental philosophy.7
Anxious not to take the conditions of validity of thought of reality again
as itself a reality, the Neo-Kantians discriminate sharply between validity
and being. Hence, they try to correct the assumed metaphysical position of
Hegel by harking back to Kant’s critical arguments. According to the Neo-
Kantians, validity and being are related to each other in such a way that,
following Kant, being has its foundation in validity, and with that ontology
in epistemology. According to their understanding of Kant’s transcenden-
tal philosophical method, philosophy has the given as its point of depar-
ture, i.e., a concrete experience, or the fact (Faktum) of culture in order to
establish its principles; as Kant would put it: the conditions of its
possibility.
For the Neo-Kantians, as presumed successors of Kant, philosophy
does not take as its theme the world in terms of a direct relation to
objects, as do the non-philosophical sciences. They do not assume the “I”
or “consciousness” to be an empirical phenomenon, nor do they take the
relation between such empirical phenomena and the world to be a
philosophical topic. Rather, philosophy aims at determining the validity
structure of experience. Time and again, Neo-Kantians criticize all kinds
of metaphysical, psychological, physiological and what nowadays are
6
See e.g. famous statements of Cohen 2001b: 27f. and Cassirer 1971: 662.
7
See on Hegel and Neo-Kantianism recently C. Krijnen, Philosophie als System: Prinzipientheoretische
Untersuchungen zum Systemgedanken bei Hegel, im Neukantianismus und in der Gegenwartsphilosophie
(Würzburg: Königshausen & Neumann, 2008).
called (neo-)structuralist and evolutionary-biological conceptualizations
of epistemology, or, more comprehensively, of the philosophy of culture.
Such attempts understand knowledge (and other human endeavors) as
an ontic relationship. According to the Neo-Kantians, these deficient
conceptualizations, including their agnostic and relativistic implications,
deprive epistemology of its fundamental theme: the validity of knowledge.
Neo-Kantians exclude the empirical subject and its anthropological and
metaphysical connotations from study insofar as they primarily focus on a
“pure subject.” This subject, in the sense of the whole of the principles of
validity (a priori structures, values, etc.), is understood as the foundation of
all that can be valid, and hence as the ground for the possibility
of objectivity. This strategy discriminates sharply between a “pure” subject
as foundation of objectivity and an “empirical” subject which is grounded
on that normative foundation. On the other hand, the Neo-Kantians try to
overcome what they consider to be certain exaggerated positions, or naïve
objectivist worldviews be they called, for example, naturalism, materialism,
psychologism, empiricism, positivism, logicism, fideism, historicism,
Lebensphilosophie, or nihilism.
As in theoretical philosophy, the relation between the unconditional
norm of the pure subject and its conditional fulfillment by the empirical
subject equally plays a central role in the philosophy of culture. This
proportional relation of validity makes clear that the duality of facticity
and meaning, of reason and reality, of “is”and “ought,” elaborated on in
the chapter on the “spiritual background,” is grounded on premises that
turn out to be false. We may illustrate this through Heinrich Rickert’s
concept of “meaning.” Meaning is conceptualized as the recognition by the
finite rational being called “man” of unconditionally valid theoretic and
non-theoretic values. The reciprocal relation of implication and the one-
sided relation of foundation between the norm and that for which the
norm is, absolute demand and finite fulfillment, principle and the
concrete, entails that the human production of meaning is characterized
by finitude. Therefore, the Neo-Kantians deny that the common duality of
subject and object, as between an empirical subject and an inner or outer
world, is fundamental to epistemology. They develop another kind of
relationship that not only turns out to be more fundamental, but also
proves to be of great importance to the development of a philosophy of
culture.
Starting philosophical analysis with given cultural phenomena, i.e.,
spheres of culture containing objective validity claims, does not imply
that the premise of the analysis is a Faktum that is stipulated dogmatically
as valid.8 Rather, the analysis takes such facta as problematic, as a validity
claim that is in need of philosophical determination and evaluation.
According to the Neo-Kantians’ understanding of the method of trans-
cendental philosophy, the original determinacy of the different spheres of
culture is to be known via an oblique, validity-reflexive disclosure of the
constituents of meaning of those spheres of culture, i.e., of the principles of
validity of those claims.
Like Kant, in order to provide a conceptual account of “the world of
man,” the Neo-Kantians take their standpoint in culture in terms of a system
of meaning. With this they aim to show reason itself to be the governing
principle of our world, of culture. Take the case of the Southwest School of
Neo-Kantianism. Unlike the Marburg School, the Southwest School does
not fall victim to the “intellectualist” narrow focus of Cohen, who initially
restricted philosophical analysis mainly to the cultural fact of scientific
knowledge, although in a later phase Natorp and Cassirer broadened the
scope of philosophy. From the start, the Southwest School takes culture in its
widest sense, striving for its philosophical comprehension.
The Southwest School conceives of culture as determined by values.
From a philosophical point of view, what is called theoretical culture
(“knowledge”) has a logical and a systemic primacy. Already in theoretical
philosophy, it turns out that theoretical culture rests on a whole of
theoretical values (a priori structures, principles) that determines the
validity of theoretical endeavors. The values that comprise the value
‘truth’ ought to be normative for the thoughts of empirical subjects in
order to assure that their thought truly is knowledge of objects, i.e., that
thought is objective. This logical relation within the realm of theoretical
culture is then transported to other spheres of culture: these too consist of
subjects who acknowledge values. In this sense, the Southwest School Neo-
Kantians propagate the primacy of “ought” (Sollen), a primacy of practical
reason in its most radical, and not just in its practical sense, namely in the
sense that it encompasses all dimensions of reason.9 They propagate a
philosophy of values as a philosophy of culture.
Numerous historical and systematic studies from the first decades of the
twentieth century make clear that the Southwest School is to be seen as a
comprehensive philosophy of values. But they also make clear that the
8
For the Neo-Kantian doctrine of the Faktum see Krijnen, Philosophie als System, 1.3.
9
See for the “primacy of practical reason,” e.g. Rickert 1899a: 44; 1909: 215f.; 1928: 309ff., 437 and
Windelband 1921: 287. See on this doctrine C. Krijnen, Nachmetaphysischer Sinn: Eine problemgeschicht-
liche und systematische Studie zu den Prinzipien der Wertphilosophie Heinrich Rickerts (Würzburg:
Königshausen & Neumann, 2001), 7.2.3.1.
concept of value is a fundamental concept: philosophy is essentially about
values. The idea of living through a metaphysical crisis fits well with this
systematic perception. After all, the exploration of values should contribute
to the overcoming of the post-German-idealist divide between values and
reality that threatened to make human orientation both practically impos-
sible and theoretically incomprehensible. The philosophy of values acts
against the culture of nihilism by showing that there are values that are
objectively valid.
Hence, the concept of “value” – and closely related concepts like “mean-
ing,” “ought,” and “validity” – has a meaning that goes far beyond its
methodological function in the constitution of the subject matters of arts
and humanities. It points to the aforementioned metaphysical dimension
that contains the grounds of our thoughts and action. The debate is thus
not so much about the validity and status of some traditional values.
Rather, against the background of the post-idealist conception of reality
as value-free and without meaning, the debate focuses primarily on the
foundations of our understanding of the world and ourselves. Values
traditionally treated by metaphysics, such as truth and morality, unity
and plurality, value and reality, function as a framework to enable our
understanding of, and dealings with, the world. Hence, the philosophy of
values operates against the background of nihilism and aims at elucidating
the principles of human existence and the world that humans live in.
In conclusion: the main schools of Neo-Kantianism take the basic
problem of philosophy to be that of the validity of our theoretical and
non-theoretical endeavors. This problem is to be solved through a deter-
mination of the principles of validity. These principles are what make up
the sphere of the “transcendental.” The transcendental domain, therefore,
is not to be confused with the psychology of an empirical subject or with
the metaphysics of an absolute reality. Far from declaring the world we live
in to be meaningless, the Neo-Kantians aim to bring to light its philoso-
phical foundations. Hence, they try to understand the rationality of our
world and its meaning. The concept of culture functions as a universal and
fundamental framework, a framework once occupied by metaphysics. This
framework is now freed from ontological premises yet is still able to
counteract nihilism.
Three Hegelian challenges

Of course, Neo-Kantianism in no way restricts itself to a preoccupation
with Kant, but seeks to tie into and build on the great tradition of idealist
philosophy. In addition to Kant, studies on Plato (e.g., by Windelband
and Natorp), Descartes and Leibniz (e.g., by Cassirer), Fichte (e.g., by
Windelband, Rickert, Lask, Bauch), and Hegel also appear. Windelband
famously pled outright for a renewal of Hegelianism, and Hegel scholars
like Hermann Glockner and Richard Kroner belonged to Rickert’s circle in
Heidelberg (Windelband 1915: 273–289). Rickert characterizes the
Heidelberg tradition of philosophy in terms of “logical education and
historical Bildung” (Rickert 1924b: iv).
Indeed, as we have seen, against the sketched spiritual background, Neo-
Kantianism was aiming for a new foundation of philosophy, appropriating
and updating Kant as the philosopher of modern culture. Hegel, for sure,
had the same kind of project. Kant’s philosophy is the common back-
ground of both philosophies. In line with their conviction of both Kant’s
relevance and the need for it to be updated in the light of a different
constellation of philosophical problems, containing the development of
German idealism too, the Neo-Kantians had to accomplish a synthesis
between Kant and Hegel. Do they succeed?
To give an answer on the timeless relevant question about a possible
synthesis between Kant and Hegel, I will concentrate on the Southwest
School and discuss three Hegelian challenges the Neo-Kantian idea of
philosophy as a philosophy of culture has to face. These challenges concern
the relation between self-formation and self-knowledge, the determinacy of
the practical, and finally the concept of realizing freedom. To discuss them, it
is illuminating to elaborate on further essentials of the Southwest School’s
philosophy of culture, taking features of Hegel’s concept of philosophy as
philosophy not of culture, but of the (absolute) idea into account. According
to Hegel’s concept of philosophy, philosophy takes the shape of a monism of
the idea that develops itself through the elements of logic, nature, and spirit.
This concept of philosophy also concerns the distinction between theoretical
and practical philosophy, a distinction that was very influential in the history
of philosophy, not at least for Kant, and became highly problematic after
Kant both in German idealism and Neo-Kantianism.
For the Southwest Neo-Kantians, as for Hegel, the traditional distinc-
tion between the theoretical and the practical is inadequate. The Southwest
Neo-Kantians claim to identify a fundamental axiological relation (axio-
tisches Grundverhältnis) in which reason itself is related to theoretical and
practical activity. Consequently, the specific link between activity and the
practical is dissolved. The fundamental discipline for this operation is the
Neo-Kantian theory of knowledge. In Neo-Kantianism, theory of know-
ledge is philosophia prima. According to Rickert, the object of knowledge
qua criterion is not some kind of being, but must be conceived as an ought,
and hence as a value. The knowing subject, then, proves to be a subject that
“takes a position” (Stellung nehmen), and consequently it is a subject who
“recognizes” (anerkennen) values: values are, from the perspective of the
subject, the “what,” the “object,” the “goal” of taking any position: know-
ing is taking a position towards values. As it is put in contemporary theories
of inferential semantics: subjects follow rules. Kant would say – and not
without reason, this is the motto of Rickert’s book about the Gegenstand
der Erkenntnis: the relation of our representations to an object is based on a
rule.10
On the basis of such insights, the Neo-Kantian activist theory of know-
ledge is made fruitful axiologically: it contains principles with relevance
far beyond the realm of knowledge; they acquire a universal function,
determining the whole system of philosophy. On the Neo-Kantian view,
“theoretical” culture contains fundamental relations which are philosophi-
cally paradigmatic insofar as they recur as basic structures in all cultural
realms. Making fundamental theoretical relations fruitful axiologically,
hence paradigmatically, leads Rickert to a relation that he characterizes as
the “starting point” and “common root” of all philosophy: the “correlation
between valid values and the valuing subject” (Rickert 1928: 438). As a
consequence, any meaningful structure (Sinngebilde), and hence culture, is
characterized by a subject that is related to values guiding his actions. By
recognizing values, the subject shapes culture. All philosophical disciplines,
then, treat values (specific orientations for subjects, more precisely: for
humans as rational beings) and their realization by subjects who recognize
them, thus producing “cultural goods” (Kulturgüter).
The theory of knowledge shows that theoretical meaning – that is,
meaning constituted by the value ‘truth’ and the subjects valuing this
value – contains an objective or noematic aspect and a subjective or noetic
aspect. Therefore, recognition can become a fundamental logical concept,
obtaining system-axiological relevance. As Hegel might have put it, the
fundamental axiological relation of Southwest Neo-Kantianism has both
a logical aspect and, on that basis, an aspect of the philosophy of reality,
more precisely: of spirit. These Neo-Kantians, like Hegel, regard logic or
theory of knowledge as an objective logic, as a logic of “comprehending
thought” (begreifendes Denken) (Hegel 1951: i, 23); they address thought as a
10
Kant 1998: [B 242]. “If we enquire what new character relation to an object confers upon our
representations, what dignity they thereby acquire, we find that it results only in subjecting the
representations to a rule, and so in necessitating us to connect them in some one specific manner.”
principle of objectivity. Nevertheless, there are significant differences
between Hegel’s view and those of the Southwest Neo-Kantians, starting
already with the logical basis of the fundamental axiological relation,
especially concerning the “idea of knowledge” and the “absolute idea,” as
we will see in what follows.
Considered from the perspective of the system of philosophy, by using
basic concepts of logic, the Southwest Neo-Kantians identify a fundamen-
tal axiological relation which consists in the self-formation of the subject:
the founding principle systemically ordering all the subject’s activities
regarding values (Wertverhalten), and hence all realms of values, is the
self-formation of the concrete subject by values that are valid “transcen-
dentally,” that is to say, their validity does not depend on the subject; on
the contrary, such values qualify the subject’s subjectivity.11 Rickert con-
ceives self-formation as a relation between subjectivity and conditional
fulfillment on the one side, and objectivity, an unconditional task, on the
other. The ordering of the system, then, is guided by various possibilities
for completing this unconditional task (Vollendung).
Hegel also affirms a fundamental logical relation with systemic implica-
tions: a relation of the subject’s self-knowledge qua self-realization of the
concept, hence of the absolute idea as the one and only content and object of
philosophy (Hegel 1951: i, 484), i.e. the “concept which comprehends itself”
(sichbegreifende Begriff ) (Hegel 1951: ii, 504), the “absolute truth and all
truth” (Hegel 1991: §236; 1951: ii, 484). This suggests a similarity between the
Neo-Kantian’s fundamental axiological relation and Hegel’s doctrine of the
idea. Whereas the Neo-Kantian analysis of knowledge presents relations
which prove to be paradigmatic for non-cognitive realms, Hegel’s idea
develops itself speculatively into itself, first in logic and then in reality,
which is founded in the idea, manifesting itself until it knows itself as spirit.12
Notice that the Neo-Kantian doctrine of the fundamental axiological rela-
tion seems to involve a distance between subject and value (i.e. the normative
11
Therefore it makes no sense to blame Southwest Neo-Kantianism for locating values in a quasi
“Platonic sky of values,” as has been charged from the beginning until now (see e.g. recently the
criticism of U. Renz, “Cassirer und der Neukantianismus,” in Konersmann (ed.), Handbuch
Kulturphilosophie, 114–119; here 116). For sure, there is a significant difference concerning the concept
of a historically relativized a priori in the Marburg School and the Southwest School; but one has to
take into account that for Southwest Neo-Kantianism, values are intrinsically related to the subject
and its intentions. As far as the historically dynamic dimension of them is concerned, there has even
been an important discussion between Bruno Bauch and Paul Natorp, focusing on the open and
closed character of the system of philosophy – which unfortunately has hardly been taken note of by
contemporary scholars.
12
See Krijnen, Philosophie als System, 4.2.3f.
factor), unlike Hegel’s logical doctrine of the idea. However, Hegel empha-
sizes this distance in his philosophy of reality, more precisely, in his philo-
sophy of spirit. Here the constellation of a “subject” which subordinates itself
to “values” and so shapes “culture” has its proper place. Within the context of
his determination of practical spirit, Hegel even comes to a positive assess-
ment of the “ought” (Sollen), which he had criticized so sharply at the level of
the logical idea. In the development of the practical spirit, it becomes clear
that and how spirit, by “giving itself its own content” (Hegel 1991: §469),
contains a “doubled ought” (gedoppeltes Sollen) in its self-determination
(Hegel 1991: §470). In his way, Hegel etches the relation of spirit to validity,
progressing from conditional to unconditional formation of the spirit.
The fundamental axiological relation is the basic constellation of the
Southwest Neo-Kantian system of philosophy; it is a relation of self-
formation. Although the Neo-Kantians regard theory of knowledge as
first philosophy, they do not systemically articulate this self-formation as
self-knowledge: self-knowledge as the fundamental constellation of the
development of the system of philosophy. Hegel, on the contrary, does so: he
achieves the required return into the concept in the absolute idea. Such a
return into the concept is indispensable, regardless of whether Hegel’s or
the Neo-Kantian conception of knowledge would require logical revision
(which would have, of course, numerous implications for the internal
relations of knowledge and the system). The required return into the
concept is part of the knowledge claim of philosophy: philosophy is the
science of “totality.” The speculative movement of the concept rules out
any Neo-Kantian, axiologized fundamental logical relation, which elabo-
rates theoretical philosophy as one cultural realm among others: Hegel
advances a progress of the self-determination of the idea, from the logical
into the real, and through the real back into itself.
This problem of the division or ordering of an idealist philosophical
system, i.e. from the perspective of Southwest Neo-Kantianism the system
of culture, is only one of many which arise when comparing Kantian and
Hegelian conceptions of normativity. Another problem concerns the
determinacy of the practical. Properly speaking, Hegel has no practical
philosophy; instead he transforms the opposition between theoretical
and practical philosophy through the absolute idea and free spirit.
Whereas Hegel’s option raises the question: how is practical philosophy
possible once it has been idealized within his philosophical system?, the
Southwest Neo-Kantian option raises a question about the particular
determinacy of the practical: what can practical philosophy be once it is
axiologized?
The fundamental axiological relation integrates “theoretical” and “prac-
tical” reason. Because knowledge includes “recognizing values,” and so is a
kind of “practical” behavior,13 Rickert thinks this constellation overcomes
the traditional position, in which theory and practice, or more generally,
theoretical and practical values, were opposed (Rickert 1928: 438; 1929a:
689). This axiologizing of the sphere of knowledge entails that traditional
concepts of “practical” philosophy are transformed axiologically, thus
becoming fundamental concepts of philosophy as such, and hence, of all
philosophical disciplines. This includes such concepts as autonomy, duty,
conscience, etc.; they turn out to concern the validity-noetic aspect of the
axiological relation, their “immanent meaning” (immanenter Sinn).14
This approach faces a problem about the particular character of practical
philosophy. Considering carefully the treatment of the practical within the
Southwest School makes clear that the project of transforming practical
reason into an axiological foundation is very difficult. It also makes clear
that the relation between theoretical and practical reason recurs in a
sublimated form – despite Neo-Kantian efforts to universalize practical
reason. Consequently, the primacy granted to practical reason loses its
particular practical character: it is replaced by a primacy of self-formation,
which appears in various kinds of self-formation (knowledge, morality,
religion, etc.), and so in various “cultural realms” (Kulturgebiete). Thus, the
systemic construction of Neo-Kantian philosophy of culture again raises
the question: Is the determination of the totality as the ground of every-
thing, i.e. the topic of philosophy, best conceived as a whole of well-
ordered constellations of self-formation, or as a whole of well-ordered
constellations of self-knowledge?
This confronts us with a third and final problem, intrinsically related to
the former two. It concerns the idea of realizing (actualizing) validity –
hence reason or freedom – within the Southwest Neo-Kantian conception
of philosophy as philosophy of culture. Their Fichte-inspired doctrine of
the primacy of practical reason leads to a fundamental axiological relation
that aims to overcome the traditional, and for Kant too, essential architec-
tonic of theoretical and practical reason. Yet, the structure “theoretical–
practical” returns in a sublated form, modeling the practical eventually as
the dimension of realizing (actualizing) validity. This model, however, falls
victim to an equivocal use of the concept of realization and, with that, of the
subject. The subject functions on the one hand as logical factor that
13
See Rickert 1914: 208; 1928: 185f., 434, 438; Rickert 1929a: 689ff.
14
See Rickert 1911: 161; 1914: 209; 1921: 309f.; 1928: 435ff.; 1929a: 690, 694; 1934: 179ff.
realizes, executes, and individuates validity (values, reason); hence, the
subject is taken here as an intentional subject. On the other hand, the
subject functions as the factor that realizes validity; hence, in this respect
the subject is taken as a person, a real, concrete, active, acting subject,
producing cultural goods.
Especially in Rickert’s philosophy of culture, this difference between
both dimensions of subjectivity – of intentionality and actualizing, realiz-
ing values (Wertrealisierung) – is smudged.15 Rickert insufficiently reflects
the distinction, which is important from a systematic philosophical per-
spective, between the subject as an intentional capacity related to values
and the subject as an actualizing capacity related to values. Therefore, the
relationship between the act of taking a position (Akt der Stellungnahme)
and action (Handlung) also remains underdetermined.
Apparently, actualizing validity implies more fundamental relation-
ships, i.e., presuppositions which remain implicit in the Southwest
model of philosophy as a system of values, instead of being made explicit.
One of these relationships indeed has been treated rather differently in
post-Neo-Kantian transcendental philosophy, and in Hegel’s too. I mean
the Neo-Kantian distinction of different levels of values or validity: intrin-
sic values, i.e., autonomous, unconditional, cultural values, objective values
(Eigenwerte, autonome, unbedingte Werte, Kulturwerte [in the narrow
sense], objective Werte), versus extrinsic values, i.e., conditional values,
values of life, values of civilization, subjective values (Bedingungswerte,
Lebenswerte, Zivilisationswerte, subjective Werte).16 According to this dis-
tinction, the extrinsic values function in essence as realizing conditions of
the intrinsic values. Therefore, extrinsic values are characterized by their
instrumental meaning: they are only means for realizing intrinsic values.
At a first glance, this influential and in those days culturally widespread
distinction looks convincing. At a second glance, however, it turns out to
be philosophically highly problematic. One of its implications is that in
Neo-Kantianism the value ‘utility’ is not only being positioned on the level
of extrinsic values, but also bears a negative connotation. Post-Neo-
Kantian transcendental philosophers like Hans Wagner and Werner
Flach, on the contrary, conceptualize the value or idea of the utility and
15
Of course, Rickert is aware of their difference, e.g. when he distinguishes action (Handlung) from the
act of taking position (Stellungnahme) (Rickert 1921: 324), as well as social personal freedom from the
free autonomous act (Rickert 1921: 331), or when he conceptualizes “real” knowledge as dependent on
the will of the “individual” to know, “real” theoretical knowledge as dependent on a real will (Rickert
1928: 309f.).
16
See, paradigmatically, Rickert’s approach: Rickert 1911; 1921: 132ff.; 1928: 235; 1934: 167ff., 170ff.
the cultural realm it constitutes, the realm of “social-economic” phenom-
ena (das Ökonomisch-Soziale) as an autonomous sphere of values.17 Hegel
conceptualizes this idea in the context of his philosophy of objective
spirit,18 hence as a figure of the free spirit and therefore rejecting the
Neo-Kantian opposition between intrinsic and extrinsic values.
On top of that, the opposition between the two value types is compli-
cated by the fact that realizing validity in Neo-Kantianism is also
conceptualized in terms of a division of the system of philosophy in a
realm of validity on the one hand, and of realizing validity on the other.
Hegel’s idealism, conceptualized as a development of the monism of the
idea that knows itself, seems to overcome the abstractness between the
realms of values (reason) which characterizes the approach of a Kantian
type of transcendental philosophy; it enables a uniform development while
at the same time leading to a differentiated concept of freedom and its
manifestations in the world we live in. With that, a sustainable concept of
realizing validity comes into sight. Undoubtedly, for Neo-Kantianism,
Kant’s architectonic of reason, dividing reason into a theoretical and a
practical usage, turns out to be deficient. It requires a grounding by more
basic constellations, leading to a fundamental axiotic relation. Still, within
Neo-Kantianism, Kant’s architectonics remains a burdensome legacy.
17
See H. Wagner, Philosophie und Reflexion (Munich and Basel: Reinhardt, 1980), § 25f.; W. Flach,
Grundzüge der Ideenlehre. Die Themen der Selbstgestaltung des Menschen und seiner Welt, der Kultur
(Würzburg: Königshausen & Neumann, 1997), 61ff.
18
See Hegel 1991: §§ 483ff.; 1955.
chapter 6
The validity of norms in Neo-Kantian ethics

Beatrice Centi
Introduction
This chapter presents the principal features of Neo-Kantian ethics, paying
particular attention to those features common to the main representatives
of both the Marburg and Baden schools. As the analysis will show, for the
Neo-Kantians, a reflection on ethics implies a reflection on the nature and
task of philosophy in general. Neo-Kantian philosophers worked exten-
sively on this issue, given that, in their time, the tumultuous development
of specialized sciences seemingly took away from philosophy every auton-
omous domain for inquiry. Thus, for the Neo-Kantians, ethics is not
merely a part or a subfield of philosophy, but rather, to the extent that
philosophy is a unitary science that cannot be divided into parts, ethics is a
way of presenting philosophy as a whole.
This unitary and systematic character of philosophy, however, is made
problematic by two seemingly opposed requirements. On the one hand,
Neo-Kantian philosophy as a systematic science sets out to grasp and
conceptualize the universal. On the other, as is well known, Neo-
Kantian philosophy wants to come forward as a philosophy of culture,
and as such it has to tackle what is inherently manifold and particular.
Ethics thus becomes a privileged field of inquiry, in that it displays in a
particularly conspicuous manner the overarching problem of mediating
between the universal and the particular, i.e., between a priori lawfulness
and determinate individual reality. Ethics for the Neo-Kantians is not a
separate theoretical field to be distinguished from the theory of knowledge.
It is rather the context in which the theoretical foundations obtained
through epistemological inquiry are put to the test and where they find
their definitive clarification. Needless to say, this very idiosyncratic angle
makes Neo-Kantian ethics appear strange and even obscure to modern
readers who are most likely used to understanding ethics as a reflection on
specific moral puzzles or dilemmas. For precisely this reason it is advisable
127
128 beatrice centi
to be particularly cautious in approaching Neo-Kantian ethics and it is
imperative to appreciate its theoretical import in the broader context of
Neo-Kantian philosophy as a whole. This means contextualizing the Neo-
Kantian reflection on ethics in the broader preoccupation with the issue of
Geltung, or validity, which constitutes the theoretical nucleus of Neo-
Kantianism as a whole. What is the relationship between the validity of
norms and the irreducible individuality of the materials that ought to stand
under these norms? How can the individual remain individual and none-
theless answer to the universal?
The Neo-Kantian solution to the apparent dichotomy of the individual
and the universal consists in exhibiting a kind of lawfulness that is both the
origin and the normative end-goal of what is individual. Individual and
universal are thus no longer construed as two separate dimensions that
philosophical reflection has to somehow reconcile, but rather as two sides
of one and the same dynamic process, which philosophy has to lay bare.
In this project, the dimension of lawfulness as the Sollen, or normativity,
presents itself in different, yet coordinated guises. The Sollen plays itself out
in concepts (understood as the cognitive tools we employ to make sense of the
world), in categories (understood as the forms in which the individual materi-
als of the world stand and through which they receive their distinctive
structure), and as ideas (understood as the norm that governs reality, and
thus guides our normative projects in the world). In other words, the Neo-
Kantians explore several dimensions: those of lawfulness, as comprising the
instruments we use to make sense of things, the categorical structure in which
things stand, and the normative ideal that indicates how things ought to be.
The Sollen, thus, consists in a kind of lawfulness that regulates objective
reality, which is knowable according to precise relations and discernible
historical and cultural configurations. This is why Neo-Kantianism as a
whole focuses on concept-formation and on the constitution of cognitive
forms through which alone reality can be “given.” Arguably, Neo-
Kantianism can be characterized as a unitary (albeit internally differen-
tiated) philosophical movement precisely by the following thesis: There is
no reality outside the cultural and cognitive forms through which we
determine reality as such, using the theoretical tools of the natural and
human sciences, including ethics.1
1
H. Holzhey, Ursprung und Einheit. Die Geschichte der ‹Marburger Schule› als Auseinandersetzung um
die Logik des Denkens (Basel and Stuttgart: Schwabe & Co. Verlag, 1986), 48–49, 126–127, 185, 202ff.
For a different perspective see J. Stolzenberg, Ursprung und System. Probleme der Begründung system-
atischer Philosophie im Werk Hermann Cohens, Paul Natorps und beim frühen Martin Heidegger
(Göttingen: Vandenhoeck & Ruprecht, 1995), 8–16, 118–120.
The validity of norms in Neo-Kantian ethics 129
In the ensuing three sections, I attempt to reconstruct the ethical inquiry
of three illustrious exponents of Neo-Kantianism: Hermann Cohen,
Heinrich Rickert, and Paul Natorp. In particular, I set out to highlight
how in each of the three thinkers the theoretical motifs outlined above are
realized. To reiterate these motifs: (1) the relationship between ethics and
philosophy as a systematic whole; (2) the problem of the relationship
between individual and universal; (3) the centrality of Geltung and its
impact on the problems of ethics.
Ought and being in Cohen

The sphere of ethics and the notion of ought (Sollen) becomes funda-
mental for Neo-Kantianism beginning with the work of Wilhelm
Windelband as well as Hermann Cohen’s Kants Begründung der Ethik
(1877). The Neo-Kantian reading of Kant shows how the various uses of
conceptual forms and the different functions of judgment delineate
different spheres of reality. The very same ideas (God, soul, and the
world as a whole), for instance, are interpreted to have a purely regulative
function when they are applied in theoretical thought, but, at the same
time, they have a constitutive function in that they determine the
distinctive objects of practical reason. In this second application, ideas
“determine” what counts as real in the sphere of ethics. The reflection on
the regulative and constitutive use of ideas thus goes hand in hand with
the systematic articulation of philosophy.2 It yields a methodological tool
to (1) articulate the various spheres of Sein (being) – those of Wirklichkeit
(reality) or Dasein, Existenz (existence) and those of Unwirklichkeit
(irreality) or Geltung, Wert (validity, value), and (2) to lead back the
multiplicity of reality to the unity of law.
Thus, for instance, Cohen’s investigation of the determining function
of judgment in Kants Theorie der Erfahrung (1871) leads him to recognize
the importance of the notion of limit in Kant’s theory of reason. This
recognition has a bearing on his ethical work, in that the reflection on the
regulative function of ideas brings to light a kind of lawfulness that opens
up a new dimension of reality. Through the regulative use of practical
reason in theoretical thinking, the limit between the conditioned and the
unconditioned becomes the condition for possibility of a domain of reality
2
For an accurate analysis see A. Stadler, Kants Teleologie und ihre erkenntnistheoretische Bedeutung
(Berlin: Dümmler, 1874). Cohen valued Stadler’s work greatly; see G. Gigliotti, Hermann Cohen e la
fondazione kantiana dell’etica (Florence: Sansoni, 1977), 37–57.
130 beatrice centi
other than physical nature that nonetheless displays a specific kind of
lawfulness and necessity. This is the theme of Cohen’s first ethical work.
Besides the realm of empirical being, the one explored by scientific
research, there lies a “realm, in which the real is what ought to be”
(Cohen 2001b: 133). The Sollen is thus taken to constitute a domain of
reality in its own right, an authentic sphere of being. The Sollen is the
reality that is not yet, but ought to be in accordance with an a priori law. In
this sense, Cohen promotes an ontological pluralism, which opposes the
reduction of reality to the theoretical domain pertaining to the natural
sciences.
Therefore, for Cohen, Plato’s “bold and ambiguous phrase ‘beyond
being’ (ἐπέχεινα τῆς οὑσίας) defines the problem of ethics in its connection
with the investigations concerning the concept of being” (Cohen 2001b:
1–2). Accordingly: “the issue of the reality, of the kind of reality of what is
moral [Realität des Sittlichen] has to undergo methodological scrutiny”
(Cohen 2001b: 11). The status of ethics must then be investigated within
the domain of being, and yet beyond the limit of what “being” is taken to
be in theoretical inquiry. The horizon of what counts as“real” or “being”
thus has to be broadened. Without a reference to some “being,” to some-
thing “real,” no object would be available to know or cognize. Therefore,
ethics investigates precisely “what the ought is. For this reason, no less than
any other branch of transcendental inquiry, ethics has a ‘being’ as its task:
ethics has to determine the being of the ought [das Seiende des Sollens]”
(Cohen 2001b: 138). The retrieval of the Kantian notion of pure a priori
normativity is thus a necessary but still insufficient theoretical step to
determine the domain of being pertaining to ethics. The foundation of
ethics that Cohen proposes is based on an original connection of the
Kantian idea of limit and Plato’s theory of ideas. The idea exerts a
normative and projective function as a condition of possibility or trans-
cendental regulation, which institutes a specific sphere of reality. The
Sollen is thus both (1) a connotation of the fundamental modal category
of necessity, which expresses lawfulness, and (2) a connotation of being.
For Cohen, thus, ethics needs to have an object of its own, belonging to
the sphere of what is real and thinkable, as he explains in Kants Begründung
der Ethik (1877). Accordingly, ethics has to be understood as an autono-
mous field of cognition parallel to logic, i.e., similarly to logic, ethics has to
be a domain structured around some kind of validity (Geltung). In a way
that anticipates Rickert’s work (as we shall see in the next section), for
Cohen validity indicates a connotation of reality that is irreducible to and
independent from the sphere of empirical existence. This is the case with
logical truth as a value, with ethical truth as relating to the good, and with
aesthetic truth as relating to the beautiful. In particular, this conception
plays out in ethics as the delineation of the limit concept of an ethical-
rational being. Such a limit concept, which is inseparable from the assump-
tion of an ethical standpoint, transcends the empirical sphere of what is
merely human. “The ethical (das Sittliche) is a problem of reality, not
exclusively of human beings, but rather of the reality of a rational being . . .
If human beings did not exist, the ethical would nonetheless be real”
(Cohen 2001b: 162), to the extent that it is thinkable as a necessary reality.
The autonomy of the sphere of ethics, however, does not rest upon a
chasm between the purely rational and universal domain of thought and
the specifically human domain, which is characterized by its historical and
cultural contingence. According to Cohen, ethics stands in a distinctive
relationship with the objects investigated by the human sciences, which he
considers as valid as the natural sciences. This position is documented in
Cohen’s later work Ethik des reinen Willens (1904). Their differences
notwithstanding, the common trait of Cohen’s two major ethical works
is that the will and concrete ethical agency as well as nature unfold
according to a law. In this sense, Cohen’s thought continues the Neo-
Kantian reflection on the different methods of the sciences and the validity
of their results. Far from disregarding human nature, ethics shows that
human beings partake of other levels of reality besides the empirical, and
that such meta-empirical levels of reality are structured according to
different degrees of generality.
Ethics thus vindicates the possibility of wedding the domain of the
absolutely universal with the domain of the irreducibly singular and
contingent. Such a combination, however, requires rethinking the notions
of the a priori, of the analytic and of the synthetic. The analytic domain
delimits formally both the theoretical and the practical use of reason and, at
the same time, it dictates its content in a constitutive fashion.3 Material and
formal aspects thus enter into a plastic relationship, whereby the formal
rule presents itself as a projective lawfulness capable of exhibiting and
grounding new sectors of reality. For example, the formal rule is the
autonomous origin of a specific mode of being, namely, the sphere of
moral agency. In particular, for Cohen, the formulation of the moral law
must be deduced from the analytical concept of pure will. In other words,
the moral law is not simply abstracted from the particular volition. Rather,
3
Karl Vorländer (1860–1928) argues on this point that the form, which sets the conditions, is also
creative and productive. Form produces its content for itself (Vorländer 1893: 52).
132 beatrice centi
the moral law displays itself as a transcendental-regulative lawfulness at
the origin of both the being and validity of ethically relevant actions. The
analytical analysis of Sollen and its prescriptive function guarantees the
autonomy of the ethical domain and establishes its irreducibility to a mere
casuistry of concrete human actions.
As stated above, however, the emergence of the ethical agent as rational
subject does not imply a complete disregard for the specifically human
domain. On the contrary, it supplies its foundation: “Only on the basis of
the ethical can the concept of the human come about” (Cohen 2001b: 11).
Both moral actions and all those behaviors inspired by culture must
translate synthetically into the concrete actions of human beings and the
mutable, empirical-historical forms of life pertaining to them. The human
sciences, and, among them, primarily jurisprudence, investigate such
forms of life. In these disciplines the historical forms of human agency
(which are per se merely factual and contingent) are made possible by two
facts: (1) the human being is also a rational being; (2) ethical validity as such
is an end in itself (Cohen 2001b: 222). By that, Cohen means that the
human being is a project, i.e., an ongoing process of self-construction and
self-transformation oriented toward ethical validity as an end in itself.
Therefore, the distinctions of a priori and a posteriori, universal and
individual do not engender any radical dichotomy. First, the human being
(in its specifically historical and empirical connotation) is never just a mere
individual. The human being is always part of a historically defined
community, which exists a priori vis-à-vis the individual. Accordingly, it
is no surprise that in the preface to the first edition of Ethik des reinen
Willens ethics is presented as the logic of the human sciences, and jurispru-
dence as their mathematics. Jurisprudence establishes the space of pure
lawfulness pertaining to human agency. Therefore, it is more analytical
than the other human sciences and, as such, can mediate between the idea
of the human being as a rational being and an autonomous agent and the
conception of the human being as culturally determined in the relationship
with other human beings.
Cohen’s ethical reflection thus draws on two distinct elements. On the
one hand, we find the analysis of Sollen, and on the other, a non-analytic
approach that considers the sphere of human agency from the perspective
of jurisprudence. Ethics is not merely a fact of reason. Rather, the justifica-
tion of the moral sphere concerns the transcendental analysis of the
conditions for the possibility of a concrete moral action, that is, an action
incarnated in the historical-cultural becoming. The transcendental idea of
pure and unconditioned will as the foundation of ethical normativity is
derived from jurisprudence, in which concrete agency is subject to the law.
The juridical subject is not a mere datum of nature, but rather a historically
determined individual emerging as such only in a normative context.
For Cohen, thus, ethics must carve out a space of lawfulness for itself
and deal with the concept of human being in its universal import, such
that, further, the will is presented as not coinciding with a private intention
as inward fact. Rather, such a private intention would be merely a moment
of the will. On the contrary, the will is first and foremost oriented out-
wardly to a communal system. In its communal dimension the considera-
tion of pure will inaugurates the branch of social ethics. Through social
agency the human being constitutes itself as a person, as a moral agent
distinct from the natural subject. Cohen’s perspective, however, is not
geared toward downplaying the natural dimension of the human being,
but rather at highlighting the specificity of a subject capable of developing
her humanity according to its end, which is first and foremost a social end.
The social nature of the human being and its belonging to a juridical-
communal context partly justifies the emergence of reason’s practical
dimension, which establishes a relationship between individual action
and norm. As a consequence, one might be tempted to think that ethical
reflection can abstract from the most idiosyncratic aspects of the indivi-
dual, i.e., those aspects that appear foreign to the social context. It would be
mistaken, however, to conceive of the universal claim of ethics and the
constitution of a rational moral subject as standing in opposition to the
sphere of human affection. On this issue, there is an interesting new factor
in the analysis of the relationship between freedom and autonomy in
Cohen’s Ethik des reinen Willens. Cohen introduces a theory of affection
as the élan of the soul, which pertains both to pure will and intellect. The
will’s being pure, in fact, does not amount to its being purely conceptual. It
involves the domain of affection, too. According to Cohen, arguing that
affection cannot be pure is itself a sheer prejudice (Cohen 2001a: 116–123).
Qua spontaneous, the soul is capable of a movement, of a tendency in
which both affection and thought partake. This tendency precedes every
representation. It excludes every determinate content other than the
tendency itself. For this reason Cohen can talk about pure affection.
Affection and thought are connected and they belong together (Cohen
2001a: 173). In this way, the action, i.e., the externalization of the will, can
move beyond thought and still remain pure.
The continuity of thinking and willing through affection makes the
agent subject more than just a simple psychological subject. It makes the
agent subject a person. In keeping with Fichte’s philosophy, Cohen
134 beatrice centi
proposes an anti-naturalistic perspective on the notion of person, accord-
ing to which the subject is capable of constituting itself through action.
Action for Cohen is an expression of thought and will alike, and it comes to
full manifestation in language. Through language, the human being
becomes a person and posits itself as both a natural and cultural being, as
both an individual and relational subject. Only within the forms of
language-regulated human consociation does the other become a Thou,
another person. This, for Cohen, is exemplified in contracts considered as
the basic form of relationship between two personal subjects, who, binding
themselves in a contract, cannot be reduced to their natural condition.
In sum, the constitution of the person qua ethical subject displays a
double directionality. From a top-down perspective, as it were, practical
reason is normative. The pure will is deduced from it. From a bottom-up
perspective, however, juridical formations, too, possess a regulative and
foundational value for ethics, which, precisely for this reason, cannot be
reduced to a mere fact of reason. Such bottom-up constitution is not
obtained via abstraction or induction from actual juridical systems.
Rather, the transcendental analysis of these socio-cultural formations
enables someone to grasp an objective and scientifically discernible law-
fulness, in light of which the human being is not only a natural individual,
but also a social subject. This social subject constitutes itself in the free
choice to obey a norm, namely, a norm that at the lowest level of generality
is merely juridical and at the uppermost level of universality coincides with
the Kantian moral law.
In general, the coexistence of freedom and a system of normativity
constitutes the core and the challenge of Cohen’s philosophy. The human
being is, as we saw, certainly a natural being subject to the laws of statistics
and causality. However, in an open critique of Nietzsche, Cohen draws on
Rickert’s thought (which will be addressed in the next section) and rejects
both a purely metaphysical and a purely naturalistic conception of the
human being. The human being is neither a mere product of nature, subject
to the determinism of natural laws, nor a purely rational being free of every
connection with contingency and necessity. The natural and the rational
dimension of the human being need not be separate, rather, their inner
connection has to be illuminated philosophically.
For Cohen such a connection exists because the law of causality is the
limit-concept of freedom. As I have tried to show, the relationship
between these two domains of law represents the key issue in ethical
theory. The primary task of ethics is namely to exhibit the possibility of
agreement between “thoroughly heterogeneous conditions: causality and
freedom . . . , two kinds of conditions, two kinds of worlds, two kinds of
entities (Wesen), two kinds of effects” (Cohen 2001b: 123–124). Such an
agreement is only possible in the ethical domain, that is, in the domain
defined by the human being’s self-constitution as an ethical subject: free
and at the same time subject to the law. The human being becomes a
person solely as an ethical subject, in that morality is the foundational law
of humanity. Ethics constitutes the highest example of a hypothesis
and foundation of the human being (Cohen 2001a: 284). This is because
ethics is driven by the regulative idea of a free human being within a
human community and by the demand of autonomous self-constitution.
Within a community, the human being exerts her freedom as self-
legislation, thereby becoming person and juridical subject (Cohen
2001a: 74–75).
Ought, validity, and values in Rickert

The fundamental theme of Der Gegenstand der Erkenntnis by Heinrich
Rickert is that thinking and willing are part of one and the same movement
and that there is a unity between logic and ethics. This theme also under-
girds Paul Natorp’s Vorlesungen über praktische Philosophie published
posthumously in 1925, which I will discuss in the next and last section.
In Rickert’s book, which was progressively expanded in the second
(1904) and third (1915) editions, the ought (and not being) is presented as
logically prior insofar as it comes first as a general a priori value-form and as
a constitutive project that can be realized in different forms. For Rickert
there is a practical component inherent in knowing, in that knowing is not
a mere representational activity but, first and foremost, the acknowledg-
ment of a value. Knowledge is an act through which the subject grasps the
object and, so to speak, takes possession of it. This practical component
underlies the sphere of ethics, too. The tight connection between knowing
and willing consists in their being fundamental kinds of position-taking
with respect to a value, that is, acknowledgments of the ought and its
validity manifested in the value.
In a manner similar to Frege, Rickert delineates a transcendent reality
(Reales) of validity that amounts to “another world” (Rickert 1928: 259).
This ‘other world’ is coextensive with the dimension of meaning, and it
grounds the multiple values that we can concretely identify. That which
Cohen considered a regulative ideal becomes for Rickert a presupposition,
an idea, an irreal dimension of being that comes first and provides orienta-
tion to our cognitive activity. Validity constitutes a Mittelreich (a middle
136 beatrice centi
realm) between the immanent real being and the transcendent irreal object.
The ought expresses validity as necessarily coming prior to both empirical
being and the manifold values in which validity plays out. In the ought the
timelessly valid value (which is independent from the subjective act of
acknowledgement) receives a form. The Sollen is the overarching form of
validity in general. When it is experienced, the necessity flowing from the
Sollen imposes itself as an imperative to the judging subject, eliciting an
affirmative judgment (Rickert 1928: 207–213). The objectivity of knowl-
edge thus depends on the ought, which encompasses the object of knowl-
edge. For Rickert, the object of knowledge is not to be conflated with the
material object. It rather consists in the connection of a form and a content,
in something that ought to be affirmed (ein Gesolltes) (Rickert 1928: 227).
In the interplay of ought and being delineated by Rickert, in which they
respectively precede and follow one another, the project of knowing, the
ought (Sollen), is on the one hand a general characteristic of value, with its
specific modality of holding valid (Gelten), and on the other hand it is
specific value (Wert). Values identify specific domains of validity, for
instance, the value of moral goodness in the ethical sphere as that through
which judgments about good actions are possible.
From these distinctions follows the possibility of a science focusing on
the forms of value in which the transcendent sense of judgment comes to
expression. Such a science is not so much a theory of cognition but rather a
pure logic capable of distinguishing the value of truth, the value of good, or
the value of beauty. It follows that the value of truth relating to the
theoretical domain has a certain priority. In fact, even in judging that
something is good, one affirms the truth of a state of affairs, namely, that a
certain action is good. Furthermore, Sollen (ought) and Wert (value) do not
coincide because “validity” expresses a mode of holding valid that rests on
itself, i.e., in this case, the being of duty regardless of the subject who
recognizes validity as a transcendent object. Taken together, ought and
value are that which makes a proposition universal and necessary, that is,
intersubjective. Ought and value render objective the reality referenced in a
proposition in its differentiation according to specific domains of validity
and its concretization in culturally determined goods (Rickert 1928: 271–
276).
Whereas the analysis and the definition of necessity can be carried out a
priori, so to speak, as investigations concerning the ought (Sollen) or value
as form, the investigation concerning the plurality of the modes of validity
is, so to speak, a posteriori, i.e., such validity has to be discovered in
different kinds of necessitation and the manners of their realization. For
Rickert, there are three forms through which the ought realizes itself; that
is, there are three forms orchestrating the transition from the ought (Sollen)
to reality as the complete object in which form and matter unite: the
transcendent norm, the category, and the transcendental form. These
forms can be found in the content of a complete judgment (fertigen
Urteilsgehalt) (Rickert 1928: 368).
The transcendent norm is the pure form of ought, or the irreal
transcendent object manifesting itself to the subject as a belonging-
together of form and content; the category is the form of the judicative
act which grasps this object via affirmation, thus enabling the taking
shape of the product of cognition (a complete judgment). For this reason
the category is an element that in the affirmation produces form; it is a
“form-giving something ‘via affirmation’ (‘aussagend ’ formgebende
Etwas)” (Rickert 1928: 366) because it carries with it the sense of predica-
tion. In this model, the subject produces the object as “real” via the
affirmative connection of the content with the appropriate form of reality
on the basis of the norm. Finally, the transcendental form is the form of the
completed, readily available judgment, of which it is believed that it
depicts (nachbilden) (Rickert 1928: 368) reality as if reality were a simply
given object. The analysis of such forms accounts for the plurality of
modes of validity and being, i.e., of necessity, in which the individual
takes on a form. In Rickert’s reconstruction, the ought (Sollen) precedes
conceptually all of reality, to the point that pre-predicative perception
itself, understood as a form of cognition, is already located in the sphere
of theoretical validity (Rickert 1928: 373).
Ethics, too, is a form of cognition, in that causality (the constitutive
form of reality according to laws and the methodological form for the
understanding of reality) also pertains to the sphere of free agency, which is
free precisely because it is ruled by a law. If we call moral in the broadest
sense of the term a will that wills what it wills only because it must will it,
or, in other words, a will that recognizes the values exclusively for their
validity, then ethical behavior comes close to theoretical behavior, and vice
versa. In fact, those who tend toward truth must subject themselves to an
ought (the value of truth) as much as those who obey their moral duty.
Actually, the theoretical concept of ought can be best explained only
through a parallel with ethical norms and their imperativeness. The theo-
retically oriented subject, too, has a kind of conscience (Gewissen), which
comes to expression in the conscious certainty (Gewissenheit) of the neces-
sity of judgment. Such conscience guides cognition in the same way in
which moral duty guides the will and action.
138 beatrice centi
Rickert, however, is far from willing to disavow all differences between
theory and ethics. He emphasizes that ethical behavior specifically depends
on a voluntary decision, which in itself is not necessary. While it is ideally
impossible to deviate from scientific truth, it is in ethics the decision to
acknowledge duty that remains in any case an individual fact (Rickert 1928:
435–437). For Rickert this asymmetry is a symptom of life’s ultimate
irreducibility to form, but also a sign of the human being’s irreducibility
to its vitalistic-naturalistic dimension. This is because humans qua rational
subjects are characterized by a spontaneous activity, in which both the will
to truth (necessary for cognition) and the will to goodness (the innermost
expression of the human capacity for personal autonomy) simultaneously
originate. Rickert considers this the key insight of Kantian ethics.
Autonomy as a constitutive ideal makes it possible to reaffirm the paral-
lelism between knowing and willing. This parallelism does not imply
insensitiveness vis-à-vis the destiny of individuals (as some critics of
Kant, e.g., Georg Simmel, would have it). Rather, the parallelism of
practical and theoretical reason opens up the possibility of an inquiry
into the epistemological relevance of ethics.
Ethics, as the doctrine of deliberately chosen action, becomes, in turn,
the foundation for the possibility of conceiving of action as rational – i.e.,
reflexive and imputable – which counts as the premise for both jurispru-
dence and the other cultural sciences. This approach, however, does not
amount to a leading back of ethics to logic, for instance, through the simple
application of the principle of non-contradiction to action, a move that
Rickert criticizes forcefully in Kant (Rickert 1914: 207). It rather means
rethinking the relationship between the universal and the particular,
between form and content, so that what is singular is not at all “removed”
from the universal. In turn, ethics is the domain in which it becomes clear
that the individual itself, as the point of intersection of a multitude of
relations, is made possible by the universal.
Considering the human being as a person, ethics lets the whole con-
stitutive potential of the relatedness of reality and value come to full
fruition. In fact, a behavior that follows duty for duty’s sake ignites a
change in both the individual, who becomes a person, and her environing
world. This holds true even when the subject is not an individual human
being but a community, which, likewise, takes on a personality. The
relational concept of person, which construes the individual as a reality
in constant relation to another individual, generates new forms of reality,
such as the forms of community that take on their own personality, thus
becoming and producing goods.
For this reason, Rickert, too, like Cohen, avails himself of the concept of
system, which has nothing to do with the famous ‘systematic spirit’ despised
by Nietzsche. Rickert’s system is not merely an ordering procedure, but
rather the instrument, through which philosophy unearths what is funda-
mental from the flow of cultural development; that is, those forms in which
values gradually come to a realization. The function of the system is to show
how the overarching reality of the ought specifies itself in different domains,
thus establishing the identity of the value-form, but within a variety of
categorially articulated structures. The invariant formal value, as the pure
expression of validity, lets different values emerge in specific cultural con-
texts. In this way, the domains of science, art, and religion (which Rickert
defines as the domains of contemplation) receive a limit and a demarcation
from the domains of the personal and social goods of activity.
Other than the logical and aesthetic values pertaining to things, ethical
values pertain exclusively to persons, that is, they belong in the domain of
autonomous action. This insight is the core of Rickert’s Vom System der
Werte (1913), in which he distinguishes six fundamental domains of values:
logic, ethics, aesthetics, the philosophy of religion in both a theistic and
mystical perspective, and the value of the person. This partition of values
correlates with the person who lives toward self-completion, stretched
between unfinished (unendlich) totality and fully completed (vollendlich)
particularity, between future and present goods (Rickert 1913: 301–304).
The system has to be open in order to do justice to the life of culture, which
is always historical and liable to change. Moreover, the one system also has
to allow for the formation of cultural systems, in the plural, and these can
be “closed” in specific cultural configurations, albeit not forever.
In the same way in which the judgment of theoretical value presupposes
the value of truth, the judgment of aesthetic or ethical value for a work of
art or a voluntary action, respectively, presupposes the validity of the
atheoretical values of beauty and morality. Here, Rickert addresses what
might strike the reader as an insurmountable problem. Is it possible to deal
with the atheoretical domain of ethics theoretically in order to determine
the truth of morality? This problem becomes solvable thanks to Rickert’s
conception of ethical value as purely formal and to its relation to theoretical
validity, which obtains at a level that precedes both the ethical and the
theoretical domain. In other words, there is a primordial relation between
the form of ethical value and theoretical validity, which precedes the
specific domains of the theoretical and the ethical.
It is the form, which depends upon reference to a value, that elevates an
otherwise ethically indifferent content to the domain of ethics, as Rickert
140 beatrice centi
points out in the lengthy essay Über logische und ethische Geltung (1914).
This does not entail that Rickert defends a view of ethics based on
intention. Intentions, in fact, must express themselves in actions and
works. However, it is not the outcome that makes an action ethical; rather,
it is carrying it out because of its ethical significance, that is, on the basis of
a decision that has to turn into an action. What we consider morally “right”
(das Richtige) is what we believe in good conscience has to be done (Rickert
1914: 191). The specific ethical character of the will consists in a kind of
necessary connection, which is anticipated and projected onto the “object”
to be realized through the will. Such an object, from a systematic point of
view, is nothing but the human being as a person, as a rational agent
oriented toward values. Thus, we want in a subjectively ethical fashion
something objectively ethical. A will that wants to be autonomous is moral
in two ways: on the one hand, it is ethically motivated, i.e., it wills on the
basis of duty alone, and on the other, it wills what is moral as such, i.e.,
consciously pursued freedom, as a value as much as truth or beauty.
The object of ethics is freedom, and for Rickert – as for Cohen – this
means the possibility of following a lawfulness that is different from, but
analogous to, the lawfulness of natural causality, which must, despite the
differences, be equally necessitating. The human being, as a thinking
subject, is able to recognize what holds valid not only in a logical-
theoretical but also in an ethical sense. This is the case even if the ethical
judgment can never be reduced to a simple judgment about being.
Nonetheless, its claim to validity as a value-judgment from a theoretical
point of view is correct: we could, in fact, say that ethical judgment refers to
the kind of being that we call “duty.”
Rickert’s painstaking analysis of the relationship between ethics and
theoretical knowledge foreshadows problems that cut across the whole of
ethics in the twentieth century. In particular, the conception of the person as
a free subject gives rise to a number of problems, particularly the fact that, in
relationship with others, particular ethical values of a different kind come
about. These are supra-personal values, such as the couple, family, right,
state, nation, and humankind. This calls for the necessary extension of the
domain of ethics to include social philosophy, taken as the study of the
possible forms of communal life, e.g., individualism and socialism, and
the conflicts pertaining thereto. Moreover, this carries with it the problem
of whether in ethics a we-subject, like the we-subject of science, is possible,
especially given that such a we-subject seems to inevitably imply heteronomy.
In spite of these ensuing problems, Rickert holds fast to the view that, in
every domain of reality, the reference to value makes subjects autonomous,
and this is to the extent that it forces subjects to posit and recognize valid
connections over and against the contents presented by mere life in its
indifferent flowing. For Rickert, there remains an inextricable connection
between factual culture and philosophy, which, however, does not imply
an end of philosophy. Philosophy never simply coincides with culturally
contingent worldviews. This connection of philosophy and culture neces-
sitates a redefinition of the task of philosophy as a unitary science vis-à-vis
the various domains of knowledge, as a science of the universal that
anticipates the forms through which the particular will possibly recognize
itself as universal.
Structure and function of the ought in Natorp

Natorp’s thought addresses the classical themes of Neo-Kantianism, which
he articulates in great detail, proposing original solutions to what he
considers the remaining problems. The notion of Sollen enables Natorp
to rethink the Kantian conception of the relationship between the
universal and the singular, to provide a new foundation for ethics, to
overcome the dualism of natural and cultural sciences, and to affirm
persuasively the unitary character of reason.
Both in the Sozialpädagogik (1925b) and in the Vorlesungen über prak-
tische Philosophie (1925a) the ought is the fundamental theme of Natorp’s
research.4 For Natorp, too, the Sollen pertains to the idea of what ought to
be and according to which matter must take on a form. The ought is thus
both the first principle and the end goal of becoming. Plato was right, then,
to call the idea on the one hand an “end” or a “goal,” and on the other a
“beginning” or “principle” (Natorp 1925b: 40). As a ύπόθεσις the Platonic
idea has no further foundation beyond itself. The idea is the unconditioned
lawfulness presupposed by every will and every object of experience. As an
end, the Platonic idea structures and regulates the development of reality.
Following this Platonic model Natorp argues that, qua both origin and
end, the Sollen pertains to the practical and the theoretical spheres alike.
The Sollen is both the element that sets the limits of cognition (as well as
the boundaries of the domains of consciousness, i.e., intellect and will)
(Natorp 1925b: 25) and the foundational principle that grounds the unity of
thinking and willing. According to Natorp there is no gap between
4
Natorp considered these works to be in connection with his Religion innerhalb der Grenzen der
Humanität (Natorp 1894) and with the essay Grundlinien einer Theorie der Willensbildung, published
in five separate articles in Archiv für systematische Philosophie 1–2 (1894–1897).
142 beatrice centi
intellect and will. They are two directions of one and the same conscious-
ness. They can only be separated abstractly (Natorp 1925b: 54).
The will is both the knowledge of the end and the striving to attain it. To
will means nothing but to posit a goal, which ought to be. In this regard the
human being differs from a merely natural animal because she is capable of
thinking herself under the idea of a goal that she has the duty to attain. In
keeping with Kant’s ethics, Natorp argues that the distinctive feature of
what he calls “practical consciousness” is its striving toward something
perceived as an ought. Like Cohen and Rickert, Natorp insists on the
ability of self-explication of what he labels Vernunftwillen, literally “reason-
will.” In the reason-will we have the unity of three factors: (1) self-
consciousness, that is, the consciousness of a rule that can be translated
into an ethical judgment; (2) the drive (Trieb) characterizing the will; and
(3) the striving (Streben) to attain one’s goals.
Like reason, the will has a projective and regulative structure founded
upon a universal norm. The will determines the particular action according
to the universality of the law. The analysis of the relationship between
particular and universal is thus the key to understanding the constitutive
work of reason, both practical and speculative. In particular, Natorp’s
inquiry focuses on the Kantian relationship between the categorial and
the individual. In the practical domain, Natorp’s approach is to show the
fundamental categorial presuppositions for action and, accordingly, for the
will in general. Like Rickert, Natorp argues that the notion of ought has to
be disentangled in order to grasp the degrees of activity in which the
categories infuse a form into the individual. Natorp avails himself of
three notions: (1) structure, (2) function, and (3) content.
These notions refer, respectively, to (1) the making possible of what
ought to be; (2) the development of what ought to be into a necessary
sequence; (3) the exposition of what ought to be in a self-contained reality
(Natorp 1925a: 94). It is in (3) that the singular and the universal finally
meet. As a matter of fact, the ought is both geared toward the singular and
grounded in the universality of the law. Natorp’s reflection on ethics can
thus be characterized as an effort to work his way back (zurückführen) from
the concreteness of the ethical action to the pure ought as that which is
eminently active, as the formative force which is both end and origin.
Natorp’s ethics aims at grasping the connection between the constitutive
activity and the categories, that is, the force of the Sollen as it unfolds
according to the categories of modality (possibility, reality, necessity) and
in the two directions mentioned above (the pure ought and the concrete
being).
The will, furthermore, is an aspect of this activity, and insofar as it is both
active in itself and capable of receiving a structure according to the order of
the categories, it can realize itself in orderly and cognizable forms (Natorp
1925a: 93). Therefore, as the form of the ought, action, too, can be led back to
three moments: project, realization of the project, and achieved result
(Natorp 1925a: 94). These three moments can also be described as “inten-
tion,” “realization,” and “completion through decision.” In the decision the
singular is connected to the universal so that the ought, qua universal,
becomes being in a specific context. The ought is namely the expression of
a law demanding that what is not yet may necessarily be. “Duty” is the form
that regulates what is possible and, at the same time, necessary.
In the system of the fundamental categories of acting and willing,
modality and relation have a special position. Relation, in particular,
does not have to be understood as a mere “standing-in-a-relation” but
rather as the dynamic formation of a connection. The transition from
modality to relation is thus analogous to the transition from possibility to
necessity and it culminates in the determination of a specific individual
substance according to a universal law. In the relation, the universal
becomes concrete in the individual substance, and the individual substance
is posited exclusively in relation to the universal. To elucidate his position,
Natorp often quotes Leibniz on the monad. That is, like Leibniz, Natorp
takes substance and will to be intimately connected. Always involved in the
living process of becoming, substance is not only the result of constitution
but can also be designated as the Ur-Sache, the originary entity, i.e., (in
plain English) the cause. Parallel to action and will, substance, too, displays
a projective moment, one of performance, and one of completion. For
Natorp, talking about the categories means talking about life and its
becoming insofar as the categories are the source of justification for the
genesis of something individual. The same goes for the domain of action,
in which ethics attempts to grasp the moment in which the will becomes a
willing of something individual (Natorp 1925a: 65–74).
In ethics the distinctive feature of agency is autonomy, i.e., the capacity
to give the law to oneself as the constitutive rule of one’s own becoming
(Natorp 1925a: 109). For this reason, Natorp accords exemplary value to
Kant’s ethics, in that it prescribes to act in a way that is purely defined by
the adherence to a universal law. In other words, Kant’s ethics requires one
to individualize one’s actions in particular cases, but so as to preserve “the
requirement of universal lawfulness as the ultimate criteria for action”
(Natorp 1925a: 110.) For this reason, Natorp criticizes harshly those ethi-
cists who protest against the one-sided demand of the law in the categorical
144 beatrice centi
imperative and against the neglect of the individual. These ethicists ignore
the mediating role of the subjective maxim between the universal and the
particular case, an aspect that Cohen already underscored. According to
Natorp, the Kantian solution remains nonetheless insufficient when it
comes to showing how the universal can become individuated in the
concrete and singular case. What is at stake here is the possibility to
show that the individual and the universal do not stand in opposition. In
general, the solution to such a problem requires a deepening of the
relationship between ought and being. This kind of inquiry must tackle
the relationship between history and structure and promote a theory of
rational praxis founded in the ultimate source of spontaneity for action,
i.e., freedom. As we pointed out above, from the standpoint of ethics this
issue pertains to the moment of the attained goal, of the completed action,
in which it becomes possible to think the singular and the universal
together.
In this way Natorp relates the problem of a foundation for ethics to two
key issues: (1) the foundation of individuals, in that the individual is the
concrete origin of the constitutive activity for the realization of the ought;
(2) the relationship of form and content in ethical actions, i.e., the dynamic
formation of the concrete sense of the ethical action. These two issues, in
turn, shed light on the two fundamental aspects of lawfulness as both
origin and end: (1) the static aspect of “structure,” according to which the
law is the form of the individual action; (2) the dynamic aspect of
“function,” according to which the law is what the individual action strives
toward. The structural aspect captures the universality of the law, while the
functional aspect accounts for the possibility of its specification. These two
aspects cannot be conceived as separate. Their unity shows the determina-
tion of content, i.e., the construction of the meaning-content of action,
which is not merely a form, but rather, the fulfillment (Erfüllung) of a
form. In turn, the form is not just the configuration (Gestalt), in which the
content presents itself. That is, Natorp interprets the form as the force that
structures its content, as a configuring force (gestaltende Kraft) following a
necessary law. Natorp goes so far as to say that such simultaneously
regulative and form-giving function characterizes law in a unique way as
individual law (individualgesetzlich). The individual is both free and
spontaneous, such that it is not rigidly predetermined by the universal; it
is rather prefigured by it (Natorp 1925a: 199).
In this way the individual, i.e., the concrete action, is lawfully consti-
tuted according to a universal form that realizes itself in it. The emergence
of a singular action is therefore justified, to the extent that one can exhibit
the universal validity of the concrete judgment, in which the subject’s
decision comes to expression. In the concrete ethical action, the individual
exhibits itself immediately in the universal and vice versa (Natorp 1925a:
200). This is the meaning of action for Natorp. In our concrete actions, the
categories exert their function of giving form and lawful structure to
reality. Ethical actions for Natorp do not merely instantiate a universal
law; they are brought into being by the form-giving force of the universal
law itself. This function, however, is a dynamic one, that is, a never-ending
process of becoming, whose ideal limit is the full realization of the universal
law in a human community. Natorp’s well-known dictum that every
factum is a fieri holds in the domain of ethics as much as it does in the
domain of theoretical philosophy. The concretely given individual action
is never just a simple given, but rather a task, namely, the continuing task
of letting the universal moral law infuse its form into the human world
(Natorp 1925a: 213).
Form and content thus stand in a close relationship. As a project, the form
puts forward some demands pertaining to the nature of the content, which
has to conform to it. For this reason, as we said, the form is both structure
and function. Natorp wishes to underscore the progression of the form, its
development and its individuation in the concrete action. The analysis of
action shows more than just a static application of the law or a merely passive
“being-moved” by the form. There emerges the concrete meaning of move-
ment, of the becoming in which the lawful connection becomes a specific,
singular, individual event. The law itself becomes mobile, flexible, versatile
and therefore oriented toward its own individuation (Natorp 1925a:
209–210). In his analysis of the form and the process of form-giving,
Natorp articulates the idea of a categorial structure of both nature and the
human being, of both thinking and acting. This structure pertains to both
human individuals and communities. In so doing, Natorp makes particu-
larly explicit the Neo-Kantian tendency to reject an interpretation of the
categories as mere logical forms that are alien to the material to which they
are supposed to apply. On the contrary, and in a way that comes very close to
Hegel, the categories must be understood as forms of the very movement of
reality in all its aspects. In this sense, nothing is alien to form, i.e., nothing is
alien to the movement of the categories. Ethical reflection sheds light on
the human capacity to act in such a way that through the categorial form
the action may take on a universal meaning, and the human being engaging
in it may become a person.
Natorp’s considerations on ethics bear important consequences regard-
ing the unitary status of science and the relationship between the natural
146 beatrice centi
and the cultural sciences. For Natorp there is a deep-seated analogy
between the sphere of nature and the sphere of praxis. To reject a natur-
alistic conception of the human being does not mean setting up nature and
culture against one another. Nature and culture follow the same dynamic
of constitution, which produces both nature as the object of scientific
cognition and the world of culture in which self-responsible human agency
takes shape. As we argued above, lawfulness presents both a structural and a
functional aspect. Thus, the structure is also a function, and vice versa. For
Natorp, the person is the logical presupposition of the human being in
ethics (Natorp 1925a: 359); it is the form that regulates an ongoing dynamic
of formation. Such a process of constitution unfolds in a continuous
interplay between the logical and the empirical, the ethical and the natural.
Thereby, the demands of the law do not oppose life, but rather, they
express life’s intrinsic necessity to take on a form.
In this way Natorp criticizes the dualism of nature and history, of value-
free and value-laden, of matter and spirit, which he imputes to
Windelband and Rickert. Such a dichotomy originates from a misunder-
standing of the concept of concrete universality, that is, of the lawfully
individuated form. All of reality has to be interpreted as actuality in a
process of becoming, as actualization.
Conclusion
In this chapter, I examined the work on ethics of three leading figures of
Neo-Kantianism. In spite of some differences in exposition and about
some specific problems, Cohen, Rickert, and Natorp agree in rejecting
an all too rigid distinction of theoretical and practical philosophy. On the
contrary, they believe that the distinctive ethical problem of reconciling the
individuality of concrete actions and the universality of the moral law sheds
light on the deep-seated unity of philosophy as a whole. Both in the
cognition of the natural world and in the evaluation of ethical agency,
philosophy unearths the dimension of validity, or law, which is the origin
and the teleological end of both activities. The unity of law and fact
achieved in concrete ethical actions becomes the paradigm to interpret
the norm-governed constitution of reality as a whole.
chapter 7
Neo-Kantianism in the philosophy

of law: its value and actuality
Jonathan Trejo-Mathys
Introduction
The Neo-Kantians have largely been forgotten. That, at least, is something
that almost everyone who writes about them can agree on. Manfred Kuehn
has recently claimed that “one of the reasons why the Neo-Kantians’
popularity has diminished is that their problems are no longer our pro-
blems,” while “Kant, on the contrary, seems to have continued relevance.”1
From the standpoint of the philosophy of law, however, some of the central
problems of the Neo-Kantians are still very much our own: for instance,
the question whether there is some third alternative in philosophical
jurisprudence between the traditional megaliths of natural law theory
and positivism, or the need to deal with fundamental challenges to estab-
lished legal concepts like the rule of law or culpability that arise from
developments in natural and social science. I will use a version of the first
problem – the dual nature of law as both ideal norm and real fact – as a
framing device in this chapter. The challenge of finding a third path besides
those of natural law and positivism, incorporating the insights of each
without their respective weaknesses and errors, is a central part of the
agenda for the Neo-Kantians and for Neo-Kantian jurisprudence in
general. I will use “Neo-Kantian” to refer to members of the Marburg
and Southwest Schools, specifically the outstanding representatives of what
I take to be the first generation of a tradition that extends into the present,
which I will also refer to as “Neo-Kantian.” After first providing a sketch of
this putative tradition and a description of the framing problem, the main
contributions of the Neo-Kantians to the philosophy of law, in particular
those of Stammler, will be discussed in some detail and then critically
1
M. Kuehn, “Interpreting Kant Correctly: On the Kant of the Neo-Kantians,” in S. Luft and
R. Makkreel (eds.), Neo-Kantianism in Contemporary Philosophy (Bloomington: Indiana University
Press, 2009), 113–131; here 128.
147
148 jonathan trejo-mathys
assessed. In closing I will turn briefly to the second and third generations,
offering a conjecture about why the second generation presents us with the
peculiar spectacle of a Kantian relativism and a hypothesis that the third
and contemporary generation of Neo-Kantian philosophy of law is in a
better position to deal with the framing problem than the most well-
known varieties of Anglophone jurisprudence.
Lineaments of a tradition
While I do not have the space to demonstrate the fruitfulness of this
perspective here, I believe that there is a discernible genealogical and
thematic connection between (at least) three generations of Neo-Kantian
philosophy of law. The first generation includes Rudolf Stammler, who is
its major representative, as well as Hermann Cohen, Paul Natorp, and
Emil Lask as a supporting cast. The second generation includes two
genuine jurisprudential giants, Gustav Radbruch and Hans Kelsen, at
least in his “classical” period.2 The third generation’s major representatives
are Jürgen Habermas, Robert Alexy, and Klaus Günther, all proponents of
a common approach called the discourse theory of law.
There are certain features that by and large every member of this
tradition shares, which I must simply state in a rather brief and dogmatic
fashion here. First, all Neo-Kantians employ the “method” or approach of
transcendental reflection on the conditions of possibility of something taken
as a given starting point that bottoms out in the discovery of something
transcendentally necessary (a form of necessity that is stronger than logical
necessity, but is weaker than physical necessity, in the sense that it rules
out more than the former but less than the latter).3 This becomes the
“constitutive” categories for the object domain of a general type of
cognition or “constitutive” presuppositions of discursive practice as
such. Second, the first generation eliminates an ambivalence in Kant by
retaining the classical philosophical orientation to the whole but in a post-
metaphysical form: the traditional content of metaphysics is demoted to the
status of “regulative” ideas, though granted job security, as it were, in
compensation. Third, all later generations share with the first a procedural,
2
It is in the second generation that the influence of the Southwestern thinkers Wilhelm Windelband
and Heinrich Rickert, along with that of Lask, bears significant fruit in the work of Radbruch, and in
that of Kelsen too.
3
Heath defines transcendental necessity as “truth at all possible worlds cognitively accessible from our
own,” and then provides a more technical “possible-worlds” formulation of the gloss I give in the
text. J. Heath, Communicative Action and Rational Choice (Cambridge, MA: MIT Press, 2001), 284.
Neo-Kantianism in the philosophy of law 149
or perhaps better processual, understanding of reason and rationality, with an
emphasis on the way, or the how and not the what, learning, not wisdom, in
the conviction that while there are enduring and describable structural
features of the process or form of reasoning, inquiry, investigation,
decision-making, justification, discourse, and so on, there are otherwise
no unrevisable contents. Fourth, by and large all Neo-Kantians share the
threefold division of human rational capacities or modes of functioning into
cognition, volition, and emotion (or thinking, willing/choosing, and
feeling) that traces back rather obviously to Kant’s three Critiques. Fifth,
it is a striking fact that every major figure across all three generations of
Neo-Kantian philosophy of law was politically committed to left-wing
projects of radical-democratic socialism or, weaker, social democracy.4 Last,
but most important for our purposes, there is a shared commitment to the
centrality of coercion and the question of the conditions of its legitimation in a
proper understanding of law and the state.
This last feature deserves some further comment given the topic of this
chapter. The centrality of coercion for an understanding of the very notion
of law was a commonplace of legal theory in the nineteenth century,
canonically expressed in the English-speaking world by John Austin in
The Province of Jurisprudence Determined (1832), and widely accepted by
most major legal theorists until a powerful avalanche of criticisms by
H. L. A. Hart in The Concept of Law (1994). After Hart, most writers
working in analytical jurisprudence simply took it as given that coercion
was not part of the concept of law as such, and hence was at best secondary
in importance to other features of law from a theoretical point of view
(however salient it might be in practice). The outstanding exception here
was Ronald Dworkin, who held that “a conception of law must explain
how what it takes to be law provides a general justification for the exercise
of the coercive power of the state that holds except in special cases when
some competing argument is specially powerful.”5 It was no doubt in part
his influential example that led the third-generation figures to maintain
this commitment in the face of massive opposition by Hart, his very
prominent student Joseph Raz, and their many capable followers. The
broader commitment to progressive political ideals also helps to explain
why coercion and the question of its (possible, if not actual) legitimation
would remain a focus of Neo-Kantian legal thought. In the Anglophone
4
See S. Schwarzschild, “The Democratic Socialism of Hermann Cohen,” in H. Holzhey (ed.),
Hermann Cohen (Frankfurt am Main: Peter Lang, 1994), 205–227.
5
R. Dworkin, Law’s Empire (Cambridge, MA: Harvard University Press, 1986), 190.
literature, there has recently been a powerful reassertion of the centrality of
coercion for legal theory. As we will see, the tight connection between
social theory and philosophy of law from the first generation right down to
the third enables the Neo-Kantian line to offer important new argumen-
tative resources to this debate.
Between facts and norms: the dual nature of law

There is an unresolved issue at the heart of legal theory on which, in my
view, a tradition going back to the Neo-Kantians has currently under-
appreciated resources to bring to bear. Jules Coleman has recently
described the problem – and the problematic state of jurisprudence – in
the following way:
There is something of a dual nature to law – its sociality and its
normativity . . . Law is indeed a normative social practice. An adequate
theory of law will explain its sociality and its normativity. It turns out,
however, that there is a major divide in jurisprudence between those who
think that the primary project of jurisprudence is to explain the normative
dimensions of law and those who think the project is to display the sociality
of law, which they take to be the requirement of showing the ways in which
law is continuous with other aspects of our social life. For some . . . the
normativity of law is . . . the normativity suitable to social organizations; this
turns out to be largely the normativity of instrumental rationality . . . But, of
course, law is not just any old social organization, just more complex. The
plans, rules, and directives of the law purport to – sometimes, if not always,
do – make a difference in what those to whom it is directed have moral
reason to do . . . It should turn out to be no surprise that those who have
focused on the difference law makes in moral space have said little about the
social dimensions of law, whereas those who have emphasized law’s con-
tinuity with social orderings more generally have been largely unconvincing
in their accounts of the normativity of law.6
Even if, as Coleman and many other leading jurists believe, the dichotomy
between natural law and positivism is now obsolete, this problem still
remains. In fact, the Neo-Kantians (1) anticipated the (putative) obsoles-
cence of this contrast by vigorously insisting on its necessity and strenuously
arguing for it and (2) did so in a path-breaking way that, if properly under-
stood and appropriated in the contemporary intellectual context, remains a
promising theoretical approach to comprehending both the social facticity
and the normativity of law. Regarding (1), the Neo-Kantians viewed
6
J. Coleman, “The Architecture of Jurisprudence,” Yale Law Journal 121 (2011), 79–80.
positivism and natural law theory as jurisprudential expressions of two
equally “dogmatic” or “uncritical” philosophies, the former the impover-
ished epistemology of psychologistic or naturalistic empiricism and the latter
the extravagant metaphysics of an unregenerate Platonism and/or
Aristotelianism. Lask’s programmatic essay of 1905, “Rechtsphilosophie,”
which would be a significant influence on the second generation, states the
goal with exemplary clarity: a critical philosophy of law must find a position
that is neither empiricist positivism nor metaphysical natural law (Lask 1923:
279–286, cf. 290f.).7 With respect to (2), the rest of this chapter must serve as
an argument for the claim.
Stammler and the three tasks of legal philosophy

For the purposes of this chapter, the contribution of the Southwest School
more or less ends with its half of the shared effort to fortify a common
frontline against enemies like psychologism, materialism, and naturalism. In
contrast, the Marburg School has been described by one prominent scholar
as the only serious attempt in Germany at a genuinely philosophical engage-
ment with the political situation of its own time since Hegel.8 The core of
this engagement was the sympathetic confrontation with Marxism. The
Marburgers must be among the first within the German academic establish-
ment to have taken Marxism seriously on both the intellectual and political
levels. This was due in part to the tremendous influence of the life and
writings of F. A. Lange, whose Geschichte des Materialismus made a profound
impact on many important figures of his generation and the next, including
Cohen and Nietzsche. The first great theoretical product of the encounter
with Marx’s historical materialism in Marburg appeared in 1896 in the form
of Rudolf Stammler’s Wirtschaft und Recht nach der materialistischen
Geschichtsauffassung (Economy and Law in the Materialist Conception of
History). This book inaugurated “social idealism” as the Marburg School’s
answer to the “social materialism” of Marx and Engels.9 At the same time, it
helped to establish Stammler’s prominence, if not his preeminence, in the
theoretical jurisprudence of his generation, a position reflected in the striking
7
See also Stammler’s criticism of natural law and the positivism of the historical school (Stammler
1906: vi–x).
8
H. Lübbe, “Die politische Theorie des Neukantianismus und der Marxismus,” in Holzhey, Hermann
Cohen,
9
See, for instance, Paul Natorp (1904; 1920). Lask also says that it was the naturalistic challenge of
Marxian socialism that helped provoke a “return to Kant” in the area of social philosophy and then
cites Cohen, Natorp, Stammler, and Staudinger as having led the charge (Lask 1905: 288).
fact that “practically every legal theorist, as well as several philosophers and
social theorists, in the German-speaking world critically engaged with
[him]” (Paulson 2002: 12n4).10 Indeed, the editorial note introducing a
translation of a lengthy article of Stammler’s published in the Michigan
Law Review in 1923 declares that “he is commonly recognized as the greatest
of the present-day jurists on the Continent” (1923a: 625 fn).11 François Geny,
himself a prominent French jurist of the time, wrote in 1915 that “the work of
R. Stammler represents without doubt the most considerable monument of
the philosophy of law that has been raised by the last generation of German
jurists”12 (2000: 493–552, 493).
Wirtschaft und Recht presented the core views about law that
Stammler would refine and develop in his voluminous later writings,
the most important of which are Die Lehre von dem richtigen Rechte
(1904), Theorie der Rechtswissenschaft (1911) and the late, encyclopedic
textbook, Lehrbuch der Rechtsphilosophie (1922). His centrality to the
Neo-Kantian tradition is further underlined by the fact that Radbruch,
looking back from 1932 to the situation in which he first published
his own legal-philosophical system in 1914, and thereby arguably
brought the Southwest School’s vision to fruition in the field of law
in the second generation of Neo-Kantianism, states that that early work
of his was “a modest contribution to a series of efforts that sought to get
the philosophy of law moving again after a decades-long standstill
during which only Rudolf Stammler held up the banner of legal philo-
sophy” (2003: 4).
According to Stammler, the three tasks of the philosophy of law are to
provide (1) an account of the concept of law, (2) a justification of law’s coercive
bindingness, and (3) an adequate formulation of the idea of law, in the sense
of a critical ideal or standard of correctness for the content of particular
laws and judicial decisions (Stammler 1906: 59).13 Legal philosophy must
solve these problems in that logical order. Without an understanding of what
law is, the remaining questions cannot be sensibly answered. However,
reflection on the concept of law reveals that law is essentially a coercive,
10
S. Paulson, “Einleitung,” in R. Alexy (ed.), Neukantianismus und Rechtsphilosophie (Baden-Baden:
Nomos, 2002), 12 n. 4. Paulson provides an impressive list of legal scholars who tangled intellectually
with Stammler’s work.
11
One’s confidence in the judgment of the authors of this note wavers a bit, however, when, after
saying that he is a “pragmatic idealist,” they go on to compare his jurisprudence to that of Oliver
Wendell Holmes.
12
F. Geny, “The Critical System (Idealistic and Formalistic) of R. Stammler,” in Stammler 2000:
493–552; here 493.
13
See Stammler 1902: 59.
non-consensual way of coordinating action. But coercion is morally proble-
matic. If a coercive medium of social organization is inherently unjustifiable,
as anarchists for instance claim, then there should be no such thing as law. If
legal coercion can be justified, however, then law has a general right to exist.
But a further question arises about what content particular legal norms
should have and what judicial decisions are correct in specific cases. This
leads to the need for general ideal standards of good or just law to guide
legislators and judges in making, adjusting, and applying law in concrete
historical contexts. According to Stammler, one of the great flaws of the
major philosophical approaches to law has been that they have not clearly
distinguished these questions and have typically sought to answer them in
one stroke (Stammler 1906: xiv–xv).14 His own sharp separation of the
question of the concept of law from the question of the substantive justifica-
tion of the content of specific legal norms, or the question of legal validity
from that of normative rightness, is reminiscent of the positivist tradition.
On the other hand, he insists on an internal connection of law and the claim
to rightness or justice, in a way that recalls natural law theory. In the attempt
to hold both of these tenets while falling into neither natural law nor
positivism, Stammler is representative of both the first generation and the
entire Neo-Kantian line.
The concept of law on the basis of a concept of society

Stammler defines law as a species of social rule or “external regulation”
(äussere Regelung). Social rules both enable and constrain the interaction
and/or cooperation of human beings as goal-seeking agents. It is the failure
of social materialism (like that of Marx) to lack a way of distinguishing such
normative relations of goal-seeking agents from the causal laws that con-
nect natural phenomena. Social idealism, in contrast, understands natural
and social science as having two different object domains, constituted
by two different “methods” for “ordering” and “unifying” the data of
consciousness. The natural phenomena disclosed to us in perceptual
experience are constituted, interrelated, and grasped in terms of causality,
while the social phenomena we confront from the standpoint of interacting
agents with wills are constitutively structured in terms of “teleology,” or
conceptual relations between means and ends, and an ordering of the
14
Stammler is referring in this context to natural law theory, the Romantic jurisprudence of the
“historical school,” and the historical materialism of Marx and Engels. He elsewhere says the same
flaw mars Kant’s definition of the concept of right (Stammler 2000: 210, 162).
pursuit of ends in a unified scheme of cooperation.15 Both cases follow a
Kantian pattern. There is some material that is constituted or synthesized
into objects of experience according to formal a priori categories.
Systematic inquiry concerning these objects is then in turn structured
and guided by a priori regulative ideas. We will return to the “social
ideal” and the regulative idea of law. For now the crucial point is that
neither society nor law can be adequately understood in terms of a science
oriented by the fundamental category of causality. Such an endeavor
overlooks the fact that societies are entities constituted by goal-seeking
agents who are interrelated by rules governing the pursuit of ends
(Stammler 1896b: 83–84 [§16]; Stammler 2000: 248).16 The shared recog-
nition of such rules is what makes a number of human beings members of a
single unit, a “society.” In contrast, every human being on the planet stands
in some definite causal relationship to every other, however minute such
influences might be, but this is not sufficient to make them members of the
same society. It is, as it were, epistemologically impermissible to assimilate
the “realm of causes” and the “realm of ends”(Reich der Zwecke). Natural
sciences based on perception and causality and social sciences rooted in
volition and purposiveness (Stammler calls them Zweckwissenschaften)
are methodologically and categorically distinct (Stammler 1922: 54–63
[§§25–29]; Stammler 1896b: 349–395 [§§63–69]).
It is this point that forms the crux of Stammler’s argument against Marx
and Engels. Even the specifically economic relations of human beings
cannot be understood apart from social and legal concepts like money,
property, and contract: “all social-economic activity presupposes a definite
legal order. It is only the working out of [that] order” (Stammler 1925: 376).
This is a point that Marxian social materialism acknowledges in its concept
of relations of production, but which it systematically disavows in its
naturalistic attempt to model social and legal change in terms of causal
laws of motion. Instead, Stammler argues that societies must be conceived
as unities of form and matter, where law is the conditioning form and the
economy is the conditioned matter or content: “the matter of social life is
cooperation directed toward the satisfaction of needs on the part of human
beings bound together in society” (Stammler 1896b: xxix). The respective
roles of the law and the economy in this account parallel to some extent the
15
For a clear statement of this epistemological or methodological dualism, see Stammler 1896b:
371–372.
16
See Stammler 1906: xxviii. See also Stammler’s (possibly unfair) criticism of Simmel for only
understanding human social interaction in terms of the natural-scientific causal concept of
Wechselwirkung (Stammler 1902: xxviii).
roles of the moral law and the satisfaction of inclinations in Kant’s moral
philosophy. In each case, the law constitutes the “formal” framework
within which various types of “material” happiness or prosperity can be
pursued. The implication, which helped to found an entire school of
political economy in Germany and places Stammler in the venerable and
recently resurgent tradition of historical institutionalism in economics, is
that “there are no universal truths in political economy” independent of
the specific legal frameworks of economic interaction (Stammler 1896b: 225
[§37]).17 For the economic dimension of a society is a concrete realization
of a selection from the possibilities made available by its legal order
(Stammler 1896b: 221).
Law as the basic form of society is distinguished from two other forms
of external regulation of action: convention and arbitrary power or
command. The former type is based on the voluntary consent of those
who obey them and only “appeal” or invite subjects to conform. Stammler
includes social customs, usages, fashion, etiquette, etc., in this class.
Arbitrary acts are, like law, non-voluntary or coercive but, unlike law,
essentially impositions of the will of one person or group of persons on that
of another for the sole purpose of satisfying the desires or advancing the
interests of the former. They do not even purport to aspire to the enduring
quality of a legal norm, but have an ad hoc situational character. Law is
superordinate with respect to both these other types of social rule. It is “an
inviolable, self-authoritative, binding will” (unverletzbar, selbstherrlich,
verbindendes Wollen) (Stammler 1922: 89), or, in a less anthropomorphic
formulation, the “inviolable, self-authoritative regulation of human social
life” (Stammler 1906: xxviii).
The inviolability of law entails that only the law determines when and
under what conditions it or some part of it shall continue to exist. A
violation of the law, say, by a person who commits a crime, does not
constitute the end of the violated norm. On the contrary, the legally
mandated sanctions for a violation of the norm are now actions authorized
by the law, typically in “violation” of, i.e., against the will of, the culprit.
The law permits it to be the case that its subjects often do not get what they
17
In fact, according to the noted sociologist Richard Swedberg, “Stammler’s ideas came to constitute
the foundation for the so-called School of Social Law in Germany, which held that society was
constituted according to legal norms and that law had priority over the economy. Economist Karl
Diehl (1864–1943) was a prominent member of this school; and Weber appreciated Diehl’s work
quite a bit.” R. Swedberg, Max Weber and the Idea of Economic Sociology (Princeton University Press,
1998), 248 n. 27. Diehl wrote a defense of Stammler against Weber. This work is cited in Swedberg,
Max Weber, 248 n. 25. At the same place, Swedberg cites Stammler’s own response to Weber in the
third edition of Wirtschaft und Recht (Stammler 1914: 670–673 n. 232).
want, but the law always gets what the law wants. It can ensure this because
of its binding, self-authoritative nature. Selbstherrlich has a decidedly
negative connotation of arrogance, haughtiness, self-aggrandizement, or
contempt for others in contemporary German. Stammler clearly had in
mind a rather literal and etymological reading of it, however. The law
recognizes no master, no rule, no Herr, above and beyond itself. This is
arguably a conceptual truth about law that is not placed in question by
dictatorships or theocracies. If such cases are genuine examples of legal
systems, then the rule by authoritative (or authoritatively interpreted)
scripture or by an individual despot is itself licensed by the system. The
rule of law in the modern sense is commonly understood to entail this
reflexive regulation of the creation, execution, and adjudication of law by
law itself.
The legitimacy of law’s coercion

For Stammler, the coercive or non-consensual nature of law is implicit in
this definition. The existence of a consensual or voluntary type of social
rule, or convention, raises the question whether it might be possible for
society to exist without law. The morally problematic nature of coercion
makes it necessary to ask whether law might be in principle an illegitimate
way to regulate human interaction: in other words, to confront the
challenge of anarchism.18 The legitimacy of law’s coercion concerns the
very form of law. If it cannot be provided, there is no point in going on to
address questions about the matter or contents of any concrete legal system.
Using a legal analogy, Stammler writes that only if law’s coercion is in
principle justified does the law even have “formal” standing in the court of
social philosophy to gain a hearing in disputes about the “material” right-
ness or justice of particular legal norms or judicial decisions (Stammler
1896b: 524f. [§92]).19
The argumentative strategy Stammler undertakes is to show that law is a
necessary means to a necessary (or justified) end (Stammler 1922: 225n.3,
226). A purely conventional, i.e., individual consent-based, entirely non-
coercive, society is a transcendental impossibility. Neither singular acts of
arbitrary power nor entirely voluntary conventions can serve as conditions
of possibility for the specifically social lawfulness (Gesetzmässigkeit) that
18
Stammler touched on this issue in almost all of his writings. For a compact statement, see his Die
Theorie des Anarchismus (Stammler 1896a).
19
The metaphor of the court of social philosophy, the “Richterstuhle der Sozialphilosophie,” comes
from Stammler 1896b: 525.
serves as a unifying parallel to the role of natural causal laws. Just as the
possibility of perceptual experience requires a basic “harmony”or coher-
ence among all possible sense-impressions, the possibility of society
requires a basic “harmony” or coherence among all possible instances of
purposive activity (Bestrebung). If the transcendental unity of apperception
grounds the claim that all possible perceptual experience must conform to
certain universally valid principles, in particular that of causality, then a
kind of transcendental unity of interaction grounds the claim that all
possible pursuit of ends must conform to certain universally valid rules
in the form of law, if anything like society is to be possible.
Stammler’s argument here is rather obscure and very different than the
one Kant offers in the Doctrine of Right. For one thing, Stammler’s seems
to be decidedly epistemic and even in a certain sense ontological in
orientation. It has no obvious normative premises, though it has premises
about the nature of normatively regulated interaction. Kant’s argument,
in contrast, is a decidedly normative one with some epistemological
aspects. His conclusion is that there is a duty to enter into or uphold
public legal order. However, I believe one may reconstruct Stammler’s
argument largely in the spirit of Kant. The argument is presented only in
a rough, abbreviated form. According to Stammler, transcendental reflec-
tion establishes that all volitions are subject to judgments categorizing
them as either right or wrong, correct or incorrect, from the standpoint of
practical reason (Stammler 1922: 61–62 [§29]). This includes acts of will
related to conventions or arbitrary commands. But individuals will
inevitably disagree in concrete situations about whether this or that
instantiation of will (be it a norm, a command, an action, or a decision)
is right or wrong. No individual will have any reason to retreat from his or
her judgment of rightness and it would be wrong to act against one’s best
judgment about what is right. The result will be that superior force will
decide whose judgment is to determine the outcome of the dispute. But
then any existing order would be effectively an equilibrium of forces in a
more or less brutally causal sense. Societies, as the special phenomena of
interest to a specifically social science, cannot be just another network of
causal forces, even if in this case the clashing billiard balls are steered by
atomistic judgments of justice. Similarly, purely conventional, voluntary
orders are going to be either ephemeral and dissolve as soon as such
disagreements arise, or be backed up by a surrounding context of com-
pulsory law, and situation-specific commands that are not authorized
either by pre-existing conventions or laws (and in that sense arbitrary) are
unable to construct or maintain a temporally enduring order (Stammler
1922: 231).20 Hence, wherever there is an enduring social order, there are
rules of action that are conceived of as inviolable and self-authoritative by
members of that order and that establish publicly recognized procedures
for the settlement of disputes about what counts as right or wrong
according to the formal normative order of society.
In addition to this argument, admittedly a speculative reconstruction
from available materials in his writings, Stammler provides another one
that is familiar from the tradition of political philosophy, in particular
Hume’s well-known criticism of social contract theory. Since the fleeting
nature of arbitrary or ad hoc commands has no society-constituting capacity,
to decide the question whether coercive legal norms are justified as a form of
external regulation, one simply has to examine the only remaining alter-
native, conventional norms, to see whether they can regulate every conceiv-
able form of human social interaction or cooperation (Zusammenwirken). If
so, there would be no unconditional need for coercive law. The answer is
that they cannot, since conventions depend on the contingent empirical
possession of the capacity to enter into voluntary agreements or contracts
(Stammler 1896b: 553f. [§96]; Stammler 1906: xxxvii). The idea that human
society could be fundamentally constructed entirely on the basis of contract-
like agreements is mistaken, akin to a pars pro toto fallacy or fallacy of
composition. Unfortunately, Stammler does not give concrete examples in
his highly abstract presentations of the argument, but he seems to have in
mind social arrangements like the family, e.g., children do not contractually
agree to their membership in the family and a fortiori to the wider society.
This recalls recent criticisms of contractualist liberal political theory to the
effect that it forgets elementary facts about human social life such as child-
hood, mental and physical disability, and senility.21 In sum, the problem
with anarchism is that it fails to understand the conditions of possibility for
the full spectrum of interactions that constitute society as a whole.
The social ideal and correct law (richtiges Recht)

After the concept of law has been defined and the coercive form of law
provided with a general justification, the third task is to show how we can
20
As Stammler points out, both conventions and arbitrary commands can be normatively right
(Stammler 1922: 229). If a fire breaks out in an office, a clear-eyed, calm worker may spontaneously
begin issuing directives to her fellow employees to organize their exit from danger. There is no
conventional agreement that she be empowered to do so, nor is there any binding law to that effect.
Nevertheless, it may be absolutely justified that she do so and others may be doing wrong if they
interfere with her attempt to lead.
21
See, for instance, MacIntyre 1999.
rationally distinguish, among the various contents of law, whether they be
statutes, customary principles, or judicial decisions, the legally correct from
the legally incorrect in accordance with some methodologically controllable
normative standard. Stammler formulates the regulative idea of law as “the
social ideal” (das soziale Ideal), a purely formal concept of an ideal society that
is supposed to guide judgment about concrete legal matters in such a way as
to make possible a principled distinction between good, right or correct
(richtig), and bad, wrong or incorrect (unrichtig), law. In other words, he
holds that the social ideal is implicit in social-practical consciousness and
that this thesis entails that the rule-guidedness or rule-structuredness
(Gesetzmässigkeit) of society – the social analogue to the lawfulness of natural
processes – is inseparable from questions of normative justification.
The social ideal is Stammler’s self-conscious socio-legal variant of Kant’s
realm of ends. His preferred formula for it is “the community of freely
willing persons” (Gemeinschaft frei wollender Menschen) (Stammler 1902:
198). As a specifically legal community, its aim is to combine legal con-
sociates into a harmonious unity of possible ends, a unity in which, while
each member pursues its own ends, all are secured treatment as ends in
themselves and not merely as means for the ends of others (Stammler 1902:
197). A given legal content has the property of rightness or correctness
(Richtigkeit) when it sufficiently “corresponds” (entspricht) to or adequately
realizes the social ideal in a given sociohistorical context (Stammler 1902:
198). As Stammler realizes, this is obviously far too abstract a notion to do
any real work in guiding legal practice. He therefore distinguishes four
stages or levels (Abstufungen) in the process of correct jurisprudence
(Stammler 1902: 277; cf. 203–204). The social ideal is the most abstract
and general level. It is not a goal to be achieved or destination to be
reached, but a “guiding star” that procedurally orients legal activity as it
navigates through the welter of contingent empirical detail (Natorp 1913:
65; Stammler 1923b: 889). As a “universally valid method” (Stammler 1902:
204) for achieving a “unified ordering” of legal contents, it is not itself a
legal proposition within any concrete legal system (Stammler 1902: 202).22
Below this are four universally valid principles (Grundsätze) that serve as a
“bridge” between the idea of correct law and particular legal issues and
ought to structure the legal system as a whole (Stammler 1902: 204).23
These divide into two principles of “respect” (Achtung) (Stammler 1902:
22
Compare the reference to “the method of uniform ordering” in Stammler 1923b: 888.
23
Just above the level of particular cases are more concrete but still general legal doctrines and
directives (Lehren und Anweisungen), for instance, concerning the limits of freedom of contract or
the interpretation of good faith clauses (Stammler 1902: 203; cf. 311–601). The latter are parts of
208; Stammler 2000: 161) and two principles of inclusive “participation”
(Teilnehmen) (Stammler 1902: 211; 2000: 163).24 Between the principles
and correct judgments in particular cases lies what Stammler calls the
“model” (Vorbild) of correct law, which serves the purpose, roughly parallel
to Kant’s schematism in the Critique of Pure Reason, of making it possible
to grasp concrete legal communities, limited in space and time, as exem-
plifications of the principles and the social ideal (1902: 277). The pure idea
of law for Stammler ultimately connotes an integral unity understood
in terms of the most harmonious possible social coordination of the
individual pursuit of ends in a given place and time, and hence as some-
thing rationally justifiable to all from the standpoint of each. It is striking
how easy it is to add to this description of Stammler’s otherwise quite
different theory the following words of the late Ronald Dworkin, expres-
sing an important part of the spirit of his own highly influential concep-
tion: “Law as integrity (we might say) is the idea of law worked pure.”25
Assessing Stammler
In Stammler’s work, for perhaps the first time in the history of legal
philosophy, the concept of law is explicitly formulated as a sociological or
social-scientific one, in this case within the framework of transcendental
philosophy. Society, as the special object of the social or purposive sciences
(Zweckwissenschaften), is an inherently normative entity. And the central set
of social rules normatively integrating society is the system of laws.26 This
entails, however, that the adequacy of Stammler’s concept of law is insepar-
able from its adequacy as a social-theoretical one. Here Stammler’s episte-
mological dualism was met with sharp criticisms, the most relevant of which
came from figures within the intellectual camp of Neo-Kantianism broadly
understood. First, Max Weber pointed out that Stammler overlooked the
presence of formidable social mechanisms that inflict sanctions on indivi-
duals who deviate from certain established usages, customs, and other social
conventions or norms, including norms of sexual behavior, fashion, gait,
everyday legal practice, which is “an art,” even if in the best case one “based on scientific insight”
(Stammler 1902: 312).
24
There are in addition four “postulates” of politics that prescribe the guarantee of legal certainty,
respect for personality, universal provision of material preconditions of individual development,
and basic distributive justice (Stammler 1902: 300f.; 2000: 229f.).
25
Dworkin, Law’s Empire, 400.
26
Even from its own point of view, the law has a normative relation to all other social rules and is
accountable for them. Existing conventions and customs, for instance, must be viewed by the law as
either permitted, forbidden or obligatory: there is no other alternative (Stammler 1902: 235–238).
address (e.g., in speech between members of different classes), and so on,
regardless of the individual’s own attitude toward them. Therefore, the
attempt to distinguish between law and social convention using the notion
of voluntary consent is sociologically useless and misguided (Weber 1978:
325ff.). Second, Kelsen argued in a compatibilist vein, using many of
Stammler’s own formulations, that the attempt to make causal-explanatory
and purpose-oriented points of view utterly incompatible with one another
was unconvincing even from the actor’s point of view since the actor must
conceive her end as a state of affairs that she can cause to occur by her own
action, and hence as something that is necessarily within the causal nexus of
events (Kelsen 1911: 58ff.). Ordinary people and social scientists each make
the same presupposition when they explain an observed action. Thus, if
purposes or reasons are irreducible elements in the explanation of human
action, they must be a species of cause. In practice, it does seem that the
social sciences by and large have never been able to avoid compatibilist
assumptions of some sort. On the other hand, Stammler seems to have been
prescient about the misguidedness of the entirely independent claim that the
social sciences should in essence adopt the natural sciences as their metho-
dological model. If the search for causal explanation has proved to be a
necessary condition of the fruitfulness of social-scientific research, the
endeavor to make it natural-scientific has generally been a sufficient condi-
tion of its stultification. Here Stammler’s emphasis on the necessarily
normative dimensions of social order is anything but passé.
From a purely legal-philosophical point of view, objections were imme-
diately raised against the claim that Stammler’s normative conclusions
about concrete questions of jurisprudence were really determined simply
by a universally valid formal method and his “pure” and “unconditioned”
social ideal and principles of correct law. Even accepting those claims, it
was difficult to see how Stammler’s answers were the uniquely “correct”
ones to draw from the principles in the specific cases he discusses, as
opposed to some among many plausible options compatible with his
highly abstract normative criteria. In fact, both the social ideal and the
principles of respect and participation seem far from being purely “formal”
and lacking in substantive normative content; rather, they are clear expres-
sions of “the basic structural principles of modern individualistic socie-
ties.”27 Indeed, if they lack substantive normative content, then they may
27
W. Kersting, “Neukantianische Rechtsbegründung: Rechtsbegriff und richtiges Recht bei Cohen,
Stammler, Kelsen,” in Alexy, Neukantianismus und Rechtsphilosophie, 58. The following argument
largely follows Kersting’s discussion.
be simply tautological re-descriptions of the bare idea of legality as invol-
ving semantically general rules, but then they would be normatively
unacceptable since they would rule out almost no factual relations
among persons given that any desired kind of subordination, discrimina-
tion, or differential treatment can be formulated in a semantically general
way. For instance, if the principles (and their mutual interaction) do not
have normatively substantive content, then the exclusion of “arbitrary”
subjection to the will of another in the first principle of respect is practically
meaningless. There is no substantively neutral way of using “arbitrary”
without depriving it of all normative force.
Furthermore, as Kersting remarks, despite continually touting the eter-
nal value and indispensability of Kantian methodology, Stammler nowhere
really presents anything like a metaphysical, let alone a transcendental,
deduction of the four pairs of pure juridical concepts that he claims
underlie, in a structural parallel to Kant’s table of categories, the four
elements of his concept of law: respectively, legality/illegality (inviolabil-
ity), legal sovereignty/legal subjection (self-authority), legal ground/legal
relation (bindingness), and legal subject/legal object (will).28 Instead, he
proceeds everywhere as if the main issue is an inductively arrived at general
classification of different types of conscious volition or norm.29 Even
Natorp, perhaps Stammler’s most sympathetic critic, reproached him for
this (Natorp 1913: 52). Natorp then tried to supply such a deduction
himself, a task that was supposedly as “indispensable” as it was “easy
to fulfill” (Natorp 1913: 55ff.). Judging from the subsequent history of
philosophy in general, and the philosophy of law in particular, most have
disagreed and viewed it as both unlikely to succeed and quite dispensable.
Oddly, Kant apparently did so too: while one finds many of the usual
Kantian devices in his writings on law and politics, e.g., antinomies,
postulates, the constitutive/regulative distinction, and so on, nowhere is
there a trace of the table of categories.
Cohen’s methodological analogy: law as the mathematics

of practical life
A different criticism of Stammler, launched at him by none other than the
father of the Marburg School, Hermann Cohen, ignited the only public
28
For presentations of the categories, see Stammler 1911: 116ff.; 1923b: 898. In German, the categories
are, in the order presented in the text prior to this note: Rechtsmässigkeit/Rechtswidrigkeit,
Rechtshoheit/Rechtsunterstelltheit, Rechtsgrund/Rechtsverhältnis, and Rechtssubject/Rechtsobjekt.
29
Kersting, “Neukantianische Rechtsbegründung,” 48.
debate between its members on legal-philosophical questions and led to
Natorp’s only publication on the topic as he attempted to mediate, to no
avail, between the quarreling erstwhile brethren.30 According to Cohen,
Stammler had severed the ties between ethics and law in a most unforgi-
vable way, assigning ethics the office of determining us an inner purity of
intention and law the entirely distinct one of externally regulating our
interaction, and leaving them without any substantive points of contact in
such a way that ethics and philosophy are “given up and surrendered”
(Cohen 1904: 214). Against this, Cohen militantly asserted that “there is no
conscience without action; no individual in the ethical sense without the
community of law” (Cohen 1904: 214). Natorp reasonably objected that
this was quite unfair since Stammler still based both ethics and law on the
same basic normative idea of an unconditional, complete harmony of ends
even though they were assigned different jurisdictions (Natorp 1913: 45). In
fact, Stammler’s demarcation of morality and legality is broadly in line
with Kant’s own.
It is unfortunate that Cohen’s Ethik des reinen Willens remains untrans-
lated, since it is probably the most important statement of the Marburg
School regarding practical philosophy. It is also a work that unequivocally
displays the marks of a certain kind of genius. The book is as full of highly
suggestive, original and often sketchily developed ideas as any of
Nietzsche’s, and as replete with inventive, rhetorically effective but ques-
tionable wordplays as any of Derrida’s. Of course, Cohen’s scholasticism
and Mandarin academic mindset distinguish him from those two figures.
For our purposes, however, the most intriguing aspect of the book may also
be its most dubious one. So far as I know, he stands alone in the list of
philosophers who have won a permanent place for themselves, at least in
works on the history of philosophy, in asserting that legal science
(Rechtswissenschaft) has a foundational role for the entire domain of the
human or cultural sciences (Geisteswissenschaften).
This daring claim, likely to excite the novice student of law and dismay
the seasoned jurist, is structured by an analogy with the place of mathe-
matics in the modern natural sciences and by implication in transcendental
philosophy understood as the critique of knowledge (Erkenntniskritik) or,
more simply, “logic.” The three-part analogy runs as follows – ethics : legal
science : sociocultural sciences :: logic : mathematics : natural sciences (Cohen
30
For a detailed discussion, see C. Müller, Die Rechtsphilosophie des Marburger Neukantianismus:
Naturrecht und Rechtspositivismus in der Auseinandersetzung zwischen Hermann Cohen, Rudolf
Stammler und Paul Natorp (Tübingen: J. C. B. Mohr, 1994).
1904: 62). Intriguing as the idea is, few if any have found this analogy
persuasive. It seems to go wrong from the beginning and to result from a
forcing of the subject-matter of law and ethics into the structural require-
ments of Cohen’s “transcendental method,” which requires philosophical
reflection to begin with the “fact” of some “science” or other. Criticizing
this piece of “Neo-Kantian scholasticism,” Lübbe has trenchantly observed
that although “human knowledge is indeed preeminently realized in the
exact natural sciences[,] moral, legal and political action are realized not
primarily in legal and political science, but rather in law and politics
themselves . . . The law’s original place is in the order of public life . . .
and to this extent it is methodologically artificial for Cohen to take the
basic concepts of his ethics from legal science, instead of the world of
ethical life, from the state and society themselves.”31
Cohen’s (again methodologically driven) criticism of Kant is indicative
of his failure to appreciate exactly this point. In an apparent bid to be
more Catholic than the transcendental-philosophical Pope, he diagnoses
Kant as having committed an “incurable error” by not seeking an appro-
priate “fact of science” out of which to develop his ethics in the Critique of
Practical Reason in a way parallel to the role (supposedly) played by
mathematical natural science in the Critique of Pure Reason (Cohen
1904: 215).32 In doing so, Cohen curiously glides over the quite different
relationship that exists in Kant’s view between ordinary common sense
(gemeiner Menschenverstand ) and the metaphysics of morals, on the one
hand, and ordinary common sense and the metaphysics of natural
science, on the other.
Kant’s own organization of the Groundwork for the Metaphysics of Morals
and the Critique of Practical Reason, as well as the famous impact on him by
Rousseau’s egalitarian and, so to speak, populist view of moral epistemol-
ogy, are further points that count heavily against Cohen’s thesis of the
uniform validity of a single “transcendental method” or philosophical
reflection about all areas of rational inquiry. In the Groundwork, Kant
sets out from common rational cognition of morality in order to seek
analytically the supreme principle of morals and then from there move
back synthetically to common moral cognition. In the Critique of Practical
31
Lübbe, “Neukantianischer Sozialismus,” 239.
32
He also anachronistically, and hence unfairly, criticizes Kant for not having a “precise concept of the
Geisteswissenschaften” in his system (Cohen 1904: 216). Since the differentiation of the
Geisteswissenschaften from other sciences was a nineteenth-century phenomenon, this is a silly
charge, comparable to accusing Descartes of not including some coverage of psychotherapy in his
Principles of Philosophy.
Reason, the subdivision of the Analytic is precisely the reverse of that in the
Critique of Pure Reason, moving from an examination of principles to
concepts and only then to the senses, instead of going from the senses to
concepts to principles. The latter point indicates that, in Kant’s view,
different procedures are necessary for dealing with distinct uses of reason
(and hence distinct disciplines or sciences). The former point indicates
that, with respect to practical reason, it is especially important to keep the
ordinary human being in focus at all times and not any putative moral
experts, much less whatever systematic moral knowledge we take ourselves
to have “as if laid out ‘in printed books’.”33
There is no space to go any further into the criticisms of Cohen’s thesis,
nor is there any reason since no philosopher of law of any importance took
it up. An overgeneralization of the relation of philosophy to natural science
led Cohen not simply to revise but, in the practical domain, more or less
throw Kant aside. As I suggest in the conclusion below, this overgener-
alization also afflicted the second generation and may help explain that
period’s peculiar juxtaposition of Kantianism and normative relativism.
Nevertheless, there is something in the idea that the law has a uniquely
privileged status among other social forms with respect to practical
reasoning. We arguably find in it the most sophisticated combination
of abstract concepts, general principles, historical understanding, and
specific factual details in the service of a concrete application of shared,
authoritative action-guiding norms, typically in order to settle a dispute.
Both constitutional law and case law are plausibly seen as repositories of
the practical knowledge and moral commitments of modern societies. To
the extent that there are any such commitments that are supposed to be
valid for all members (regardless of the views of any particular member),
they are reflected by and primarily embodied in the legal system. If there
is a definition of moral personality and its rights and duties that expresses
the view of the society, it is found in the basic rights, duties, and
principles of the constitutional order. Further, such elements of the
legal structure are precipitates of long social struggles involving elements
of argumentation, bargaining, conflict, moral learning, political
negotiation, and violence, a mix that ensures that the legal system is
never completely “pure,” but also never merely an appearance of power.
33
S. Edgar, “Hermann Cohen,” Stanford Encyclopedia of Philosophy (Winter 2012), sec. 10, http://plato.
stanford.edu/entries/cohen/. Edgar is alluding to Cohen’s famous claim that Kant’s transcendental
method seeks not “the ultimate formal elements of our thinking” in a psychological sense, but rather
“the highest principles of experience as actualized in printed books,” i.e., scientific works (die obersten
Grundsätze einer in gedruckten Büchern wirklich gewordenen Erfahrung) (Cohen 1877: 27).
Lastly, much work in the philosophy of law on the nature of legal rights
(including some of the best recent work) has made significant use of a
highly formal scheme of mutually interdefinable relations that was
unearthed by Wesley Hohfeld in an analysis of case law in an extraordina-
rily influential pair of essays.34 In fact, Hohfeld’s seminal scheme has been
taken up and generalized into a theoretical “logic” of normative positions.35
To a significant extent, it looks as if a theory that has some claim to be a
foundational conceptual framework for ethical reflection about the moral
relations between agents in potentially any domain was indeed discovered
by sifting through the facts of jurisprudence. While no doubt not identical
to the intended meaning of Cohen’s original thesis, it nevertheless serves as
a partial vindication of Cohen’s intuition, as well as a further confirmation
of the fruitfulness of his thought.
Conclusion: a conjecture and a hypothesis

I would like to close with a brief glance forward at the second and third
generations. In hindsight, one of the most striking things about the second
generation is that its greatest representatives, Radbruch and Kelsen, were,
despite being clearly in an important respect Neo-Kantians, self-described
relativists. Yet the bare notion of a Kantian relativism seems oxymoronic,
so this raises a pressing intellectual question concerning, so to speak, the
conditions of possibility of such a position. While many factors must have
been at work, including broader social and political circumstances and the
influence of Max Weber, I offer a conjecture about two factors present
within Neo-Kantianism itself.
Firstly, there is what I will describe without much elaboration as its
transcendental scientism, something we saw in the discussion of Cohen’s view
of the methodological importance of legal science for ethics and the cultural
sciences (Geisteswissenschaften).36 No doubt the Neo-Kantians intended to
raise the core concerns of all areas of culture to what they saw as the dignity
of “scientific” (wissenschaftlich), i.e., genuine objects of philosophy. But in
doing so they also raised expectations of rational justification so high and took
34
M. Kramer, “Rights Without Trimmings,” in A Debate Over Rights (Oxford University Press, 1998),
7–113; H. Steiner, “Working Rights,” in A Debate Over Rights (Oxford University Press, 1998),
233–303; R. Alexy, A Theory of Constitutional Rights (Oxford University Press, 2010), 132ff.
35
L. Lindahl, Position and Change: A Study in Law and Logic (Dordrecht: Springer, 1977).
36
While I disagree, on some readings even Habermas, and hence much of the third generation, have
not escaped the baleful consequences of this legacy. See D. Rasmussen, Reading Habermas (Oxford
and Cambridge, MA: Blackwell, 1990).
those of methodological form in such one-sided directions that they were
almost bound to be disappointed in the contentious domain of morality, law,
and politics. Hence the second generation could confidently dismiss Natorp’s
own confident insistence on the supra-empirical, unconditional, normative
“Anhypotheton” of Platonic–Kantian transcendentalism and let it be sub-
merged by the moment of “Hypothesis” that supports conditioned empirical
cognition (“scientific” in the narrower sense), and thereby distort Cohen’s
own understanding of hypothesis (Natorp 1912b: 216).37
Secondly, there is a deep structural tension going back all the way to the
Sage of Königsberg himself: neither Kant nor his followers succeeded in
fully articulating the distinctive nature of the Kantian view of practical
reasoning and moral psychology in a schematic form whose clarity and
distinctness could compete with the traditional view shared by both
classical philosophy and modern philosophy since Hume. Both of these
latter, otherwise distinct and competing, approaches treat practical reason-
ing as if it could be exhaustively modeled using a correlated pair of dualistic
practical schemata: means/ends and beliefs/desires (in Hume’s terms, reason
and the passions). Kant had already implicitly added an additional element
in his invocation of “maxims” and “laws,” the former being a subjective
and the latter an objective aspect of this putative third dimension of
practical reasoning. But he never found a way to express this element of
his view in a sufficiently perspicuous way.
Stammler’s social and legal philosophy displays this unstable position
in dramatic fashion. His main “transcendental” argument is effectively
that social order is impossible without a shared commitment to common
rules that is not dependent on any empirically contingent desires or ends
of the sociated agents. Yet he defines the sciences founded on this insight
as “sciences of ends” (Zweckwissenschaften) and the social ideal he thinks
is implied in it in terms of a harmonious unity of ends. Perhaps he is
spellbound by the apparent parallelism between rules connecting causes
and effects in the domain of nature and rules connecting means and ends
in the domain of culture.38 In any case, the conceptually distinct moment
of the norm disappears entirely, despite the fact that Windelband had
earlier claimed in an influential essay that “norm” is the central concept
of the Critical philosophy as a whole (Windelband 1911: 59–99, 88).
37
For Kelsen’s simultaneous dismissal of Kant’s (and Neo-Kantian) practical philosophy and adoption
of (Neo-)Kantian “transcendental method,” see Kelsen 1998. On Cohen’s notion of hypothesis and
Kelsen’s legal theory, see G. Edel, “The Hypothesis of the Basic Norm: Hans Kelsen and Hermann
Cohen,” in Kelsen 1998: 195–221.
38
See Natorp 1913: 47.
Unfortunately, the same occlusion occurs in the Southwest School’s
ultimate decision to give the axiological concept of value pride of place
instead of Windelband’s earlier uses of deontological vocabulary. For
the relation of values to each other in a rank-ordering can easily seem to
be entirely dependent on contingent empirical factors, if only by default
in the absence of some clear, self-evident, a priori criterion for ranking
them. Since there is no such ranking principle, the logical result is
Radbruch’s systematic, value-theoretical relativism, in which the striving
for correctness (justice) is transcendentally necessary but the standard of
achieving it (the determination of its content) is relative to irrational
value choices (Radbruch 1914; 1950: 47–244).39 In any case, values are just
highly abstract and general descriptions of sets of more concrete and
specified ends.
The twofold picture of practical reasoning and the motivation of
action in terms of means (beliefs about causal connections between
available actions and resulting states of affairs) and ends (desires for the
realization of certain of those states of affairs) sits awkwardly with the
threefold division of rational capacities into thinking, willing, and feeling
and dimensions of culture into science, law, morality, and art. The
cultural institutions of science and art can both be plausibly seen as
specializing in the production and evaluation of cognitive claims to
truth (beliefs, means) and aesthetic claims to beauty or desirable
experiences (possible ends). But, while Cohen, Stammler, and
Radbruch could offer rightness or justice as the value (or validity-claim)
in which morality, law, and politics specialized, they could not consis-
tently offer a clear place for this third dimension in the picture of
practical reasoning itself. Norms could then easily look derivative or
instrumental, i.e., become hypothetical imperatives or useful only rela-
tive to certain ends, as opposed to indispensable constitutive elements of
practical reasoning. The ever-present Neo-Kantian trope or concept of
unity cannot provide a satisfactory substitute for the function of norms as
direct preferences (or dispreferences) over actions in contrast to desires as
preferences over states of affairs. Indeed, the notion of unity tends to
become regulative, i.e., an end to be striven for, and in the practical
domain it is commanded that one strive for it (it is an “obligatory end,”
another paradoxical concept). Such a construction also tends to compel
39
Lask’s discussion of law, which heavily influenced Radbruch, is also entirely dominated by the
concept of “end” (Zweck). For instance, see his discussion of whether law serves only as means to pre-
social individual ends or an intrinsically social sort of value or end (Lask 1923: 292).
one to represent all action-coordination as cooperation for some
common shared end. But that is almost surely a false model.40
Nevertheless, Stammler in particular (and in his own way Cohen too)
was on the right track in focusing on the concepts of society and interaction
and on the institution of law. It would be left to further developments in
social theory in Durkheim and Parsons, and more recently Habermas,
game theory and evolutionary anthropology to both establish the consti-
tutive, irreducible contribution of norms and normative motivation to
social order and provide a way of explicitly incorporating it into a formal,
theoretical model of practical rationality.
This leads me to the hypothesis with which I will conclude. The
discourse theory of the third generation (Habermas, Günther, Alexy),
supported by some of the cited recent work in theoretical social science,
has in my view demonstrated that sociality itself is unavoidably constituted
by forms of normativity that are irreducible to instrumental rationality, in
particular the normativity involved in the use of language in communica-
tion and in the coordination of action. This means that even if one grants
that the determinants of legal content are ultimately to be found in “social
facts” alone, as positivists assert, a connection to the normativity of reason
and morality is inescapable, for social facts are themselves in part norma-
tively constituted.41 However, unlike Stammler, it has not identified this
fundamental, socially integrating form of normativity with law. This
allows an even greater degree of contingency, relativity, and facticity to
enter into the law than Stammler envisaged and most importantly avoids
absolutizing the kind of validity claimed by law in the way that Stammler
and arguably Kant before him did. On their construals, the flouting of law
was tantamount to the undermining of social order as such. And while it
attributes to law and jurisprudence an utterly crucial role in providing a
functional backstop or safeguard for the integration of society (Habermas
1997: chs. 1–4, 7–8), and hence also gives to jurisprudence, like Cohen, a
role in constituting and making explicit the normative structures that hold
the various social spheres together (Habermas 1976: chs. 5–6, 9), it does not
try to locate the “mathematics” of the human sciences in legal science, nor
does it root law and modern legal systems per se in an extra-societal moral
or rational foundation as natural lawyers do.
40
It also encourages Hobbesian and Humean views of society as cooperation for mutual benefit. The
normative order of society is then a means to that “shared” end.
41
For the positivist “sources thesis,” see Joseph Raz, The Authority of Law: Essays on Law and Morality
(Oxford University Press, 2009).
And even if the traditional opposition between natural law theory and
legal positivism is more or less obsolete, as Raz recently objected against
Alexy,42 by embedding the theory of law in a comprehensive theory of
society, Habermas and his discourse theory have given Neo-Kantian
philosophy of law a promising platform for addressing the task of
adequately capturing the peculiarly dual nature of law as described by
Coleman, its position between other forms of sociality and morality, or as
Habermas puts it, between social facts (Faktizität) and normative validity
(Geltung) (Habermas 1996). But the original ones who saw that this latter
opposition loomed beyond the tired and time-worn antagonism of natural
law and positivism were the Neo-Kantians. I submit, then, pace Kuehn,
that their value remains quite actual.
42
On the obsolescence of the distinction, see J. Raz, “The Argument from Justice, or How Not to
Reply to Legal Positivism,” in G. Pavlakos (ed.), Law, Rights and Discourse: The Legal Philosophy of
Robert Alexy, ed. George Pavlakos (Oxford: Hart, 2007) and the literature cited therein. For Alexy’s
uncompromising reply, see R. Alexy, “An Answer to Joseph Raz,” in Pavlakos, Law, Rights and
Discourse, 37–58.
chapter 8
Neo-Kantianism and the social sciences: from

Rickert to Weber
Gerhard Wagner and Claudius Härpfer
The legacy of the Southwest German School

of Neo-Kantianism
The legacy of the Southwest German School of Neo-Kantianism seems
evident.1 A century after Wilhelm Windelband’s inaugural address on
“History and Natural Science” (Windelband 1894, 1980) and Heinrich
Rickert’s treatise on The Limits of Concept Formation in Natural Science
(Rickert 1896, 1902, 1986), their distinction between nomothetic and
idiographic (Windelband), or generalizing and individualizing (Rickert),
sciences remains entrenched.2 Even the prestigious Gulbenkian
Commission on the Restructuring of the Social Sciences used this “antinomy”
to reconstruct the history of the social sciences and make a case for the
reconfiguration of these disciplines in the direction of the natural sciences.3
The contemporary conception of this distinction is superficial and does
not do justice to either Neo-Kantianism or the social sciences. Even though
the Neo-Kantians understood the distinction as an antinomy, it holds
true – as Rickert himself pointed out – only in a very limited sense: in
the case of mechanics, the most fundamental natural science, and history,
insofar as it is concerned with personalities. All other disciplines, from
physics, chemistry, biology, and psychology to the social sciences, the
cultural sciences, and many of the historical sciences, are defined as hybrid
forms, located in a continuum between these poles.4
1
Thanks to Tom Kaden, Peter-Ulrich Merz-Benz, Kai Müller, Angelika Zahn and – especially – Guy
Oakes who helped us to get over the limits of our concept formation.
2
See A. Tucker, Our Knowledge of the Past: A Philosophy of Historiography (Cambridge University
Press, 2004).
3
I. Wallerstein et al., Open the Social Sciences: Report of the Gulbenkian Commission on the Restructuring
of the Social Sciences (Stanford University Press, 1996), 10, 95.
4
A. Zijderveld, Rickert’s Relevance: The Ontological Nature and Epistemological Function of Values
(Leiden and Boston: Brill, 2006).
171
172 gerhard wagner and claudius härpfer
The social sciences employed the antinomy far less frequently than is
generally supposed. Especially for someone like Max Weber, who studied
Rickert’s works intensively and employed Rickert’s ideas in his own writings,
the differences between generalizing and individualizing sciences were “not
in themselves as fundamental as they might look at first glance” (Weber 1982:
173; 2012: 115). Weber developed a social science that is grounded in Neo-
Kantian premises and proceeds in either a generalizing or an individualizing
direction, depending on its scientific objective. By individualizing, social
science becomes a cultural science. By generalizing, it becomes sociology,
which is emancipated from its roots in the philosophy of history. It is worth
noting that both kinds of social science are already oriented in the direction of
the natural sciences, as proposed by the Gulbenkian Commission. This is due
to the fact that the cultural sciences must adopt a natural scientific orienta-
tion in order to achieve their objective.
Natural sciences and historical sciences

According to Rickert, both the natural sciences and the historical sciences are
empirical disciplines: both investigate reality as an object of sense perception
(Rickert 1902: 31–47, 336–370; 1986: 78–99). The empirical world consists of
an infinite multiplicity of concrete perceptual things that Rickert terms
individuals. The multiplicity is extensively infinite because there is an infinite
number of individuals. It is intensively infinite because no single individual
can be exhaustively described. Thus the empirical world, regarded indepen-
dently of some conceptual apparatus, is unknowable. It becomes an object of
knowledge only by virtue of simplification, such as ordinary language achieves.
Limitations of linguistic reference solve the problem of the extensive infinity of
reality, and limitations of linguistic meaning or content solve the problem of
intensive infinity. Scientific knowledge, therefore, is possible only by forming
concepts, the terms of which limit the referential range and content of reality.
The natural sciences conceive of reality in general terms in order to
arrive at knowledge of laws (Rickert 1902: 47–146, 228–248; 1986: 36–45).
They define the scope of a concept by using it to refer to all things in time
and space that have at least one specific property in common. Such a
concept has the logical form of a general concept of things. For example,
the concept body. This concept refers to all things that have, for instance,
the property mass. The natural sciences define the content of a concept by
formulating judgments (propositions) which articulate their knowledge of
the laws concerning the investigated things, properties, and their relations.
These judgments hold general validity for all the things, properties, and
Neo-Kantianism and the social sciences 173
relations that the judgments cover. They have the logical form of general
relational concepts in which general concepts of things designate the
properties. An example of such a general relational concept is the proposi-
tion that all bodies attract each other mutually due to a force of attraction.
This force is defined as the product of their masses, divided by the square of
their distance and multiplied by a constant. Since Isaac Newton, this
general relational concept is known as the law of gravity.
Thus, the natural sciences begin with general concepts of things and
attempt to arrive at general relational concepts. To achieve their objective
of nomological generality, they dissolve things progressively into relations
of their parts and their properties. This reduction eliminates the concrete
perceptuality and qualitative diversity of a thing. As a result, concepts of
things are translated into relational concepts. In the final analysis, things
consist of relations between so-called ultimate things, which cannot be
further reduced or analyzed. These imperceptible, indivisible atoms are not
only absolutely equivalent; they are also unalterable. All diversity and
variability of concrete empirical reality are a consequence of the quantita-
tively definable (i.e., measurable) motions of this fundamental substrate of
ultimate things in time and space. These motions are governed by ultimate
laws expressed in ultimate general relational concepts.
Rickert charts a system of the sciences, locating the disciplines of the
natural sciences in a continuum by considering the degree to which they
have analyzed things into relations of their parts and their properties, as well
as the translation of concepts of things into relational concepts (Rickert 1902:
268–289). At one end of this continuum is sociology, which had not yet
succeeded in formulating general relational concepts. Adjacent to sociology
is psychology followed by biology, chemistry, and physics. These disciplines
progressively established laws and formulated them in general relational
concepts without using concepts of things. At the other end of the con-
tinuum is mechanics, which Rickert calls the ultimate natural science.
Mechanics has transcended the concrete perceptuality of the world entirely,
conceptualizing an abstract world of atoms and their motions, to which all
concrete phenomena are reduced. Of all the natural sciences, mechanics
comes closest to their ideal of knowledge. It employs no concepts of things,
with the exception of the concept of ultimate things, and attempts to
establish knowledge of ultimate laws, expressed in a system of general
relational concepts formulated in mathematical equations.
As the title of his treatise The Limits of Concept Formation in Natural
Science suggests, exhaustive knowledge of reality is not possible by con-
sidering only the general and the nomological (Rickert 1902: 248–264;
1986: 45–60). Indeed, Rickert is convinced that the natural sciences can
make a claim to knowledge of reality only in a very limited sense. As things
are progressively reduced to abstract relations, what defines them as real –
namely, the perceptual qualitative diversity that Rickert calls individual-
ity – disappears. There is a profusion of things that interest us not from the
standpoint of nomological generality, but because they are relevant for us
as individuals. The natural sciences are limited to what does not take place
at a specific point in space and time, but holds everywhere and always.
Knowledge of what actually exists at a specific point, i.e., only here and
there, now or then, Rickert calls the historical sciences. They consider
reality with reference to the particular, things in their concrete perceptual
individuality, and thus reality as such. In consequence, they are the sciences
of reality in the proper sense. The natural sciences, on the other hand,
proceed from concrete reality but abstract from it, in which case they
qualify only as sciences of concepts.
The historical sciences define the scope of a concept by reference to a
single thing at a specific position in space and time that is of general
relevance because of at least one of its properties (Rickert 1902: 336–391;
1986: 78–107). According to Rickert, this thing is related to a general value,
which makes it a so-called historical individual for everyone. Such a
concept follows the logical form of an individual concept of things.
Consider the concept Bismarck. This concept refers to a man, who is
related to the value of the nation, because of his property Founder of the
German Empire. The historical sciences define the content of a concept by
formulating judgments (propositions) which articulate their knowledge of
the investigated thing in its significant individuality. Such judgments make
a claim to general validity and have the logical form of individual relational
concepts. They do not express laws in general propositions, but rather
unique relations in singular propositions that implicate individual con-
cepts of things; for example, the fact that Bismarck provoked France’s
declaration of war against Germany by revising the Ems Dispatch on
July 13, 1870. Such individual relations can be unequivocally causal. The
natural sciences identify the necessary connection between a cause and an
effect with the necessity of a law. By doing so, however, they presuppose
an individual effect of physical or psychic properties (Rickert 1900: 63–64,
81–82; 1902: 129, 307, 412–436; 1986: 63). It is this kind of individual
causality – for example, the causality of an intentional decision – that
makes it possible to contextualize a historical individual historically; in
other words, to situate it in a historical nexus, where it originates and where
its meaning develops.
As we have seen, the natural sciences have no interest in the concrete
perceptual individuality of things. They cover them with general concepts
of things, reducing them to relations of their parts and their properties and
transforming general concepts of things into general relational concepts.
Mechanics comes closest to this ideal. The historical sciences have an
interest in things in their concrete perceptual individuality. They grasp
them by means of individual concepts of things and define their properties
with individual concepts of things that are implicated in individual
relational concepts. The sphere of the historical sciences concerned with
personalities comes closest to this ideal. All other disciplines – from
physics, chemistry, biology and psychology to the social sciences, the
cultural sciences, and history – are hybrid forms that Rickert locates in
the – now expanded – continuum of his system of the sciences (Rickert
1902: 289–304).
Rickert designated the general concepts of things that are implicated in
general relational concepts but not yet translated into relational concepts as
historical components of the natural sciences. This is because they include
residual elements of concrete perceptuality (Rickert 1902: 264–289; 1986:
113 n.9). Conversely there can be components of the natural sciences in the
historical sciences (Rickert 1902: 480–530; 1986: 113–116). Rickert considers
general concepts of things and generalizations, which can be included in
individual relational concepts. He also notes the possibility of the general
meaning of a historical individual that has its origins in a property that is
found in several individuals that are otherwise different. These individuals
are both parts of a unique whole as well as instances of a general concept in
the sense of the natural sciences; for example, the soldiers in a battle or the
citizens of a town. Therefore, the distinction between the natural sciences
and the historical sciences is relativized.
Therefore, by highlighting a fundamental difference between the
natural sciences and the historical sciences, Rickert simply means that
there are two logically different ways to consider reality; either with
regard to the general or with regard to the particular (Rickert 1902: 28–
29, 227–228; 1986: 28, 33–34). His claim that reality becomes nature
when conceived with reference to the general and history when con-
ceived with reference to the particular should be understood in this sense
(Rickert 1902: 254–255; 1986: 53–54). However, Rickert stresses that the
expression of this circumstance differs vastly from the reality of academic
life. Reality obviously is not completely indifferent towards these
two epistemological interests (Rickert 1902: 29). There are things that
have to be investigated only by the natural sciences and other things
that have to be investigated only by the historical sciences. Finally, there
are things accessible to the concepts of both types of science.
Just as an investigation of the atom by the historical sciences is out of the
question, the same holds for an investigation of the Ems Dispatch by the
natural sciences. But the development of a form of life can be considered
both from the standpoint of its singularity as well as laws of development.
This is why Rickert distinguishes a natural scientific from a historical
biology (Rickert 1902: 291; 1986: 113 n.9). The same point holds for
human society. Its parts can be considered from a general as well as a
particular aspect (Rickert 1902: 293–295). Thus sociology attempts to
produce a natural science of society, while the cultural and historical
sciences are interested in historical description. Rickert maintains that
sociology can never replace the science of history. There is no conceptual
space for an investigation of individual personalities in the natural sciences.
Astronomy, of all sciences!

In Rickert’s system of the sciences, astronomy has a special position, where
the extremes meet (Rickert 1902: 285–286, 444–449, 508). On the one hand,
astronomy is a historical science. Its field is absolutely historical, insofar as
the celestial bodies are individuals in the strict sense; they are individually
related, and these individual relations are stages of an individual develop-
ment. The fact that astronomy designates celestial bodies and their config-
urations with proper names emphasizes their eminently historical character.
On the other hand, astronomy is a natural science, since it formulates general
relational concepts with regard to these individuals. These concepts are
paradigm cases of the concept of an absolutely general law.
Since Rickert’s project was to define the limits of concept formation in
natural science, it is a delicious irony that both ideals of scientific knowledge
coincide in astronomy. At least since the publication of Pierre Simon de
Laplace’s Philosophical Essay on Probabilities in 1814, celestial mechanics was
considered as the paragon of what a science should be (Laplace 1902: 3–5).
Based on the principle of sufficient reason, which holds that every being is
caused by another, Laplace considered the current state of the universe as an
effect of a former state and as a cause of a later one. He formulated the
hypothesis of a supreme intelligence, which – at a given time – could
comprehend all the forces that are effective in nature and the respective
positions of the entities that compose it, an intelligence that could embrace
in a single formula the movements of the greatest celestial bodies as well as of
the smallest atoms. For this intelligence, everything would be certain. The
past as well as the future would be present to its eyes. According to Laplace,
the human mind offers a feeble reflection of this intelligence in the perfec-
tion it has given astronomy. Its discoveries in mechanics and geometry
enable the human mind to comprehend the past and the future states of
the system of the world. By applying the same method to other objects, it has
succeeded in deducing observed phenomena from general laws and in
predicting other phenomena that will occur, given certain conditions. As a
result, astronomy comes closer to approximating this all-embracing intelli-
gence, from which, nevertheless, it will always be infinitely remote.
The Gulbenkian Commission has noted the immense influence of
astronomy, not only on the modern natural sciences but also on some of
the leading figures in nineteenth-century social science, with Auguste
Comte and John Stuart Mill leading the way. According to Comte,
celestial mechanics was the most fundamental of all empirical sciences,
and not only because it was the first to overcome theology and metaphysics
and to formulate a positive science of nomological generalizations, such as
the law of gravity, based on observation and computation. In addition, all
other phenomena lie in its sphere: the cosmos (Comte 1903: 157, 164, 166).
Comte constructed a hierarchy of the empirical sciences, which he called
an encyclopedic law. It proceeds from astronomy via physics, chemistry,
and biology to sociology, in which it culminates. Sociology investigates
societal phenomena, in the same spirit as astronomy, physics, chemistry,
and biology: with a view to establishing laws. Comte even claimed to have
discovered one of these laws – the so-called law of three stages, according to
which mankind progresses with absolute necessity from a theological state
via a metaphysical state to a positive state.
Mill agreed that since the time astronomy had become an exact science,
it was the model for other disciplines (Mill 1974: 844–848, 875–878). In its
infancy, astronomy could only reduce the general course of planetary
motions to its causes, but not their perturbations. Later it discovered
laws that comprehend all the causes by which celestial phenomena are
influenced, assigning to each effect its precise cause. In contrast, the science
of human nature and social science are not exact sciences. However, this is
not due to some inherent impossibility. It is a result of the extreme
difficulty of ascertaining all the causes that act upon human beings. In
astronomy, the causes are few, and they change little. In the latter sciences,
the causes are innumerable and they change perpetually. These sciences
could attain ideal perfection if they enabled us to predict how persons
think, feel, and act throughout their life, with the same certainty that
astronomy predicts planetary positions and eclipses. However, they will
never attain this ideal. The multitude of causes is so great as to defy our
limited powers of calculation. Mill also makes the point that the impossi-
bility of quantifying the facts of psychology and the social sciences
produces an insurmountable impediment to exact prediction, even if the
powers of the human intellect were otherwise adequate to the task.
Although these sciences fall far short of the standard of precision realized
in astronomy, Mill claims, they are sciences in the same sense that this was
true of astronomy before the seventeenth century.
Rickert made no objection to sociology as a natural science that analyzes
the general and nomological aspects of societal phenomena (Rickert 1902:
287–288, 293–294; 1986: 133, 191–192). In his view, the science with the
unhappy name sociology had attained corresponding unhappy results.
Thus he debunked Comte’s law of three stages as a relic of a naturalistic
philosophy of history (Rickert 1902: 16–20, 608–612; 1986: 12–13, 22–25,
181–184). Oddly, he barely mentioned Mill, whom he introduced as merely
another proponent of the universal method of the natural sciences (Rickert
1902: 219–220, 225; 1986: 129). Instead, he responded to Emil du Bois-
Reymond, whose position he regarded as an advance over Comte’s positiv-
ism (Rickert 1902: 611; 1986: 183). His interest in this position is not
surprising, since it posed a direct challenge to a science of history.
In 1872, the physiologist and physician du Bois-Reymond delivered his
widely praised address on “The Limits of our Knowledge of Nature”
(du Bois-Reymond 1874: 18, 20, 29). Proceeding from Laplace’s hypothesis
of an all-embracing intelligence from which the human mind – due to its
accomplishments in astronomy – differs only incrementally, he repre-
sented astronomical knowledge as a model of all scientific knowledge.
However, scientific knowledge is limited because it cannot grasp the
essence of matter and force or the emergence of mental processes from
physical conditions. The Ignorabimus controversy opened up the
question of these limits. Rickert took exception to du Bois-Reymond’s
assumption that the Laplacean spirit had insights into history (Rickert
1902: 445). In lunar equations, the astronomer needs to give a certain
negative value to time in order to determine whether the sun as seen
from Piraeus was eclipsed when Pericles embarked for Epidaurus. So,
according to du Bois-Reymond, the mind imagined by Laplace could, by
a suitable application of its universal formula, tell us who was the Man in
the Iron Mask, or how the President was lost. Just as astronomers can
predict the day on which a comet reappears, so the Laplacean mind could
derive from its equations the day on which the Greek orthodox cross will
glitter from the mosque of St. Sophia or when England will have consumed
its last lump of coal. If the Laplacean mind plugged in t = – ∞ into its
formula, it could discover the primeval condition of all things. Conversely,
if it let t increase to infinity, it could experience the final state of the
universe. Since the human mind does not differ fundamentally from the
Laplacean mind, for du Bois-Reymond there is no argument in principle
against an astronomic knowledge of history, even though it could never
attain the ideal of astronomical knowledge.
Rickert responded to this challenge with the argument that astronomy
may indeed be concerned with individuals in the strict sense. These
individuals may be individually related and their relations may be states
in a series of an individual development. However, astronomy abstracts
from these individuals in a way that is typical of the natural sciences
(Rickert 1902: 444–449). According to Rickert, general relational concepts
cover only the quantitative aspects of celestial bodies. The distinctiveness
of all their qualitative aspects remains incomprehensible. This is widely
disregarded, since the qualitative aspects of celestial bodies are either
unknown or of no interest to a science that operates with mathematical
relational concepts. Apart from individual data on space and time, every-
thing in astronomy is general. Although celestial bodies are given proper
names, they are merely instances of general concepts and can be replaced by
any other body with the same quantitative properties; i.e., the same size,
weight, density, spatial location, etc., although they might differ in their
qualitative aspects. To write the history of our solar system, for instance, it
would be necessary to consider the qualitative differences of its celestial
bodies. However, it is impossible to deduce this qualitative diversity from
general relational concepts that are concerned with the origins and devel-
opment of solar systems. According to Rickert, the ideal of astronomical
knowledge is limited to astronomy itself. In their full concrete perceptual-
ity, individuals have both quantitative and qualitative aspects. The latter
cannot be derived from general relational concepts.
Social sciences
It is widely known that Weber associated himself with Rickert’s theory of
scientific knowledge. He agreed that reality is an extensively and inten-
sively infinite multiplicity of concrete perceptual things. In order to
become a possible object of knowledge, reality must be subjected to
generalizing or individualizing concept formation. As early as 1903, in his
article on Wilhelm Roscher, Weber made it clear that, apart from pure
mechanics and certain parts of the historical sciences, none of the
empirically existing sciences can form its concepts solely from one or the
other of these two points of view. In this respect also, he attempted to
follow Rickert’s position faithfully (Weber 1982: 6–7; 2012: 4–6).
In view of the special position that astronomy has in Rickert’s system of
the sciences, it is not surprising that Weber was also concerned with this
science. But in contrast to Rickert, he did not see astronomy exclusively as a
natural science. On the contrary, he recognized in astronomy the model of
an individualizing social science that falls within the cultural sciences.
Andrea Albrecht has called attention to this connection.5
In 1904, in his programmatic essay on “The ‘Objectivity’ of Knowledge
in Social Science and Social Policy,” Weber mentioned a “leading natural
scientist,” who believed he might be able to deduce even knowledge of
historical individuals from a system of general relational concepts. This
scientist described the ideal objective of such a treatment of cultural reality
as “astronomical ” knowledge of the events of life (Weber 1982: 172; 2012:
114). Undoubtedly, Weber had du Bois-Reymond in mind and thought it
might be worthwhile to consider the status of astronomy. On the one
hand, he took Rickert’s position against du Bois-Reymond: in principle,
reality in its individuality cannot be deduced from any general relational
concept. On the other hand, he emphasized against Rickert that the kind
of astronomy that du Bois-Reymond obviously had in mind is not a
knowledge of laws. Rather it draws general relational concepts from
other disciplines, such as mechanics, and employs them as premises. He
emphasized that astronomy has an interest in the question: which indivi-
dual result is caused by the effects of laws on an individual constellation of
celestial bodies? That is because such individual constellations are signifi-
cant for us.
Weber regarded astronomy as a science for which general relational
concepts are not theoretical objectives but conceptual tools. To establish
knowledge of a constellation of celestial bodies in its individuality, astron-
omy must refer with relational concepts to the laws that contribute to the
origin and development of this constellation. This consideration is not
altered by the fact that each individual constellation is causally explicable
only as an effect of a former individual constellation. Admittedly, reality
retains the same individual character. Therefore it remains impossible to
deduce reality from general relational concepts, no matter how far one goes
5
A. Albrecht, “Konstellationen: Zur kulturwissenschaftlichen Karriere eines astrologisch-astronomischen
Konzepts bei Heinrich Rickert, Max Weber, Alfred Weber und Karl Mannheim,” Scientia Poetica:
Jahrbuch für Geschichte der Literatur und der Wissenschaften 14 (2010), 104–149.
back into the mists of the most remote past. Nevertheless, the laws that are
valid in reality always have a role in the constitution of the individual.
According to Weber, an individualizing social science is concerned with
individual constellations as long as they are significant for us (Weber 1982:
172–174; 2012: 114–115). The same situation that we have in astronomy –
where it is seen as a limiting case – is also present here, but to a specifically
higher degree. In astronomy, celestial bodies only engage our interest with
regard to their quantitative relations that can be measured with precision.
An individualizing social science, however, investigates qualitative aspects,
especially because the social sciences are concerned with mental processes.
Understanding such processes is a task that is distinctively different from
the objectives that can be achieved by employing the formulas of the
natural sciences. However, Weber claims that these distinctions are not
in themselves as fundamental as they might appear to be at first glance.
Apart from pure mechanics, the exact natural sciences also cannot elim-
inate qualities. The question depends on how narrowly or widely the
concept of law is understood: whether regularities qualify as laws even if
they cannot be quantified. Weber notes that the inclusion of mental events
in a science does not exclude the formulation of rules of rational action. In
this time, psychology had been credited with establishing not a mechanics,
but at least a sort of chemistry of the psychic bases of social life by analyzing
its complex phenomena into their psychic causes and effects and reducing
the latter to the most elementary psychic factors.
In this context, Weber seems to be thinking of a method that corresponds
to the analysis of things into the relations of their parts and properties in
natural science (Weber 1982: 174–175; 2012: 115–116). Suppose that psychol-
ogy, by imitating the Laplacean spirit, succeeded in analyzing all observed
and imaginable causal interconnections of human co-existence into elemen-
tary ultimate factors. And suppose it could also systematize this result
exhaustively in a vast casuistry of concepts and rules that have the strict
validity of laws. Even given these assumptions, the reality of social life could
still not be deduced from such a system. But at least important preliminary
results for social science would be available. In this sense, the formulation of
such hypothetical factors and laws would be the first step to gaining know-
ledge of an individual constellation. The second step – which would be
performed by employing these results even though it would constitute a
completely new and independent task – would be to analyze and order the
historically given individual grouping of these factors and the significance of
its concrete interaction, also elucidating the reason and type of this signifi-
cance. The third step would be to retrace historically the significant
individual peculiarities of these groupings as far as possible into the past,
explaining them historically on the basis of earlier individual constellations.
The fourth and final step would be to assess possible future constellations.
For the last three steps, knowledge of these hypothetical laws as a theoretical
tool would not only be of great value, but indispensable.
Weber was convinced that an individualizing social science requires
general relational concepts for the causal explanation of individual con-
stellations (Weber 1982: 178–180; 2012: 118–119). A valid imputation of
concrete individual effects to concrete individual causes – for example,
the investigation of western capitalism as a distinctive phenomenon in its
historical development and cultural relevance (Weber 1982: 174; 2012:
115–116) – is utterly impossible without the knowledge of causal laws.
The question of whether a single individual element of a nexus is causally
important for the effect can be ascertained only by estimating the effects we
generally expect of it, as well as other elements of the complex that are
relevant for the explanation (Weber 1982: 179; 2012: 118). The certainty of
the causal connection depends on how comprehensive our knowledge of
causal generalizations is, even if these generalizations do not satisfy criteria
of precision employed in the natural science, but rather constitute adequate
causal connections expressed in rules.
The above considerations raise the question of who is responsible for this
knowledge of generalizations. According to Weber, astronomy is in the
lucky situation of drawing its general relational concepts from other scien-
tific disciplines, for instance, mechanics. For astronomy, such concepts are
not objectives, but rather cognitive tools. An individualizing social science is
not in this happy predicament. Weber was skeptical of the possibility of a
chemistry of the psychic foundations of social life. He imposed on social
science the burden of forming the general relational concepts it requires for
the causal explanation of individual constellations. Weber rejected the posi-
tion that the formation of abstract generic concepts, the knowledge of
regularities, and the formulation of connections determined by laws have
no scientific justification in the cultural sciences (Weber 1982: 178–179; 2012:
118). However, just like using general relational concepts drawn from other
scientific disciplines, the formulation of such concepts is not the aim, but an
instrument of knowledge (Weber 1982: 179; 2012: 118). Even though know-
ledge of social laws is not knowledge of social reality in its concrete percep-
tuality, it is one of the various cognitive tools that our intellect needs for that
purpose (Weber 1982: 180; 2012: 119).
Since knowledge of generalizations is a pre-condition for the causal
explanation of individual constellations, an individualizing social science
finds itself in the paradoxical situation of beginning with generalizations.
Before it can analyze and order groupings and interactions of the factors of
an individual constellation in its cultural significance, in order to make
inferences from the past and extrapolate into the future, an individualizing
social science must take a first step by acquiring knowledge of the laws that
have effects on this constellation. In other words, it has to perform the
same function as the generalizing social science that Rickert included in his
system of the science and ascribed to sociology: an account of society based
on the conceptual apparatus of the natural sciences.
Constellations
When Weber published his essay on “The ‘Objectivity’ of Knowledge in
Social Science and Social Policy” in 1904, he was not writing about sociology
but rather the exact political economy of Carl Menger, whom he regarded as
the first thinker to make the distinction between generalizing and indivi-
dualizing sciences (Weber 1982: 187; 2012: 123; Menger 1963). Weber identi-
fied in Menger’s theory a special kind of a generalizing concept formation
that he named “ideal type” (Weber 1982: 190; 2012: 125). By all accounts
Weber took the concept “ideal type” from the physicist and physiologist
Hermann von Helmholtz who had used it in 1871/3 in his theories of optics
and aesthetics (Helmholtz 1995: 281, 307–308);6 Weber’s use of the theory of
objective possibility and adequate causation, which Johannes von Kries – a
physiologist and student of Helmholtz – had developed, also shows that
Weber was not reluctant to establish contact with the natural sciences
(Weber 1982: 179, 269–270, 288; 2012: 119, 171–172, 181–182).
Therefore it is not surprising that Weber followed the ideal of
astronomical knowledge. However, it is remarkable that he employed the
astronomical concept of constellation that neither du Bois-Reymond nor
Rickert had used. In other methodological writings, Weber also used this
concept, including variations such as psychic or historical constellations,
constellations of interests, power or exchange, etc. (Weber 1982: 45, 48, 50,
57, 58, 59, 78, 80, 130, 133 [Roscher and Knies], 164, 166, 169, 172, 174, 175, 178
[Objectivity], 220, 221, 231, 250, 251, 268, 273, 280, 285, 286 [Critical Studies],
295, 338, 339, 350 [Stammler], 398 [Marginal Utility], 425, 426 [Energetical
Theories], 428, 430, 433, 452, 460 [Categories], 528 [Value Freedom]; 2012:
30, 32, 34, 37, 38, 50, 52, 84, 86 [Roscher and Knies], 110, 113, 114, 116, 118
6
See G. Wagner and C. Härpfer, “On the Very Idea of an Ideal Type,” Società Mutamento Politico 5
(2014), 215–234.
[Objectivity], 142, 149, 160, 161, 170, 173,7 177, 181 [Critical Studies], 187, 212,8
213,9 220 [Stammler], 250 [Marginal Utility], 268 [Energetical Theories],
273, 275, 276, 288, 293 [Categories], 327 [Value Freedom]). The same holds
true for Weber’s empirical studies, especially for Economy and Society (Weber
1968). Readers will search the indexes of these works in vain for the concept
of constellation. This is somewhat surprising. Dieter Henrich, who had
published a study on the unity of Weber’s methodological writings in
1952, had drawn on Weber’s use of this concept in order to develop a method
for the history of philosophy that he called “Research on Constellations.”10
Following du Bois-Reymond, Weber used the concept of constellation
as a metaphor. However, this is not a cause for criticism. The natural
sciences have also employed metaphors. In 1904, the same year Weber
published his essay on “The ‘Objectivity’ of Knowledge in Social Science
and Social Policy,” Hantarō Nagaoka published a study in which he
developed a model of the atom that was inspired by the planet Saturn
with its rings. He conceived the atom as a large positively charged ball
circled by negatively charged electrons (Nagaoka 1904). Nagaoka’s model
was used by Ernest Rutherford, whose model of the atom is known as the
planet model (Rutherford 1911). Nagaoka himself had relied on a study by
James Clerk Maxwell (Maxwell 1859), who also used metaphors and did
not hesitate to transgress the boundaries of scientific disciplines in order to
use Adolphe Quetelet’s social physics for his explanation of the behavior of
gas molecules (Maxwell 1995 [1873]). Weber was in good company.
It was Mary Hesse who first elucidated the distinctive explanatory power
of metaphors.11 According to Hesse, the two phenomena that are connected
by a metaphor (e.g., “Man is a wolf”) are two systems associated with
meanings. She regards the first system (“man”) as an explanandum
7
Mistaken translation: “unter einer konkreten politischen Konstellation” is translated as “under a
concrete combination of political conditions.”
8
Mistaken translation: “Konstellationen” is translated as “combinations.”
9
Mistaken translations: “Abschätzung der Konstellation” is translated as “assessment of the distribu-
tion of the cards”; “die Konstellation z” is translated as “the distribution [of the cards] is Z.”
10
See D. Henrich, Die Einheit der Wissenschaftslehre Max Webers (Tübingen: J. C. B. Mohr Siebeck,
1952); D. Henrich, Konstellationen: Probleme und Debatten am Ursprung der idealistischen Philosophie
(1789–1795) (Stuttgart: Klett-Cotta, 1991); D. Henrich, “Konstellationsforschung zur klassischen
deutschen Philosophie: Motiv – Ergebnis – Probleme – Perspektiven – Begriffsbildung,” in
M. Mulsow and M. Stamm (eds.), Konstellationsforschung (Frankfurt am Main: Suhrkamp, 2005),
15–30; M. Mulsow, “Zum Methodenprofil der Konstellationsforschung,” in Mulsow and Stamm
(eds.), Konstellationsforschung, 74–97.
11
M. Hesse, “The Explanatory Function of Metaphor,” in Revolutions and Reconstructions in the
Philosophy of Science (Brighton: Harvester Press, 1980), 111–124. See M. Rentetzi, “The Metaphorical
Conception of Scientific Explanation: Rereading Mary Hesse,” Journal for General Philosophy of Science
36 (2005), 377–391.
explicated in the light of the second system (“wolf”) by transferring the
meanings associated with the second system to the first. Such a “metaphoric
redescription of the domain of the explanandum” is also at stake in
translating the meanings of the model of celestial bodies into the atomic or
social spheres. In these domains, too, we have not only similarities and
differences that are more or less evident, but also neutral analogies that
enable us to formulate new questions, hypotheses, and theories.
This brings us to our conclusion. It should be clear that justice cannot be
done to either Neo-Kantianism or the social sciences by classifying
scientific disciplines on the basis of an antinomy: nomothetic and
idiographic or generalizing and individualizing sciences. This is precisely
what the Gulbenkian Commission has done. Rickert’s system of the sciences
is much more complex. It makes it possible for the social sciences to pursue
both cognitive objectives, in the course of which the methodological
orientation of the natural sciences is always present. On closer inspection,
even Rickert’s strictly individualizing historical science, which is concerned
with personalities, is not possible without nomological knowledge.
Weber’s individualizing social science, which he conceives as a cultural
science, must also be anchored in a generalizing social science.
According to Rickert, sociology is such a generalizing social science.
However, Weber’s candidate for this science was Menger’s exact political
economy – obviously, since at that time he was investigating economic
issues, and sociology had little to offer. Nevertheless, it is quite plausible to
designate the theory that Weber developed by following Menger, von Kries,
and others as sociology. This point becomes clear in a remark by Ludwig von
Mises, a student of Menger: “When Max Weber held that in the causal
explanation of cultural phenomena ‘knowledge of laws of causation’ cannot
be ‘the aim but only the means of investigation’, one must agree completely.
For historical science, sociology is an auxiliary science, but of course an
indispensable auxiliary science” (Mises 1929: 496). In fact, Weber later
turned towards sociology, replacing his individualizing cognitive aim with
a generalizing objective. In the final analysis, the social science that he
wanted to pursue was an interpretive sociology. However, its foundations
lie in this earlier auxiliary science.
chapter 9
Simmel’s Rembrandt and The View of Life

Karen Lang
Introduction
Rembrandt’s pictures in Braunschweig seem to have transfixed Georg
Simmel in 1913. Moved, maybe even overawed, by the “magnificent
Rembrandts,” he sensed that writing a scholarly monograph on the artist
would require a certain amount of courage. For the moment, he wrote on
May 4 to Margarete von Bendemann, he busied himself with unlocking
the backdoors and side alleys of his topic, for he lacked the courage [Mut]
to storm the gateway. By June 26, however, he had charged through to
become “absorbed in Rembrandt.”1 Simmel’s Rembrandt: An Essay in the
Philosophy of Art was published by the Kurt Wolff Verlag in Leipzig in
1916.2 Rembrandt was Simmel’s last monograph and his most popular
work. The context of the First World War did not disrupt the book’s
circulation, and a second edition appeared as early as 1917. Between the
book’s appearance and his death in September 1918, Simmel prepared The
View of Life, a summation of his thinking in “four metaphysical essays.”
Simmel considered The View of Life his intellectual testament. Although
he would not live to explore the raft of questions it opened, in Rembrandt
one finds the view of life from the standpoint of life, “the theme of life as a
culture-generating force,” and the meaning of ethics and values which
shape his final theory.3
1
Georg Simmel Gesamtausgabe, vol. 23. Hereafter cited as GSG, followed by volume number.
Uta Kösser (“Simmels ‘Rembrandt,’” Simmel Studies 13/2 (2003), 457) quotes the letters from
Simmel to Margarete von Bendemann. She also provides a thorough account of the lead-up to the
Rembrandt book in Simmel’s published writing and university lecture courses. Whenever possible,
references are made to English translations.
2
M. Ermarth (ed.), Kurt Wolff: A Portrait in Essays and Letters (University of Chicago Press, 1991)
offers a fascinating account of the activities in Germany of the Kurt Wolff publishing house, and of
its predominant role in establishing “new” ideas and new authors. This chapter will cite from
G. Simmel, Rembrandt: An Essay in the Philosophy of Art (New York: Routledge, 2005).
3
Simmel 2010: xiv; originally published as Lebensanschauungen: vier metaphysischer Kapitel (Leipzig:
Duncker & Humblot, 1918). Simmel returned to ideas throughout his life so it is not surprising to find
186
Simmel’s Rembrandt and The View of Life 187
Simmel’s writing in context
Simmel engaged with topics in art throughout his writing life. Considering
the remarkable breadth and versatility of his scholarship, this is not
surprising. It makes sense, too, when we remember that Simmel “came
to intellectual maturity with the appearance of modernism in Germany,”
with Impressionist painting and the founding of the Berlin Secession in
1898, and with the bursting onto the scene of Expressionist art and
literature during the first decade of the new century.4 Berlin was the perfect
setting in which to witness modern culture and to experience life in the
modern metropolis. From Simmel’s birth in Berlin in 1858, the city had
transformed from a provincial capital to a leading industrialized metropo-
lis, the fifth largest city in the world by 1910.5 As a student at the University
of Berlin, Simmel studied philosophy and history, but he combined this
with courses in Italian Renaissance language and literature (his minor
field), folk psychology (Völkerpsychologie), and the history of art.6 In 1885
he began as a private instructor (Privatdozent) in the faculty of philosophy,
under Wilhelm Dilthey, and his lecture courses included the philosophy of
art. In 1900, at the age of forty-two, with six books and more than seventy
articles to his credit, he was finally promoted to Extraordinarius.7 Simmel
collected Persian carpets, far eastern ceramics, paintings, and drawings. He
married Gertrude Kniel, a painter who had studied at the Académie Julien,
the only art academy in France to allow female students. He was
acquainted with the artists Auguste Rodin and Max Liebermann, and
this book based on earlier publications. The first chapter, “Life as transcendence,” was freshly written to
provide a cohesive introduction to the metaphysical concept of life which underpins the book; the
remaining three chapters are based on essays published since the 1890s in the distinguished humanistic
journal Logos. Andrews and Levine (Simmel 2010: xxvi–xxvii) point to the “important but largely
subterranean role” of The View of Life in twentieth-century German philosophy, especially in
Heidegger’s Being and Time.
4
H. Liebersohn, Fate and Utopia in German Sociology, 1870–1923 (Cambridge, MA: MIT Press,
1988), 140.
5
D. Frisby and I. Boyd Whyte (eds.), Metropolis Berlin: 1880–1940 (Los Angeles: University of
California Press, 2012).
6
Simmel studied philosophy with Eduard Zeller and Friedrich Harms and history under Theodor
Mommsen and Heinrich von Treitsche (who was known publicly as an anti-Semite). Simmel learned
Völkerpsychologie, the psychological study of human society and of adaptive forms of human culture,
from Moritz Lazarus and Heymann Steinthal, the founders of the field. He studied ethnology with
Adolf Bastian. David Frisby (“Georg Simmel and Social Psychology,” Journal of the History of
Behavioural Sciences 20 (1984), 107–127) reveals the origins of Simmel’s sociology in the psychological
theories of the 1880s. See also K. C. Köhnke, Der junge Simmel – in Theoriebeziehungen und sozialen
Bewegungen (Frankfurt am Main: Suhrkamp, 1996), 337–355.
7
R. H. Weingartner, Experience and Culture: The Philosophy of Georg Simmel (Middletown, CT:
Wesleyan University Press, 1962), 5. Weingartner notes that this came after a failed petition of 1898 –
this petition also an “unusually long time” in coming.
188 karen lang
with the poets Rainer Maria Rilke and Stefan George. Rilke and the
playwright Paul Ernst came to hear him lecture; Georg Lukács, Simmel’s
student, and the philosopher Ernst Bloch joined the private seminar at his
home; he met Max and Marianne Weber and he corresponded with Henri
Bergson.8
Although Simmel mingled with highbrow intellectual and cultural
figures, his consideration of art was wide and inclusive. He commented
in print on the Berlin Trade Exhibition of 1896.9 In 1899, he published the
first of many occasional pieces in Jugend, the satirical Munich journal.
Wryly titled “Momentbilder [snapshots] sub specie aeternitatas,” these
instant images step back from the new art to consider it from the perspec-
tive of eternity. To be sure, the light-hearted irony of these pieces con-
trasted with the objective tone of scientific philosophy. (Simmel, who was
also publishing in the main academic journals, signed them with his
initials.) Yet it can be argued that Simmel’s viewpoint sub specie aeternitatas
exemplified his general philosophical point of view.10 At the same time as
he stood back and surveyed, he recognized that his point of view was finite
and particular – a snapshot, even if a deliberated one, conditioned by his
own moment, his own time of modernity. Simmel’s drawing together of
the universal and the particular cannot but help have been prompted by the
modern culture he devoted himself to understanding. In modernity, “all
that is solid melts into air,” as Karl Marx famously said, and “Simmel
delighted in showing how simple truths gave way to ironies under modern
conditions.”11 He also enjoyed showing how ironies could rest on simple, if
unexpected, truths.
Simmel’s varied topics and his penchant for publishing simultaneously
in scholarly periodicals, art journals, daily newspapers, and the Sunday
feuilleton, garnered him the reputation of an aestheticist – the opposing
figure to the forbiddingly grounded German university scholar. It did not
help that he marshaled footnotes rarely or that as early as 1894, the
positivist and social Darwinist assumptions of his scientific writing
retreated in favor of an aesthetic point of view inspired by Nietzsche’s
8
As recounted in Liebersohn, Fate and Utopia, 126, with additions.
9
G. Simmel, “The Berlin Trade Exhibition,” trans. Sam Whimster, Theory, Culture & Society 8/3
(1991), 119–123; originally published in Die Zeit (Vienna), July 25, 1896.
10
Simmel’s occasional pieces in Jugend are collected in GSG, vol. 17. Simmel’s snapshots are con-
textualized in O. Rammstedt, “On Simmel’s Aesthetics: Argumentation in the Journal Jugend,
1897–1906,” Theory, Culture & Society 8/3 (1991), 125–144; and D. Frisby, Sociological Impressionism:
A Reassessment of Georg Simmel’s Social Theory (London: Routledge, 1992), 102–131ff.
11
Liebersohn, Fate and Utopia, 131.
perspectivism.12 Like Nietzsche, Simmel engaged with “the problem of
‘meaning’ within ‘subjectivist’ culture,” which is also to say, with “the
problem of modern ‘perspectivism.’”13 In the 1896 article, “Sociological
Aesthetics,” Simmel captured society as a form in which part and whole
existed in meaningful relation, and he defined the attitude of the indivi-
dual to society in aesthetic terms, such as “distance” and “self-
sufficiency.”14 In published essays, he considered the aesthetic signifi-
cance of the face, the picture frame, and the handle in light of their
aesthetic properties of form, symmetry, and purposiveness.15 These essays
demonstrated Simmel’s aestheticism and his reliance on a Kantian notion
of organicism. They also highlighted his search for meaning in form.
In the late nineteenth century, the German academy existed as the
conservative counterpart of an increasingly repressive society.16 The uni-
fication of Germany in 1871 was achieved from above and an authoritarian
state grew up around an increasingly disaffected middle class. Adding to
the repression and to the mood of despair, from 1878 to 1890, working-class
political organization was outlawed. Within the academy, the German
mandarins sided with the government politically. At the same time, these
established scholars lamented the condition of the university itself, for the
cherished ideal of unified knowledge upon which the University of Berlin
had been founded by Wilhelm von Humboldt was giving way to increasing
specialization. The German sociologists Ferdinand Tönnies, Ernst
Troeltsch, Max Weber, and Simmel, reflected on the political, cultural,
and religious fragmentation of the country in studies that challenged the
status quo. At a time when the German cultural ideal of Bildung (of self-
cultivation and unified knowledge) seemed in danger of dissolving,
Simmel’s unorthodox scholarly methods, like his disciplinary border cross-
ings, could have easily been misconstrued as symptoms of fragmentation
12
Lawrence A. Scaff (“Weber, Simmel, and the Sociology of Culture,” in D. Frisby (ed.), Georg
Simmel: Critical Assessments (London: Routledge, 1994), 257) pinpoints the date for Simmel’s turn to
Nietzsche to the summer semester of 1894.
13
Scaff, “Weber, Simmel, and the Sociology of Culture,” 259.
14
M. Betzler, “Georg Simmel,” in M. Kelly (ed.), The Encylopedia of Aesthetics (Oxford University
Press, 1998), vol. i, 288. Simmel’s article “Soziologische Ästhetik” was originally published in Die
Zukunft, 17 (1896), 204–216.
15
These essays are translated into English in D. N. Levine (ed.), Georg Simmel: On Individuality and
Social Forms (University of Chicago Press, 1971); and K. H. Wolff (ed.), Georg Simmel, 1858–1918
(Columbus, OH: Ohio State University Press, 1959).
16
For a full consideration of this context, see Liebersohn, Fate and Utopia; A. Mitzman, Sociology and
Estrangement: Three Sociologists of Imperial Germany (New York: Knopf, 1973); F. K. Ringer, The
Decline of the German Mandarins: The German Academic Community 1890–1933 (Cambridge, MA:
Harvard University Press, 1969); and F. Stern, The Politics of Cultural Despair: A Study in the Rise of
Germanic Ideology (Berkeley and Los Angeles: University of California Press, 1989).
190 karen lang
and decadence rather than as signs of intellectual creativity and methodo-
logical renewal. To make matters worse for Simmel, contemporary
German society was teeming with anti-Semitism, and his status as a
converted Jew of an upper-middle-class mercantile family played its part
in stalling his academic career.17 In spite of the range and quality of his
publications, and even in the face of his extraordinary popularity as a
charismatic lecturer, Simmel would wait until 1914 to obtain a regular
professorship, at the University of Strasbourg, distant from the vital center
of German intellectual and cultural life.18 Like other assimilated Jews,
Simmel resided at once inside and outside of the academy, in a position
in reality that dovetailed with the point of view he adopted in theory. That
he eventually championed the view of life as all-encompassing (for life
included a standpoint within and beyond itself) was in keeping with his
personal advocacy for assimilation as the only viable path toward healthy
European nations.19
Modern society and the aesthetic point of view

Like Kant’s philosophy, Simmel’s philosophy and sociology are systematic,
not scientifically complete.20 Akin to his forebear, Simmel tackled pro-
blems of knowledge, morality, and aesthetics, and he formulated a philo-
sophy of history. In contrast to Kant’s reliance on reason and the
transcendental point of view, however, Simmel’s early thought unfolded
through shifts in viewpoint. For Kant, the French Revolution was the
defining historical event of the age, a “historical sign” of the capacity of
humankind to use the courage of its own reason to make progress in
society. For Simmel, the Industrial Revolution and the processes of
modernization and modernity it unleashed gave rise to new forces and
17
In Germany, the 1880s “marked a decisive turning point” in the escalation of anti-Semitism, and in
the history of Zionism. In the words of Roland Fernandez (Mappers of Society: The Lives, Times, and
Legacies of Great Sociologists (London: Praeger, 2003), 111), “Zionists looked to Palestine or Argentina
as a possible shelter from the storm of hate, while Georg Simmel, fortune in hand, never wanted to
leave Berlin.”
18
Simmel’s publications reflect the range of his university lecture courses. The reminiscences of his
students are gathered in Buch des Dankes, 1993.
19
Pierre Birnbaum provides a revealing account of Simmel’s “cultural embrace” of assimilation and his
rejection of Zionism after the first Zionist Congress in Basel in 1897. P. Birnbaum, Geography of
Hope: Exile, the Enlightenment, Disassimilation (Stanford University Press, 2008), 123–168ff.
20
At the beginning of Sociology, a book of 1908, Simmel “noted that the idea of scientific completeness
was a self deception,” and that “the individual can reach completeness only in the subjective sense of
communicating everything that he has been able to see.” G. Simmel, Soziologie: Untersuchung über
die Formen der Vergesellschaftung, GSG, vol. xi, 31; cited in Levine and Silver, Simmel 2010: xviii.
This chimes with Simmel’s penchant for perspectivism.
experiences that tantalized, but also frayed, the individual. Where Kant
could champion reflection over reaction, the “viewing spectator” of the
French Revolution over the active participant, Simmel grappled with a
modern condition of distraction and with the consumer as a new social
type. Under the economic conditions of modernization and the social
conditions of modernity, the concept of distance loses the flavor of
human achievement it had held in Kant’s philosophy, marking, there,
the difference between the human and the animal, between the reasoning
subject and the enchanted participant. In the modern metropolis, where
distraction eroded the space of reflection, the individual had no choice but
to participate, at least to some degree. In keeping with these new condi-
tions, Simmel sought a theoretical point of view that could accommodate
and encompass the fragmentary condition of cultural life. At the same
time, he sought a means of distancing himself from the onslaught of
sensations that marked modern life. Just as Charles Baudelaire had fixed
modernity in an image of the evanescent and the everlasting in 1863,21 so
did Simmel’s aesthetic viewpoint encompass the momentary and the
eternal, the unique and the typical, the fleeting and the essential. Not
surprisingly, Heraclitus is the first philosopher to appear in “Sociological
Aesthetics,” and indeed Simmel contemplated and interpreted modern
culture under the sign of flux.
Peering under the blinking contrasts of cultural life, Simmel saw
exchange value of course, but also money. From an economic, a philo-
sophical, and an aesthetic perspective, money perfectly symbolized the
relativity of all values that epitomized modern life.22 From an aesthetic
point of view, Simmel noted money’s “absolute lack of individuality.”
Like God, money was supraindividual and all-powerful, an ideal concept
and a circulating coin or bill. And yet, in the last analysis, money was
“nothing but the relation between economic values” (Simmel 1990a: 125).
In the 1880s, Nietzsche had announced that “God is dead” and he
heralded the transvaluation of values. At the turn of the century, in his
landmark book The Philosophy of Money, Simmel pronounced money
“the secular God of the world” and he described how money, “the
absolute means,” had been “elevated to the psychological significance of
21
Charles Baudelaire’s essay “The Painter of Modern Life” originally appeared in installments in Le
Figaro, November and December, 1863.
22
Fernandez, Mappers of Society, 122. On this point, see also M. S. Davis, “Georg Simmel and the
Aesthetics of Social Reality,” Social Forces 51 (1973), 320–329; and H. Böhringer, “Die Philosophie des
Geldes als ästhetische Theorie: Stichwörte zur Aktualität Georg Simmels für die modern bildende
Kunst,” in Georg Simmel und die Moderne, 178–201.
192 karen lang
an absolute purpose” (Simmel 1990a: 232), into “a completely engrossing
final purpose governing our practical consciousness” (Simmel 1990a:
238). Kant may have sounded the death knell for metaphysics when he
put human reason on a par with Being, but he did not live to see the
world in which money replaced the old God and the new reason, to take
up value as absolute means and final purpose. The ascendency of eco-
nomic value changed the ethical value of the human being from a
Kantian end in itself into a means to an end. In The Philosophy of
Money Simmel reported that “we are alone in a world where the means
deliberately foster objectivity, impersonality, and devastating indiffer-
ence to the ultimate ends of human life” (Simmel 1990a: 469–470). In his
most well-known essay in English, “Metropolis and Mental Life” of 1903,
Simmel described cynicism and the blasé attitude as psychological
defenses against the onslaught of society, and the decline in the estima-
tion of human life and human worth.23
Simmel’s aesthetic standpoint captured the many facets of his subject.
Often, as in The Philosophy of Money, his perspective brought new aspects
to light and it exposed what received ideas and concepts had covered over.
The flexibility of the aesthetic standpoint also mirrored the modern point
of view, since the human being, especially the denizen of the big city, now
had to deal with the perceptual and psychological consequences of rapid
contrast. If money had replaced God in the new economy, it wove its
charms through incessant production. Simmel’s broad definition of culture
included its forms of religion, science, and art, and its industrial produc-
tions, namely, its commodities. In his account of the Berlin Trade
Exhibition, for example, he described the “press of [visual and aural]
contrasts” and “the shop window quality of things,” meaning the way in
which the industrial goods were exhibited in a “tempting display” that also
highlighted their utility. Here, he suggested, values were revalued, jumbled
and floating, for the division between aesthetic value and use value had
been erased in the exhibition’s “tempting displays.” At the same time, the
human being confronted this expanded swathe of aesthetic value with
diminished means, for the “differentiation of modern work” had denied
the human soul a “unified development,” turning the psyche instead into
an “impatient flux of forces.” In the attempt to overcome its shattered state,
the psyche, Simmel observed, “seeks to come alive” through “diversity of
23
G. Simmel, “The Metropolis and Mental Life,” in Levine, Georg Simmel, 324–339; originally
published as “Die Großstädte und das Geistesleben,” in Die Großstadt: Vorträge und Aufsätze zur
Städteausstellung, ed. Th. Petermann, Jahrbuch der Gehe-Stiftung 9 (1903), 185–206.
consumption and enjoyment.”24 Not surprisingly, this tactic offered only a
temporary cure. Production proceeded apace but as objective culture
advanced, it perfected its forms “at the expense of personal accessibility.”25
In 1911, Simmel called the inability of the individual to incorporate or adapt
social forms the “tragedy of culture,” and in 1916, he pronounced it a
serious crisis.26 What Simmel termed objective culture (the exterior culture
of society) and subjective culture (the appropriation of cultural forms by
the individual) were forces at odds.
Simmel and Neo-Kantianism

According to Simmel, society was formed through interaction
(Wechselwirkung) and interaction enabled the human being to flourish,
both as an individual and as an individual in society.27 The concept of
interaction originated in Kant and it featured in the contemporary theories
of Gustav Fechner, Hermann Lotze, and Dilthey.28 In the Critique of Pure
Reason, Kant tackled the problem of how we can presume unity in our
knowledge of the world (since we perceive the heterogeneity of all
substances) and he presented interaction as one of three principles which
guide our judgment in the presumption of unity. The principle of inter-
action presumed unity by claiming that at any given instant, all substances
exist in a state of mutual causal influence. Simmel’s principle of interaction
24
Simmel, “The Berlin Trade Exhibition,” 122; translation modified. Simmel developed these ideas in
the essay “Metropolis and Mental Life.”
25
A. Salomon, “Georg Simmel Reconsidered,” in G. D. Jaworski (ed.), Georg Simmel and the American
Prospect (Albany, NY: SUNY Press 1997), 111. “As far as I can see,” Simmel wrote in 1909, “the reason
for the apparent pessimism of the majority of philosophical minds regarding the present state of
culture is the widening gulf between the culture of things and personal culture.” G. Simmel, “The
Future of Our Cuture,” in P. A. Lawrence, Georg Simmel: Sociologist and European (Sunbury-on-
Thames: Thomas Nelson & Sons, 1976), 250; originally published in the Frankfurter Zeitung.
26
G. Simmel, “On the Concept and the Tragedy of Culture,” in K. P. Etzkorn (ed.), Georg Simmel:
The Conflict in Modern Culture and Other Essays (New York: Teachers College Press, 1968), 27–46;
originally published as “Der Begriff und die Tragödie der Kultur,” in G. Simmel, Philosophische
Kultur. Gesammelte Essais (Leipzig: Werner Klinkhardt, 1911), 245–277. G. Simmel, “The Conflict in
Modern Culture,” in Etzkorn (ed.), Georg Simmel, 11–25; originally published as Der Konflikt der
modernen Kultur (Munich: Duncker & Humblot, 1918).
27
Simmel relates the Kantian principle of interaction to society in the 1908 essay “How is Society
Possible?” See Levine, Georg Simmel, 6–22. Originally published as “Exkurs über das Problem: Wie
is Gesellschaft möglich?,” in G. Simmel, Soziologie: Untersuchungen über die Formen der
Vergesellschaftung (Munich: Duncker & Humblot, 1908). See also K. Schrader-Kleberg, “Der
Begriff der Gesellschaft als regulative Idee: zur transzendentalen Begründung der Gesellschaft bei
Georg Simmel,” Soziale Welt 19 (1968), 97–117.
28
Simmel would have been acquainted with all this work, though this has often been overlooked. (The
situation is not helped of course by Simmel’s notorious lack of footnoting.) Frisby, “Georg Simmel
and Social Psychology,” provides a useful account of interaction in this wider context.
194 karen lang
registered the incessant flux between elements (individuals); it also implied
the unity of social forms, since social forms were underpinned by the
principle of interaction. Importantly, Simmel’s use of interaction charted
a middle course between the conception of society as an autonomous or
reified value (“the nation” or “the Volksgeist”) and “a thoroughgoing
individualistic foundation of society” in which “social reality is reduced
to isolated atoms.”29 Within this framework, Simmel captured the richness
and variety of social interaction in particular cultural contexts, writing with
a firm grasp of the facts of history and the novelist’s eye for detail.
Simmel’s particular conception of interaction distinguished his thought
from Neo-Kantianism in Berlin. Lotze, Zeller, and Friedrich Theodor
Fischer used psychological concepts to extend Kant’s categories. By equat-
ing Kant’s a priori with the operations of brain physiology, they each
developed the new “psychologism.” Dilthey also sought to follow Kant’s
critical path by “founding an empirical science of the human mind,” but he
did not seek to ground psychology in unchanging laws of perception.30
Dilthey approached psychology through the logic of language (since con-
sciousness inhered in language) but he laid the basis of the cultural sciences
in the changing form and content of language in history. After 1883, and
the publication of Einleitung in die Geisteswissenschaften, Dilthey wavered
in the belief that psychology could provide regulative principles for the
cultural sciences, and he turned to Verstehen, a hermeneutic mode of
inquiry in which the interpreter entered into the past in order to under-
stand its meaning. For Dilthey, Vestehen promised complete historical
recovery – the reconstruction of culture and of the mental life of historical
actors – through self-reflection. Simmel rejected the search for psycholo-
gical laws “in the natural scientific sense,”31 but he believed in bringing the
interpreter into the account. Still, Simmel considered the full recovery of
the past a self-deception. His conception of interaction revealed that no
29
Frisby, “Georg Simmel and Social Psychology,” 110.
30
In the case of the cultural sciences (Geisteswissenschaften), Dilthey outlined the task of Neo-
Kantianism somewhat earlier, in his inaugural lecture at the University of Basel in 1867. “The
task of our generation is clearly before us: following Kant’s critical path, but in cooperation with
researchers in other areas, we must found an empirical science of the human mind. It is necessary to
know the laws which rule social, intellectual, and moral phenomena.” Cited in M. Ermarth,
Wilhelm Dilthey: The Critique of Historical Reason (University of Chicago Press, 1978), 74.
31
Simmel believed that the complexity of psychological phenomena prevented psychology from
arriving “at any laws in the natural scientific sense.” “Indeed so many processes simultaneously
play a part in our psyche,” he wrote, “so many forces are simultaneously effective in it, that the
establishment of a causal connection between simple psychological concepts . . . is always completely
one-sided.” Simmel 1890b: 7; cited in Frisby, “Georg Simmel and Social Psychology,” 109. Simmel
suggested that only the concept of interaction would be able to account for this complexity.
similarity existed between cause and effect; interaction was a function, not
a substance that could be recovered. Moreover, the confrontation of
individual and society included extrasocial elements that were particular
to the individual personality, and these eluded scientific analysis and
interpretation. This was also true because the meaning of forms often
exceeded the understanding of the maker.32 Where fellow academics
busied themselves in the description and classification of social psycholo-
gical processes, Simmel aimed to reveal hidden causes and constellations
of social reality through an approach that combined rational thought,
intuitive understanding, and feeling.
The turn to the philosophy of life (Lebensphilosophie)

Simmel found in Nietzsche the penetrating depth of psychological analysis
absent in “psychologism.” Nietzsche’s writing betrayed great learning, yet
his method was altogether different from the “scientific bureaucracy”
which provoked Simmel’s disparagement. Nietzsche’s ideas were the
expression of a turning inward, of “artistic intuition.”33 At the same time,
Nietzsche offered a guide for strong individualism in the wake of moder-
nity. Simmel had published four major articles on Nietzsche before his
book, Schopenhauer and Nietzsche, appeared in 1907. For Simmel,
Nietzsche was a moralist who recognized the necessity of distance and
distinction (Vornehmheit) in modern culture. On the exterior, distance
from “the herd” and from objective cultural forces that threatened to
swamp or annihilate the individual was imperative for the preservation of
the self; on the interior, Nietzsche championed an “ethic of responsibility,”
a process of self-overcoming in which one continually strived to be one’s
highest self. In Nietzsche’s psychological philosophy, the self was taxed and
hardened, but nothing less would do for experience – experience of
singularity, of the nobility of the soul, of life. Like countless others of his
generation, Simmel praised Nietzsche’s “morality of nobility” for placing
the self in service to life. If Nietzsche’s transvaluation of values had
destroyed a stable and shared sense of social value, his “morality of
32
“In some senses it is valid to say that the weaver doesn’t know what he weaves. The finished effort
contains emphases, relationships, values which the worker did not intend.” G. Simmel, “On the
Concept and the Tragedy of Culture” [1911], in Etzkorn (ed.), Georg Simmel, 41–42.
33
G. Simmel, “F. Tönnies: Der Nietzsche Cultus,” Deutsche Literaturzeitung (October 23, 1897):
column 1649. Cited in Frisby, “Georg Simmel and Social Psychology,” 113. Simmel’s opinion of
contemporary method is more fully explored in “Tendencies in German Life and Thought Since
1870,” The International Monthly 5 (1902), 93–111 and 166–184; rpt. in Georg Simmel, Critical
Assessments, 5–27.
196 karen lang
nobility” showed how value could be created from within the self, and how
meaning could be generated amidst conditions of fragmentation. Around
1911, Simmel developed a philosophy of life (Lebensphilosophie) that drew
on Nietzsche and Bergson.34 As he reported with obvious delight in a letter
to Marianne Weber, he had “set sail in search of an unknown land.”35
Inspired by Goethe’s life and work, in a book of 1913 Simmel contrasted
the condition of the contemporary individual with “qualitative individual-
ity,” the capacity to stamp one’s unique individuality on life.36 Akin to
Nietzsche’s “morality of nobility,” Goethe’s sense of “qualitative individual-
ity” made life the purpose of life. When Simmel now lowered a plumb line to
the most essential aspect of form, he not only found interaction, he also
perceived life itself. As early as the 1895 essay on Böcklin’s landscapes,
Simmel had defined the modern aesthetic in relation to modernity: like
the individual in modern society, the artist had to create form out of sharply
differentiated external and internal contexts. Great artists created form and
they achieved unity of form above and beyond necessity, above and beyond
endless decisions over this or that, either/or.37 Moreover, like actual land-
scapes, Böcklin’s painted views were self-sufficient, and Simmel prized this
quality of self-sufficiency for separating art from the viewer and bringing the
beholder into art. Like the individual in society, the artist in creation wrested
form from the elements, and Simmel considered the ability to appropriate
forms and to shape them to one’s own ends essential for individualization,
and even for human life itself. In this regard, Simmel’s Lebensphilosophie
differed crucially from Bergson’s emphasis on life as unformed, temporal
duration. Simmel conceived life and form in interaction. As Goethe’s notion
of “qualitative individuality” revealed, individuality could overcome life
when life received the stamp of individual form.
Simmel’s Rembrandt and the individual law

When Simmel gathered his courage to approach the monograph on
Rembrandt, he was ready to forge his ideas into a philosophy of art.
Rembrandt was a suitable topic for such an enterprise, not only on account
34
On Simmel and Bergson, see G. Fitzi, Soziale Erfahrung und Lebensphilosophie: Georg Simmels
Beziehung zu Henri Bergson (Constance: UVK, 2002).
35
“Ich setze nun die Segel um und suche ein unbetretenes Land.” Georg Simmel to Marianne Weber,
December 9, 1912. Reprinted in Buch des Dankes, 1993, 132.
36
Simmel, Goethe, chapter 5. GSG, vol. xv. In his letter to Marianne Weber, Simmel said that he
wanted to capture “the whole Goethe.” His book is dedicated to her.
37
G. Simmel, “Böcklins Landschaften,” Die Zukunft 12 (1895), 272–277.
of the unquestioned quality of his art but also because of its current
popularity in Germany. The Rembrandt revival, which began in France
in the mid-nineteenth century, brought the artist into the popular
imagination, as Rembrandt’s biography was shaped to fit the myth of the
misunderstood, penniless (often helpfully mad) artist, who was wedded to
his enterprise above all else.38 Shaping Rembrandt to fit the myth of the
modern artist played havoc with the facts of history; however, it made the
artist “modern,” and in nineteenth-century France, Rembrandt’s work was
thought to be a forerunner to the latest impulses in art. In mid-nineteenth-
century Germany, by contrast, Rembrandt was received as an appealing
“romantic” whose art was suffused with “Germanic” spirit and imagina-
tion.39 Julius Langbehn, whose anonymously published book of 1890,
Rembrandt as Educator, became a runaway bestseller, made the political
undertones of the romantic-germanic Rembrandt explicit when he
proclaimed the artist the embodiment of the Volksgeist, the spirit of the
German people.40 In Langbehn’s hands, Rembrandt stood for the revival
of the “race,” for a reactionary anti-modernism rooted in the Volk, which
took aim at modern artists and Jews alike. Rembrandt also resonated with
the modern German artists Langbehn despised, first in the 1870s, with the
circle around Wilhelm Liebl in Munich, and then around 1900, with the
artists of the Berlin Secession, including Liebermann, Max Slevogt, Lovis
Corinth, and Emile Nolde. For these artists, Rembrandt was at once
historical and contemporary. Rembrandt’s stature also flourished through
publication, by academic and museum-based art historians, and through
38
A. McQueen, The Rise of the Cult of Rembrandt: Reinventing an Old Master in Nineteenth-Century
France (Amsterdam University Press, 2003). Catherine B. Scallen develops this context of connois-
seurship in Rembrandt, Reputation, and the Practice of Connoisseurship (Amsterdam University Press,
2004). It should be noted that this cult of Rembrandt included the artist’s painting and etching, for
after the 1840s, etching was gaining acceptance as an art form in its own right, against the dictates of
the Academy, where only copper plate engraving was taught.
39
J. Stückelberger, Rembrandt und die Moderne: Der Dialog mit Rembrandt in der deutschen Kunst um
1900 (Munich: Wilhelm Fink Verlag 1996), 31.
40
Langbehn’s book went through 39 editions. It was naturally praised by conservative members of the
right, but also by members of the art historical profession, for its championing of the German spirit.
Suffice it to say here, this praise must be understood in the context of mounting anxiety over the
influence of French art on modern German art, which included protest against the purchase of
modern French art by leading museum curators in Berlin, Hamburg, and Bremen. Simmel reviewed
Langbehn’s book in 1890. This review, in Vossische Zeitung, contains the first mention of Nietzsche
in Simmel’s published work. Langbehn’s title, Rembrandt als Erzieher, was meant to invoke
Nietzsche’s “Schopenhauer as Educator,” and Simmel lambasted Langbehn for conceiving of
Nietzsche as a prophet of the Volk. J. le Rider (“Rembrandt de Langbehn à Simmel: du clair-
obscur de ‘l’âme allemande’ aux couleurs de la modernité,” Société 44 (1994), 145–155) suggests that
Simmel’s book was a reply to Langbehn’s.
198 karen lang
keen museum acquisition.41 Simmel’s Rembrandt arose within this rich
context. For progressive artists of the Berlin Secession, and for Simmel,
Rembrandt’s art simultaneously preserved its distance from the present and
evaporated the distance between past and present.
The immediacy, in our present moment, of Rembrandt’s works of art and
the ability of the beholder to grasp “the whole Rembrandt” is the subject of
Simmel’s Rembrandt. The book is a philosophical essay “in the strict sense of
the term”42 because Simmel isolated the artist from biography and historical
context, on the one hand, and from art-historical investigations of
Rembrandt’s “intention” and the circumstances of his art production, on
the other hand. At the point where these two paths diverged “is the
experience of art,” Simmel began, “the primary, indivisible fact that the
work is present and exercises a direct effect upon the recipient.” Seeking, in
his book, to carry out “the essential task of philosophy. . . on the phenom-
enon of Rembrandt” (Simmel 2005: 1), Simmel lowered “a plumb line
through the immediate singular, the simply given, into the depths of
ultimate intellectual meanings . . .” (Simmel 2005: 3). Nietzsche’s perspecti-
vism had alerted Simmel to the idea that an object could be considered from
many points of view, and he demonstrated this for the reader of Rembrandt.
Still, the purpose of philosophy was to delve beneath the surface in pursuit of
ultimate meaning, and, in any case, lowering a plumb line was vital since,
according to Simmel, meaning was not a substance, it was function: meaning
existed in the interaction between elements, in thousands of threads imper-
ceptibly spun, as well as in the elements themselves. In Rembrandt, “in the
depths of intellectual meaning,” Simmel discovered that meaning was elusive
to form and to intellectual means, for here the ultimate meaning, the pearl
Simmel brought to the surface, was “life” – and life spun the net in which it
was captured.
For Simmel, the organic unity of Rembrandt’s art existed in crucial
contrast to differentiated, mechanized, instrumentalized contemporary
life. To that end, Rembrandt’s art preserved its distance from the present.
Simmel unpacked this contrast through a running comparison between
the Italian Renaissance portrait and the portraits of Rembrandt. The
Renaissance portrait aimed “to capture the self-contained being, the
timeless qualitative essence of the individual,” and it achieved this through
a construction of the painting, part by part, according to a timeless,
41
This context is wonderfully rendered in Stückelberger, Rembrandt und die Moderne, 1996. See also
Knüppel, “Rembrandt und kein End.”
42
Scott and Staubmann, introduction to Simmel, Rembrandt, Simmel 2005: xii.
trans-historical or spiritual idea of form. The metaphysical–physical dual-
ism of the Renaissance portrait – its timelessness and its individuality –
removed the sitter from the stream of time, even as the portrait depicted its
subject at a moment in time. Simmel attributed the marked “reserve” and
“typical enigma” of the “classical” Renaissance portrait, of the Mona Lisa,
and of certain portraits by Botticelli, Giorgione, and Titian, to this
dualism. Reserve and enigma were contravened in this case, however, by
the depiction of the figure within describable, individual concepts of
“content or character,” and by the use of naturalism and conventionalism.
The concepts and codes of the Renaissance portrait could be analyzed,
separated out in perception, just as the unity of the painting itself, built up
part by part, could be unraveled to reveal the “atemporal logical connection
of contents” which underpinned its structure. The analysis of the
Renaissance portrait was amenable to scientific means: in other words,
the portrait’s meaning could be captured in the net of Kantian categories of
experience. In Rembrandt, Simmel diagnosed “something of the tragedy of
the German spirit” in the fact that the most popular Rembrandt sheet in
Germany at the time, The Hundred Guilder Print, was “Italianate” in
character. According to Simmel, this was the very work in which “the
pure Rembrandtian spirit” was “least evident” (Simmel 2005: 9). In its
“geometrically clear construction,” in its depiction of “this one concep-
tually graspable life moment” (Simmel 2005: 10), a little like “the living
picture,” The Hundred Guilder Print approximated “the Classical form” to
which the European artistic sensibility had “adjusted” (Simmel 2005: 13).
All this had been “acquired,” Simmel wrote, in tones dripping with the
money economy, “at the price that the darkly flowing totality of life is
reduced to it” (Simmel 2005: 168 fn. 4).43
If the Renaissance portrait exhibited a metaphysical–physical dualism,
Simmel contended that Rembrandt’s art, especially after 1634, gave form to
the totality of life. Instead of limning the sitter at a moment in time which
was also timeless, Rembrandt captured the historical dimension of the
depicted individual. Past, present, and future inhered in Rembrandt’s
portraits, Simmel said, for they were “interwoven with long running
strands of fate” (Simmel 2005: 14). Arising from the totality of life, these
portraits presented the sitter in time, as a living individual rather than as
tableau vivant, and they did so by drawing “the third dimension” into art.
Where naturalism and conventionalism related the work of art back to the
43
On the latter, see also G. Simmel, “Germanischer und klassisch-romanischer Stil,” in Simmel 1957:
160–167; originally published in Der Tag, March 2, 1918.
200 karen lang
actual world or to the world of artistic concepts, Simmel maintained that
the aesthetic quality of the artwork did not reside in a correspondence
between depiction and reality.44 Rather, the work of art existed for its sake
alone, and, when successful, it drew the beholder into its world. In what
Simmel described as “artistic procreation,” the artist created “the third
dimension” – the dimension of life, of art – which enlivened the work of
art. Simmel consequently dismissed intention, for the artist did not follow
a preconceived idea, from beginning to end. He suggested, instead, that the
work of art unfolded in the manner of a “germ cell” (Simmel 2005: 27–32).
Simmel’s conception of the germ cell presumed a functional relation of
beginning and end, yet it assumed that, in the process of making, the artist
followed the work, rather than the other way around. Taking on a life of its
own, the work of art had the capacity to surprise the artist, who would have
latched onto those unknown, imperceptible threads which spin themselves
in the creation of the work. If, in the Renaissance portrait, life and time
were construed as a sequence, in the Rembrandt portrait, the unity of life
and time could not be divided, bit by bit, but only intuited, experienced, as
the riverbed which underlay the time of the work of art. According to
Simmel, who was, himself, not immune to the chords of nationalism, the
Germanic disposition included the capacity to feel the unity of organic
being, and Rembrandt shared this spirit (Simmel 2005: 45).
For Simmel, Rembrandt grasped the human being as a “pure indivi-
dual” rather than as a type. In the individual portrait, as in the series, the
artist succeeded in presenting “the entire being.” Although one could say
that Rembrandt had stamped these portraits with his style, what was
essential for Simmel was their communication of the artist’s inner life,
not as intention or biography, but in the sense of the individual of
Rembrandt. In the words of the author, the artist and the work were
“indivisible” (Simmel 2005: 41–60). Rembrandt’s portraits communi-
cated the totality of life, and especially in the self-portraits of old age,
the artist’s view of life from the standpoint of life. Where Simmel had
taken inspiration from Goethe’s notion of “qualitative individuality,” he
now defined life as relentless and more powerful than the human capacity
to stamp life with form. Life exceeded the Kantian categories, it burst
the dam of conceptual thinking (Simmel 2005: 17, 20, 36, 69). If
Rembrandt’s art contained something of the quality of music for
44
See, for instance, G. Simmel, “On the Third Dimension in Art,” and “The Dramatic Actor and
Reality,” in Etzkorn (ed.), Georg Simmel, 86–97. This is also the reason for Simmel’s critique of
photography.
Simmel, this was because these works could not be captured in the net
of conceptual thinking. Simmel widened the Kantian categories of
experience by grounding the experience of the work of art in felt value
(Wertempfindungen) (Simmel 2005: 1).45 Rembrandt presented “the
depiction of the totality of life as a problem in painting.” Rembrandt
did not represent totality or depict psychology; instead, he brought
totality into representation (Simmel 2005: 10, 12, 21). As we, too, are of
this totality, Rembrandt’s art closes the gap in time, dissolving the
distance between past and present.
The View of Life

The theme of individualization and freedom is the leitmotiv of Simmel’s
oeuvre.46 Individualization and freedom are entwined in Simmel’s concep-
tion of the individual law. When Rembrandt discovered the individual law
in the creation of his art, he and his art became indivisible, and his art took
on a life of its own. Simmel also shows how, in his late self-portraits,
Rembrandt viewed his life from the standpoint of life, and how this view –
this immanence of transcendence – guided the making of his works of
art.47 Simmel dedicated a chapter to the individual law in The View of Life,
his metaphysical testament of 1918. In this final work, the Kantian Ought is
brought to the inside, to become internal to the process of individual
life. In Simmel’s hands, the Kantian categorical imperative is thus
reconstructed, “from a doctrine of universalizability to an imperative of
authentic individuality.” In opposition to the degradation of the individual
in modernity, Simmel’s transposition fortifies the individual as an end
in itself. Simmel’s reconstruction restores meaning and value to the
individual, and it reconceives truth, from a pragmatic conception in
which, in a manner similar to the workings of the money economy,
truth is defined as what is useful and what is useful is defined as true, to
a conception of truth as “a manifestation of the entirety of life.” As in his
theory of social interaction, so here the Ought is conceived as a regulative
principle bound to individual action, to individual making and remaking.
Guided by the whole life, by life itself, the individual law is also objective
45
After all, Simmel wrote, “the primary experience of the work of art” was “not to be determined with
objective clarity” (Simmel 2005: 3).
46
Levine and Silver, in Simmel 2010: xxiii.
47
In The View of Life, Simmel enfolded the individual law within the encompassing dialectic of “life”
and “more than life.”
202 karen lang
and moral, for here life guides us in the responsibility of every act while
every act is responsible to the whole life.48
As ever, Simmel’s theory was intertwined with his times. In 1916, and
especially by 1918, Simmel conceived of life as the riverbed from which
form arose, and from which form was reclaimed. By 1918, and the end of
the First World War, Simmel defined life “as a struggle in the absolute
sense of the term,” for the present was “too full of contradictions to stand
still,” and “the bridge between the past and future of cultural forms”
appeared to be “demolished.” As “we gaze into an abyss of unformed life
beneath our feet,” it remained unclear whether life, “which encompasses
the relative contrast between war and peace,” would ever reveal the divine
secret of “absolute peace.”49 From his outpost in Strasbourg, Simmel
perceived that, in expanding and consolidating its force in the objective
form of “the nation,” Germany had lost sight of “the idea of Europe.”50
Like life, the idea of Europe could not be attached to any one form, to any
one nation. Although the “essence” of life, and of Europe, could not be
pressed into the service of a concept, Simmel wagered that this did “not
diminish” the “momentum” of life, and of Europe, as ideas.51
48
Levine and Silver, in Simmel 2010: xxviii. My discussion of Simmel’s transposition of the Kantian
categorical imperative is much indebted to Levine and Silver.
49
Simmel, “The Conflict in Modern Culture,” in Etzkorn (ed.), Georg Simmel, 25.
50
G. Simmel, “The Idea of Europe,” in Lawrence, Georg Simmel, 267–271; originally published in
Georg Simmel, Der Krieg und die geistigen Entscheidung (Leipzig: Duncker & Humblot, 1917).
51
As Simmel put it in regard to the idea of life: “The fact that the possibilities of expression of the
essence of life are so limited does not diminish the momentum of life as an idea.” Simmel, “The
Conflict in Modern Culture,” in Etzkorn (ed.), Georg Simmel, 26; translation modified.
chapter 10
The binding of Isaac and the boundaries of reason:

religion since Kant
Peter E. Gordon
Introduction
In his 1793 work Religion within the Boundaries of Mere Reason, Immanuel
Kant refers to the Akedah, the biblical story of the binding of Isaac (Genesis
22: 1–19) in order to illustrate the principle of rational precedence.
According to this principle, reason should dismiss as invalid any apparently
divine command if such a command contravenes human standards of
morality, since such violation indicates that the command could not
have had a genuinely divine origin: “If something is represented as com-
manded by God in a direct manifestation of him yet is directly in conflict
with morality, it cannot be a divine miracle despite every appearance of
being one (e.g., if a father were ordered to kill his son who, so far as he
knows, is totally innocent)” (Kant 1999: 6:87).1 In the biblical passage to
which Kant refers, God demands that Abraham take his son Isaac to
Mount Moriah and bind him on an altar as a sacrifice to the Lord. At
the last minute, when Abraham has already drawn the knife, an angel of the
Lord calls out to him and insists that he leave his son unharmed. A ram is
sacrificed in Isaac’s stead. The angel then praises Abraham for his readiness
to obey the divine command and Abraham is assured that his seed will be
multiplied “like the stars of the heavens, and like the sand on the seashore.”
Kant finds this story provocative insofar as it suggests we should see
Abraham’s readiness to undertake the sacrifice as morally praiseworthy.
Kant himself considers it morally offensive. To the modern reader there is
indeed something unsettling about a scenario where a father who claims to
hear the voice of God could feel authorized to murder his son: For what if
Abraham is mistaken about the provenance of the command? Nor does it
1
All references are to The Cambridge Edition of the Works of Immanuel Kant. All pagination references
make use of the standard system, which gives the pagination to the standard German edition of
Kant’s works, Kants Gesammelte Schriften, edited by the Royal Prussian Academy of Sciences.
203
204 peter e. gordon
help to say that pious obedience to God must always be absolute since that
is just what piety is about. For individual believers may differ as to what
revelation requires. Kant’s solution to this difficulty was to insist on the one
standard of rational morality that is shared across all human beings. Even
when faced with the apparent commandments of God himself human
reason must not cede its authority. The principle of rational precedence
tells us that no religious revelation can be genuine if it violates the moral
principles that are intelligible to human reason. A command that directly
conflicts with morality cannot be divine in origin. In cases where reason
and revelation appear to be at odds, human reason must have the last word.
As a philosopher of the German Enlightenment, Kant must have found
the biblical scenario of the binding of Isaac especially troubling.
Enlightenment is defined as the human being’s “emergence from his self-
incurred minority” (Kant 1996: 8:33–8:42). For Kant and his contemporaries
the term “minority” (Unmündigkeit) held a precise legal meaning as desig-
nating an age of legal dependency. The irrationality of a parent’s authority
over his dependants therefore served as a perfect illustration of the condition
Kant believed humanity as a whole must leave behind. Nor could the
resemblance between religious and political absolutism have escaped his
attention. A religiously inspired patriarch who wields incontestable power
over his children is not unlike a despot who claims unquestionable authority
over his subjects by divine right. Indeed, for Kant himself such an analogy
would have held a special and very personal significance: following the death
of Frederick the Great (who was known for his tolerance) and the accession
to the throne of the more conservative Friedrich Wilhelm II, Kant came into
serious conflict with the Prussian censors for his unorthodox views on
matters of religion. The king even wrote to Kant to insist that he cease
writing on religious matters.2 But although he yielded temporarily to the
state’s prohibition he remained in principle committed to the view that
nothing in religion can with right remain immune from rational scrutiny. In
The Conflict of the Faculties (1798) he declared that although theologians
might have a duty to uphold biblical faith, this could not inhibit “the
freedom of the philosophers to subject it always to the critique of reason”
(Kant 1999: 287). For Kant the binding of Isaac was therefore a paradigmatic
tale insofar as it illustrated the general interpretative principle that, in matters
of religion and religiously grounded authority, the judgment of human
reason alone takes precedence.
2
On Kant’s 1794 conflict with Friedrich Wilhelm II, see Manfred Kuehn, Kant: A Biography
(Cambridge University Press, 2001), 378–382.
Religion since Kant 205
Later philosophers in the Kantian tradition who have addressed the
question of religion have remained more or less faithful to the idea of
rational precedence. But they have construed the idea in various ways. For
the purposes of this chapter we can distinguish between two basic strategies
of interpretation: on the first reading, the principle of rational precedence
serves as an instrument of clarification that permits us to separate the
religion of reason from its historically contingent forms. Here we should
think of the relation between them as internal: the religion of reason is
contained within historically revealed religion. On the second reading, the
principle of rational precedence means that reason possesses its very
own non-religious criteria for assessing the merits of revelation. Here the
relation between revelation and reason is not internal but external, and it is
therefore conceivable that reason might consider any and all religion as
temporary and extraneous to human morality.
The first reading finds strong support in Kant’s own texts. In Religion
within the Boundaries of Mere Reason Kant distinguishes between the
“historical faith” and the “pure religious faith” and he expresses his
philosophical conviction that over time the former will yield to the latter:
“Historical faith,” he writes, is “ecclesiastical,” and as such it “needs a holy
book to guide human beings.” But any ecclesiastical structure will inhibit
religion from reaching true universality. It is the ultimate fate of historical
faith that eventually it will “pass over into a pure religious faith which
illumines the whole world equally.” Kant therefore enjoins the reader to
work with diligence for “the continuous development of the pure religion
of reason out of its present still indispensable shell” (Kant 1999: 6:135,
note). It is important to recall that for Kant human morality points toward
or motivates certain religious ideas even if those ideas do not qualify as
genuine items of theoretical knowledge. The concept of God, for example,
is a postulate of practical reason and as such it retains a certain dignity
within the bounds of a purified and rational faith.3 On this interpretation
even an enlightened philosopher will therefore retain some commitment
to religion even if the kind of religion in question is a faith that has been
purified of nearly all its historical and particularistic markings. This is
because the historical faith given in revelation contains the pure religion of
reason – a relation Kant likens to a relation between concentric circles. The
philosopher must remain within the “inner circle” and must therefore
“abstract from all experience.” But on this view revealed religion and
3
See, e.g., A. Wood, Kant’s Moral Religion (Ithaca: Cornell University Press, 1970); and A. Wood,
Kant’s Rational Theology (Ithaca: Cornell University Press, 1978).
206 peter e. gordon
rational reason nonetheless bear an internal relation to one another and the
philosopher should abstain from treating historical forms of religion as
nothing more than irrational. To treat historical religion in this manner
unfairly sees it as essentially incompatible with rational religion, as if they
were oil and water: “[T]he two would have to be often shaken up together
that they might, for a short time, combine . . . [but] they would soon have to
separate again and let the purely moral religion . . . float to the top” (Kant
1999: 6:13). On this reading, then, Kant believes that revealed or historical
religion contains the religion of reason, and, as advocates of enlightenment,
we should work to purify historical faiths until we reach their rational core.
On the second reading, there is nothing distinctive in religion that we
must retain since all of our requisite moral concepts can be found within
the bounds of a purely non-theistic moral philosophy. From the rational
point of view not only revealed religion but all religion proves to be
dispensable in the end since non-theistic morality itself possesses all that
we require as moral agents. This second reading finds rather less support in
Kant’s own texts because Kant seems to have really believed in the lasting
importance of the religion of reason. But Kant nonetheless left consider-
able room for this reading insofar as he occasionally implied that religion
satisfies human needs that are wholly extrinsic to our moral requirements.
In Religion within the Boundaries of Mere Reason he tells us that in principle
religion does not require any “assertoric knowledge” even regarding God’s
existence. Faith requires “only the idea of God which must occur to every
morally earnest (and therefore religious) pursuit of the good, without
pretending to be able to secure objective reality for it through theoretical
cognition.” Kant concludes that religion only requires “a minimum of
cognition,” that is, the mere concession that “it is possible that there is a
God” must suffice. Furthermore, from the rational point of view no
religion can require any acts beyond those required by human morality
itself: “There are no particular duties toward God in a universal religion;
for God cannot receive anything from us; we cannot act on him or for him”
(Kant 1999: 6:154, note). But if this is the case then human morality might
appear to be largely or even wholly self-sufficient. Even the celebrated
notion of religious postulates that Kant had introduced in the Critique of
Practical Reason would not seem to be strictly entailed by the arguments of
Kant’s moral philosophy. Kant tried to retain a place for religion in his
moral system with the twofold suggestion that in our moral conduct
we require (a) the thought of everlasting life (such that our moral progress
would not be cut short without full realization) and (b) the idea of
Providence (since our ultimate happiness must be proportionate to
our moral desert). The pursuit of the highest good therefore serves as a
justification for our belief in God (Kant 1996: 5:124–132). But one might
think that such arguments support a conception of religion that remains
vanishingly thin. In the opening lines of the preface to Religion Kant
admits that “on its own behalf morality in no way needs religion” (Kant
1999: 6:3). This second reading of rational precedence therefore suggests
that religion may ultimately prove to be dispensable.
Broadly construed, Neo-Kantian philosophies of religion divide up into
those that endorse the containment thesis and those that endorse the
dispensability thesis. Among the most ardent champions of the containment
thesis was the late nineteenth- and early twentieth-century German-Jewish
philosopher Hermann Cohen, an expositor of Kant’s philosophy who is
commonly remembered as the co-founder of the “Marburg school” of
Neo-Kantianism. However, for a variety of reasons (chiefly pertaining to
the non-theistic character of the contemporary philosophical profession), the
containment thesis no longer enjoys broad acclaim. By contrast, the dispen-
sability thesis remains perhaps the single most dominant stratagem for
thinking about the relation between reason and revelation. Among the
many proponents of this thesis we may count two of the most consequential
philosophers of the twentieth century, John Rawls and Jürgen Habermas. In
what follows I will briefly address each of these three philosophers in turn, in
order to cast some light on their distinctiveness and their underlying
commonality.
Cohen and the religion of reason

Hermann Cohen was a German-Jewish philosopher who played a central
role in the Neo-Kantian revival of the later nineteenth century. During his
years as a professor at the Philipps-Universität in the Hessian town of
Marburg in western Germany, Cohen laid out a new and radically anti-
metaphysical interpretation of Kant’s critical philosophy in a succession of
three major works which he then followed with three no less innovative
contributions of his own that still bore a recognizably Kantian imprint.
Known for both his socialist political commitments and his passionate
attachment to the Jewish religion, Cohen was a figure of singular impor-
tance not only in the history of modern philosophy but also in the
development of modern Jewish religious thought. Upon his retirement
from Marburg in 1912 he moved to Berlin’s Hochschule für die
Wissenschaft des Judentums, where he lectured on Jewish themes until
his death in 1918. His final work, Religion of Reason out of the Sources of
208 peter e. gordon
Judaism (Religion der Vernunft aus den Quellen des Judentums), appeared
posthumously in 1919.4
It has been a frequent conceit in the scholarly literature on Cohen’s
intellectual career that his posthumous contribution to the philosophy of
religion marks a sharp break from his earlier contributions to Neo-Kantian
epistemology, ethics, and aesthetics.5 But this is an exaggeration. At least
some of the themes developed at greater length in the posthumous study of
religion are already anticipated at least in embryo in Cohen’s earlier,
independent ethical work, the Ethics of Pure Will (1904), and in The
Concept of Religion in the System of Philosophy (1915). Already in his Ethics
Cohen explains that from the ethical perspective the human being can only
make its appearance as the subject of universal law and must therefore
remain an exemplar of the human species in general. To discover “the
emergence of the individual” as such we must look instead to religion.
Only from the religious point can the exemplary human being be
“detached” from the species; it is only in religion that we first conceive of
the human individual as a creature who possesses a soul and is susceptible
to sin.6 Although these themes are developed and extended in Cohen’s 1915
work on the concept of religion and its relation to a philosophical system,
the themes are to reach their fullest expression only in Religion of Reason.
Cohen’s task in Religion of Reason was to develop a rationalist conception
of religion that drew inspiration from classical Jewish sources, chiefly the
Hebrew Bible. The title of the work conveys the author’s conviction that
Judaism does not enjoy any exclusive claim upon the religion of reason:
other monotheistic religions have their share in this philosophical religion.
But in Cohen’s view Judaism nonetheless figures historically as a “primary
origin” and it lays down the original pattern for the other monotheistic
faiths. It is crucial to note, however, that a religion of reason represents a
philosophical reconstruction of religion in accordance with rational criteria.
A religion of reason must therefore emerge from reason alone and not from
the senses or from the elements of our affective or animal life. “The religion
of reason turns religion into a general function of human consciousness; it
makes consciousness human” (Cohen 1995: 7).
4
Cohen 1966. I also cite the English translation: Cohen 1995.
5
See, e.g., Franz Rosenzweig’s Introductory remarks in Cohen’s collected Jewish writings:
“Einleitung,” reprinted in Rosenzweig 1937: 299–350. For an overall assessment of this idea of a
rupture in Cohen’s thought, see P. Gordon, Rosenzweig and Heidegger: Between Judaism and German
Philosophy (Berkeley: University of California Press, 2003), 39–81.
6
On this theme see N. Rotenstreich, “Religion within Limits of Reason alone and Religion of
Reason,” in Yearbook of the Leo Baeck Institute 17 (1972), 179–187; here 180.
Whereas in ethics the human self is grasped only as a “member of
humanity,” religion permits us to grasp the concept of the human being as
an individual. This is because the concept of the human being as it appears
within religion is the concept of an individual who can experience suffering
(Leiden) and who can awaken to compassion (Mitleid) (Cohen 1995: 17).
This is according to Cohen the “turning point” at which religion first
distinguishes itself from ethics. For in the relation of compassion the self
comes to understand its own identity: “I discover my own I through
the Thou” (Cohen 1995: 20). The original model for this so-called
“correlation” between the I and Thou is given in the correlation between
the individual and God. But this distinctive correlation also serves as the
prototype for the general bond of compassion between God and the suffer-
ing of humanity writ large. According to Cohen the poor are the symbol of
human suffering as such. Thus the religion of reason awakens a distinctively
political and social attitude of compassion for the poor and promotes a
program of social betterment to alleviate their suffering. Because the purely
ethical conception of the human being cannot awaken compassion, religion
alone serves as a supplement to ethics; in the correlation between God and
humanity we can discern religion’s distinctive contribution:
The religion of the unique God had to let this ray of hope rise up for men
out of the tribulations of social misery, out of the innermost contradictions
of political justice. Ethics remains unchanged in its basic theoretical value,
according to which its method has to guide the determination of human
worth; but religion has discovered objective insights and derived them from
its own principles of the concept of God and the concept of man, which
remained closed to the method of ethics. (Cohen 1995: 23)
It is important to note that Cohen did not believe this argument strayed
beyond the proper domain of reason. His conception of God was merely that
of an idea: it implied no experiential or mystical commitments. On the
contrary, Cohen hastened to explain that the essential correlation between
God and humanity is forged through reason alone: “Man’s spirit is based on
God’s spirit,” he explained, “not only as a living creature, or only as an
intellectual creation, but his reason, which in an eminent way is moral reason,
is also derived from God” (Cohen 1995: 87). This idea of “created reason”
served a crucial purpose in Cohen’s system insofar as it justified the idea that
religion and reason were in no sense opposed: human reason could grasp
God because, in a very literal sense, reason and God were presumed to be
identical. In this respect Cohen remained a partisan of a long-enduring
tradition of “philosophical religion” that spanned the millennia from Plato
210 peter e. gordon
through Maimonides and possibly beyond.7 The identity of God and
reason, moreover, provided the crucial justification for Cohen’s philosophi-
cal project: like Kant before him, Cohen did not wish to see the relation
between reason and revelation as one of stark opposition. But Cohen
arguably went even further than Kant: where Kant only wished to see a
religion of reason as contained within revelation, Cohen actually insisted on
their thoroughgoing identity.
Rawls on comprehensive doctrines

The American philosopher John Rawls ranks amongst the most important
political thinkers of the twentieth century, and he is also indisputably the
most distinguished of the many political philosophers in the English-
speaking world who have remained faithful to the Kantian legacy in spirit
if not in letter. Although as an adult Rawls was not himself religious – he
lost his faith following his service as an infantryman in the Pacific theater of
the Second World War – he was nonetheless remarkably knowledgeable
about religious doctrine. His Princeton senior thesis, “A Brief Inquiry into
the Meaning of Sin and Faith,” serves as a reminder of the sincere
religiosity of the young Rawls in the years before he adopted a purely
secular political philosophy (Rawls 2009).
In the major work that Rawls wrote in the earlier phase of his career, A
Theory of Justice (1971; revised 1999), religion plays only a quite limited role.
The counterfactual scenario of the original position from which we are asked
to imagine the basic contours of a just society abstracts us from all thick
features of worldly identity, including religious affiliation. By means of this
thought-experiment Rawls believed that members of a social order would
grasp the basic principles of justice that underwrite a liberal polity. Justice
would entail that “All social values – liberty and opportunity, income and
wealth, and the social bases of self-respect – are to be distributed equally
unless an unequal distribution of any, or all, of these values is to everyone’s
advantage” (Rawls 1999: 54). Rawls felt that the lessons of the original
position were such as to require assent from any reasonable person regardless
of other commitments – including religious commitments. He therefore
considered religious obligations a matter of indifference from the point of
view of political justice: we cannot ask those of another faith to share our
own religious perspective and for this same reason they cannot place their
7
On this theme see C. Fraenkel, Philosophical Religions from Plato to Spinoza: Reason, Religion, and
Autonomy (Cambridge University Press, 2014).
own religious obligations above the demands of political justice (Rawls 1999:
182–183). “The state can favor no particular religion and no penalties or
disabilities may be attached to any religious affiliation or lack thereof.” But it
does not follow that religious liberty is absolute: “Liberty of conscience is
limited . . . by the common interest in public order and security.” But on the
other hand the government “has neither the right nor the duty to do what it
or a majority . . . wants to do in questions of morals and religion. Its duty is
limited to underwriting the conditions of equal moral and religious liberty”
(Rawls 1999: 186–187). Rawls insisted that this doctrine of governmental
neutrality with regard to religious identity did not imply any kind
of metaphysical bias against religion. Liberty, he wrote, does not imply
philosophical skepticism or indifference toward religion (Rawls 1999: 188).
It implies only that a certain criterion of liberty agreed upon by all reasonable
citizens takes precedence over any other criterion in laying the groundwork
for a just social order.
In his later work, Political Liberalism (1993; revised 2005), Rawls had a
great deal more to say concerning the place of religion in modern public
life. More specifically he tried to develop a framework for explaining how
citizens who may subscribe to different and distinct “comprehensive
doctrines” (i.e., metaphysical or religious worldviews or fundamental
philosophical perspectives) might nonetheless engage with one another
in a common medium of “public reason.”8 Rawls now recognized that a
modern democratic society cannot assume uniform agreement on
comprehensive doctrines. It must instead develop principles of public
reason that take cognizance of the fact that its citizenry may subscribe to a
wide variety of comprehensive doctrines that are “incompatible yet
reasonable.” Because a liberal government cannot expect that all citizens
will come to agreement on such basic matters, it must develop a frame-
work for reasonable deliberation that allows for this pluralism and does
not privilege any one of them over the others. In fact a theory of political
liberalism will see that such a plurality is “the normal result of the exercise
of human reason within the framework of the free institutions of a
constitutional democratic regime” (Rawls 1993: xvi).
Here we can see that Rawls deploys the instrument of rational precedence
as a criterion for public deliberation. In the course of our ongoing public
discourse with our fellow citizens within a liberal order, public reason takes
precedence, whereas religious commitments are permissible only so long as
they do not conflict with the fundamental and reasonable commitments of
8
See especially Rawls 2005.
212 peter e. gordon
the democratic order itself: “Since justification is addressed to others, it
proceeds from what is, or can be, held in common; and so we begin from
shared fundamental ideas implicit in the public political culture in the hope
of developing from them a political conception that can gain free and
reasoned agreement in judgment” (Rawls 1993: 100–101). Rawls was there-
fore ready to accept the possibility that citizens could appeal to their
comprehensive doctrines in the course of public debate. But he was adamant
that this permission was to be granted only on the proviso that “in due course
proper political reasons” be given as well.9
The Rawlsian proviso is supposed to guarantee that no particular
comprehensive doctrine that is shared by only one subset of the citizenry
can take precedence over the common language of public reason that is
shared in common by all citizens. And even if all citizens were to agree
upon a single comprehensive doctrine (a circumstance that given the fact of
modern pluralism may strike us as highly unlikely), it would still be the
case that public reason must in principle leave itself open to citizens who do
not share the religious (or whatever) identity of the majority. It is impor-
tant to see, however, that Rawls did not believe the proviso somehow
imposes an unfair or discriminatory barrier against citizens who happen to
profess religious views. On the contrary, the proviso allowed for the
possibility that citizens might also share their religious reasons with one
another. The only stricture was that religious reasons must eventually be
supplemented by non-religious reasons that all citizens could share regard-
less of their religious commitments. In fact Rawls even accepted the idea
that citizens may derive crucial motivation (spiritual, affective, moral, and
so forth) for democratic commitment from their comprehensive doctrines.
Reasonable citizens could acknowledge one another’s religious and non-
religious commitments precisely because they would recognize that the
democratic order draws its strength from religious and non-religious
sources of inspiration. “We may think of the reasonable comprehensive
doctrines that support society’s reasonable political conceptions as those
conceptions’ vital social basis, giving them enduring strength and vigor”
(Rawls 2005: 463). But Rawls was not naïve. He recognized that any
democratic order would have to impose limits on what kinds of compre-
hensive doctrines would be allowed. Here, too, reason would enjoy a clear
precedence:
9
“[R]easonable comprehensive doctrines, religious and nonreligious, may be introduced in public
political discussion at any time, provided that in due course proper political reasons – and not reasons
given solely by comprehensive doctrines – are presented that are sufficient to support whatever the
comprehensive doctrines introduced are said to support” (Rawls 2005: 462).
Unreasonable doctrines are a threat to democratic institutions, since it is
impossible for them to abide by a constitutional regime except as a modus
vivendi. Their existence sets a limit to the aim of fully realizing a reasonable
democratic society with its ideal of public reason and the idea of legitimate
law. This fact is not a defect or failure in the idea of public reason, but rather
it indicates that there are limits to what public reason can accomplish. It
does not diminish the great value and importance of attempting to realize
that ideal to the fullest extent possible. (Rawls 2005: 489)10
Rawls may be said to follow Kant in arguing that, when assessing the
relation between religion and reason, it is reason that must enjoy priority:
reason presides as judge over all matters that are fundamental to the just
arrangement of our life in common. Like Kant, Rawls was therefore ready
to admit that reason imposes limits upon what religion may demand. For
Kant this meant that the sometimes arcane and authoritarian elements of
historical religion would eventually be required to yield in favor of the pure
moral faith. Unlike Kant, Rawls was chiefly concerned with the role of
religion in reasonable political deliberation: “public reason” would have to
remain neutral vis-à-vis all comprehensive doctrines. In this sense Rawls
went further than Kant. Kant believed that our morality itself turns us
toward the recognition of certain religious postulates. Rawls, however,
drafted his own principles of political liberalism precisely so as to mark
their neutrality in relation to any and all religion – even the religion of
reason. The principle of rational precedence did not serve to clarify religion
but rather stood as an external check upon what religion might demand.
Habermas on religion in the public sphere

The German social theorist and philosopher Jürgen Habermas has devoted
much of his career to promoting the ideal of rational critique first articulated
by thinkers of the Enlightenment. Coming of age in the immediate after-
math of the Second World War, he remains keenly aware of the historical
obligation that Germans in the postwar era must bear for overcoming the
political and cultural legacies of fascism. A student of the Frankfurt-school
neo-Marxist philosophers Theodor W. Adorno and Max Horkheimer,
Habermas is widely considered the most prominent living representative
10
Rawls also writes: “There are limits, however, to reconciliation by public reason (e.g.,) those
[conflicts] deriving from irreconcilable comprehensive doctrines . . . Political liberalism concerns
primarily the first kind of conflict. It holds that even though our comprehensive doctrines are
irreconcilable and cannot be compromised, nevertheless citizens who affirm reasonable doctrines
may share reasons of another kind, namely, public reasons given in terms of political conceptions of
justice” ( 2005: 487).
214 peter e. gordon
of critical theory. Unlike his teachers, however, Habermas embodies a
philosophical sensibility that brings Western Marxism into a fruitful
partnership with Neo-Kantianism (though these are hardly the only two
intellectual traditions from which he has borrowed). A distinctively Kantian
inheritance was already evident in his 1962 habilitation, The Structural
Transformation of the Bourgeois Public Sphere, and Kantianism has remained
a persistent presence throughout his major works such as The Theory of
Communicative Action (1981) and Between Facts and Norms (1992). Especially
in the last two decades Habermas has turned his attention more decisively to
the question of religion, and, partly thanks to his engagement with John
Rawls, he has sought to refine his views regarding the role that religion can
play in the public sphere.
It is worth noting, however, that in his earlier work Habermas typically
thought of religion as an authoritarian medium of social solidarity
that modern society would ideally leave behind. In The Theory of
Communicative Action Habermas draws upon Weber’s sociology of
religion to suggest that, in primitive communities, religion served to bind
social agents together into an undifferentiated whole, whereas in modern
societies this phenomenon of social binding is secularized into the binding
power of a rationally achieved intersubjective consensus. This transition
demonstrates a change in the sense of a social bond: the term religio derives
from the Latin root for ligare (to bind or connect). But a religious bond
connects man to God with holy compulsion; it is not something that is
merely formed through rational deliberation. Habermas claims that this
original sense of sacred compulsion is preserved or “sublimated” into our
secular experience of intra-human obligation to the better argument. This
transformation is due in part to the gradual relaxation of the holy bond as
revelation yields to discursive analysis: once a human being can engage
with God in a conversation or even a rational argument, the sacred
character of religious compulsion begins to slacken, and human rationality
will assume an increased role in the constitution of social norms. Habermas
calls this process the “linguistification of the sacred.”
[T]he socially integrative and expressive functions that were at first fulfilled
by the ritual practice pass over to communicative action; the authority of the
holy is gradually replaced by the authority of an achieved consensus. This
means a freeing of communicative action from sacrally protected normative
contexts. The disenchantment and disempowering of the domain of the
sacred takes place by way of a linguistification of the ritually secured, basic
normative agreement; going along with this is a release of the rationality
potential in communicative action. The aura of rapture and terror that
emanates from the sacred, the spellbinding power of the holy, is sublimated
into the binding/bonding force of criticizable validity claims and at the same
time turned into an everyday occurrence. (Habermas 1985: 77)
It is important to notice that, in this earlier phase of his social-philosophical

writing, Habermas seemed content to assume that from a normative point of
view the process of secularization would reach completion: the authority of
the holy would be “replaced by” the authority of a rationally secured
consensus. In his earlier work Habermas thus appears as a convinced
secularist who goes well beyond Kant in his assumption that revealed religion
must gradually yield its authority, and that it will be transformed into not
just a religion of reason but into a non-religious social practice of rational
argument.
In his more recent work, however, Habermas appears to have drawn
back from the normative premise of thoroughgoing secularization. In his
autumn 2001 acceptance speech for the Frankfurt Booksellers’ Peace Prize,
Habermas adopted a view that suggests a diminished confidence in the
independence of profane reason. Resisting those who were calling for a
militant defense of secular society against its religiously inspired enemies,
Habermas instead declared that religion, if properly understood, might
play a vital and even irreplaceable role in modern society by providing
citizens with the normative guidance they require. This role could only be
fulfilled, however, if religious and non-religious citizens could work
cooperatively so as to agree upon a “salvaging formulation” by which to
translate from the language of religion to the religiously neutral language of
the public sphere: “Those moral feelings which only religious language has
as yet been able to give a sufficiently differentiated expression may find
universal resonance once a salvaging formulation turns up for something
almost forgotten, but implicitly missed. The mode for nondestructive
secularization is translation” (Habermas 2005: 236). In his insistence on
translation Habermas seemed to endorse the Rawlsian idea of a proviso.
The method of translation would be “non-destructive” insofar as it would
permit the normative content of religious revelation to survive for the
purpose of moral-political instruction while also protecting secular society
from the intrusion of non-universal and non-rational elements. But
Habermas also seemed to go well beyond Rawls in suggesting that secular
society might actually need the normative instruction that perhaps only
religious tradition can provide. In his 2006 essay, “Religion in the Public
Sphere,” Habermas argued that “The liberal state has an interest in
unleashing religious voices in the political public sphere, and in the
216 peter e. gordon
political participation of religious organizations as well. It must not
discourage religious persons and communities from also expressing
themselves politically as such, for it cannot know whether secular society
would not otherwise cut itself off from key resources for the creation of
meaning and identity” (Habermas 2008: 131).
This argument represents a dramatic departure from the premise of
thoroughgoing secularization or linguistification that Habermas had
presented in The Theory of Communicative Action. His new argument
seems to endorse at least the possibility that religion is uniquely equipped
to serve as the reservoir for normative insight rational modernity cannot
generate on its own. Already in the 1993 essay, “Themes in Post-
metaphysical Thinking,” Habermas claimed that “Philosophy, even in its
postmetaphysical form, will be able neither to replace nor to repress
religion as long as religious language is the bearer of a semantic content
that is inspiring and even indispensable, for this content eludes (for the
time being?) the explanatory force of philosophical language and continues
to resist translation into reasoning discourses” (Habermas 1993: 51). And in
his 2009 essay on the nature of post-secular society he reiterates this claim
as a justification for permitting religious citizens an active share in public
deliberation, even when those citizens cannot refrain from articulating their
moral-political insights in religious terms:
On the one hand, those who are neither willing or able to separate their
moral convictions and vocabulary into profane and religious strands must
be permitted to participate in political will formation even if they use
religious language. On the other, the democratic state should not over-
hastily reduce the polyphonic complexity of the range of public voices, for
it cannot be sure whether in doing so it would not cut society off from
scarce resources for generating meanings and shaping identities. Especially
regarding vulnerable domains of social life, religious traditions have the
power to provide convincing articulations of moral sensitivities and
solidaristic intuitions. (Habermas 2009: 76–77)
This new démarche in Habermas’s reflections on the role of religion in
modern society has intriguing implications for the idea of rational pre-
cedence: The new view suggests that modern society cannot leave religion
behind because in doing so it may lose an irreplaceable resource for moral-
political guidance.11 Habermas admittedly hedges his bets on this point: his
11
On this theme see P. Gordon, “Between Christian Democracy and Critical Theory: Habermas,
Böckenförde, and the Dialectics of Secularization in Postwar Germany,” Social Research, special
issue: “Political Theology?” 80/2 (2013), 173–202.
language suggests that the independent stabilization of wholly profane
modernity remains an “open question” and that the work of translation
from religious to non-religious discourse must continue at least for the time
being. But this pragmatic or agnostic perspective on the possibility of
secular society already signals an important shift in Habermas’s thinking.
Indeed it suggests that Habermas has moved considerably closer to Kant’s
original idea that reason itself contains or points to a universally valid
religion of reason. For if it is really true that modern society cannot leave
religion behind, then the relation between revelation and reason cannot be
extrinsic as Rawls (and other secular Kantians) presumed. Reason would
still need to commit itself to the task of rationally purifying the historical
sources of revealed religion so as to strip away the normative insights from
their historical shells. But reason could not be assured that it could arrive at
a station of complete secularity untied from the religious past. The bond
with religion would remain secure. The most it could hope for would be a
constant process of rationalizing translation. This would return us to the
Kantian insight that, although we live in an age of enlightenment (Zeit der
Aufklärung), we are not yet and may never be in a fully enlightened age
(aufgeklärte Zeit) (Kant 1996: 8:33–8:42).
part iii
The theory of knowledge
chapter 11
The philosophy of the Marburg School: from

the critique of scientific cognition
to the philosophy of culture
Sebastian Luft
Introduction
The Marburg School of Neo-Kantianism is, besides the Southwest School,
the most widely known and most influential school formation to emerge
within the broad movement of Neo-Kantianism in the latter third of the
nineteenth century.1 The school consisted of the twin stars Hermann
Cohen (1842–1918) and Paul Natorp (1854–1924), with the younger Ernst
Cassirer (1874–1945) as their satellite. Perhaps more than in any other
philosophical grouping, the term “school” used to designate this cluster
of philosophers merits attention. For its core members insisted that their
project consists, essentially, of a unified weltanschaulich outlook on life
whose basic commitments its followers must share. This was the manner in
which the school was organized by its founders and by which they granted
entrance to novices. As a tightly knit philosophical community, it has
become paradigmatic for other groupings in the twentieth century, such as
the Phenomenological Movement, the Vienna Circle or the Frankfurt
School, who all tried, despite their own tendencies, to emulate this “school
spirit.”
Yet, the core tenets of this school are for the most part misrepresented.
This is not surprising given the plethora of their writings and the lack of a
real “manifesto.”2 The contours of an overarching vision can only be
gleaned through an overview of the publications stemming from the
1
To deflect possible criticism from the start, when I speak of “the Marburg School,” I mean essentially
Hermann Cohen, Paul Natorp, and Ernst Cassirer, and I exclude other philosophers working at the
University of Marburg, such as Nicolai Hartmann or Hans-Georg Gadamer, who wrote his
dissertation under Natorp.
2
Natorp’s short Philosophie (Natorp 2008) can serve as an introduction to the main aims of the
Marburg School. However, published in 1911, it comes rather late in the life of the school and at a
time in Natorp’s life when his own paradigms had begun to shift.
221
222 sebastian luft
Marburg School, and not just its most popular writings, which lie in the
philosophy of science. Given the predominant reception of the writings in
this area, to introduce the philosophical principles and intentions driving
the Marburg School, one has to confront a widespread misunderstanding
head-on. In most presentations, the (Neo-)Kantianism of the Marburg
School is treated as a narrowed form of Kant’s philosophy. Allegedly,
it reduces Kant’s transcendental philosophy to a theory of natural-
mathematical science, thereby relegating philosophy to a “handmaiden
of the sciences.” Kant’s critical system becomes whittled down to an
epistemology of natural-scientific cognition. A main concern in presenting
this school must be a rejection of this erroneous reading, which is under-
standable only in light of the writings that were most widely read.
While it is true that the theory of scientific cognition the Marburgers
formulated is an important aspect of this school, it would be misleading to
see their main intentions exhausted in a theory of science. Instead, the main
intention of this school was from the outset a broadening of the critique,
both in method as well as scope. Formulaically, “the critique of reason
becomes the critique of culture,” as Cassirer announces in the first volume
of his Philosophy of Symbolic Forms in 1923 (Cassirer 1954: i: 11). But it is
mostly overlooked that when Cassirer issues this “battle cry,” he merely
repeats the sentiment that inspires the school as a whole, starting from
Cohen and Natorp. Such a critique, however, cannot do without a look at
the effort of the sciences that work on different aspects of cultural life. As
Natorp emphasizes, what is sought is “a truth which, armored with the
impenetrable steel of the most genuine, most durable science, should at the
same time be fit to satisfy not calculating reason, but to answer to the most
secret, innermost doubts and questions of the soul” (Natorp 2008: 22).3
This concept of science is the target of criticisms that claim that science is
an alienation from life and that culture is nothing but the indifferent field
of inauthentic existence.
To sketch a grand vision of this school is the task of this chapter. The
school’s scope is, in effect, so great that it goes beyond that of Kant’s
architectonics, such that the designation “Neo-Kantian” is inadequate. It is
no exaggeration to say that this school presents one of the most encom-
passing philosophical visions to arise in the tradition of classical German
philosophy. Having modified and updated Kant’s critical philosophy,
having taken in Hegel’s philosophy and the dimension of the historical,
having witnessed the dominance of scientific positivism in the latter half of
3
All quotations from the German original are translated by the author.
The philosophy of the Marburg School 223
the nineteenth century, and having observed other Neo-Kantian tenden-
cies attempting to counter the dominance of such a positivism, the
Marburg School, in the late Cassirer, self-consciously situates itself
between Kant and Hegel. It is one of the last great synthesizing attempts
in the Enlightenment tradition before the advent of postmodernism. But,
as contemporary thought has moved beyond such a radical stance, there are
reasons to believe that there will be more sympathy for this “Marburg
vision” today. It has the power to rival or complement other attempts that
consider modernism an “unfinished project” rather than a quaint dream
that is ausgeträumt.
As these preliminary reflections make clear, I treat the Marburg School
as a unified philosophical vision. This is not to say that all members agreed
on every aspect. Hence, it will be necessary to trace the steps that led, first,
from Kant to Cohen, who laid out the first systematic position that defined
the Marburg School. Yet, both Natorp and Cassirer were dissatisfied with
Cohen’s position the moment it went beyond a philosophical justification
of the exact sciences. The subtle moves on the part of Natorp away from
Cohen are a story of their own,4 though Natorp remained in strong
outward unity with Cohen. This “closing of ranks” is explicable only
through the intellectual and political landscape of the time. Cassirer,
who never lived in Marburg and was, for that reason, often not considered
part of this school, was unconcerned with such political games and
expanded Cohen’s scope, while remaining within the general framework
of the Marburg School. Thus, the full-blown shape of the Marburg School
as a philosophy of culture can be seen in Cassirer.
Hence, in the first section of this chapter, I discuss the general frame-
work of Kant’s question as to the conditions of the possibility of synthetic
a priori cognition and how it takes on the concrete task of philosophical
work with respect to the existing factum of the sciences. The latter is the
starting point for the Marburgers. To repeat, the most important aspect of
their philosophical efforts is their expansion of Kant’s critique of reason
into a critique of culture. A critique of scientific culture is but the begin-
ning. But, to understand this transformation of the Kantian project into
that of Marburg Neo-Kantianism, I discuss some key moves that occurred
in Cohen in order to bring about this novel project. In the second section, I
will trace the move from Cohen’s “critical” idealism to Cassirer’s
4
See the seminal study by H. Holzhey, Cohen und Natorp, 2 vols. (Basel and Stuttgart: Schwabe, 1986),
who details the philosophy of Cohen and Natorp both in their collaboration as well as in their
differences (including a great amount of archival material).
224 sebastian luft
“symbolic” idealism. Here, too, I confine myself to the main moves
occurring in Cassirer in order to reconstruct his philosophy of the
symbolic. In the third section, I will unfold the full-blown philosophy of
culture as it is laid out in Cassirer’s Philosophy of Symbolic Forms, including
his teleological vision for humankind. In a conclusion, I will spell out what
I believe could be the legacy of the Marburg School for contemporary
efforts in philosophy and culture writ large. This legacy has lived on, in
peculiar ways, in “Neo-Neo-Kantians,” but it has been forgotten that it
flourished in Marburg a century ago.
From transcendental to critical idealism: the factum

of the sciences
Kant’s Critique of Pure Reason is an exercise in theoretical philosophy
(epistemology) and in this function an investigation into the nature,
capacity, and limits of knowledge. Kant also dubs this investigation a
“tractate on method” since the task of critique involves concrete steps to
establish the fact that we have true cognition. “Real” rational cognition is,
as Kant is proud to have discovered, synthetic a priori, as our reason is
capable of expanding our knowledge without the aid of experience. Hence,
the task Kant sets for himself is to establish how this cognition comes
about. In the First Critique, this presentation is synthetic, in that Kant
presents step by step the contributions of the two stems of knowledge,
sensibility and understanding. The possibility of sensibility is explained,
in the Aesthetics, through the doctrine of space and time as forms of
intuition. The Analytic discusses the manner in which the understanding
applies concepts (categories of pure reason) to intuitions. The crucial step
beyond an explanation of this fact (quid facti) is the question as to how we
are justified in making a priori claims about objects of experience (quid
facti). In the Prolegomena, however, Kant opts for a different manner of
presentation, the analytic one, starting out from the factum of existing
cognition and inquiring into the conditions of the possibility of its coming-
about. Both manners of presentation supposedly reach the same desired
goal; their difference is merely heuristic.
It is the analytic path of the critique that the Marburgers preferred. The
reason for this is the acknowledgment of the factum of synthetic a priori
cognition, not as it is established in the abstract, but as it is applied in the
mathematical exact sciences. As factum, as the result of the process of being
made (facere) by scientific thought, it is to be explained as already existing.
The working out of this factum was carried out in what Cohen calls Kant’s
“transcendental method,” a term that would become the method of the
Marburg School, though this phrase is not found in Kant. One of Cohen’s
key interpretive claims regarding Kant’s philosophy was, accordingly, that
“the transcendental method was conceived in a meditation on Newton’s
Philosophiae Naturalis Principia Mathematica” (Cohen 1987: 94). For
Cohen, Kant’s achievement lay not so much in the abstract possibility of
synthetic a priori cognition; instead, this achievement was fueled by the
fascination that this type of cognition was being obtained in mathematical
natural sciences, insofar as mathematics is applied to nature, thereby
enabling and effectively creating a priori cognition. How this was possible
and how to justify this fact was, according to Cohen, Kant’s main concern.
Hence, the factum of reason in the abstract was concretely worked out in
the factum of science (das Faktum der Wissenschaft). This insight is not
Kant’s discovery, however, but a restatement of the Western tradition.
The factum of science is the key to understanding the Western project of
philosophy, to Cohen, and Kant merely took up this thread that began
with Plato, who placed philosophy on the secure foundation of science.
This entire tradition can only be adequately appreciated when seen
through the prism of the constant proximity of philosophy and science
beginning with Plato’s idealism, which is “the methodological generator of
science” (Cohen 1987: xvii). Science, however, is an ongoing process by
which thought, through the production of new ideas and hypotheses,
conquers reality. What reality is can only be obtained under this assump-
tion, that we can only comprehend what we, as Kant defines transcen-
dental idealism, “lay into things” through our reason. Hence, the
connection between science and philosophy as idealism is a necessary one:
The factum of science is the basic assumption that philosophy makes and
without which it cannot begin. Therefore, this factum is not dogmati-
cally assumed but is rather the methodological presupposition. If all
thinking, as production of ideas, unfolds and constructs itself in hypo-
theses, then it is necessary to understand as the first hypothesis: that of
science itself. (Cohen 1987: 41)
Thus, while scientific progress is ongoing, what remains constant is the
production of hypotheses, which are confirmed or falsified. In this activity,
the sciences are in effect idealistic, as they bring, in broad generality, reality
under ideas (concepts, theories). Only then can one truly speak of experi-
ence of reality. Cohen’s original reading of Kant’s Theory of Experience (the
title of his first work on Kant), in his attempt to “lay a new ground for the
Kantian doctrine of the a priori” (Cohen 1987: ix), is that the experience
226 sebastian luft
Kant means is the experience on the part of the scientist, when she
conceives the numbers and laws under which nature stands. This is an
ongoing activity of thought, not in the generational sequence of research-
ers, but in terms of the logical content of the scientific achievements
(Cohen is a fierce enemy of a psychologistic reading of scientific progress).
The factum is, as Natorp says, a fieri, something being made, a making on
the part of thinking subjects with regard to the logical progress (Natorp
2013: 39). As fieri, this progress is never-ending. Reality as fully logically
penetrated is the unattainable “thing in itself.”
Looking back at Kant’s transcendental idealism, what has thus
happened here? Transcendental idealism is the claim that we undertake
the Copernican experiment to view objects insofar as they conform to our
cognitive capacities and that we never experience the thing in itself but
always only insofar as it appears to us. In the Marburg reading, modern
natural science is already enacting this idealism, unbeknownst to itself.
The phenomena that science ascertains are the laws of nature that we
impose based on our reason. Philosophy, then, has no task of its own, it
can have no territory of its own where it can perform its sui generis work. All
it can do is reconstruct the work in the sciences, not attempt to do
something over and beyond it. It is, thus, a critique of the factum of the
sciences; the critique is a critique of reason as it is enacted in scientific
progress. As such, Cohen conceives transcendental idealism as critical
idealism with the express purpose of critiquing thought as it becomes
enacted in the sciences. This confirms the close tie between philosophy
and science that is the signature of Western thought. Kant merely brought
this insight to the clearest expression.
How does philosophy do this, concretely? What remains as the task of
philosophy (the “transcendental method”)? Cohen describes the latter
thus:
Critique, thus, means first and foremost the warning: not to identify or
place on equal footing philosophy with mathematics or natural science.
Philosophy’s task is not to create things or – as the seductive and infamous
saying goes, borrowed from mathematics – to ‘construct’ them, but instead
merely to understand and to test how the objects and laws of mathematical
experience are constituted. But the critique yields, along with this warning,
at the same time the insight and the consolation that mathematical natural
science does not merely rest on mathematics and experience, but itself
partakes in philosophy. The critique teaches to see and explore this partak-
ing, and the philosopher exploring this feels in the object of his critique the
spirit of his own spirit. (Cohen 1987: 734f.)
What happens now with one of the central demands of Kant, the establish-
ment of a priori cognition, if science is an ongoing process that never ends?
Did not Kant claim to have discovered the totality of categories and the
principles by which we apply the former to nature? Here we find what is
perhaps the most original idea developed by the Marburg School. If science
is an ongoing process, then the original categories that Kant discerns will not
suffice for the purpose of giving expression to rational cognition. But the
option of simply dropping the demand of a priori cognition cannot simply
be accepted, since this would open the door to relativism. The solution can
only be to reconceive the a priori. As Cohen states, “new problems will require
new presuppositions. The necessary idea of the progress in science has as a
necessary presupposition . . . the idea of the progress of pure cognitions” (Cohen
1977: 396). The a priori becomes, thus, dynamic,5 insofar as new insights
require new concepts that are necessary for the time being but which can be
modified or expanded (or perhaps rendered obsolete) as the progress ensues.
This is the attempt to reconcile the claim to objective knowledge with the
dynamic progress of scientific cognition, which is rational, not merely
empirical. As Cassirer states:
The “fact” of science is and will of course remain in its nature a historically
developing fact. If in Kant this insight does not yet appear explicitly, if his
categories can still appear as finished “core concepts of reason” in number
and content, then the modern development of critical and idealistic logic
[i.e., Cohen] has made this point perfectly clear. The forms of judgment
mean for it the unified and active motivations of thought, which course
through the manifold of its particular formations and are continually put to
use in the generation and formulation of new categories. (Cassirer 1994a: 18)
In Cassirer’s rendition of this concept, one can distinguish, more precisely,
two levels of a priori, a strict and a dynamic one, as he says in a letter to
Schlick:
I would call “a priori” in the strict sense merely the idea of “unity of nature,”
that is, of the lawfulness of experience as such or perhaps, more concisely,
the “distinctness of attribution” [Eindeutigkeit der Zuordnung] . . . But this
principle of distinctness itself is, to me, indeed, more than just a “conven-
tion” or an “inductive generalization”: it is, to me, an expression of “reason,”
of Logos itself. (Cassirer 2009a: 50–51, from October 23, 1920)
Beyond the strict sense of a priori, there is a dynamically evolving web of
concepts that are necessary for a given phenomenon and for the time being,
5
On this conception and its defense see M. Friedman, The Dynamics of Reason (Stanford: CSLI
Publications, 2001).
228 sebastian luft
the “very best bets” we have at a given time, with the knowledge that we
might have better or different bets at a later stage of scientific development.
This conception of the a priori retains the transcendental nature of Kant’s
philosophy (necessary conditions of possibility of cognition), while
acknowledging that science makes progress and ever-expands its bound-
aries. The idea of scientific progress is, however, purely “logical,” concern-
ing the progress of rational cognition, not a matter of empirical discovery.
But to the Marburgers, new empirical discoveries are only then real
(“experienced”) when they are understood rationally. Everything else
would be a fallback into a naïve realism that purports to be in touch with
the “things themselves.”
Now if cognition can only be attained in the sciences, what is left for
philosophy to do other than to “attest to” (beglaubigen) the latter, as Cohen
interprets the justificatory aspect of critique? At this point the critique of
philosophy as the “handmaiden of the sciences” seems to stick. Can
philosophy do anything but “follow carrying the train of lady science”
(Cassirer 2004: 358)? And if so, does this not mean that philosophy as a
discipline of its own has become obsolete? The Marburgers were, of course,
well aware of this reproach. I will return to this point in the conclusion. For
now, we can utilize the proximity between philosophy and science to set
out the broad lines of the Marburg School, returning to the project of a
philosophy of culture.
Critical idealism is most carefully worked out in Cohen’s theory of
scientific cognition. But, according to the original intention, the transcen-
dental method should be put to work in all areas of culture. Culture is
defined as “the entire work of humanity in which the latter produces and
forms ever higher [hinaufbildet] what is peculiar to humanity itself”
(Natorp 2008: 42). The areas of culture, such as what is made by human
beings – facere does not only take place in science – follow the Kantian
canonical distinction into epistemology, ethics, and aesthetics. The
philosophy of culture, then, is the logic of each cultural formation,
where the transcendental method reconstructs the logics that produce
each cultural region. Such a particular logic can, to Cohen, only recon-
struct the logical structure of that cultural region, not its empirical or
material elements. But just as in epistemology, the logical concepts under-
pinning ethics and law can only be derived from an existing science. The
paradigm of starting out, in each region, from the “factum of the sciences,”
means that in ethics one must start out from legal science, jurisprudence,
which is the point of crystallization of legal affairs. The grounding of
ethics, hence, can only occur through a reconstruction of the concepts
and theories of an existing legal theory. Ideally, there is, as an ideal natural
science, an ideal legal science as factum that is constantly under way, yet
with the status of its laws and procedures as a priori.
Where it is plausible that natural science works towards an ideal status,
such a claim might strike one as less than convincing in the realm of
jurisprudence. While Cohen might have had in mind a universal doctrine
of universal human rights, there are undeniably very different legal systems
based on different legal traditions, which are unlikely ever to converge. But
the transcendental method becomes even less convincing when Cohen
declares that art history is the scientific factum from which one has to derive
the logic in the realm of aesthetics. Such a rather artificial approach raises
the question whether such a search for a logical structure starting from a
scientific factum can do justice to the wealth and multiplicity of culture.
When the Marburg School, hence, is chastised for having a “scientistic” or
“logicistic” outlook on culture, this critique is justified, the moment one
goes beyond a theory of scientific cognition. It is at this point that Cassirer
departs from his teacher and moves into his own philosophy of culture,
centered on the concept of the symbolic.
From critical idealism to symbolic idealism: the ubiquity

of the symbolic
Cassirer accepted Cohen’s critical idealism with respect to the logic of
science, but clearly regarded this approach as inadequate when it came to
culture as a whole. This insight was itself reached, interestingly, in his early
work in the philosophy of science, Substance and Function, of 1910.
Cassirer’s main philosophical contribution after his historical work on
the development of modern science in Das Erkenntnisproblem (volumes i
and ii appeared in 1906 and 1907) was his recognition that a profound
paradigm shift had occurred in modern science, more precisely in the
process of concept formation. Scientists, in antiquity, believed that con-
cepts mirrored things in the world. What underlay this assumption, as well
as the concomitant conception of objects, was Aristotle’s substance ontol-
ogy. Accordingly, concepts were substance concepts. This Aristotelian
manner of concept formation has endured into modernity. Gradually,
however, this process was paralleled and subsequently replaced by a differ-
ent one, which conforms to the “transcendental” reading of modern
science according to the Marburgers. Modern science has already been
performing the Copernican turn insofar as the theories are not read off of
the things (themselves) but are a rational creation; they are what we lay into
230 sebastian luft
nature through reason. Accordingly, concepts scientists use to express
lawful structures under which nature stands are mental creations, not
read off of substances “out there” (recall a concept such as “atom”).
These concepts are themselves the reflection of a different ontology.
Such an ontology cannot be about substances existing independently of
us. Instead, things are what they are to us to the extent that we construct
them through our rational labor. This labor consists in the creation of
theories and, more essentially, concepts, which do not stand as substances
in and of themselves. Instead, concepts are relational, and the process of
concept formation in modern science is the creation of a string (Reihe) of
relations. Cassirer concludes his historical overview of this process:
Thus it becomes clear that all concept-formation [in the new paradigm] is
bound to a certain form of string-formation [Reihenbildung]. We call an
intuited manifold grasped and ordered conceptually when its members do
not stand alongside one another in isolation [beziehungslos], but when it comes
forth according to a creative basic relation from a certain beginning member in
necessary relation . . . according to a principle. (Cassirer 1994b: 19f.)
This standing-in-relation according to a principle of order is captured in the
mathematical concept of function, f (x). The concept in modern science is a
“rule for the connection of the individual” (Cassirer 1994b: 25), whereby the
“generality of a string-principle is the characteristic moment of the concept”
(Cassirer 1994b: 26). Gradually, in modern science, the “logic of the
mathematical functional principle” (Cassirer 1994b: 27) comes to confront
and replace the concept of substance. The functional concept can also be
called a symbol:
The basic concepts of every science, the means by which it poses its
questions and forms its solutions, no longer appear as passive representations
of a given entity, but as self-created intellectual symbols. (Cassirer 1954: i: 5)
To the extent that Cassirer explains the modern use of concepts as formed
through a functional principle, rendering them symbols, he is adding
greater detail to Cohen’s basic doctrine. But Cassirer, by the time he has
developed his philosophy of the symbolic in the 1920s, goes beyond Cohen
and his own analysis of 1910 with the claim that this symbolism is at work
not only in the realm of scientific cognition, but in all functions of
“spiritual life.” Spiritual life writ large is symbol-creating, and the symbolic
is ubiquitous:
Every genuine basic function of spirit has in common with cognition this
one trait that an original-formative, not just imitative, power is inherent in
it. The latter does not just express passively what is there, but contains a
spiritual energy of its own through which simple entities of intuition receive
a certain “meaning,” a peculiar ideal content. This goes for art as it does for
cognition, for myth as well as religion. They all live in peculiar image-worlds
in which not merely something empirically given mirrors itself, but which
they produce according to an independent principle. Each of them creates
its own symbolic forms which are, not identical to, but in their own spiritual
origin on a par [ebenbürtig] with intellectual symbols. (Cassirer 1954: i: 9)
The ubiquity of the symbolic, beyond scientific concept formation, is the
basic idea of the philosophy of symbolic forms. Cassirer’s conception of the
symbol-creating spiritual energy is the antithesis of any representationalism,
according to which (as Rorty claims) the human mind is the mirror of
nature. In Cassirer’s use of the mirror image, the opposite holds; reality is the
mirror of the human mind, which is not exhausted in producing cognition
alone. Cognition with its scientific method is one form of spiritual energy;
other spiritual energies, those at work in myth, religion, art, have their own
functional structures, generating their own symbolic forms. The task of
philosophy is to perform the “transcendental method” on them, not in
Cohenian fashion searching for each respective factum of science, but in
reconstructing the functional principle by which each spiritual energy creates
and shapes the functional nexus governing each form. This is a task philo-
sophy cannot do without leaning on empirical research in these different
areas, but this is different from choosing a factum of science of these cultural
realms for a reconstruction of their “logics.” The functional principle at work
in different forms of culture cannot be reduced to a logic; “function”
displaces Cohen’s rigid conception of logic.
The term “spiritual energy” is not to invoke some form of mysticism; it is
the result of a reconstruction starting regressively from the different spheres
of meaning that art, myth, etc. are. They are the facta into whose “conditions
of possibility” must be inquired. But these different spheres of meaning are
not simply there. They are created, they are the results of different types of
intuition, of which Kant merely discerned the abstract forms of space and
time. Intuition is not passively receiving but actively forming. There is a
plurality of experiencing the world, and each type of viewing sees something
different. Mythical space is different from the space of modern physics, as it
is different from the space of art or religion. This is not an empirical
statement about the psychological capacities of the human species, but a
transcendental statement concerning the culture-forming capacities of the
mind. Kant’s Transcendental Aesthetics becomes pluralized, but each
account of a particular intuition remains thereby transcendental, clarifying
232 sebastian luft
the constitutive principles governing each functional nexus. Cassirer illus-
trates this plurality of seeing with an example:
I grasp in [a serpentine line] the character of a certain ornament, which is
linked up for me with a certain artistic meaning and an artistic
signification . . . Once again the form of the observation can change, insofar
as that which at first presented itself to me as a pure ornament, can reveal
itself as the bearer of a mythical-religious meaning. . . . And to this form of
seizing and internal acquisition we can juxtapose, with deliberate sharpness,
another one . . . While the aesthetically contemplating and savoring indi-
vidual gives himself over to the intuition of the pure form, where to the
religiously touched person a mystical meaning is disclosed in this form; the
form that stands before one’s eyes can also serve for thought as an example
of a purely logical-conceptual structural nexus . . . Where the aesthetical
direction of viewing perhaps saw Hogarth’s Line of Beauty, the mathema-
tician’s gaze sees the image of a certain trigonometric function, e.g., the
image of a sine curve, while the mathematical physicist sees in the same
curve the law of a periodic wave. (Cassirer 2009b: 97f.)
The concrete task of the philosophy of the symbolic is to describe the
“symbolic logic” in each case of symbolic formation. An individual thing is
construed as a symbol; it makes sense only in a context of other things;
hence the symbol is the “throwing together” (sym-ballein) of the individual
and the general. Its meaning is contextual and has a different meaning in
different symbolic gazes. This amounts to the universalization of Kant’s
transcendental idealism: the phenomenon becomes the symbol, and to
account for it means to reconstruct the “logic,” the functional principle in
each symbolic form. Myth, art, religion, science all have their own struc-
tural principles and manners of functioning.6 Thus we arrive, from critical
idealism, at symbolic idealism.
The sum total of these symbolic forms is culture; it is the totality of the
deeds on the part of humans, “for the content of the concept of culture
cannot be isolated from the basic forms and directions of spiritual produ-
cing: ‘being’ is always only to be grasped in ‘doing’” (Cassirer 1954: i: 11).7
Culture itself, then, is itself a functional concept that can be defined only
through the deeds that bring it about, and not through a formal definition,
which would render it a substantial “thing.” This is why the philosophy of
the symbolic can never be a finished “system.” Rather, the systematic
6
What they have in common is what one can call the triad of the symbolic: the function of impression,
expression, and presentation, but these functions work differently in the different forms; see Cassirer
2009b: 70–73.
7
One of Cassirer’s favorite authors is Goethe. Recall that Faust translates “logos” as “Tat” (deed); see
Faust i, v. 1237.
character of this philosophy must be demonstrated in the method of
reconstructing the functional nexuses and in distinguishing the basic
nexuses from one another. The concrete method of this philosophy is,
thus, phenomenological.8
Cassirer was not able to complete his “system,” since his fleeing from the
Nazis forced him to change plans for publishing the Philosophy of Symbolic
Forms. After he had moved from Germany to England, from there to
Sweden and finally to the United States, his relocations and encounters
with different intellectual milieus caused his interests to shift. His
philosophical plans were far from completed when he died in 1945. This
can be seen from the many additions and novel ideas that Cassirer produces
after 1933. There is no denying that Cassirer left many critics wanting and
many questions unanswered. Cassirer never doubted the truth of the
project of a critique of culture as an analysis of the symbolic in its manifold
expressions. By way of discussing some critical questions, I turn to the
metaphilosophical significance of a critique of culture.
The complementaristic plurality of culture and humanity’s

self-liberation in culture
The purpose of the philosophy of symbolic forms as a critique of culture is
to give a rich account of the plural expressions of culture while keeping
such an account within the boundaries of transcendental philosophy. The
philosophy of symbolic forms is no empirical science, though it can and
must rely on material provided by scientific disciplines. One will never find
Cassirer discussing a problem without reference to scholarship; e.g., lin-
guistics in the case of language, and anthropology in the case of mythical
consciousness.9 However, the claim is a philosophical one, to provide the
basic functional structure of each symbolic form, its particular logic, which
is the respective condition of the possibility of viewing, thereby creating,
the world, “its” world. While there can be no system that would claim to be
complete – culture is an ever-evolving process – the question remains, what
is the overall aim of this philosophy? If it was said that the concrete method
is phenomenological (descriptive), one can point out that no description is
8
Cassirer oftentimes confirms his alliance with phenomenology; see Cassirer 2009b: 7, 98–99. For a
comparison between Cassirer and Husserl see S. Luft Subjectivity and Lifeworld in Transcendental
Phenomenology (Evanston: Northwestern University Press, 2011), 235–266.
9
See Cassirer 1979: 80. “We have no other way to find [the rules governing each form] than to ask the
special sciences, and we have to accept the data with which we are provided by them.”
234 sebastian luft
naïve and without presuppositions. What is, thus, the guiding clue under-
pinning Cassirer’s analyses?
The plurality and richness of culture cannot be subjected to the bench-
mark of science. Conversely, the philosophy of culture makes it clear that a
merely empirical account of culture will not suffice, because such an
account is prone to a cultural relativism. But what is the status of the
symbolic forms themselves in the framework of a transcendental account?
It is curious that Cassirer, when it comes to the number of symbolic forms,
rarely gives a “deduction” or a justification of why he mentions the ones he
does and not others. Moreover, his enumerations vary and sometimes
include other forms besides language, myth, religion, art, and cognition.
And it is even more curious to note that he devotes systematic studies only
to language, myth, and cognition, and in this order. What is the rationale
for his procedure? Cassirer is not consistent and fully clear here, and there is
no denying these systematic gaps.
Part of this lack of full exposition can be chalked up to Cassirer’s
inability to finish his philosophy due to his biography. The fact that he is
not consistent in enumerating the symbolic forms – in later years, he
mentions technology and economy – can be countered by reminding us
of the functional nature of culture, that culture cannot be defined from
above and that it is possible that new forms arise in the process of culture.
Indeed, such a discussion about the arrival of new forms of culture can
be an interesting exercise in cultural philosophy (is, for instance, the
Internet a new symbolic form?). As he says, “human civilization neces-
sarily creates new forms, new symbols, new material things in which the
life of man finds its external expression” (Cassirer 1979: 139). But there is
one important point where Cassirer’s wavering gives rise to a more serious
concern.
On the one hand, Cassirer insists that he wants to give an account of the
richness of “spiritual activity” where each spiritual power has its sui generis
“logic” that cannot be compared to others, and it would be a metabasis if
one symbolic form were measured by the standard of another (as Cohen
did when measuring all other forms with the standard of science). Thus,
the symbolic forms are irreducible to one another. The philosophy of
symbolic forms is an account of these symbolic forms without such an
overreaching from one form to another. They are, in this scenario, ordered
horizontally, displaying no hierarchy. Rather, they complement one
another, together yielding a richer sense of culture (Cassirer quotes
Hegel, das Ganze ist das Wahre). It is the task of the philosophy of the
symbolic to “spell out of phenomena different symbols and, so to speak,
different alphabets of thought that do not contradict each other but
complete one another” (Cassirer 1979: 76).
On the other hand, statements as to the relation of the different forms to
one another in terms of a hierarchy are not absent. For one, Cassirer is
unfailingly clear on one point, the systematic locus of myth. Myth is the
first form of spiritual and cultural development, which is primitive (though
not irrational) and is overcome through higher forms of cultural expression.
Religion, language, and science “conquer” myth and relegate it to a form of
unenlightened pre-cultural existence. It is a stage of human development to
which one may not return in a developed culture. The fact that myth is
made to re-enter the arena of culture is his critique of modern fascism
(cf. Cassirer 1946). This marks him as standing in the tradition of the
Enlightenment. Conversely, Cassirer speaks of science as the highest
expression of human spirit and as the purpose and end-goal of culture.
While all symbolic forms “possess their own distinctive type of ‘universal
validity’ . . . the clearest and best example of such ‘universal validity’
continues, in good Marburg style, to be given by the language of mathe-
matical exact science.”10 These passages make him vulnerable to the charge
that despite the emphasis on all cultural forms being equal, they are, at the
end of the day, subordinate to scientific theory and its logical ideal of
universal validity in a lingua universalis, and to the critique that he remains
aloofly disrespectful of our facticity and our finitude. In this way, one may
summarize Heidegger’s position against Cassirer during the Davos
standoff.
There is, as mentioned, no denying that one finds conflicting passages in
this respect. A solution can only be to spell out what Cassirer should have
unambiguously said in light of this contradiction. Should scientific conduct
be interpreted as the highest form of human culture? Surely a scientific
positivism or unalloyed belief in scientific progress is not Cassirer’s senti-
ment. Despite his admiration for modern science, and his intimate know-
ledge thereof, Cassirer is nonetheless more aptly placed on the
“humanistic” side of the “two cultures.” Or perhaps better, Cassirer’s
position is best described as “synthetic” in wanting to do justice to both
traditions, in that the humanistic aspect of culture should not be left
unaccountable to reason and rational critique, and in that natural-scientific
reason should not operate in the way of a cold-hearted technology.
Thus, while the philosophy of culture was described as proceeding by a
10
M. Friedman, A Parting of the Ways: Carnap, Cassirer, and Heidegger (Chicago: Open Court,
2000), 152.
236 sebastian luft
descriptive method, the normative aspect of critique must not be over-
looked. To subject culture to critique, then, must not be understood in the
way of a cold rational look at culture, and hence the exclusive search for
logical structures, but in the way in which Natorp describes rationality, as
answering to the deepest questions of mankind.
The “critique of culture” may thus not be seen as the call to rationally
“scan” the different cultural forms as to their “deep” rational structures.
Rather, what must be emphasized is that culture is the correlate of the
common humanity that unites us; in all cultural forms we find “ultimately
the ‘same’ human being that we always continually encounter in the
development of culture, in thousands of manifestations and in thousands
of masks.”11 This insight amounts to acknowledging the emancipatory
power of culture as the self-liberation of the human being in and through
culture. The task of the critique of culture is to make understandable that
culture, in all of its forms (starting centrifugally from myth), is the gradual
acquisition of freedom. This critique “rejects the conception that mind
submits to an outward fate. Mind must realize and actualize its own
freedom in order to possess it, and the whole work of culture is this very
process of self-realization” (Cassirer 1979: 89). Yet, mind is to be found, not
in an absolute state, but in the refractions of the various forms of culture,
which are forms of our own making and spaces where alone we can be
human beings. Thus, the description of the symbolic forms is at the same
time a prescription of the common core of humanity to which we all ought
to belong. The animal rationale is, thus, defined in a more encompassing
sense as animal symbolicum and the ethical ideal of this vision is that
freedom can only be obtained in culture; that culture is not the inhibitor
of freedom, but its condition of possibility. Cassirer characterizes the
“promise and hope” (1979: 90) of the distinctly Marburg-infused philoso-
phy of culture thus:
It hopes to come to a sort of grammar and syntax of the human mind, to a
survey of its various forms and functions, and to an insight into those
general rules by which they are governed. By this we may be able to under-
stand in a better way the koinon kosmon of humanity, that common world in
which each individual consciousness participates and which it has to recon-
struct in its own way and by its own efforts. (Cassirer 1979: 89)
The philosophy of culture, thus, has no task over and above this realiza-
tion, it is not a symbolic form of its own, but it is the logic of the symbolic
11
Quoted in Friedman, A Parting of the Ways, 154.
in its richness of expression. It makes us understand that there is not and
ought not to be any life outside of culture, and that culture is both our
space and ours to create responsibly.
Conclusion. The legacy of Marburg: the collaboration

of philosophy and science; the emancipatory function of culture
Regarding the legacy of the Marburg School, let me mention one aspect in
conclusion, where I believe its efforts bear resemblance to some of today’s
philosophical concerns.
Let me return to the critique of philosophy as the “handmaiden of
the sciences.” The critique consists in the claim that, with this all-
encompassing ideal of culture and the insistence that philosophy cannot
do without the individual sciences dealing with the different forms of
cultural expression, philosophy is indeed relegated to nothing other than
the former’s handmaiden. I think it is fair to say that many contemporary
philosophers would bristle at this notion of philosophy. In order to
correctly assess what philosophy is to accomplish, one can compare the
Marburg School with another form of Neo-Kantianism, namely Robert
Brandom’s inferentialism. I cannot delve deeper into one of the most
impressive philosophies of our day, nor do I claim that Brandom is the
only contemporary philosopher who could serve as example here. I only
wish to point out one contemporary parallel to the Marburg position.
In philosophy’s task of “making explicit” the discursive commitments
we undertake when we engage in the “social practices of giving and asking
for reasons” (Brandom 2001: 92), Brandom rejects the role of philosophy as
being the “queen of the sciences” (“philosophy is at most a queen of the
sciences, not the queen” (93)). Indeed, philosophy plays no “foundational
role with respect to other disciplines” (ibid.) and in this sense, Brandom
expressly embraces the image of the “handmaiden” (ibid.), and he describes
the task of philosophy as follows:
For what we do that has been misunderstood as having foundational or
methodological significance is to provide and apply tools for unpacking the
substantive commitments that are implicit in the concepts deployed
throughout the culture, including the specialized disciplines of the high
culture. Making those norms and inferences explicit in the form of claims
exposes them for the first time to reasoned assessment, challenge, and
defense, and so to the sort of rational emendation that is the primary process
of conceptual evolution. But once the implicit presuppositions and
consequences have been brought out into the daylight of explicitness, the
238 sebastian luft
process of assessment, emendation, and evolution is the business of those
whose concepts they are, and not something philosophers have any author-
ity over or expertise regarding. (Brandom 2001: 92–93)
This definition of the task of philosophy has interesting similarities with
that of the Marburg School. The Marburgers would agree that philosophy
cannot, and should not, provide an ultimate foundation, it cannot be
“first” philosophy, since it cannot make itself independent of the sciences
which discover the concepts and functions of the regions they are sciences
of. As Cohen says, philosophy is no Grundlagenwissenschaft, a science of
foundations. But as handmaiden, it also develops the tools for “rational
emendation,” which precisely is critique, but necessarily critique of some-
thing that functions implicitly through norms, and this is culture. Directly
addressing the issue of whether philosophy is the torch- or train-bearer of
the sciences, Cassirer maintains that there can be a third alternative.
Philosophy’s task cannot consist in
mediating the inner battles that always again arise in science and to silence
them through hasty solutions. Rather, it stands in the midst of these battles,
it cannot and wishes not to be anything but the fellow combatant
[Mitstreiterin] in these. Instead of overcoming the oppositions through the
command of thought or attempting to reconcile them through a mere
compromise, it must rather make these visible in their full seriousness and
gravity. (Cassirer 2004: 358)
As combatant in these battles, philosophy equally has the task of making
explicit the functional principles guiding the symbolic forms, which is not
something individual sciences can do on their own, but which is not some-
thing entirely above and beyond them either. Philosophy merely has the task
of making explicit what goes on implicitly in the cultural activities. What is
made explicit in the essentially self-reflective intellectual activity philosophy
is the fact that culture is the expression of the common humanity that we
share. Surely, the philosophy of symbolic forms “makes things explicit” in its
own manner, as explained in this essay, but in this division-of-labor concept
of philosophy, “Marburg” bears striking similarity to “Pittsburgh.” But
understood in the right way, this is not a reproach that should put philoso-
phers working in this mode on the defensive. Instead, it is the healthy
balance any philosophy must strike when it acknowledges that it cannot
operate in a vacuum but in a culture rich with creative work, part of which is
carried out in the sciences, while not ceding its position to naturalism.
The moral demand is that this process of culture becomes ever-
expanded, keeping barbarism at bay, while knowing that the human
being is a “crooked piece of timber” that can never be made straight.
Emancipation from “self-incurred tutelage” can only come through
partaking in culture, which is not anybody’s private achievement, but the
product of “spirit.” Thus, the Marburg School situates itself, in Cassirer,
consciously between Kant and Hegel.
chapter 12
Natorp’s psychology
Daniel O. Dahlstrom
Introduction
In 1912 Paul Natorp published the first book of his General Psychology
according to the Critical Method. The book’s seemingly innocuous title,
“the object and method of psychology,” betrays the central challenge of
the work, since psychology’s object and method are like no other. What
Natorp means by “psychology’s object” is its subject matter, since the latter
is, he contends, generically distinct from any object or thing, properly
so-called, including those that form the subject matters of other sciences.
So, too, he argues that the method of psychology must be sui generis among
sciences, since sciences are generally in the business of objectifying, while
psychology’s method, given its subject matter, cannot take the form of
objectification in any sense.1 Even from this brief précis, it is evident that
Natorp’s concern with securing the proper way of investigating human
subjectivity has affinities with Husserl’s efforts at elaborating phenomenol-
ogy’s subject matter and method. These affinities were well recognized by
both thinkers, perhaps explaining why they both appreciated their differ-
ences so acutely.2 Indeed, in his General Psychology Natorp works out the
finer points of his psychology through criticisms of Husserl’s phenomenol-
ogy circa 1912 (a year before the publication of Ideas I ). Not surprisingly,
these criticisms elicit responses not only from Husserl but also from
Heidegger in his earliest Freiburg lectures.
1
Natorp 1912a. Hereafter Natorp’s work is referred to as “AP ” followed by the page numbers of the
1912 edition. When a word or phrase is quoted without an immediate reference to the corresponding
page numbers, the reference for it is to be found in the next parenthetical reference or footnote within
the same paragraph.
2
See S. Luft, “Natorp, Husserl und das Problem der Kontinuität von Leben, Wissenschaft und
Philosophie,” Phänomenologische Forschungen 6 (2006), 104f.; “Reconstruction and Reduction:
Natorp and Husserl on Method and the Question of Subjectivity,” in R. Makkreel and S. Luft
(eds.), Neo-Kantianism in Contemporary Philosophy (Bloomington and Indianapolis: Indiana
University Press, 2010), 59–91.
240
Natorp’s psychology 241
The aim of this chapter is to consider the influence of Natorp’s
psychology on Husserl’s and Heidegger’s developing phenomenologies
prior to 1920, i.e., the time when the record of their critical engagement
with his psychology is well established. To this end, the first section of the
chapter reviews Natorp’s account of psychology’s object and method, in
the interest of demonstrating the account’s intrinsic merits and setting
the stage for examining his criticisms of Husserl’s phenomenology. The
second section addresses two of these criticisms. In conclusion, I suggest
the likely impact of Natorp’s criticisms on Heidegger’s own early efforts
to develop a phenomenological method.
Psychology’s subject matter and method

Natorp’s General Psychology is intended, not as a work in psychology
simply, but as a “philosophy of psychology,” designed “to procure clarity
regarding its logical foundations,” above all, the question of its subject
matter and method. The work’s orientation coincides, at least in
general, with his 1888 Introduction to Psychology (Einleitung in die
Psychologie)3 and its opposition to the “dominant, objectivistically
oriented psychology.” This opposition is shared by prominent contem-
poraries, including Husserl, who, in fact, commented critically on parts
of this early work in his Fifth Logical Investigation. When Natorp pub-
lished his General Psychology, he planned a two-volume work, comprising
a “general phenomenology” of consciousness in its systematic structure
(the so-called “ontic” dimension) and a second volume on “the logical
basic framework for the sequence of levels of unities of experience” (the
so-called “genetic” dimension). Natorp never delivers on the planned,
second volume. This point deserves noting, not least since in the General
Psychology he criticizes Husserl precisely for failing to attend to the
genetic, dynamic dimension of consciousness. Whatever the merits of
the criticism (which Husserl in fact takes to heart), Natorp does not
himself manage to fulfill the demand for a genetic account of subjectivity
(AP i–iv). At the same time it is noteworthy that Natorp describes the first
part of his psychology as a phenomenology, an association that Husserl
flatly eschews a year later in Ideas I.
3
Natorp 1888. At the outset of his General Psychology, Natorp notes his basic agreement with the
content of this early work, albeit with the recognition that it needs to be presented in a different way
(Natorp 1912a: v).
242 daniel o. dahlstrom
The subject matter: the content of consciousness
Natorp identifies subjectivity or consciousness as the exclusive subject
matter of psychology, in contrast to all other sciences. Whereas everything
that is an object falls under other sciences, psychology alone examines
subjectivity as what objects are for and as something with its own standing.
More precisely, consciousness includes a threefold structure:
(1) a content (that of which one is conscious, purely in the sense of being
correlative to consciousness),
(2) the ego, the subject (who is conscious), and
(3) the relation between the content and the subject of consciousness.
While this relation is “neither capable of nor in need of any explanation
or reduction more,” a necessary reciprocity as well as a certain asymmetry
obtain between the relata, i.e., (1) and (2). That is simply to say that, while
there is no ego without content and vice versa, the ego is conscious of the
content but not vice versa. So, too, Natorp contends, the content is given
to us, but consciousness as such is not. The ego is nothing other than the
point of reference for some object of consciousness, precisely in the relation
that “consciousness” signifies for both the ego and the object. We can
present neither being-an-object for consciousness nor the conscious ego to
ourselves without making them into objects – in which case, he adds, we
have already ceased to think of them as such (Natorp 1912a: 25–31).
Natorp acknowledges how paradoxical these claims are, namely, that
the ego and the relation to it can in no way be made an object. Indeed, he
is quick to qualify the claims by adding, a bit more circumspectly, that
none of the assertions of existence, factuality, and so on can be said of it
“in the same sense as these assertions are otherwise understood” (Natorp
1912a: 29). They cannot because subjectivity (the ego and the relation to
an object that makes up consciousness) is always their presupposition.
Just as appearing cannot be called an appearance in the same sense as
what appears, i.e., just as appearing does not itself appear, so subjectivity,
the original consciousness of an object, cannot itself be an object. To
construe it as an object is to invert its nature, to convert it into its
opposite. In other words, consciousness cannot be conscious of itself in
the way that it is conscious of objects, since that would entail an identity
of it as subject and as object. However, there is no such identity, Natorp
contends. “An act like presentation, knowledge, consciousness in general
cannot possibly be thought in such a way that subject and object of this
act in every respect, content-wise and numerically, would be the same”
(Natorp 1912a: 30).
If Natorp were to stand unqualifiedly by his claim that there is no
consciousness of subjectivity, he would, of course, be shutting the door
on any psychology. His considered position is rather to accept the need
to speak of and reflect on one’s subjectivity as an “object” or “theme” –
albeit at times deliberately employing, not Objekt, but Gegenstand and
Vorwurf respectively (Natorp 1912a: 30). Yet, while there can be no
question of forbidding objectification, what he wants to insist on is the
“derivative” or “representative” character of what one then entertains. It
speaks volumes that he employs the analogy of a person’s reflection in a
mirror to convey this derivativeness and representativeness. On the
one hand, subjectivity is, “as it were, one’s mirror image,” that is to say,
“the object of the act that we call ‘self-consciousness’ is necessarily no
longer the original I, but a derivative one.” On the other hand, the
metaphor contains a mirror in which I can catch sight of myself, even if
“only in an inauthentic way.” “What is then the mirror, in which the ego
is reflected? Apparently, the content, the something of which one is
conscious” (Natorp 1912a: 32).
In this way Natorp identifies the content of consciousness as the
actual subject matter of psychology, a content that entails but is not to
be confused with the act or subject of consciousness or its relation to the
content. Presumably, we can find our likeness as subjects mirrored in
that content, though this likeness is never to be confused with our
authentic subjectivity. In other words, through the content we can find
what we are like, but an identity as such eludes us. While we can
recognize mirror images of ourselves, we do not identify ourselves
(or even our appearances to ourselves, to the extent that they are
available to us) with those images. Still, those images are typically the
only access that we have, for example, to our faces. “All the ways by
which one tries in any case to describe the ego are obviously borrowed
from the content” (Natorp 1912a: 31).
From these considerations, i.e., from the fact that the original, pure ego
is the ground of all phenomena but not itself a phenomenon, Natorp infers
that “pure egohood, likewise the pure relation to the ego,” is not what
psychology investigates. Once again, only the first element of the threefold
structure of consciousness – the content of consciousness – forms the
proper subject of psychology, while the other two elements are not given,
but simply presupposed as its ground. More specifically, the ego is pre-
supposed, “not [merely] as a unity [of the contents], not as the unity of
connection in each case, but as the unity, as the ultimate ground of being of
the unification, and thereby as the possibility of unity, in a word, as the
ground, in general, of unity.”4 Nor is it at all objectionable that this ground,
this ultimate unity of consciousness, does not belong to the phenomen-
ological data of consciousness (what is phenomenologically given in
consciousness), since, if it were necessary to recognize the experiencing as
a fact, i.e., to experience it as a fact, the same would hold of that experience,
and so on, leading to an infinite regress. Such a regress is out of the question
here, since the experiencing is actual, i.e., does come to pass.
Of course, Natorp has the pre-Ideas Husserl in his sights and this same
criticism may lose its force when it comes to Husserl’s later conception
of the fundamental, unifying level of consciousness, i.e., inner time-
consciousness. In the latter, one might argue, the requisite structure for
initiating the regress, the apprehension/object structure, is not at hand. But
this development, too, in Husserl’s thinking may well be another indica-
tion of the reach of Natorp’s criticism.
The subject matter of psychology for Natorp is, as noted, the content
and not its relation to the ego. Only the former counts as given, not the
latter; just as appearing does not itself appear, so the presupposed ground
of givenness, the full, threefold complement of “being-conscious,” is not
given. Hence, the focus of psychology is not that ultimate unity of
“being-conscious,” but the “relative unity” of consciousness, i.e., the
particular way that the content of consciousness displays itself in unity,
encompassing a form and materials formed (i.e., a unity and a manifold).
As for those who object to the idea that psychology is restricted to the
content, either they fail to see that the content is not simply materials but
also the unity of these materials or they are thinking of the ultimate, but
inscrutable unity (Natorp 1912a: 41).
This conception of psychology’s subject matter runs counter to the
frequent deployment of the distinction between the act and the content
of consciousness. The distinction is misguided, Natorp maintains, even if
it is legitimately motivated by an attempt to maintain rigorously the
subjectivity of consciousness in the face of naturalistic accounts. The
emphasis on acts of consciousness appears to be rooted in the suspicion
that any account of content collapses or at least runs the risk of collapsing
into some objectifying, physical account (as when, for example, sounds are
tied to the effects of vibrational disturbances in the air on auditory nerve
pathways). Thus, for example, on this view psychology is concerned with
the act of hearing, not the sound heard, since the latter is a matter for
acoustical physiology. Yet this focus on alleged acts of consciousness is
4
Natorp 1912a: 36.
completely misplaced, Natorp argues, since we do not have a sensation of
our sensing, but only of its contents.5
It is also misplaced, given the fact that the contents have their own
standing as part of subjectivity, quite apart from any neurological objects.
At the same time, while it is possible to consider, in abstraction from one
another, the content and the I for whom it is content, neither is actually
separable from the other. Nonetheless, the content enjoys a certain pri-
macy, phenomenologically, since it alone is given. Though the content is
always content for a consciousness (for an ego), being-content “for” con-
sciousness (or “in” consciousness) is not itself given thereby – or so Natorp
contends. In other words, we have no direct consciousness, he maintains,
of the so-called activity of consciousness. When we experience a sensation,
we sense, not our sensing, but instead the content of the sensation, the
color red, the sound of G, and so forth. “I do not hear my hearing; I hear
only the sound; but it is precisely I who hear it, i.e., it is there for me, it is an
integral part of my . . . consciousness, and thereby lines up into the
concrete unity of this my experiencing; that alone is the ‘subjective’ side
of the fact in question” (Natorp 1912a: 46). Natorp does not dispute that
there are diverse ways that the components of the content fit into the unity
of experience, but contends that they are only demonstrable in (terms of)
the content (Natorp 1912a: 47).
It is noteworthy that, in the course of criticizing the use of the act/
content distinction, Natorp remarks that Husserl formerly employed the
distinction but no longer does so. Moreover, much of what Natorp says on
these pages regarding content is mirrored in Husserl’s account of the
noema in Ideas I. Nor would it be easy to gainsay the fact that the noema
enjoys a certain priority in Husserl’s exposition. Nevertheless, whether
these similarities with Natorp’s psychology are serendipitous or not,
Husserl parts ways with Natorp, at least expressis verbis and no doubt
quite deliberately, by insisting on a noetic phenomenology in addition to
a noematic phenomenology.
Perhaps the central distinctiveness of Natorp’s account of psychology’s
subject matter, i.e., the contents of consciousness, lies in its supposedly
fluid relation to what is considered objective. He observes how, on a naïve
level of knowledge, one might believe that one can grasp the color red as
5
The sensing is also not part of what we are conscious of, i.e., the Bewußtheit, though, of course, what
we sense (see, hear, etc.) is only in consciousness (its relation to an ego) and for an ego. Natorp
accordingly rejects the contention that the acts of consciousness are simultaneously reflexive. Prima
facie, his view is thus at odds with the Aristotelian conception, revived by Brentano; see Aristotle, De
anima, iii, ch. 2 (425b17f.); Brentano 1874: 179f.
an object and not as something that is only for someone perceiving it. He
does not explain the observation, but the fact that we use a common term
for the color, one that allows us to identify and communicate it, might
lead us to believe that it is objective. However, it is evident, he submits,
that on a supposedly higher level it is something merely subjective,
present only for my perception, and, indeed, that it cannot be tenably
conceived as an object in a strict sense. Again, Natorp is short on
explanation here, but such a claim may be based upon the unique
conditions of each sighting of the color, rendering it available only to
the respective perceiver and thus subjective. But a further objective
determination, he submits, is possible here as well, namely, the sort of
scientific determination that might be given in terms of a velocity of
oscillations of light.
This account leaves much to be desired, not least given the different
meanings of “objective” and “subjective” at the different levels. Natorp
neither sorts out these differences nor explains their common features nor
why we should assume that they bear on the same color. He seems satisfied,
however, if he succeeds in illustrating the main thesis that every objective
determination that we arrive at “can and must” be viewed in turn as
“subjective,” and how every level of consciousness (subjectification)
displays an objectification, i.e., an identification of some sort. The process
of knowing can proceed endlessly in these two diametrically opposed, yet
corresponding, directions, the paths of subjectification and objectification
(Natorp 1912a: 67f.).
The matter is complex; not only are the objective and the subjective
dimensions correlative, but they also are only relatively objective or
subjective. That is to say, what is objective on one level may be subjective
on another level and vice versa. In other words, there is no fixed objective
dimension and no fixed subjective dimension, i.e., no object in itself and
no subject in itself, but always only something relatively objective or
relatively subjective. While we legitimately take objects to obtain indepen-
dently of us, we also suppose them at the same time to have this or that
property, without yet knowing that this is the case. This supposition is
subjective but once the hunch is confirmed (in Husserl’s sense, once the
empty intention is fulfilled), i.e., once the supposed object and property are
given, this subjective content now becomes part of the object. Even if the
object with this property is not given, we anticipate that it should be given.
That is to say, the connection of the property with the object is imple-
mented in thought, “even if not at once with the claim of being valid
knowledge” (Natorp 1912a: 66). In this way, Natorp concludes, the
consciousness in all knowledge directed at objects constantly expands,
tackling with each step what was not tackled in a previous step.
The difference thereby is always that of the direction of the considera-
tion: objective insofar as it is a consciousness of a law, subjective insofar as
it is a living, concrete fulfillment. Natorp has in mind something akin to an
infinite continuum where, at any point on the continuum, one moves in
one direction to the objective side and, in the opposite direction, to the
subjective side. Starting from either side, objective or subjective, there is an
endless, albeit nonetheless progressive series of steps to be taken to come to
fuller understanding of the experience.
Viewed in this light, subjectivity is always a task of its own, since
knowledge first pursues only its original direction, toward the object,
thereby taking leave of the lower level. Yet the possibility always remains
of returning to, i.e., inquiring into, the subjective, precisely as that on the
basis of which the objective was recognized.6 As discussed in the next
section, Natorp draws various methodological consequences from this
fluid, bi-directional character. His emphasis on this character forms the
backdrop of one of his objections to Husserl’s phenomenology (the
objection reviewed below).
The method: reconstructing subjectivity

For Natorp, the subject matter of psychology is the content of conscious-
ness or, equivalently, the immediacy of experience, what is immediately
given in experience. From this account of psychology’s subject matter, he
infers that its method must be “fundamentally different” from that of all
knowledge that proceeds by way of objectification.
If it is the task of psychology to know [zur Erkenntnis bringen] the
subjective [dimension] of consciousness, apart from all objectification,
to know the ultimate immediacy of it . . . clear of all mediations, [i.e.,] the
full, concrete content of what is experienced beyond any mere abstraction,
then what is not suited for such knowledge is precisely the way of
proceeding which, as far as its aim and possible accomplishment is
concerned, is aimed much more at securing concepts and mediating a
knowledge of an object. (Natorp 1912a: 91)
After drawing this conclusion in the opening paragraph, he immediately
acknowledges that psychology historically mimics object-oriented sciences,
6
See the striking image of outward conquest and taking full possession (from within or in the interior)
of the already conquered land (Natorp 1912a: 70).
a mimicking that is perfectly understandable, given the successes of those
sciences. Yet nothing is “more certain” (nichts gewisser), he contends, than
that this way of proceeding, far from getting closer to the immediacy of
experience, increasingly distances itself from that immediacy. While all
objectifying knowledge draws on the content of what is experienced, it
necessarily breaks (zerteilt) that content down into parts, putting aspects
and individual relationships “in the place of the original, undivided whole
of experience” (Natorp 1912a: 92).
Concrete objective sciences in Natorp’s sense encompass both natural
and cultural sciences, including knowledge of what is and knowledge of
what should be, and they are all grounded in the various pure objective
sciences of law or method, namely, logic, ethics, aesthetics, and philosophy
of religion. Just as the concrete sciences search for and construct the laws of
the phenomena in their domain, so the pure sciences construct the laws
governing that procedure; that is to say, those pure sciences treat “the
objective grounds of the prevailing concepts and laws on which the entire
procedure of the concrete objective sciences rests” (Natorp 1912a: 94). In all
these objective sciences the method is accordingly constructive. Yet just as
none of these disciplines is grounded in psychology, so psychology is not
grounded in them.
To be sure, Natorp concedes, many of his contemporaries recognize that
investigation of the immediacy of experience is psychology’s task, but the
recognition is more a matter of instinct than that of “a consciousness with a
secure scientific method” (Natorp 1912a: 96). The lack of the proper
method is evident from the fact that, despite this insight, psychologists
still adopt objectifying, natural science as their model (Vorbild) and
accordingly “strive to reduce the given experiences to universal laws and
causal laws” (Natorp 1912a: 97). Even the best psychologists of the day,
among whom he counts Husserl and Lipps, do not extract the subject as
such, but only “objectifications of various levels” (Natorp 1912a: 97f.).
Natorp identifies the character of assertions as a culprit here. Without
denying their necessity, he adds that they willy-nilly prepare the way to
abstraction and universalizing, “more or less distancing themselves from
the immediately experienced that is always determined concretely and
individually” (Natorp 1912a: 99). Assertions objectify by expressing
immediate experience through some lower level objectification, where,
for example, the color-predicate in the assertion “I see red” designates
something generically and recognizably seen. “Red” stands, not for the
subjective experience of it, but for what is seen by anyone in a situation
comparable to my own. Traditional psychology mistakenly takes what is so
signified for something subjective and given, when in fact it is “an objecti-
fication of merely a lower order” (Natorp 1912a: 100). With remarkable
naivete, researchers take such objectifications (e.g., types of sense data) for
what is subjective, thereby taking what is given for what first needs to be
sought. “The purely subjective is, precisely as such, never given but what
first needs to be sought” (Natorp 1912a: 101).
Natorp stresses the direness of the situation (Notlage) psychology finds
itself in. If psychology is supposed to grasp immediate experience in itself
apart from any assertion, any judgment, any belief (since all the latter are
objectifying), how can it do so, he asks, without isolating it from what is
experienced, severing it from what is subjective, and thereby making it
willy-nilly into an object?
Thus, in the end, one apparently never apprehends the subjective, as such,
in itself. Instead, in order to grasp it scientifically, one first has to strip it
of its entire subjective character. One slays subjectivity in order to dissect
it and presumes that the life of the soul is on display in the dissected
remains! (Natorp 1912a: 103)
The situation thus seems hopeless, Natorp adds, making it understandable,
at least to a certain extent, why psychologists pursue a wholesale objecti-
fication of the subjective (Natorp 1912a: 103–105).
Nevertheless, such pursuits rest upon a fundamental mistake. The mistake
is the supposition that objective and subjective aspects lie next to one another
as co-ordinated domains of appearances – as though there were, for example,
two sorts of appearances, those of consciousness and those of external nature
(Natorp 1912a: 113). The mistake, in other words, is to fail to see how
everything appearing is at once subjective and objective. The appearing itself
is subjective, while what thereby appears is an object at some level of
objectification.7 A single experience, numerically and specifically the same,
thus has a subjective character (someone experiencing) and an objective
character (something experienced).8 Between these two characters there is
not simply a co-ordination of two mutually necessary sides (“no object
without subject, no subject without object”), but a reciprocity, like the
relation between right and left (“just as there is a left, only seen from a
right, a right only seen from a left”). This reciprocity entails a “complete
7
Much like Husserl whom he cites in this connection, Natorp notes the “correspondence and
correlativity” of content and object; but he also emphasizes how their relation is relative “in the
progression from stage to stage” (Natorp 1912a: 110f.).
8
Natorp construes his account accordingly as “the only tenable monism, that of knowledge itself,” a
monism that encompasses in itself the only tenable dualism, “that of the two directions of knowledge,
correlative to one another, the direction of objectification and subjectification” (Natorp 1912a: 112).
identity” between the characters, “when taken ideally,” i.e., when one thinks
of the infinite process of knowing. In the case of any actual knowledge of
experience, however, we stand in the middle of this process, and the
difference between the two characters is “always clear and unambiguous”
(Natorp 1912a: 107).
All other sciences, proceeding by way of analysis and abstraction, con-
struct objects from appearances, objects that “save” those same appearances –
even if, in the interest of objectivity, the sciences typically ignore the
subjective experience of those appearances. Consider, for example, how
mathematics takes such abstract objects as numbers, points as the basic
data of the science, omitting any consideration of the appearances or
experiences of them. A similar sort of objectification is no less at work in
pre-scientific and non-scientific (including technological) representations of
“things” (Natorp 1912a: 195f.). Moreover, just as there are objects of imagina-
tion, dreams and dream world, and art, so, too, ethics, aesthetics, and
religion constitute further domains of objectification (Natorp 1912a: 197f.).
Yet it is always possible to retrace these processes of objectification back
to the appearances, leading the discrete objects respectively back to the
continua of appearances out of which they were extracted, and in this way
“to restore, to a certain extent, the concrete context of what was originally
experienced.” The method of psychology consists precisely in this process
of “theoretically reacquiring the original life of consciousness,” a process of
restoration that amounts to inverting the scientific and non-scientific
method of objectification. “While this method makes objects out of
appearances, psychology reconstructs the appearances from the objects” –
or, at least it does so, as just noted, “to a certain extent.” Natorp iterates this
qualification, as he insists that the abstraction involved in the scientific
construction of an object is the only means of “conceiving the concrete life
of consciousness in any way at all or at least approximating such a concep-
tion.” Psychology approximates such a conception through its method of
reconstructing what is immediate in consciousness from the objectifica-
tions that were formed from it by science and, prior to any science, “by the
everyday manner of presenting things” (Natorp 1912a: 192f.).
This account of psychology’s method iterates the correspondence between
objectification and subjectification (the two-sidedness or bi-directionality of
knowledge), but it also underscores the dependency of psychology on
objectifying knowledge. While that correspondence entails the irreducibility
of each side in ordo essendi, psychology is dependent in ordo cognoscendi upon
objectifying knowledge, since “it can reconstruct nothing at all, where
objectifying knowledge has not previously constructed something.” For
similar reasons, there is no specter of psychologism, since the sort of
subjective grounding (Begründung) that psychology provides is secondary
to grounding “in the genuine, objective sense” (Natorp 1912a: 200f.).
The critical differences with Husserl’s phenomenology

As noted earlier, in Husserl’s Logical Investigations he criticizes certain
themes drawn from Natorp’s 1888 work, Introduction to Psychology.9 In
the course of introducing the sense in which consciousness is the subject
matter of psychology, Natorp takes exception to those critical remarks.10
In this context he raises the first of two objections to Husserl’s phenom-
enology, explicitly criticizing the latter’s early conception of the ego.
The criticism is significant for the ways it illuminates how Husserl’s
developing conception of the ego, while moving closer to Natorp’s
conception, remains opposed to it. The second objection, directed at
Husserl’s conception of phenomenology as a kind of first philosophy,
raises questions about the abstract focus of Husserl’s early phenomenol-
ogy, aimed at a more or less fixed slice and form of consciousness.
Criticism of Husserl’s early conception of the ego

Reviewing Husserl’s explicit criticisms of his early work, Natorp begins
by noting that Husserl is plainly mistaken to suppose that he (Natorp)
finds something “phenomenologically given” in the relation of the
experienced to the experiencing consciousness beyond the “unity of the
connection” (Verknüpfungseinheit) of experiences. Nor does he contend,
as Husserl appears to suggest, that being-an-object (for an I) cannot be an
object of acts in any sense or that the different relations to the ego in
different acts do not have a bearing on the content. Natorp insists that he
never contended as much, albeit not without adding that he did question
the notion that “being-an-object (for an I)” could be an object of the same
act (Natorp 1912a: 34f.).
Beyond registering what he takes to be Husserl’s misrepresentations of
his views, Natorp criticizes Husserl’s own early account of the ego’s
relation to the content of consciousness, at least in terms of the “ultimate,
absolutely original meaning” of that relation. Natorp takes exception
9
Husserl 1968: 359–363.
10
In Natorp 1912a, chapter 5, §9 Natorp also responds to other remarks by Husserl in the Logical
Investigations, that Natorp reckons could be construed as objections to his account.
specifically to Husserl’s early equation of the ego merely with the juncture
or collection of parts of the content (Fügung der Inhaltsmomente). On such
a view, Natorp contends, the statement “I am conscious of a” implausibly
signifies merely that a fits, as one such part, into a specific entire set
(Gesamtheit) of such components, where that entire set allegedly makes
up the ego. Yet we are conscious, not simply of individual parts (a, b), but
also of the entire set, e.g., S (a, b). It would be “quite artificial,” Natorp
observes, to regard the consciousness of the entire set as a part of the set.
Instead one set, e.g., S (a, b) is a component of a second set, e.g., S (a, b, c),
the latter of a third set, e.g., S (a, b, c, d) and so on. “According to Husserl,
one would have to be able to say that the entire set of contents is conscious
of the individual content, but there is no consciousness of the former at all”
(Natorp 1912a: 35). By contrast, Natorp submits, every consciousness, as
individual, supposes an ego that is conscious of the content, whether
individual or collective, but is thereby itself precisely not content.
One might, like Husserl himself at times, be tempted to say that the
difference between contents and the ego amounts to that between the
manifold contents and the unity of the combined contents. But this will
not do, Natorp contends, since the ego is conscious of that unity, the form
in the Aristotelian sense, no less than the manifold contents. Hence, the
ego does not coincide with the “unity of the connection” among contents,
even though this unity alone, as the “what is given phenomenologically”
(der phänomenologische Befund), corresponds to the ego, and even though
this unity can and must represent it for psychological inquiry, “since it can
be concretely apprehended in nothing else (as I have stressed again and
again)” (Natorp 1912a: 36).
Natorp adds what he takes to be a decisive argument for his contention
that “the ultimate, overreaching unity of consciousness” is not part of the
content of consciousness or even “a ‘phenomenological given’ of its
own” (einem eigenen ‘phänomenologische Befund’ ). “Were it necessary that
‘experiencing,’ in order for it to be recognized as a fact, be itself experienced
in turn (as ‘content’), then the same would hold with like necessity of
experiencing the experiencing and of experiencing this experiencing in
turn, and so on ad infinitum” (Natorp 1912a: 36f.). The regress is untenable
yet inevitable when one presupposes, as the early Husserl does, that the ego
is in some way part of the contents of consciousness.
In sum, Natorp charges that Husserl’s early construal of the ego-relation
as a component of the contents of consciousness is untenable for two,
complementary reasons. First, it mistakenly presumes that we experience it
as the juncture of contents or even as the unity of their combination, and,
second, it fails to appreciate how that presumption entails an infinite
regress. According to Natorp, the truth of the matter is that being-
conscious – or, equivalently (not identically), the ego that is conscious of
the content – is necessarily supposed by but not itself part of the content.
In Ideas I Husserl explicitly distances himself from his early account of the
ego, i.e., the account that Natorp contests. Since Husserl does so in the
course of referencing Natorp’s remarks, there is reason to think that he
regards the criticism or a version of it as trenchant.11
Still, even if this sort of criticism took effect, a considerable difference
remains, at least in Husserl’s view and that of commentators. Despite
identifying the “ego-pole” of his phenomenology with the ego as Natorp
conceives the latter, Husserl regards the ego as something that can be
intuitively grasped, and, indeed, in different ways in different sorts of
experiences. Yet the difference may be less than meets the eye. While
eschewing an intuition or, correlatively, a givenness of the pure ego,
Natorp by no means discounts the diverse manners of subjectification
and their accessibility through reconstruction. At the same time, to the
extent that both thinkers share a contention regarding the grounding
character of the ego, one has to wonder what the evidence or basis of it is
and whether they can shed more light on that grounding character itself.
There are at least two weighty questions here that neither thinker seems to
have addressed adequately: In what sense is the ego the ground of the unity
of content of consciousness? How is that ground identifiable, given its
difference from what it grounds?
Criticism of phenomenology as first philosophy

The criticisms of Husserl’s early conception of the ego, glossed in the
foregoing section, can be found in the second chapter of Natorp 1912a: ii,
§6. These criticisms, however, are mitigated, if not, indeed, nullified, by a
remark that Natorp makes in the eleventh, penultimate chapter of the
book, where he focuses on Husserl’s phenomenology (Natorp 1912a: xi,
§§11–14). There he remarks that apparent differences about consciousness,
acts of consciousness, and the like rest for the most part on “misunder-
standings and mere differences in terminology.” While this puzzling
remark may undermine the force of the earlier criticisms, it does serve to
highlight what Natorp takes to be “the essential difference” and “the only
11
By the time of Ideas I and the second edition of the Logische Untersuchungen (Husserl 1968: 361),
Husserl comes around to positing a pure ego; see Husserl 2014: 106 n.16.
serious, substantive difference” between his psychology and Husserl’s
phenomenology (Natorp 1912a: 282).
In the General Psychology Natorp iterates his complete agreement with
Husserl’s criticism of psychologism and demand for a “rigorously objective
justification” of logic and knowledge in the Prolegomena to the Logical
Investigations (Natorp 1901). This agreement is hardly surprising, given
the extent to which, as Iso Kern has shown, Husserl’s case against psycho-
logism’s presuppositions rests upon positions already elaborated by
Natorp, if not in print, then in personal correspondence with Husserl.12
Nevertheless, precisely given their agreement on this score, Natorp finds it
puzzling that, instead of directly proceeding to that objective justification,
Husserl’s first move is to provide a phenomenological, i.e., subjective,
justification. Natorp does not deny the efficacy of a subjective justification;
indeed, he understands his psychology to be doing just that. But he insists
that, in keeping with his reconstructive method, this justification can only
follow upon the objective one.
This objection is, on the face of it, puzzling. While Natorp may be right
to insist that the analysis of subjective states depends upon some account of
the relevant kind of object involved, he has to concede, on his own
principles, that any account of the object involved presupposes a subject
for whom it is an object (and how it is such). So construed, the objection
appears trivial, since we can only do (i.e., start with) one thing at a time.
However, there is much more to the objection; indeed, it goes to the
heart of the difference between Husserl’s phenomenology and Natorp’s
psychology. According to Natorp, Husserl starts with phenomenology
rather than objective investigations because he construes it as a purely
descriptive, non-theoretical study that grounds theoretical studies.
Natorp understands part of the motivation for this construal, namely,
the presumption that objectification and theory (knowledge of laws)
go hand-in-hand. The construal of phenomenology as descriptive
psychology, bent on capturing the facts of consciousness in a way distinct
from an explanatory science of laws governing causal connections, aims at
establishing an investigation of consciousness, of subjectivity, that is not
objectifying (Natorp 1912a: 155f., 189f.).
Yet this construal, Natorp claims, is misguided, since “description and
theory belong undeniably together; description is no less objectification
than theory is” (Natorp 1912a: 280). Natorp provides several reasons
for this claim. First, description classifies what is described, inevitably
12
See Kern, Husserl und Kant, 321–325.
subsuming it under universal concepts, reducing the individual to a token
of a type. Second, descriptions have a scientific meaning and function only
in the context of knowledge of explanatory laws. They serve, for example,
as points of departure, premises, or the conclusion to scientific reasoning,
aimed at determining objective states of affairs. Third (and closely
connected to the second point), it is untenable to segregate knowledge of
facts and knowledge of laws, since to understand facts is to understand the
laws governing them and to understand laws is to understand the facts that
they govern. Laws such as those determined by natural science merely
assert what is generally factual (Natorp 1912a: 156, 189f.).
One might well object that Natorp runs two distinct claims together
here, namely, the theory-ladenness of any description and its objectify-
ing character. Even if it is uncontestable that theory depends upon
some descriptions and vice versa, more needs to be said to show that
every description is a function of theory or, at least, is so in a way that
somehow undermines any independent efficacy or trenchancy of the
description, given the nature of what is described. For example, a
description made in the context of some practice may have a theoretical
basis without undermining the practical purposes and colorings of the
description.
Still, in Natorp’s defense, it may be noted that practical disciplines can
be no less object-oriented (or thing-oriented) than objective sciences.
Moreover, the present discussion is concerned with, not a practical
discipline, but psychology. In this context, he is contesting the integrity
and fidelity of an allegedly non-theoretical description of subjectivity,
where the description objectifies it, even if the description is not made in
the interest of theoretically explaining subjectivity.
Husserl is obviously not oblivious to this problem. Indeed, Natorp cites
Husserl’s own acknowledgement that a “secondary” reflection converts the
psychic experience into an object, albeit in a sense that is new and
completely different from the normal sense. In contrast to many psychol-
ogists and modern philosophers, Husserl shares Natorp’s view that the act
of positing an object is the primary act of consciousness. Thus, Husserl
notes that it is impossible to describe the intending acts without having
recourse to the intended matters – an observation that, Natorp remarks,
comes very close to his position.13 Content and act, noema and noesis are
only severable in reflection and abstraction.
13
Natorp cites Husserl 1968: 11: “Es ist schlechterdings nicht möglich, die meinenden Akte zu
beschreiben, ohne im Ausdruck auf die gemeinten Sachen zu rekurrieren.”
Nonetheless, the “essential difference” between them, as Natorp sees it, lies
in Husserl’s cavalier attitude toward the process of objectifying subjectivity.
“According to Husserl,” Natorp writes, “subjectivity is obviously a second
subjectivity, generically alike and coordinated with the first objectivity, the
one usually meant” (Natorp 1912a: 281). Yet subjectivity just is not, he stresses,
“a second objectivity, that of the acts of consciousness.” The only possible
access to it is indirect, namely, via its reconstruction, and, indeed, “just as the
counterpart of objectivity, a negative, as it were, of the latter” (Natorp 1912a:
281). Psychology’s task consists in reconstructing how an objectification had
to have been consciously implemented and, indeed, implemented for one
consciousness. In principle, all objectifications, pre-scientific as well as
scientific, are fair game here. Yet, although we are commonly presented
with “things” that are no less constructions, i.e., that are “of essentially the
same sort” as those of science, we tend to be more conscious of what we are
doing, every step of the way in science (Natorp 1912a: 196f.). Hence, it is from
sciences and, ultimately, the philosophical sciences – logic, ethics, aesthetics,
and religion – that psychology takes its bearings. Only in the wake of the
primary, objective justification of these disciplines, Natorp submits, can
psychology go about its task of reconstructing their subjective justification
or explanation (Natorp 1912a: 195f., 200f., 279).
By contrast, Husserl resists conceiving the secondary act, the objectifica-
tion of the subjective, as completely dependent upon the primary act of
objectification – as Natorp contends he must. Indeed, in Husserl’s view,
the deeply rooted habit of implementing the primary, objectifying acts
merely impedes the reflection on subjectivity. Those primary objectifica-
tions intrude and impose themselves on the secondary objectifications
(objectifications of “the acts of the primary objectifications”), threatening
to falsify them. In order to ward off such falsifications, stringent measures
are required, and, for Husserl, they take the form of “construing the subject
(those acts) purely in themselves,” something that Natorp finds it impos-
sible to do (Natorp 1912a: 282). Husserl’s confidence in doing so explains
precisely why he regards phenomenology as philosophy’s first task, whereas
Natorp regards his reconstructive psychology as its final task, dependent
upon the foregoing establishment of the constructive, objective philoso-
phical disciplines.
This difference between Husserl and Natorp is an impasse, sealed by their
differing conceptions of access to the pure ego. For Husserl, as noted above,
the ego is and must be capable of being given and intuited. It cannot merely
be supposed as the ground and ultimate source of unity of the content of
consciousness. By contrast, for Natorp, the ego, while never part of the
content of consciousness, must be supposed as the ground of the unity of
consciousness in general, and, indeed, supposed, not as a concept but as the
primordial experience (Urerlebnis) (Natorp 1912a: 39). Commenting on this
very passage, Husserl writes: “I cannot suppose anything as necessary that I
do not see in terms of its possibility.”14
To a certain extent, to be sure, this difference may appear more cosmetic
than substantive, since the phenomenological attitude, in bracketing the
natural attitude, retains its content, indeed, as a guide to analysis.15 Yet the
difference between their approaches is far more than cosmetic. Natorp
regards the methods of objectification and subjectification as correlative
ways of considering (knowing) the same thing, namely, appearances, and
they are ways that exactly correspond to one another (even if, as suggested
earlier, Natorp does not appear to have a way to demonstrate the supposi-
tion of that identity or this exact correspondence). The object-oriented
claim that “something appears to me” and the subject-oriented claim that
“I am conscious of something” are equivalent, albeit not identical, referring
to the same state of affairs (Natorp 1912a: 210ff.).
By contrast, in Ideas I Husserl explains that “consciousness (experience)
and real being are anything but types of being of the same order” (Husserl
2014: 89). As far as its sense is concerned, the entire spatio-temporal world
possesses “merely intentional being, that is to say, the sort of being that has
the merely secondary, relative sense of a being for a consciousness” (Husserl
2014: 90). Moreover, already in the Logical Investigations Husserl construes
phenomenological research as the source of the basic concepts of pure logic
and objective knowledge in the sciences. As he puts it: “Pure phenomen-
ology presents a domain of neutral investigations in which various sciences
have their roots” (Husserl 1968: 2f.). So, too, on the final pages of Ideas I,
while recognizing that ontological disciplines, both formal and material,
supply a “guide” to phenomenology, Husserl is quick to add quite “expli-
citly” that they by no means provide a justification for it (Husserl 2014:
308f.). To the contrary, whether it is a matter of formal disciplines or the
constitution of things of various regions in material disciplines, he insists
that they presuppose what is only made clear in phenomenological analysis
(Husserl 2014: 293, 305). In other words, none of these object-oriented
studies are, as Natorp conceives them, self-sufficient.
14
Kern, Husserl und Kant, 363.
15
Kern makes and documents this point trenchantly; see Kern, Husserl und Kant, 331–335. Note, too,
that Natorp’s reference to “natural consciousness” bears similarities to Husserl’s crucial invocation of
a “natural attitude”; see Natorp 1912a: 194; see, too, the naturalness of objectifying knowledge
(Natorp 1912a: 200).
More would have to be said to weigh these and other considerations as
ways of rebutting Natorp’s objection to Husserl’s conception of phenom-
enology as a kind of first philosophy, especially since the objection rests so
squarely upon a quite different conception of consciousness and metho-
dological access to it. In any case, for the purposes of the present chapter it
may suffice to note that the objection does not appear to have taken effect
in Husserl’s case. The same cannot be said, however, of an issue raised by
Natorp in the course of making that objection. Indeed, the issue eventually
motivates, at least to a degree, Husserl’s subsequent attempts to give
phenomenology a new, genetic direction.
Natorp introduces the issue by way of considering the question of the
relation between “the ‘descriptive’ content and ‘intentional’ object of
knowledge, a distinction that for Husserl means, to be sure, nothing less
than the foundation of scientific epistemology and psychology in general”
(Natorp 1912a: 282). Natorp acknowledges that there is a radical difference
between “having” and “meaning,” parallel to the difference between the
immediacy of what is subjectively experienced, and the mediated character
of what is objective. However, invoking his own account, Natorp adds that
this difference “resolved itself, rather deepened itself” into the difference
between two corresponding, reciprocal directions that do not stand fast
once and for all, but continue indefinitely in unbroken succession on both
sides, such that the subjective content may become the object and vice
versa, depending respectively upon the level of attained or foreseeable
knowledge (Natorp 1912a: 283). In other words, what is content and
what is an object changes – something that a static analysis of them fatally
overlooks.
Natorp contends that Husserl fails, on the one hand, to recognize that
being, in the objective and consequently also in the subjective sense, has
this character of a process, and, on the other hand, that the two sides (the
objective as well as the subjective) coincide in the concept of “appearance.”
Accordingly, after distinguishing the “genetic” from the “static” aspect of
knowledge (or, equivalently, “fieri [becoming]” from “factum”), Natorp
stresses that the coincidence of the subjective and the objective is only
intelligible in the sense of the genetic aspect. Whereas objectification’s
method of construction differentiates and interrupts the process, subjecti-
fication’s method of reconstruction integrates, carrying us back to the
“continuous function” of consciousness.16
16
Natorp 1912a: 285.
The ultimate basis of the fundamental difference between Husserl’s
phenomenology and Natorp’s psychology can be traced, Natorp submits,
to the fact that Husserl focuses merely on the specific, individual stage of
knowledge, while he (Natorp) considers the whole of human knowledge in
its continuous development from the lowest to the highest levels. In the
course of this development a subjective dimension may become something
objective and an objective dimension may become something subjective.
In other words, there is nothing objective in itself or subjective in itself. As
a result, there can be no analysis of the subjective dimension in itself that
serves as a kind of first philosophy, grounding objective sciences.
Just as Husserl holds fast to the intuitability of the pure ego, he also
never ceases to regard phenomenology as a kind of first philosophy. Yet if
Natorp helped him change his mind about the status of a pure ego,
Natorp’s explanation of the fundamental difference between them also
struck a chord with Husserl. Particularly in the wake of Husserl’s thorough
study of General Psychology in 1918, Natorp’s work is of “decisive signifi-
cance” for the genetic phenomenology introduced by Husserl shortly
before 1920.17
Concluding remarks: Natorp’s influence on Heidegger’s

phenomenology
There are some patent ways in which Heidegger’s hermeneutic phenomen-
ology and his criticisms of Husserl’s phenomenology echo the challenges
already advanced by Natorp.18 (1) Heidegger, too, calls into question what he
regards as Husserl’s pretension of a theory-free or presupposition-less
description (albeit without by any means signing on to Natorp’s psychology).
Husserl’s descriptions are formulated as propositions that serve as potential
premises in a theoretical science of experience. (2) Heidegger also follows
Natorp by challenging what they both regard as Husserl’s objectification
of subjectivity. Indeed, while Heidegger regards “being-in-the-world” as
17
For a demonstration of the parallels between Husserl’s genetic phenomenology and Natorp’s
reconstructive psychology, see Kern, Husserl und Kant, 350–356 and D. Welton, The Other
Husserl: The Horizons of Transcendental Phenomenology (Bloomington: Indiana University Press,
2002), chapter 9.
18
For studies of Heidegger’s critical reception of Natorp’s psychology, see M. Bowler, “Heideggerian
Reflections on Natorp,” in Heidegger and Aristotle: Philosophy as Praxis (New York: Continuum
2008), 55–75; D. Zahavi, Subjectivity and Selfhood: Investigating the First-Person Perspective
(Cambridge, MA: MIT Press, 2005); and S. Arrien, “Natorp et Heidegger: une science originaire
est-elle possible?” in S. Jollivet and C. Romano (eds.), Heidegger en dialogue 1912–1930: Rencontres,
affinités et confrontations (Paris: J. Vrin, 2009), 111–129.
equivalent to a radical conception of subjectivity, he assiduously avoids a
construal of it as something objective (even though, a few years after his
Natorp lectures, he espouses fundamental ontology as a theory and, indeed,
an objectification). (3) Just as any turn to the subjective follows upon some
objectification for Natorp, i.e., just as the preoccupation with objects is
already in place when we turn to the forms of pre-occupation themselves,
so Heidegger’s analysis of being-in-the-world begins with the ever-abiding
facticity of its involvement with things within-the-world. (4) Like Natorp,
Heidegger distinguishes human subjectivity from “things” presented in
everyday, pre-scientific practices and from “objects” of scientific theory.
Yet if Natorp’s influence on Heidegger is evident from his adoption of
Natorp’s criticisms of Husserl, it is no less evident in the very objections
that Heidegger raises against Natorp’s psychology. Reconstruction remains
itself a kind of construction and, hence, no less subject to the same
criticisms of fatally objectifying subjectivity that Natorp raises against
Husserl and others. Objectification is, as Heidegger puts it, “foundational”
(grundlegend) for subjectification which, for its part, exactly and thus fatally
corresponds to objectification (Heidegger 1987: 104, 107). To be sure, all
construction or objectification is only a starting point, but Natorp provides
no criterion for moving from that starting point to the original, pre-
reflective subjectivity. Indeed, he has a fatal preconception (Vorbegriff )
that prevents him from doing so: he continues to think of subjectivity in
terms of how it constitutes an object. As a result, he is unable to entertain
key aspects of the pre-theoretical sphere (notably, the ways we relate to our
worlds, particularly our environment, and the eventfulness of doing so).
To put it in the terms of Being and Time, he lacks a conception of being-in-
the-world. Not least, given his initial attempts to reform Husserl’s
phenomenology, Heidegger takes issue with Natorp’s contention that
phenomenology’s use of language and conception of intuition inevitably
objectifies what it addresses and intuits (Heidegger 1987: 111–112).
These criticisms are substantial but they do not gainsay the persisting
impact of Natorp’s challenge to Husserl’s phenomenology. In the light
of Heidegger’s pursuit of a fundamental ontology, a transcendentally
conceived phenomenology, it may seem that he rejects Natorp’s criticism
of Husserl’s conception of phenomenology as a first philosophy. Yet, if he
does, his subsequent move away from any such project may also be
regarded as a confirmation of the trenchancy of Natorp’s criticism.
chapter 13
Cassirer and the philosophy of science

Massimo Ferrari
The legacy of Marburg

At the beginning of the twentieth century, Ernst Cassirer was a prominent
spokesman of Marburg Neo-Kantianism, one of the most influential
philosophical movements in Germany at that time. His early work, starting
in 1899 with his PhD dissertation on Descartes and his brilliant Leibniz’
System (1902), is deeply connected to the legacy of Cohen’s and Natorp’s
systematic philosophy and their outstanding interpretation of Kantian
thought. Nevertheless, according to the received view, Cassirer would
have rapidly abandoned his philosophical origins in favour of a philosophy
of culture breaking with the Marburg tradition and moving closer to Hegel
or to Peirce’s semiotics. A similar view implies that the philosophical
adventure of the young Cassirer is of merely historical interest, whereas
his “greatest” achievement should be identified with his late Philosophy of
Symbolic Forms. Moreover, the importance attributed to Cassirer as a
philosopher of culture has often led scholars to underestimate the role
played by science in the framework of Cassirer’s philosophy. Until
recently, the significance of the historical development of scientific knowl-
edge for his thought has been largely neglected; in particular, many critical
studies devoted to Cassirer seem to forget that he offered profound inter-
pretations of Goethe, Schiller, and Humboldt, but also of Einstein’s and
Heisenberg’s physical theories.
In order to rectify this misunderstanding, it seems useful to highlight,
first of all, some fundamental aspects of his early philosophical enterprise.
Long before Cassirer would be the celebrated author of Substance and
Function, he had already developed a “functional” conception of scientific
knowledge in his juvenile studies about Descartes and Leibniz. Though
many recent interpretations of Cassirer’s philosophy have marginalized the
fundamental importance of his apprenticeship in Marburg, it seems quite
impossible to comprehend Cassirer’s own insights into science and its
261
262 massimo ferrari
relationship to philosophy without considering the influence exercised on
him by the major exponent of Marburg Neo-Kantianism, namely,
Hermann Cohen.1
Cohen’s Kants Theorie der Erfahrung first appeared in 1871 and in a
second enlarged version in 1885. The main feature of Cohen’s interpreta-
tion is the account he gives of Kant’s concept of experience as scientific
experience and of a priori as a pure form enabling experience, without
any reference to the psychological content of the mind. Cohen thus broke
with the interpretations of Kantian philosophy that characterized the
early “back to Kant” movement, as promoted by Friedrich Albert Lange
and Hermann von Helmholtz. The first important consequence of this
attempt to renew the “Kantian doctrine of a priori” consisted in a refusal
to consider the transcendental aesthetics as an autonomous stage within
the development of a priori structures of knowledge. More precisely,
Cohen argued that sensibility is not defined by its receptivity, but is
always the product of the synthetic activity of pure forms of intuition
(space and time), which in turn are rooted in an intellectual synthesis.
Cohen suggested that no veritable gap exists between a priori forms of
intuition and a priori forms of intellect. On the contrary, an essential
connection obtains between them such that the Transcendental Analytic
can be seen as the “organic fulfilment” of the Transcendental Aesthetics
(Cohen 1987: 61). For Cohen, but also for Cassirer, the question is thus
not to ascertain the a priori determination of material contents, but
rather to anticipate the general form of experience according to the
“plan of reason” (Cassirer 1981: 167). In his Das Prinzip der
Infinitesimal-Methode (1883), Cohen offered a kind of “application” of
this view to infinitesimal calculus (unfortunately neglected by Kant) as
grounded on intensive magnitudes, that is – in Kantian terms – on the
Principle of the Anticipation of Perception. According to Cohen, this is
just the fundamental “tool” able to “determine the object” not only of
infinitesimal calculus, but also of mathematical science in general: “the
tool for knowledge of nature understood as an organ producing and
constituting natural objects” (Cohen 1984: 133).
The final outcome of this recasting of Kant’s theory of knowledge is
Cohen’s Logik der reinen Erkenntnis (1902), which represents a definitive
overcoming of Kant’s original insights. In this work, the productive power
1
See Cassirer’s own recollections in his late article “Hermann Cohen 1842–1918” (1943). A more
detailed reconstruction of Cassirer’s early intellectual biography is available in M. Ferrari, Il giovane
Cassirer e la scuola di Marburgo (Milan: Franco Angeli, 1988), 117–149.
Cassirer and the philosophy of science 263
of pure thought (Erzeugung des reinen Denkens) is considered as the origin
(Ursprung) of any possible knowledge as well as any form of reality.
Accordingly, nothing can be given to pure thought prior to its own
“production” (Cohen 1977: 13). In this context, there is no more room
for Kant’s pure forms of sensibility. Moreover, the Kantian difference
between “knowing” and “thinking” is also effaced. The extremely wide
extension given to pure thought in the construction of scientific knowledge
was not, however, completely shared by other members of the Marburg
School (including Cassirer), although the leading view of knowledge as
based on the “origin,” i.e. on pure intellectual activity, remained at the core
of Marburg Neo-Kantianism.
It is noteworthy that the “given” (in a very peculiar sense) is always the
starting point of transcendental inquiry. Mathematical science is indeed a
“fact” (Faktum), historically determined and steadily changing, that
requires an analysis that moves back to its conditions of possibility, its
synthetic principles and epistemological foundations. Cohen thus sug-
gests that the standard definition of “theory of knowledge” is misleading
since the proper meaning of Kant’s reformulated project should be “the
critique of knowledge (Erkenntniskritik)” (Cohen 1984: 4–6). In this
manner, transcendental philosophy deals neither with the constitution
of the human subject nor with its ability to know, but rather with a
“meta-level” reflection – philosophical and epistemological – on the a
priori conditions of scientific knowledge. As Cohen writes: “The trans-
cendental refers to the possibility of a concept whose a priori or scientific
justification supports its validity” (Cohen 1984: 7). According to Natop,
the only “dogma” defended by Marburg Neo-Kantianism is this “meta-
reflexive” level, which constitutes proper philosophical inquiry into the
structure of scientific “paradigms” and, more generally, form of cultural
objectivity (Natorp 1912a: 193–194).
For his part, Cassirer repeatedly praised Cohen’s notion of the trans-
cendental method as “the core of scientific philosophy” and as a meth-
odological principle for an interpretation of Kant’s thinking. Cassirer
considered that Kant’s theory of experience should be interpreted from
the premise that Kant himself treated scientific experience within the
framework of transcendental conditions and investigated the possibility
of the “fact” of mathematical natural science in primarily referring to “the
lawfulness and the logical structure of experience” (Cassirer 1995: vol. i,
662). In agreement with Cohen, Cassirer also holds that the a priori is the
“supreme concept (Oberbegriff )” of any “possibility of experience”
(Cassirer 1912: 259). More precisely, Cassirer considers the main problem
264 massimo ferrari
of Kant’s critique as illuminating the “stable structure of empirical objects”
by means of transcendental analysis. It was for this reason that Cassirer
dedicated one of the most original sections of his admirable work The
Problem of Knowledge in Philosophy and Science of Modern Time to a
historical-systematic presentation of the long road leading “from Newton
to Kant.” In this work, Cassirer tracks the transformations during the
eighteenth century of the fundamental notions of space, time, matter,
law, and force, which formed the backdrop for the growth of the modern
natural sciences and Kant’s critical philosophy (Cassirer 1995: vol. ii,
391–582). The final goal of this “phenomenology of scientific mind” is a
transcendental point of view capable of demonstrating the “different forms
of objectivity” through the “mediation of a specific form of cognition.”
Hence, as Cassirer notes, “the matter to which it refers is . . . always already
in some way formed” and ‘reality’ is accessible to knowledge not per se, but
only through the “medium of geometry and mathematical physics”
(Cassirer 1981: 154).
This Neo-Kantian interpretation of Kant does not at all signify that
Cassirer wished to conceal his divergence from Cohen (Cassirer 1981: 3).
This divergence primarily concerns the radically dynamic way in which
Cassirer transformed Cohen’s motto “begin with a fact in order to inquire
into its possibility.” To be sure, Cohen based his Logik der reinen
Erkenntnis on the “developing fact [Werdefaktum] of mathematical natural
science” (Cohen 1977: 76); moreover, Cohen was convinced that the
categories of “pure thought” are involved in the historical process of
scientific knowledge, thus undergoing continuous change and develop-
ment (Cohen 1977: 396). Yet, it was Cassirer who first grasped the historical
and mutable dimensions of this “developing fact” in its veritable signifi-
cance for a Kantian epistemological project. In so doing, he bound, more
deeply than his Marburg teachers, the fate of critical philosophy to its
relation to the development of the exact sciences. Cassirer thus located the
sole enduring task of a critical inquiry based on the transcendental method
with the “continually renewed examination of the fundamental concepts of
science, . . . which simultaneously implies a thorough subjective self-
examination of the critique itself” (Cassirer 1907: 1). As Cassirer writes in
his Introduction to the first volume of Erkenntnisproblem (1906):
The “fact” of science is and will of course remain in its nature a historically
developing fact. If in Kant this insight does not yet appear explicitly, if his
categories can still appear as finished “core concepts of reason” in number
and content, so the modern development of critical and idealistic logic
[meant here is Cohen’s Logik der reinen Erkenntnis] has made this point
perfectly clear. The forms of judgment mean for it the unified and active
motivations (Motive) of thought, which course through the manifold of its
particular formations and are continually put to use in the generation and
formulation of new categories. (Cassirer 1995: vol. i, 18)
Cassirer’s impressive reconstruction of the problem of knowledge is there-
fore the result of his Neo-Kantian apprenticeship and the highest proof of
his original approach to an epistemological reflection on the scientific
Faktum. On the one hand, Cassirer’s main idea is that science and philo-
sophy must be mutually connected. Modern philosophy and modern
science constitute a unique whole and, more precisely, the understanding
of the problem of knowledge must consider philosophers such as Descartes
and Leibniz as well as scientists such as Galilei, Kepler, and Newton. On
the other hand, Cassirer intended to continue the ambitious project the
young Natorp had sketched in his early book on Descartes’ theory of
knowledge (1882), namely, to outline the prehistory (Vorgeschichte) of
Kant’s critical philosophy through a philosophical and historical examina-
tion of its sources in Descartes, Galilei, Kepler, and Leibniz.
Not accidentally, Cassirer’s early works – the dissertation on Descartes’
(1899) as well as the very extensive book on the scientific foundations of
Leibniz’s “system” (1902) – are conceived as “a prehistory of critique of
pure reason” aiming at following the long path from Leibniz to Kant
(Cassirer 1902: ix). Cassirer thereby exhibits his deep commitment to
Cohen and Natorp. Yet, it is particularly interesting that Cassirer already
stresses in these early studies that the concept of function was for the first
time elaborated by Descartes and Leibniz. Although Descartes’ mathema-
tical and scientific views are strikingly in contrast with his “dogmatic”
metaphysics of substance, it seems clear – according to Cassirer – that
Descartes perfectly understood the concept of mathematical function and
its importance for grasping an empirical fact as a variable depending on “a
series of relatively simpler presuppositions” (Cassirer 1902: 64). This
modern conceptualization of mathematics and the mathematical sciences
is equally central in Leibniz. For Cassirer, it was Leibniz’s great merit to
have criticized an Aristotelian notion of the concept as a generalization or
abstraction from empirical cases – to which even Kant remained indebted
with his contrast between intuition and concept. On the contrary, Leibniz
argued through his analysis situs that the concept of space is by no means an
“abstract concept” in the Aristotelian sense; it is rather a construction
through pure thought (Cassirer 1902: 146). The consequence of Leibniz’s
ingenious standpoint is a relational theory of concept. In particular, the
mathematical concept becomes a “paradigm” for any conceptualization:
266 massimo ferrari
the expression, more generally, of the “concept of function” (Cassirer 1902:
117–119, 215).
Cassirer defines the concept of function as the “logical model” leading to
the development of modern thought: it is the result and goal of the
mathematical science of nature and its epistemological justification
(Cassirer 1995: vol. i, 363). In 1906–1907 Cassirer outlines masterfully –
in his Problem of Knowledge – a “teleological” history based on the intel-
lectual conflict between a metaphysical concept of substance and a new
ideal of knowledge as endorsed by modern science – moving “from things
to relations” (Cassirer 1995: vol. i, 402). As a genuine Marburg Neo-
Kantian, Cassirer underlines once again that Kant’s greatest achievement
in the history of philosophy is his subordination of empirical data to the
pure and constitutive functions of knowledge (Cassirer 1995: vol. ii, 762).
The road to Cassirer’s outstanding Substance and Function was opened.
From substance to function: a new paradigm

of scientific knowledge
Substance and Function (1910) is not only the standard work of Neo-
Kantian epistemology. It is also a profound and fruitful debate with (the
then) contemporary philosophy of science. At the beginning of the twen-
tieth century, Cassirer was perfectly acquainted with the main topics in
both the philosophy of mathematics and the philosophy of physics. His
brilliant analysis devoted to the structure of abstract concepts and to a
functionalistic interpretation of scientific theories is founded on an admir-
able knowledge of the works (among others) of Gottlob Frege, Richard
Dedekind, Bertrand Russell, Henri Poincaré, Ernst Mach, Karl Pearson,
Federigo Enriques, Emile Meyerson and, last but not least, Pierre Duhem.
Moreover, Cassirer’s own “style of thought” emerges clearly from these
pages in a very attractive way thanks to the way in which his epistemolo-
gical analysis is connected to an account of historical development of exact
sciences. The philosophy of science and the history of science are tied
together, thus showing how the concept of function runs through “the
most important fields of scientific investigation” (Cassirer 1923: iv).
In order to comprehend the basic structure of this fundamental book we
must first bring into focus the general framework within which Cassirer
situates his detailed epistemological inquiry. On the one hand, Cassirer
outlines a general theory of scientific experience or, more precisely, the
form for every kind of scientific theory. The “fact” of science is always
changing, but what is not changing is its “ultimate constant form,” such
that a critical theory of experience is nothing other than a “universal
invariant theory of experience.” As Cassirer writes:
The procedure of the ‘transcendental philosophy’ can be directly
compared . . . with that of geometry. Just as the geometrician selects for
investigation those relations of a definite figure, which remain unchanged by
certain transformations, so here the attempt is made to discover those
universal elements of form, that persist through all change in the particular
material content of experience. The ‘categories’ of space and time, of
magnitude and the functional dependency of magnitudes, etc., are estab-
lished as such elements of form, which cannot be lacking in any empirical
judgement or system of judgements. (Cassirer 1923: 288–289)
As seen from this crucial passage, Cassirer’s conception of critical analysis is
not centred on a “system of categories” gained by means of a transcendental
deduction in the strictly Kantian sense or through the increasing “produc-
tion” (Erzeugung) of pure thought, as Cohen and, to some extent, Natorp
had envisaged.2 Of course, Cassirer does not deny that the “form” of
scientific invariant (or, in a broader sense: categories) derives from an
“origin” in Cohen’s sense. Yet, Cassirer’s main question is how such a
critical inquiry can lead to “the ultimate common element of all possible
forms of scientific experience” in the historical development of modern
science. The conditions of any scientific theory cannot be achieved once for
all since they are not bound to a single stage of scientific development.
Nevertheless, Cassirer argues that scientific thought aims at a final goal in
terms of a final system of invariants, or what Marburg Neo-Kantianism
understood as “the infinite task” (unendliche Aufgabe) of knowledge. As
Cassirer remarks:
Only those ultimate logical invariants can be called a priori, which lie at the
basis of any determination of a connection according to natural law. A
cognition is called a priori not in any sense as if it were prior to experience,
but because and in so far as it is contained as a necessary premise in every
valid judgement concerning facts. (Cassirer 1923: 269)
“Facts” are not a mere aggregate of perceptions, but a whole based on
specific conceptual assumptions, i.e. on general forms or categories
enabling knowledge as the “validity of certain relations and . . . their
superordination or subordination.” A general system of relations represents
in this sense the core of Cassirer’s original interpretation of Marburg
2
Natorp 1921: 7–11. See also J. Pulkkinen, Thought and Logic: The Debates between German-Speaking
Philosophers and Symbolic Logicians at the Turn of the 20th Century (Frankfurt am Main: Peter Lang,
2005), 248–252.
268 massimo ferrari
Neo-Kantianism, aiming to justify the transformations of scientific know-
ledge through a “liberal” and relativized view of the a priori conditions of
scientific experience. The elaboration of such a theory of relation has to be
found in modern logic and mathematics; on the other hand, its final goal is
to open “a path” to reality or, more precisely, to mathematical and
scientific knowledge of nature (Cassirer 1923: 270).
Let us go back briefly to Cassirer’s seminal essay “Kant und die moderne
Mathematik” (published in 1907), where he gives a stimulating account of
the recent debate about philosophy of mathematics, with particular regard to
Russell’s “Logicism.” Here, too, he deals with a new “fact” requiring critical
analysis: the “fact” of the logic of relations put forward by Russell in his
Principles of Mathematics as the foundation of mathematics, which is conse-
quently an “application” of such a “universal” logic (Cassirer 1907: 4–7).
Nevertheless, and contra Louis Couturat’s interpretation of Russell’s
Logicism, Cassirer contends that Russell does not understand the basic
principles of relational logic as simply analytic. Identity, difference, part,
whole, function are indeed “fundamental relations” that are not derived
from the principle of contradiction. They are “new and particular positions
and forms of connection created by thought” and contents of thought
resulting from a constructive procedure (Cassirer 1907: 36). Accordingly,
Cassirer suggests a synthetic interpretation of Russell’s logic, because its
presuppositions are precisely posed by thought, i.e. by pure thought in
the Marburg sense. At the same time, Russell’s great merit consists for
Cassirer in his overcoming of traditional, predicative logic in favour of a
logic of relations, i.e. in favour of a more powerful conceptual tool for the
foundation of mathematics according to the new perspective opened by
Gottlob Frege. In Cassirer’s mind Russell’s logic confirms the idealistic logic
of Marburg Neo-Kantianism. Pure thought is here, according to Cassirer,
evidently at work and, what is even more important, the activity of
pure thought consists fundamentally in establishing relational connections
(for instance the concept of sum in arithmetic) without any reference to
sensible content (Cassirer 1907: 41). This point is crucial. Once again the
ground problem of Marburg Neo-Kantianism – the revision of Kant’s theory
of pure sensibility – becomes the central question of every attempt to
“modernize” Kantian philosophy in keeping with the wide-ranging devel-
opment of mathematics. Beyond Cohen’s conception of infinitesimal mag-
nitudes as a “product” of the Kantian “anticipation of perceptions,” Cassirer
recognizes that the fundamental transformation of logic, mathematics, and
exact science in general during the nineteenth century unavoidably urges one
to rethink the structure of Kant’s Critique of Pure Reason. This means, on the
one hand, to interpret Kant’s doctrine of pure intuition as still belonging to
his pre-critical period and, on the other hand, to assimilate Kant’s transcen-
dental logic to Russell’s Logicism, insofar as the Kantian categories may be
considered as logical relations enabling not only mathematics, but at once
mathematics applied to natural science – the very decisive aspect which
Russell, nevertheless, leaves aside (Cassirer 1907: 31–32).
Cassirer’s early attempt to “translate” modern logic into the language of
Neo-Kantian philosophy constitutes the starting point of Substance and
Function. Two different problems are approached and represent what
Cassirer calls “the general schema and model according to which the
modern concept of nature has been moulded in its progressive historical
development” (Cassirer 1923: 21). This “general schema” is the concept of
function we have already encountered in Cassirer’s early work on Descartes
and Leibniz. The first step to grasp this “schema” consists in the refusal of
any theory of abstraction. An abstract concept is no longer, and quite
differently from Aristotelian tradition, a generalization of the common
features and properties of single empirical facts, but a general rule accord-
ing to which we can generate an order of elements. As Cassirer states, “a
sensuous manifold is conceptually apprehended and ordered, when its
members do not stand next to one another without relation”; the veritable
meaning of “abstraction” is by contrast a procedure enabling one to
generate a necessary sequence of members “according to a fundamental
generating relation” (Cassirer 1923: 15). The peculiarity of abstraction as
depending on a “fundamental generating relation” is thus gained through
the view of conceptualization in mathematics, i.e. by assuming that single
cases – numbers – are only conceivable as grounded on “the universal
validity of a principle of serial order” (Cassirer 1923: 20).
In its pars destruens, Cassirer’s theory of abstraction is similar to
Husserl’s. In his Logische Untersuchungen (to which Cassirer refers)
Husserl also argues that the conception of abstraction endorsed by empiri-
cism is untenable. For Husserl, as for Cassirer, we need a general, abstract
concept (“redness”) if we want to classify something as “red,” that is, as a
variation of an essence (in Husserl’s terminology, the “unity of species”)
(Husserl 1984: 112, 120, 155). Nevertheless, an important difference remains
between Cassirer and Husserl. Cassirer does not agree with Husserl’s view
of the possibility of grasping conceptual essences (or ideal meaning species)
through intuition. On the contrary, for Cassirer, a general, abstract concept
has to be understood as a principle of generation for a series; more
precisely, for example, a mathematical concept is a functional concept
grounded on the “generating relation” (erzeugende Relation). As he writes:
270 massimo ferrari
The meaning of the law that connects the individual members is not to be
exhausted by the enumeration of any number of instances of the law; for
such enumeration lacks the generating principle that enables us to connect
the individual members into a functional whole. (Cassirer 1923: 26)
We can now understand the position Cassirer assumes within the debate
about the foundations of mathematics, and especially about arithmetic,
inaugurated by Frege. To sum up, we can say that Cassirer endorses
Richard Dedekind’s concept of number. Before considering Cassirer’s
appraisal of Dedekind’s fundamental contribution to this debate with his
Was sind und was sollen die Zahlen, let us briefly recall the main reason why
Cassirer distances himself from Frege’s and Russell’s Logicism. According
to Cassirer, the concept of class by which Frege and Russell attempt to find
the basis for a logical deduction of the concept of number is compromised
by the use of a traditional concept of abstraction. Aside from Cassirer’s
conviction (similar to Husserl’s, Kerry’s or Poincaré’s) that Frege and
Russell do not evade a vicious circle postulating the equivalence of classes
as grounding the concept of number (Cassirer 1923: 50), the essential
objection Cassirer opposes to logicism is that the notion of class and,
consequently, of number gained from equivalence between classes (or
classes of classes) involves that “class” is something given, something
similar to a “thing” not produced by thought. So the number becomes,
to some extent, the generalization of class members. The only solution in
agreement with Cassirer’s theory of numeral concepts is thus Dedekind’s
construction of numerical sequence through “generative relation.” For
Dedekind, Cassirer remarks, “there is a system of ideal objects whose
whole content is exhausted in their mutual relations. The ‘essence’ of the
numbers is completely expressed in their positions” (Cassirer 1923: 39).
This kind of convergence between, on the one hand, Cassirer’s Neo-
Kantian insight into the essential function of pure thought as origin of
“generative relations” and, on the other hand, Dedekind’s mathematical
structuralism is undoubtedly one of the most interesting points of
Cassirer’s “functional” thinking. And, last but not least, such a view of
numbers and arithmetic in Dedekind’s spirit represents – as Cassirer
recognizes – a further reason for revising Kant’s own standpoint concern-
ing the relationship between time and number, given that the sequence of
numbers of a numerical series no longer has a connection with the
intuition of pure time (Cassirer 1923: 40).
Cassirer’s functional theory of concept and number is the starting point
for the interpretation of modern science in Substance and Function. In fact,
Cassirer’s transcendental constructivism is extended, first of all, to
geometrical knowledge, breaking here with the privileged role Kant attrib-
uted to the pure intuition of space. In his account of the development of
geometry in the nineteenth century, from Poncelet to Klein, from
Grassmann to Hilbert, Cassirer stresses over and over again a common
feature of modern geometry: “the inclusion of the spatial concepts in the
schema of pure serial concepts is here finally accomplished” (Cassirer 1923:
87). Cassirer’s “functionalism” is thus the background for finding a meth-
odological unity for the substantial development of geometrical theories
after Kant. This by no means signifies that Cassirer intends to refuse the
fundamental assumptions of Marburg Neo-Kantianism. For instance, and
surprisingly enough, he considers Hermann Grassmann’s “theory of exten-
sion” (Ausdehnungslehre) as grounded on the principle of “origin” in
Cohen’s sense. Grassmann’s audacious mathematical project confirms,
according to Cassirer, “the tendency . . . to subordinate the ‘given’ ele-
ments as such and to allow them no influence on the general form of proof”
(Cassirer 1923: 99). The relational, conceptual, and logical status of
geometrical knowledge seems thereby to be the “invariant” of any concept
of geometrical space and geometrical system. In this way Cassirer gives at
the same time an answer to the widely discussed question concerning non-
Euclidean geometries and their compatibility with Kant’s thought.
Cassirer’s position here is quite similar to Poincaré’s.3 According to
Cassirer, Poincaré correctly argued that it is entirely useless to inquire
into a supposed “essence” of space; the question is instead how different
geometrical systems, that is, systems of relations not founded on experi-
ence, are possible and why Euclidean geometry seems at first glance to be
the privileged one. Cassirer answers as Poincaré did: because it is simpler
than other systems (Cassirer 1923: 108–109).
Cassirer’s main effort is to emphasize in any field of scientific knowledge
not only the functional role played by concepts as rules forming a series,
but also that scientific knowledge in general depends on conceptual
assumptions or theoretical presuppositions. There is no “given” or
“pure” experience. Facts as such are not “mute,” but always need a logical
function in order to become objective facts beyond subjective sense-
perceptions. Interestingly enough, Cassirer quotes in this context one of
his philosophical mentors, the writer and poet Goethe:
From a new side, it is seen how far, as Goethe said, all that is factual is
already theory; for it is only the thought of the necessary determinateness of
3
Alan Richardson, Carnap’s Construction of the World (Cambridge University Press, 1998), 115.
272 massimo ferrari
phenomena, that leads us to arrest a single transitory observation, and
establish it as a fact. (Cassirer 1923: 243)
For this reason Cassirer calls into question Mach’s empiricism while
praising Duhem’s holism. Mach’s empiricism is unable to understand
the mathematical and logical structure of scientific experience; moreover,
it falls into an evident contradiction since his “mental experiments”
(Gedankexperimente) stand in contrast to the assumption according to
which scientific knowledge consists in pure phenomena (Cassirer 1923:
238–239). On the contrary, it is Duhem’s “excellent exposition” (Cassirer
refers here to La théorie physique) that offers an exact account of the mutual
relation between physical facts and the physical theory. Two main aspects
of Duhem’s epistemological approach seem to Cassirer particularly apt in
outlining a critical framework according to “logical idealism” (although
Duhem had nothing to do with the Kantian tradition). Firstly, Duhem has
the great merit to have understood the holistic character of scientific
theories, so that their empirical confirmation can be gained only when
individual concepts are considered as members of a “theoretical complex”
(Cassirer 1923: 146–147). Secondly, Cassirer particularly admires Duhem’s
profound insight into the unavoidable intellectual function in the forming
of scientific experience. As he writes: “The intellectual work of under-
standing, which connects the bare fact systematically with the totality of
phenomena – so Cassirer sums up Duhem’s account – only begins when
the fact is represented and replaced by a mathematical symbol ” (Cassirer
1923: 147 [my emphasis]).
Considering the problem of scientific knowledge as regarding the sym-
bolical status of concepts, too, Cassirer opens thereby a new path to the
Kantian problem with which Substance and Function deals: “how is knowl-
edge possible?” And he says consequently:
To know a content means to make it an object by raising it out of the mere
status of givenness and granting it a certain logical constancy and necessity.
Thus we do not know “objects” as if they were already independently
determined and given as objects, – but we know objectively, by producing
certain limitations and by fixing certain permanent elements and connec-
tions within the uniform flow of experience. (Cassirer 1923: 303)
The relativized a priori

In contrast to other Neo-Kantian “immunization strategies,” which
aimed at showing how Kant’s theory of knowledge does not require any
modification due to the advent of modern physics, Cassirer concentrated
on the attempt to “liberalize” Kant’s a priori, so that Einstein’s physical
worldview could be compatible with its fundamental function and the
unavoidable acknowledgement that “the relation between philosophy
and exact science has . . . changed fundamentally.”4 Kant, as Cassirer
especially urged, had offered “nothing but a philosophical circumlocu-
tion” of the presuppositions of Newtonian physics. Nonetheless, the
veritable aim of Kant’s Critique of Pure Reason was not to connect
philosophy once for all with “a fixed dogmatic of concepts,” but to
consider it within the development of science for which “there can be
only relative, not absolute, stopping points” (Cassirer 1923: 355).
According to the methodological principle already endorsed in his earlier
essay on Kant and philosophy of mathematics, Cassirer points out, first
and foremost, the necessity to understand relativity theory as a “fact”
offering “a new scientific problem by which the critical philosophy must
be tested anew” (Cassirer 1923: 355). Thus it was necessary for Cassirer to
justify epistemologically the theory of relativity (and later quantum
mechanics) as based on the conception of the a priori as a “process.”
Though Cohen, as Cassirer noted, took Kant’s epistemology to be a
systematic reflection on Newtonian natural science, the Critique of Pure
Reason is not to be understood as a finished system, the fate of which is
exclusively linked to that of Newtonian physics. In other words, the
transcendental method is not bound to a particular form of the natural
sciences: it rather demands an epistemological reflection on the problems
emerging from the development of science, going beyond Cohen’s
“Newtonian” interpretation of Kant’s relationship to science (Cassirer
1923: 355).
This also means that the essential contribution of modern physics is a
purely functional or relational concept of objectivity that is not in the least
anthropomorphic (as Max Planck would have said), but resting merely on
the form of science, i.e. “the real logical framework of nature in general”
(Cassirer 1923: 374). The concept of form is here central. Einstein’s theory
of relativity is a theory grounded on principles (e.g. the constancy of light
4
Cassirer, Einstein’s Theory of Relativity Considered from the Epistemological Standpoint (1923: 353, 394).
For the historical and philosophical context see Klaus Hentschel, Interpretationen und
Fehlinterpretationen der speziellen und der allgemeinen Relativitätstheorie durch Zeitgenossen Albert
Einsteins (Berlin: Birkhäuser Verlag, 1990), Thomas Ryckman, The Reign of Relativity. Philosophy in
Physics 1915–1925 (Oxford University Press, 2007), and Massimo Ferrari, “Neukantianismus und
Relativitätstheorie,” in C. Krijnen and K. W. Zeidler (eds.), Wissenschaftsphilosophie im
Neukantianismus: Ansätze – Kontroversien – Wirkungen (Würzburg: Königshausen & Neumann,
2014), 177–192.
274 massimo ferrari
velocity), whereas the issues of empirical measurements represent, as
Goethe would say, “only half of the experience.” The reference to experi-
ence depends on the “ideal form” the physical theory is able to construct in
order to show how experience is possible. In spite of its “revolutionary”
contribution to modern physics, this is exactly the form of Einstein’s
theory and, broadly speaking, of physical knowledge in general. On the
one hand, Cassirer stresses the essential continuity characterizing the devel-
opment of physics and, more precisely, he considers the theory of relativity
not as a break with previous physical theories, but rather as “one of those
triumphs of the critical and functional concept over the naive notion of
things and substances, such as are found more and more in the history of
exact sciences” (Cassirer 1923: 405). On the other hand, Cassirer makes it
clear that experience depends to some extent on theory or, as we say today,
is “theory-laden.” Against a widespread interpretation of relativity theory
as grounded on Mach’s empiricism, Cassirer’s gives a Duhemian account of
Einstein’s physics, according to which it does not depend on any kind of
experimentum crucis. It is rather a pure conceptual interpretation of facts
that constitutes the basis of relativity theory. An “holistic” concept of form
and the conviction that “facts” have to be considered only through con-
ceptual elaboration represent Cassirer’s Neo-Kantian insights into
Einstein’s physical theory. As he writes:
We never measure mere sensations, and have never measured with mere
sensations, but in general to gain any sort of relations of measurement we
must transcend the “given” of perception and replace it by a conceptual
symbol, which possesses no copy in what is immediately sensed. (Cassirer
1923: 427)5
Cassirer thus emphasizes the role of co-variance in the General Theory of
Relativity by noting that “it collects all particular systematic principles into
the unity of a supreme postulate, in the postulate not of the constancy of
things, but of the invariance of certain magnitudes and laws with regard to
all transformation of the system of reference” (Cassirer 1923: 404). If the
theory of relativity can also be considered within a universal invariant
theory of experience as well as purely relational structures of scientific
thought, one must therefore assume that space and time serve as concep-
tual tools or simple “schema[s] of connection” (Schema der Verknüpfung)
(Cassirer 1923: 412). Cassirer thus clearly agrees with the Marburg
5
Cassirer explicitly refers here to Duhem’s La théorie physique in order to characterize the symbolic
status of our conceptualization of the given. See Pierre Duhem, The Aim and Structure of Physical
Theory (Princeton University Press, 1974: 175).
“correction” of Kant’s Transcendental Aesthetic (Natorp 1912a: 196) in
considering space and time as productions of pure thought and its catego-
rical apparatus (Cohen 1977: xii, 12, 26–27, 150–152, 188–192). Following
this typical Neo-Kantian interpretation, Cassirer had no difficulty in
acknowledging the “crisis of intuition” in modern physics and the neces-
sary overcoming of the Newtonian paradigm. In his opinion, it was
sufficient to prove the “objective significance” of space and time in the
construction of a physical theory and to view both concepts not “as rigid,
but as living and moving [forms].”6 This, however, entails that the a priori
character of space and time is thereby reduced to a pure form of alignment
(Reihenform) of following-alongside or following-after; as such, they are
“ultimate ideal determinations” that owe more, it seems, to the Leibnizian
idea of an “order of phenomena” than to the Kantian notion of intuition
(Cassirer 1923: 418–419). Such forms of “functional thinking” are concei-
vable only as a priori forms – hence the a priori of space contains “no
assertion concerning any definite particular structure of space in itself”,
rather it deals only with “that function of ‘spatiality’ in general, that is
expressed . . . quite without regard to its character in detail” (Cassirer 1923:
433). The question of the “real” (Euclidean or non-Euclidean) structure of
space therefore loses all relevance, as the new task at hand is to ascertain the
a priori validity of spatiality as a general relation, and not to define it in
terms of empirical measurement.
The constitutive function of this “moving” Leibnizian relational a priori
is not called into question. As Cassirer remarks in his polemical correspon-
dence with Moritz Schlick regarding the epistemological interpretation of
relativity theory, he is convinced that Kant is in need of a “revision,” since
Kant tied the a priori forms of space and time to determinate contents in a
one-sided and absolutely fixed manner. In a letter to Schlick of October 23,
1920, Cassirer remarks that their main disagreement concerns the concept
of the a priori, which is for him not “a steady and definitively established
6
See Cassirer, Einstein’s Theory of Relativity Considered from the Epistemological Standpoint (1923: 411,
412). According to Friedman, the interpretation offered by Cassirer and, more generally, by the
Marburg School consisting in the tendency “to minimize or downplay the role of Kantian faculty of
pure intuition” is “a profound mistake.” See Michael Friedman, “Ernst Cassirer and Thomas Kuhn:
The Neo-Kantian Tradition in the History and Philosophy of Science,” in R. Makkreel and S. Luft
(eds.), Neo-Kantianism in Contemporary Philosophy (Bloomington: Indiana University Press, 2010),
183, as well as A Parting of the Ways: Carnap, Cassirer, and Heidegger (Chicago: Open Court, 2000),
91. Friedman seems to pose this question correctly from the point of view of Kant’s own conception
of knowledge and of his a priori foundation of possible experience, but he does not consider the very
systematic and historical grounds that lie at the core of Cassirer’s (and more generally of the Marburg
School’s) attempt “to minimize” the role of pure intuition.
276 massimo ferrari
complex of material intuitions or concepts,” but only a function. This kind
of functional a priori can be conceived as a “form” and “direction” capable
of “the most varied developments in the progress of knowledge.” In this
sense, the a priori means in general the lawfulness of experience, whereas its
particular specifications are connected to the principles and the presuppo-
sitions emerging from the advances of scientific thought. As Cassirer adds:
Therefore I admit . . . that Kant has not distinguished enough between the
general principle and its particular, concrete realization. His theory needs
from this point of view a revision; nevertheless, the principle of uniqueness
is for me something more than a mere “convention” or an “inductive
generalization”: it is according to me “an expression of reason, of logos
itself.” (Cassirer 2009a: 50–51)
In refusing the “orthodox” view of a priori endorsed by Schlick, Cassirer
tried to elaborate a relativization of a priori principles as connected with the
“progress of knowledge.” Quite differently from this perspective, Schlick
formulated a drastic objection to Cassirer’s logical idealism: the existence of
synthetic a priori judgements, he says in his 1921 review of Cassirer’s book,
depends on the possibility of exhibiting a scientific theory resting on uni-
versal and necessarily valid principles. Unfortunately, the theory of relativity,
together with Euclidean geometry, by no means demonstrates that principles
of this kind really exist. Accordingly, Cassirer does not provide a “convincing
evidence of how we may heal the wound dealt to the original Kantian
viewpoint by the overthrow of Euclidean physics” (Schlick 1979: 327–328).
Schlick thus argues that Cassirer’s reinterpretation of Kant’s thought is based
on a “general formulation” that is “unclear” (Schlick 1979: 323).
For his part, Cassirer is convinced that the only remaining principle of
lawfulness of experience in general must be considered, in this very broad
constitutive framework, as an “expression of reason” or, otherwise, as a
synthetic presupposition of scientific inquiry in order to formulate more
specific, empirically determined laws of nature. Cassirer was in this sense
quite near to and, at the same time, not in agreement with Reichenbach’s
proposal.7 Cassirer believed that it was appropriate to take into account the
constitutive role of a priori principles without interpreting them as uni-
versally valid; nevertheless, he maintained that Kant was not compromised
by a conception of a priori incompatible both with Reichenbach’s analysis
7
See Hans Reichenbach, The Theory of Relativity and A Priori Knowledge (Berkeley: University of
California Press, 1965), 87. This topic has been stressed by Michael Friedman’s illuminating studies:
see Reconsidering Logical Positivism (Cambridge University Press, 1999), 59–70, and Dynamics of
Reason (Stanford: CSLI, 2001), 71–82.
of scientific theories and with Einstein’s theory of relativity. In a letter to
Reichenbach of July 7, 1920 Cassirer pointed out his affinity with
Reichenbach’s view, yet argued that Kant was still compromised by a
psychological conception of a priori that was inconsistent with its epistem-
ological significance.8
Cassirer was thus engaged in defending a notion of the relativized
a priori which represents, on the one hand, a rigorous alternative to
Schlick’s diagnosis of the failure of synthetic a priori as a consequence
of Einstein’s relativity theory and, on the other hand, a far-reaching
invitation to take better care of Kant’s own insights, defending him
from possible misinterpretations.9 Cassirer’s interpretation of a priori
forms as “living,” “moving,” and “flexible” is not confined to the epistem-
ological context of relativity theory alone. In the final pages of his book
on Einstein, he shows that a plurality of dimensions or directions can be
ascribed to the a priori forms of space and time. For instance, it seems
suitable to distinguish physical from psychological space, mythic from
aesthetic space, as well as physical from psychological time, mythic from
aesthetic or historical time. The possibility then arises to define different
modalities within a common a priori, i.e. form-giving ordering function,
or – using the terminology of the late cultural theory Cassirer sketches
here for the first time – to anchor the different “symbolic forms” in their
symbolizing function. In this regard, physics is not a separate realm of
philosophical reflection, but is interconnected with other cultural
domains. As Cassirer writes:
It is the task of systematic philosophy, which extends far beyond the theory of
knowledge, to free the idea of the world from [any] one-sidedness. It has to
grasp the whole system of symbolic forms, the application of which produces for
us the concept of an ordered reality . . . Each particular form would be
“relativized” with regard to the others, but since this “relativization” is through-
out reciprocal and since no single form but only the systematic totality can serve
as the expression of “truth” and “reality,” the limit that results appears as a
thoroughly immanent limit, as one that is removed as soon as we again relate
the individual to the system of the whole. (Cassirer 1923: 447)
8
This letter is now available (number catalogue 284) in the DVD-Edition sämtlicher bislang aufgefun-
denen Briefe von und an Ernst Cassirer, enclosed in Cassirer’s Ausgewählter wissenschaftlicher
Briefwechsel (2009).
9
On this topic see Massimo Ferrari, Ernst Cassirer. Stationen einer philosophischen Biographie
(Hamburg: Felix Meiner Verlag, 2003), 123–135. As Ryckman states, “the claim that general relativity
sounded the death knell of ‘the Kantian position’ follows only if, as Schlick did, one ignored
important neo-Kantian developments of Kant’s thought as well as many of the most significant
developments in relativity theory in the period 1915–1925” (The Reign of Relativity, 5, 53–59).
278 massimo ferrari
Lost determinism?
The book on Einstein’s theory of relativity does not represent the last
answer Cassirer gave to the challenges posed by modern physics for a Neo-
Kantian philosophy of science. He was convinced that the mathematical
and natural sciences were by no means a simple annex to a philosophy of
culture and a philosophy of symbolic forms; they represented the core of
human intellectual development (Cassirer 1944: 207). In the last years
of his life, Cassirer was still deeply interested in contemporary science. It
was during his Swedish exile that Cassirer composed not only the fourth
and last volume of his impressive Problem of Knowledge, but also his
admirable inquiry into the revolutionary developments of quantum
mechanics. Determinism and Indeterminism in Modern Physics (1936) is
one of the most influential contributions to the debate flourishing during
the 1930s concerning the philosophical implications of the “new” physics
and, more precisely, the supposed decline and fall of determinism and
causality. Philipp Frank, an eminent exponent of the Vienna Circle, wrote
in his review of Cassirer’s book that the analysis offered by Cassirer rested
upon “a thorough knowledge of the subject.”10 Moreover, Frank was
convinced that “some of the basic conceptions of Cassirer are in remarkable
manner related to, or almost coincident with, the conceptions formed by
the logical empiricism of science.”11 To be sure, Frank did not agree with
Cassirer’s idealistic theory of knowledge, which he labelled – in his book on
causality published in 1932 – a kind of “scholastic” philosophy that
conceived science as a mere “system of symbols and not as a picture of a
‘true world’.”12 Nonetheless, Frank praised Cassirer’s “highly successful
attempt” to comprehend the increasing progress of science.13
Frank’s positive review of Cassirer’s work is paradigmatic not only
because of the affinities Frank underlined between Cassirer and Logical
Empiricism. The great merit of Cassirer’s last impressive work on the exact
sciences consists in the attempt to combine a profound knowledge of
quantum mechanics with the original Kantian point of view he had already
sustained in his book on Einstein’s theory of relativity. Once again, the
critical philosophy aims to provide an account of a new “fact” of science,
10
P. Frank, Modern Science and its Philosophy (Cambridge, MA: Harvard University Press, 1949), 176.
11
Frank, Modern Science and its Philosophy, 174.
12
Frank, Modern Science and its Philosophy, 268.
13
Frank, Modern Science and its Philosophy, 185. See also P. Frank, The Law of Causality and its Limits
(Dordrecht: Springer, 1997), 12. In a letter to Cassirer (without date) Frank suggests that Cassirer is
quite close to the Vienna Circle in putting into the foreground, more than “whichever other
philosopher,” “the logical constitution” of science (Cassirer 2009a: 176).
which involves the reformulation of some fundamental principles of Kant’s
theory of knowledge. To emphasize this aspect signifies by no means to
consider Cassirer as a Kantian philosopher who simply poses the question
about the extent to which quantum mechanics is compatible with Kant’s
theory of knowledge. Just as with the case of theory of relativity, Cassirer is
engaged in a more sophisticated epistemological project. Neo-Kantianism
is for Cassirer only a mode of thought, not necessarily tied to the single
achievements of Cohen’s and Natorp’s systematic philosophy (Cassirer
1956a: xxiii–xxiv).
Let us consider, first of all, Cassirer’s view of the relationship between
the process of physical knowledge and its a priori structure. This is a point
linking together the books on relativity and quantum mechanics:
When in the face of the new factual material and the new theoretical task
which it was facing, physics extended and transformed its conceptual
apparatus, it did not simultaneously give up its general character and
structure. It only made evident the fact that this structure is to be
thought of not as rigid but dynamic, that its significance and efficacy
do not rest upon its substantial rigidity established once and for all but
precisely upon its plasticity and flexibility. The a priori that can still be
sought and that alone can be adhered to must do justice to this flexibility.
It must be understood in a purely methodological sense. It is not based
on the content of any particular system of axioms, but refers to the
process whereby in progressive theoretical research one system develops
from another. This process has its rules, and these rules provide the
presuppositions and foundation for what we may call the “form of
experience.” (Cassirer 1956a: 74)
According to Cassirer, the new, unexpected challenges posed by quantum
mechanics receive a thorough answer only within this epistemological frame-
work. Hence, the “methodological sense” of the a priori is both the pre-
supposition and the consequence of a critical concept of objectivity. Not
only classical or relativistic physics, but also quantum mechanics testifies that
every form of “naive realism” is no longer an adequate standpoint for
physical knowledge. Instead of the common-sense concept of reality, we
are compelled to consider physical reality as well as physical objectivity as the
consequence of the concept of law, which is “prior to that of object, whereas
it used to be subordinate to it” (Cassirer 1956a: 131). This is a very crucial
point. For Cassirer, scientific laws are not simply read off from empirical data
or observations and measurements. It is precisely the formulation of laws
that allows for the right comprehension of what exists and has to be
investigated through scientific theories. As he writes: “Objectivity or
280 massimo ferrari
objectivity reality, is attained only because and insofar as there is conformity
to law (Gesetzlichkeit) – not vice versa” (Cassirer 1956a: 132).
Cassirer’s interpretation of quantum mechanics rests on this fundamen-
tal assumption. In particular, the law of causality must be considered as a
general form of lawfulness “not drawn from nature” – that is, as a general
statement regarding not the real events or things, but only the conceptual
connection constituting scientific statements and making possible scien-
tific experience (Cassirer 1981: 167). After Hume and Kant this remains the
unique way to approach the problem of causality in a very “critical” sense
(Cassirer 1956a: 20–21). Nevertheless, recent developments in physics have
led, as Cassirer recognizes, to a deep transformation of the causal schema
(Cassirer 1956a: 75). However, despite this undeniable transformation, the
main question remains for Cassirer to clarify the extent to which “science
must henceforth change its direction and radically alter his methodology”
(Cassirer 1956a: 114). To give an answer to this question requires a pre-
liminary inquiry into the epistemological structure of a physical theory as
such.
Cassirer distinguishes three levels of scientific statements, following in this
sense the theory of logical types proposed by Alfred North Whitehead and
Bertrand Russell in Principia Mathematica. According to Cassirer, physical
knowledge starts with statements regarding the results of measurements,
then formulates in a second step statements of law and, finally, statements of
principles. All three levels are interconnected with each other in a more
general conceptual structure, albeit the principles, in particular, do not stand
at the same level of laws, i.e. of statements concerning specific phenomena of
physical experience and embracing statements of measurements. As Cassirer
writes: “Principles are not themselves laws, but rules for seeking and finding
laws. This heuristic point of view applies to all principles” (Cassirer 1956a:
52). The heuristic or, in Kantian terms, regulative function of principles can
be understood as a general framework of human thought, which guides and,
at once, permits the search for ever more general laws. Accordingly, the
causal law itself is nothing but a regulative principle, and in this sense it is
given a priori, it is a transcendental law “for a proof of it from experience is
not possible” (Cassirer 1956a: 65).
Starting from this basic characterization of causality as “the general
requirement of conformity to law,” Cassirer is able to answer to the
challenge posed by quantum mechanics and, particularly, by Werner
Heisenberg’s at first glance “revolutionary” contribution to its develop-
ment. For such a general requirement seems not to be in conflict with
Heisenberg’s uncertainty relations that no longer represent “an exception”
(Cassirer 1956a: 123). On the contrary, according to the “law of continuity”
(in Leibniz’s sense), we may argue that “the uncertainty relations have by
no means demolished the bridge between quantum theory and classical
physics, or between the former and the causal principle” (Cassirer 1956a:
124). The main issue in Cassirer’s interpretation is that precisely this
“bridge” constitutes the standpoint requested in order to give an extensive
account of the significance of quantum mechanics. First of all, the general
form of atomic physics is in no way something different from the form of
physics in general. The statistical statements are rigorous and mathematical
statements, so that the logical determination of the theory is not lost,
although in this context it no longer deals with dynamic, classical laws,
but only with statistical ones (Cassirer 1956a: 118–119). Secondly, quantum
mechanics does not renounce a basic feature of every physical theory: the
function of “invariants” or “universal constants” constituting the essential
structure of a theory, in this case, the universal constant or, in other terms,
the “fixed frame” of Planck’s elementary quantum of action (Cassirer
1956a: 121–122). Thirdly, quantum mechanics confirms once again that
“no path any longer leads back to that ‘naïve realism’ which used to be
recommended as the natural and adequate standpoint for physics”
(Cassirer 1956a: 131). And finally it is exactly this impossibility of coming
back to old forms of realism or to substantialistic assumptions that repre-
sents a further achievement of the functionalistic way of thinking Cassirer
had already professed since 1910 (Cassirer 1956a: 130–131).
In 1936, the Neo-Kantian Cassirer was also convinced that his early
epistemological insights needed to be enlarged and, to some extent, mod-
ified, maintaining nevertheless the same essential features as twenty-five
years ago. Obviously, the quest for a re-examination of the status of
Kantian causality was the more compelling and difficult task in the face
of quantum mechanics. Yet, Cassirer intended in no way to follow the
direction in which Kant had moved in order to specify how the category of
causality is applicable empirically through the sensuous schemata and the
pure forms of space and time. These schemata, Cassirer contends, have lost
“their universal significance,” such that a “transcendental logic can thus no
longer be connected with or be dependent on transcendental aesthetics, as
was the case in Kant’s system” (Cassirer 1956a: 166).
No wonder, one may remark, that a Marburg Neo-Kantian such as
Cassirer was so ready to modify some fundamental aspects of Kant’s theory
of knowledge. To be sure, this was the unique way still open in the attempt
to reconcile a “revision” of Kant’s a priori knowledge with the increasing,
and even astonishing, transformation of modern physics. As Cassirer
282 massimo ferrari
suggested, the essential nature of causal law in a broad Kantian sense
remained, however, “untouched” and “defined only by the demand for
strict functional dependence” (Cassirer 1956a: 118). Nonetheless, this over-
view of the development of quantum mechanics seems to be too “irenic” as
it neglects the most important “oppositions” within the scientific commu-
nity at that time. An outstanding physicist such as Max Born pointed out,
in a letter written on March 19, 1937 after reading Cassirer’s book, his
excessive attitude in stressing the continuity and “normal” character of
quantum mechanics as compared with classical mechanics. Such a denial of
revolutionary breaks in history of physics had as a consequence, according
to Born, a view of unchangeable “invariants” in experience, whereas Born
was convinced that some essential “departures” (Abweichungen) from the
linear process of knowledge were essential. This was, in Born’s opinion, the
only way to formulate a correct assessment of Kant’s philosophy, that is,
both by avoiding the a priori (this “fatal word”) and by accepting the
“critical” standpoint (Cassirer 2009a: 161). Paradoxically enough, the main
question is thus whether Cassirer’s interpretation of causality was so
broadly conceived that he was compelled to refuse the causal law of classical
mechanics on which Kant’s understanding was based. In other words, the
doubt – expressed in particular by Carl Friedrich von Weizsäcker –
regarded the too flexible notion of causality as a priori principle that
Cassirer had defended in order to justify from a critical standpoint the
new physics: a generous concession that the modern physicist himself had
no reason to request from a Kantian philosopher (Weizsäcker 1937:
860–861). It is precisely for this reason that the Kantian-inclined physicist
Max von Laue asked Cassirer in 1937 whether it was still possible to call
“Kantian” a point of view not considering as a mainstay of the Critique of
Pure Reason the doctrine of categories: in his mind, the “existence of the
whole theory of knowledge” depends on this crucial point, which
Cassirer unfortunately had overlooked in his book (Cassirer 2009a:
164–166).
Surely Cassirer was no longer (and in fact he had never been) the
spokesman of a strict Kantianism. However, as already in the case of
relativity theory, he was an acknowledged authority within the epistemo-
logical debate about modern physics, capable, at the same time and quite
differently from other protagonists of this debate, grasping the widest
philosophical implications of exact sciences. As in the book on Einstein’s
theory of relativity, the inquiry into quantum mechanics ends with a more
general account of its philosophical significance. Cassirer was rightly con-
vinced that the overall discussed term of “indeterminism” was ambiguous
and to some extent dangerous (Cassirer 1956a: 89).14 The opinion accord-
ing to which the world has to be seen as totally indeterminate and
dominated rather by chance is only the fruit of misunderstandings unfor-
tunately often accepted “in popular discussions” (Cassirer 1956a: 3–25, 117–
118). On the contrary, Cassirer is particularly sensitive to emphasize the
ethical consequences of the presumed indeterminism dominating our
physical world. A veritable indeterminism cannot even be found in the
domain of human action and volition. All the great philosophers, from
Plato to Spinoza and Kant, have never conceived freedom as “indetermi-
nateness,” but rather “as a certain form of determinability.” In this sense, it
is simply meaningless to speak of an opposition between freedom and
causality, as we have well understood since Kant’s distinction between the
realm of causal experience and the realm of liberty, according to which the
theoretical world and the ethical world are complementary (Cassirer 1956a:
203, 205). As with the case of the theory of relativity, Cassirer here attempts
to show that the scientific knowledge of nature is always connected, to
some extent, with the world of man. Science has long abandoned the hope
of an exhaustive explanation of the whole natural world based on a unique
system of symbols. But it is exactly this “manifold of perspectives” opened
by the new situation of exact sciences which allows us to draw important
conclusions for our image of the world and, more precisely, of human
freedom in a cultural dimension. From this standpoint, the final conclu-
sion of Cassirer’s Determinism and Indeterminism is still a Kantian one; and
what is at stake is his firm belief that the “history of physics is not so much a
history of detection of new facts as the detection of new means of
thinking.”15
To be sure, this is one of the most interesting features of Cassirer’s
philosophy of science. Furthermore, his attempt to consider philosophy
and the history of science as two faces of a single process developing through
various historical phases or, as Thomas Kuhn would say, through the
succession of different “paradigms,” represents an essential character of
Cassirer’s all too little acknowledged “modernity.” However, as I have
shown elsewhere,16 Cassirer was firmly convinced that the concept of
14
See also Cassirer’s letter to Einstein on February 8, 1937, where Cassirer underlines the “strange
conclusions” some “philosophical circles” have drawn from the supposed indeterminism of modern
physics (Cassirer 2009a: 156).
15
H. G. Dosch, “Renormalized Quantum Field Theory and Cassirer’s Epistemological System,”
Philosophia Naturalis (28), 111.
16
M. Ferrari, “‘Wachstum’ oder ‘Revolution’? Ernst Cassirer und die Wissenschaftsgeschichte,”
Berichte der Wissenschaftsgeschichte 35 (2012), 113–130.
284 massimo ferrari
“scientific revolution” was valid solely with regard to Galilei’s foundation of
the mathematical science of nature. Nevertheless, according to a widespread
view in the history of philosophy, Cassirer belongs to the past and may be
considered – at most – as a prestigious forerunner to our present philoso-
phical understanding. The story of this highly disputable view regarding
Cassirer’s contribution to the philosophy of science in the twentieth century
is undoubtedly a very intricate one. But nobody will deny that the recon-
struction of the relationship between the philosophy of science and the
history of science is deeply connected to the Neo-Kantian tradition (Kuhn
1997: 11–12, 107–109, 147), for which its most eminent figure is surely the
author of Substance and Function: a philosopher until recent times neglected,
but nowadays not accidentally rediscovered.
chapter 14
Kant and the Neo-Kantians on mathematics

Luca Oliva
Introduction
An important feature of late nineteenth- and early twentieth-century
philosophy has been the emergence of mathematical questions.
This trend can be easily traced back to Kant’s theoretical
philosophy, especially his account of mathematical knowledge based
on intuitions. The approach of Neo-Kantians consists in offering
variations on the Kantian notion of intuitions. Roughly put, they
introduce intuitive elements into mathematical reasoning in order, for
instance, to make sense of numerical compositions. These intuitions
have recently been seen in a similar way, namely as carrying a logical
connotation.
In the First Critique, Kant argues that the mathematical concepts
are constructions pertaining to the synthesis of non-empirical
intuitions (A 716/B 744). These constructions are supposed to provide
evidence without involving experience; and such evidence is
supposed to derive from the exhibition of concepts in concreto (A
711/B 739). I believe that these suppositions are ultimately justified by
the nature of intuitions, which I see as consistent with the logical,
mathematical characterizations set forth by Hintikka and Parsons.
Building on these references, my chapter aims to show that Kant’s
intuitions behave like free variables which allow for abstract construc-
tions such as numbers and basic operations on numbers, like
addition.
I’ll support my thesis by examining the parts of the Transcendental
Doctrine of Method (A 709–738/B 738–766) which explicitly address the
representation of the universal in the individuality of intuitions; and the
parts of the Transcendental Aesthetic (A 22–36/B 37–53) along with those of
the Lectures on Metaphysics (28:506–561) which characterize these intuitions
as pointing at part–whole relation and magnitude. I’ll then derive a
285
286 luca oliva
consistent account of numbers which will show how intuitions evolve into
homogeneous quanta mereologically combined in proportions. Such an
evolution remains unconceivable unless we acknowledge the logical import
that intuitions bear from the start.
Further evidence for this claim can be found in the Neo-Kantian
account of numbers, which has thoroughly assumed the logical character
of Kant’s intuitions. On the one hand the Neo-Kantians do not recognize
intuitions as a faculty independent of the understanding, on the other
they introduce all the properties of spatiotemporal intuitions into their
mathematical constructions. I’ll prove this by carrying out a detailed
examination, previously unattempted, of Rickert’s account of logical
and mathematical objects. Although Cassirer’s and Natorp’s logicism was
ultimately rejected by Rickert, Neo-Kantians share Dedekind’s construc-
tionism, in which the theory of numbers replaces temporal intuitions with
relational properties. This variation also supports my thesis.1
Mathematical propositions
In his Transcendental Doctrine of Method, Kant characterizes mathematics
as construction in intuition. He argues that mathematical concepts can be
exhibited a priori by means of non-empirical intuitions (A 713/B 741). The
significance of this part can hardly be underestimated, for Kant is giving a
direct answer to the main question of his Critique, namely “How are
a priori synthetic judgments possible?” (B 19).
The argument derives from the analysis of geometrical demonstrations.
Kant notices that the truth-value of Euclid’s propositions runs from one
claim to another “through a chain of inferences guided throughout by
intuition” (A 716/B 744). Any passage is both synthetic and evident. What
really strikes Kant is that the evidence of those inferences does not come
from experience, as is supposed to be in the case of synthetic claims. Even
more surprising is that the synthesis carries on strict and not merely
comparative universality. It hardly bears repeating that for Kant a strictly
universal proposition states that Ps “cannot be otherwise” (A 1/B 3),
namely, it is true in all possible worlds. A comparatively universal proposi-
tion, by contrast, states that it has never been disproved before, as is
the case, for instance, with a valid generalization or with Hume’s
1
I would like to thank James Garson, Robert Hanna, and Paolo Parrini for all the corrections and
critical comments. Many suggestions have come from Michael Friedman, Charles Parsons, Lisa
Shabel, Daniel Sutherland, and Paul Teller. For the language I am grateful to Thomas Behr and
Sharon Joyce.
Kant and the Neo-Kantians on mathematics 287
custom-induced inference. Thus instances are used to provide synthetic
claims with empirical evidence. In this sense, Euclid’s demonstrations
genuinely challenge the classic distinction of analytic and synthetic that
Kant thoroughly takes for granted.
In all judgments in which the relation of a subject to the predicate is thought,
this relation is possible in two different ways. Either the predicate to the subject
A, as something which is (covertly) contained in this concept A; or outside the
concept A, although it does indeed stand in connection with it. In the one case I
entitle the judgment analytic, in the other synthetic. (A 6–7/B 10)
Using current terminology, all judgments are reducible to categorical
propositions of either A (All Ss are Ps) or E (No Ss are Ps) type. Venn
diagrams respectively show the relation of S-class and P-class in terms of
inclusion and exclusion. Hence, A and E type propositions, respectively,
correspond to analytic and synthetic ones; for if S is part of P their union is
equal to P (S⊆P=S∪P=P), whereas if the intersection of S and P is empty
(S∩P=Ø) their union is equal to their symmetric difference (S⊕P=(S∪P)\Ø)
and is therefore included in a third class (Q⊆(S∪P)). All propositions require
some evidence to establish their truth-values. Analytic propositions are true
by definition since the S-class is already part of the P-class, while synthetic
propositions need something else since their composition is claimed rather
than show. In this latter case, Kant appeals to intuitions.
Geometrical passages are hardly reducible to mere identities in terms of
analytic propositions. We “can analyze and clarify the concept of a figure
enclosed by three straight lines, and possessing three angles,” but we “can
never arrive at any property not already contained in these concepts”
(A 716/B 744). Consistently, mathematical propositions like those of
Euclid’s Elements are properly explained by synthesis rather than analysis.
The former is the only way to add a new property to a concept.
Hence, mathematical concepts are synthetic propositions which carry
non-empirical evidence. Kant’s account of concepts is not original, however.
Basically, it amounts to a variation of the classic bundle theory of Hume.
Accordingly, any concept is a collection of properties or, more precisely, a
subsumption of properties under a common class. Kant puts a strong
emphasis on the function of synthesis which is a priori assumed in any
concept formation; in this sense, concepts are a priori synthetic unity of some
manifold (A 78–79/B 104–105). In fact, “concepts rest on functions” which
are meant as “the unity of the act of bringing various representations under
one common representation” (A 68/B 93). The distinguishing factor of
mathematical concepts thus pertains to the manifold of representations
288 luca oliva
and not to their synthesis. The latter is a logical operation which mathe-
matical propositions share with all synthetic propositions whatsoever.
Kant has previously argued (A 19–21/B 33–35) that any representation of
objects is ultimately an immediate and individual intuition. Hence, to
“pass beyond [a concept] to properties which are not contained in this
concept, but yet belong to it . . . is impossible unless I determine my object
in accordance with the conditions either of empirical or of pure intuition”
(A 718/B 746). Therefore, the nature of intuitions eventually decides all
synthetic propositions, geometrical inferences included. In fact, a synthetic
claim about the world is composed of empirical representations, and its
truth-value is ultimately justified by experience, e.g. “the apple is green”
(Ga) is true or false as long as it can be observed. As true the relation
of S-class and P-class claims a composition which my perceptions
corroborate. If the claim is not about the experiential world, then other
evidence is required.
Hence, mathematical propositions are syntheses of non-empirical
intuitions. Kant does not say how they are supported by evidence, but he
gives precise indications regarding how they are to be intended. From the
latter we can easily derive the former. In what follows I’ll try to characterize
non-empirical intuitions accordingly.
Intuitions and mathematical concepts

Kant primarily addresses intuitions in the Transcendental Aesthetic. He
thinks of them as conditions for everything we may perceive or, more
generally, represent (A 19–20/B 33–34). Perceiving and representing over-
lap only empirically since representations can formally be viewed as
entirely pure, like in the case of the forms of intuitions (space and time).
In the Transcendental Doctrine of Method he further argues that non-
empirical intuitions are employed in mathematical constructions (A 713/
B 741). The two parts are clearly connected: the former accommodates a
metaphysical distinction of intuitions which fits into the mathematical
account of the latter. If intuitions were only empirical, mathematics would
be consistently abstracted from experience and thus derived via inductive
reasoning. Hence, it would be hardly necessary or universally valid.
Such abstractionism, strongly defended by Mill, has been unani-
mously rejected by the Neo-Kantians, especially by Cassirer. In his
Substance and Function (1910), he systemically criticizes “the process of
comparing things and of grouping them together according to similar
properties” since it is still “placed upon thing-concepts” rather than
“relation-concepts.” This sensualistic approach which also characterizes
Husserl’s Philosophy of Arithmetic (1891) ultimately fails to make sense of
mathematical constructions. In order to avoid that abstractionism, Kant
explicitly refers mathematical constructions to non-empirical intuitions.
He argues that constructing in intuition amounts to an exhibition of
concepts in concreto – something which involves non-empirical intuitions
in a significant way.
For the construction of a concept we therefore need a non-empirical intui-
tion. The latter must, as intuition, be a single object, and yet none the less, as
the construction of a concept (a universal representation), it must in its
representation express universal validity for all possible intuitions which fall
under the same concept. Thus I construct a triangle by representing the
object which corresponds to this concept . . . The single figure which
we draw is empirical, and yet it serves to express the concept, without
impairing its universality. (A 713–714/B 741–742)
Non-empirical intuitions behave like individual representations which
stand for all other representations; they combine individuality and uni-
versality by instantiating the latter under the former. In his First Critique,
Kant has thoroughly argued for that, as at the beginning of the Analytic of
Principles when he says, “transcendental philosophy has the peculiarity that
besides the rule . . ., which is given in the pure concept of understanding, it
can also specify a priori the instance to which the rule is to be applied”
(A 135/B 174). Non-empirical intuitions exhibit a priori the concept in a
concrete instance. As Kant says, “in mathematics . . . the concepts of reason
must be forthwith exhibited in concreto in pure intuition” (A 711/B 739),
therefore “to construct a concept means to exhibit a priori the intuition
which corresponds to it” (A 713/B 741).
The instantiation of universality sharply distinguishes the mathematical
concepts from the philosophical ones, and shows them to be an alternative way
of conceptualizing. As Kant puts it, “philosophical knowledge considers the
particular only in the universal, mathematical knowledge the universal in the
particular, or even in the single instance, though still always a priori and by
means of reason” (A 714/B 742). That corresponds to, “philosophical
knowledge . . . has always to consider the universal in abstracto (by means of
concepts), mathematics can consider the universal in concreto (in the single
intuition) and yet at the same time through pure a priori representation”
(A 734–735/B 762–763). Non-empirical intuitions clearly realize the idea that a
single object or individuality may stand for a manifold of objects or univers-
ality, which is exactly the idea of a free variable (x, y, z), for instance in first-
order logic. Thus, an intuition is intended “as containing an infinite number of
290 luca oliva
representations within itself,” while a concept is thought of “as a
representation which is contained in an infinite number of different possible
representations (as their common character), and which therefore contains
these under itself” (A 25/B 40).
Hintikka notices this perfectly by noting that Kant’s mathematical
method is “based on the use of constructions” which consists in “introdu-
cing particular representatives of general concepts and carrying out
arguments in terms of such particular representatives, arguments which
cannot be carried out by means of general concepts.”2 Similarly, Parsons
argues that “the reasoning involving constructive operations” is carried on
“as reasoning with singular terms,” and “Kant clearly understood this
reasoning as involving singular representations.”3 “Free variables, and
terms containing them, have the property that Kant requires of an intui-
tion constructing a concept, in that they are singular and yet also ‘express
universal validity’ in the role they play in arguing for general conclusions.”
The same position is held by Brittan,4 Friedman,5 and Thompson.6
If we assume that a concept is a collection of properties (individual
representations) and each of them is variable or constant (non-empirical or
empirical intuition, respectively), then Kant is saying that any philosophi-
cal concept unifies individual constants which are derived a posteriori,
whereas any mathematical concept unifies individual variables (like a set
containing sets) which are available a priori.
The logical character of intuitions also makes sense of the semantic
problem, namely “concepts which relate a priori to objects” (A 57/B 81).
Since any statement containing individual constants, such as “Fa,” allows
for an existential generalization such as “(∃x)Fx,” for any object of experi-
ence there is something a priori which allows for its conceptualization.
Otherwise put, in order to represent something as X, the representation of
2
J. Hintikka, “Kant on the Mathematical Method,” The Monist 51/3 (1976), 352–375.
3
C. Parsons, “The Transcendental Aesthetic,” in P. Guyer (ed.), The Cambridge Companion to Kant
(Cambridge University Press, 1992), 62–100; here 78. Parsons also points out that the algebraist’s
“manipulating symbols according to certain rules [requires] analogous intuitive representation of his
concept,” and that “the symbolic construction is essentially a construction with symbols as objects of
intuition.” See C. Parsons, “Kant’s Philosophy of Arithmetic,” in C. J. Posy (ed.), Kant’s Philosophy of
Mathematics (Dordrecht: Kluwer Academic Publishers, 1992), 43–79; here 65.
4
Already Brittan noticed the difference between intuitions seen as variable or as constant; however, he
saw them as two ways of constructing in intuitions rather than as pure and empirical intuitions, as I
myself do. See G. G. Brittan, “Algebra and Intuition,” in C. J. Posy (ed.), Kant’s Philosophy of
Mathematics (Dordrecht: Kluwer Academic Publishers, 1992), 315–339.
5
M. Friedman, “Kant on Concepts and Intuitions in the Mathematical Sciences,” Synthese 84 (1990),
213–257.
6
M. Thompson, “Singular Terms and Intuitions in Kant’s Epistemology,” in C. J. Posy (ed.), Kant’s
Philosophy of Mathematics (Dordrecht: Kluwer Academic Publishers, 1992), 81–107.
X must be presupposed (A 250); as Rickert suggests, anything we perceive
implies something logical as a priori structure. Hence, the logical-
transcendental conditions concerning the possibility of objects of
experience ultimately rest on such non-empirical intuitions which behave
like free variables. Along with the logical-analytical conditions (concerning
the non-contradictoriness of concepts) they make definite sense of Kant’s
epistemological paradigm as discussed by Parrini.7
Quantifying over intuitions

The construction in intuition discussed above gets further specified as Kant
distinguishes between ostensive and symbolic constructions. Respectively,
they represent the geometrical and the mathematical way to employ non-
empirical intuitions in a logical fashion; namely as individuals which stand
for universals (and not universals which subsume individuals). Both ways
significantly involve quantification.
Accordingly, construction in intuitions evolves into construction of
magnitudes. This looks quite consistent with the nature of such intuitions.
In modern logic binding a variable which ranges over a domain of dis-
course is called quantification. Free variables are intended to be quantified.
Consequently, non-empirical intuitions are named as quanta, namely as
quantified properties.
But mathematics does not only construct magnitudes (quanta) as in geo-
metry; it also constructs magnitude as such (quantitas), as in algebra. In this
it abstracts completely from the properties of the object that is to be thought
in terms of such a concept of magnitude. (A 717/B 745)
Quanta represent the building-blocks of ostensive and symbolic con-
structions, which are therefore differently quantified. Kant proceeds from a
less abstract to a more abstract idea of quantification:8 “quanta . . . can be
7
P. Parrini, “Kant’s Theory of Knowledge: Truth, Form, Matter,” in P. Parrini (ed.), Kant and
Contemporary Epistemology (Dordrecht, Boston, and London: Kluwer Academic Publishers, 1994),
195–230.
8
Brittan bestows priority on the arithmetical way of quantifying, while Shabel maintains that
symbolic constructions are ultimately species of the ostensive ones: “such as the variable ‘x’ can be
used to represent concretely constructible entity, such as a line segment . . ., the variable symbolically
constructs the concrete object by symbolizing the ostensive construction of that object” (L. Shabel,
“Kant’s Philosophy of Mathematics,” in P. Guyer (ed.), The Cambridge Campanion to Kant, 2nd
edn., Cambridge University Press, 2006, 101). She correctly offers a genealogical suggestion, for the
argument derives from examining Euclid. However, logically speaking if x stands for everything,
including y which stands for geometrical figures, then y may replace x iff x is such a figure: if (y ⊃ x),
then (y → x) but not (x → y).
292 luca oliva
exhibited a priori in intuition, that is, constructed, either in respect of the
quality (figure) of the quanta, or through number in their quantity only
(the mere synthesis of the homogeneous manifold)” (A 720/B 748). And he
finally attributes logical homogeneity to the latter in order to accommodate
mereological combinations.
In the metaphysical lectures on Leibniz’s principle of indiscernibles
(Kant 2001: 29:839), that homogeneity emerges from the lack of qualitative
(or specific) difference; this latter connotes a compositum with heterogene-
ity instead.9 The notion of quantity is primarily, although not exclusively,
characterized by that lack. In fact, it also requires a combination of parts,
which accordingly are viewed as logically homogeneous. Thus quantity has
everything to do with the homogeneity of parts and the way in which they
are combined together (zusammengesetzt), namely synthesized. Kant’s
leading idea is that combining parts (the manifold) which are homoge-
neous leads to magnitudes.
Like Euclid, Kant holds that magnitudes are a combination of parts that
are homogeneous. Such combining is to be understood in terms of propor-
tions (ratios). In fact, it is not enough just to belong to the same genus or
kind, as the word ‘homo-geneous’ suggests; the parts must also stay in
proportion, namely they must be either bigger or smaller or equal.
Any proportion entails homogeneity, but homogeneity does not entail
proportion. In order to be in proportion (bigger, smaller, equal) two parts
must be conceived as inside of one another, that is, as part and whole. Kant
makes sense of it by saying, “A > than B if a part of A=B; in contrast A < B,
if A is equal to a part of B” (28:506, late 1780s) or “something is larger than
the other if the latter is only equal to a part of the former” (28:561).
This explains quantity in terms of both part-whole relations and homo-
geneity in a way which is fully consistent with the above definition of
“synthesis of the homogeneous manifold.” Not just an homogeneity with
respect to a more comprehensive concept, but a strict homogeneity, namely
that which is restricted to proportions or ratios. As Sutherland points out,
“homogeneous magnitudes stand in comparative size relations,”10 because
“one magnitude will be larger than another as long as a part of it is equal in
9
Quantum and compositum are magnitudes containing a plurality, but only the former accepts
homogeneous parts. The latter allows for “an aggregate of heterogeneous parts,” while the former
accepts only homogeneous parts: “[a] composite differs from quantum, and the many would in that
case be able to be a variety, every quantum contains a multitude but not every multitude is a
quantum; rather [it is one] only when the parts are homogeneous” (Kant 1997: 29:991).
10
D. Sutherland, “Kant on Arithmetic, Algebra, and the Theory of Proportions,” Journal of the History
of Philosophy 44/4 (2006), 533–558; here 538.
size to the other”; something “which in principle allows the ordering of any
two magnitudes of the same kind according to their size,” and this by
means of their proportions (in symbols: >, <, =) which behave like “invar-
iants under all equimultiple compositions.”11
In the Aesthetic, the pure representations have been already characterized
as intuitions of single wholes, and these wholes as purely representing
infinite magnitudes (A 24–25/B 39–40 and A 31–32/B 47–48). For we
represent things as placed in different spaces and as related simultaneously
or successively; things can only be thought in intuitions. Hence, intuitions
have something to do with places, and when the same places do not contain
anything (existential or empirical) they are still something, namely (ideal or
pure) placeholders. It is quite clear that non-empirical intuitions represent
wholes containing parts, and that each of these parts can be replaced by
anything empirical since its representation is intended as free variable.
Numbers
Kant primarily discusses numbers in the Schematism of his First Critique
and in his Lectures on Metaphysics L2. In what follows I’ll try to make sense
of their main thesis, namely that numbers are homogeneous parts com-
bined in succession. To accomplish that I’ll explain two intertwined
notions, the schema of magnitude and the addition of discrete quanta.
Both rely on the logical nature of non-empirical intuitions.
“The subsumption of [objective] intuitions under pure concepts” (A 138/
B 177) follows certain rules, which are called by Kant schemata. He restricts
such intuitions to time only, and he holds that “the schemata are . . . a priori
determinations of time in accordance with rules” (A 145/B 187) – a quite
misleading move, unless it is properly recognized that spatial intuitions are
thoroughly presupposed in any schematization. Despite that, Kant is quite
consistent in deriving temporal rules for each group of categories.
What is a schema, exactly? Kant contrasts schemata with images. An
image instantiates concepts, e.g., “if five points be set alongside one another,
thus . . . . . have an image of the number five”; whereas a schema generalizes
concepts, e.g., “a number in general, whether it be five or a hundred, . . . is
rather the representation of a method whereby a multiplicity . . . may be
represented in an image in conformity with a certain concept, than the image
itself” (A 140/B 179). A schema represents a higher order of universality
11
D. Sutherland, “Kant’s Philosophy of Mathematics and the Greek Mathematical Traditions,” The
Philosophical Review 113/2 (2004), 157–201; here 181, 171.
294 luca oliva
in contrast to (and in connection with) “some specific universal concept”
(A 141/B 180). Kant’s arguments implicitly reply to Locke’s fictionalism
regarding the nature of general ideas.12
Indeed it is schemata, not images of objects, which underlie our pure sensible
concepts. No image could ever be adequate to the concept of a triangle in
general. It would never attain that universality of the concept which renders it
valid of all triangles, whether right-angled, obtuse-angled, or acute-angled; it
would always be limited to a part only of this sphere. (A 141/B 180)
If it is not an image, what is a number in general? Although he definitely
restricts his analysis to natural numbers, Kant is looking for a higher order
concept whose universality ranges over all numbers – something like the
concept of triangle, whose validity regards all triangles.
. . . the pure schema of magnitude (quantitatis), as a concept of the under-
standing, is number, a representation which comprises the successive addition
of homogeneous units. Number is therefore simply the unity of the synthesis
of the manifold of a homogeneous intuition in general, a unity due to my
generating time itself in the apprehension of the intuition. (A 142–143/B 182)
My best guess is that the concept of number is precisely a concept that
does not collect things but intuitions which stand for things (namely pure
intuitions), and that collects them increasingly (namely in addition) by
means of some quantification. If I’m right, then numerical collections
(numbers) are additions of quantified non-empirical intuitions. Kant’s
philosophy of mathematics stands or falls on this.
As Rickert correctly points out, additions are not simple conjunctions.
While an apple added to another makes two apples, a number added to
another does not make two but one, increased number. However, succession
alone is not enough to make sense of addition either. An event may follow
another without ending up in any union, whereas any number is comprised
in the next higher one. Thinking of pure intuitions as placeholders may help
to understand their addition. If pure intuitions behave like free variables
which stand for any constant, additions may regard logical places instead of
empirical things; and the succession in time may be only my way to represent
these places. If that is true, then addition may be something which turns the
spatial intuitions which temporally follow one another “{Ø}, {Ø}, {Ø}” into
their unification in extension “{{{Ø}}}”; however, the number three follows
the number one and two because it mereologically includes them and not
12
E. Carson, “Locke and Kant on Mathematical Knowledge,” in E. Carson and R. Huber (eds.),
Intuition and the Axiomatic Method (Dordrecht: Kluwer Academic Publishers, 2006), 3–20.
simply because it comes after them in conjunction. I believe that Kant shares
Benacerraf’s Ernie theorem,13 “for any two numbers, x and y, x is less than y if
and only if x belongs to y and x is a proper subset of y” vs Johnny’s “given two
numbers, x and y, x belongs to y if and only if y is the successor of x” (54); and
that he could number three in set-theoretic way: 0=Ø, 1={0}={Ø}, 2={0, 1}=
{Ø,{Ø}}, 3 ={0, 1, 2}={Ø,{Ø},{Ø,{Ø}}}. For a subset can easily be seen as a
part, as indeed Lewis sees it.14 This certainly deserves a further analysis,
involving Frege and Russell along with the relations between set-theory and
mereology.
I believe that Kant’s mereological characterization of numbers suggests
this conclusion, which also makes sense of his above definition. For what
characterizes a number is that it comprises the homogeneous units by
means of successive addition. The central point seems to me precisely the
addition, which Kant clearly intends in mereological terms. The addition
of units is carried out by means of comprising parts, and the succession (the
time-series, A 145/B 184) merely derives from such comprising parts which
enlarges wholes. I also believe that Kant is closer to Cantor and Frege than
we used to think.15
My evidence relies on Metaphysics L2 (1790–1791: 28:560–561), where
Kant discusses discrete quanta (another name for number). He holds that
“each quantum is as a multitude consisting of homogeneous parts” and
that, as such, “each quantum can be increased or decreased.” This goes
through combining its parts, “the parts that, connected with each other,
make a number concept.” In this mereological connection “something is
larger than the other if the latter is only equal to a part of the former” in
fact, “for something to alter into a larger is to increase, and for something
to alter into a smaller is to decrease.” As Kant says:
A quantum continuous in itself is one in which the number of parts is
indeterminate; a quantum discrete in itself is one in which the number
of parts is arbitrarily determined by us. Discrete quantum is therefore
called number. Through number we represent each quantum as
discrete. (28:561, 1790–1791)
The former represents an undetermined multitude (or manifold) of parts
because “it does not consist of individual parts.” While the latter is made of
13
P. Benacerraf, “What Numbers Could Not Be” (1965), in P. Benacerraf and H. Putnam (eds.),
Philosophy of Mathematics (Cambridge University Press, 1983), 272–294.
14
D. K. Lewis, Parts of Classes (Oxford: Blackwell, 1991).
15
See C. Parsons, “Arithmetic and the Categories,” in C. J. Posy (ed.), Kant’s Philosophy of Mathematics
(Dordrecht: Kluwer Academic Publishers, 1992), 135–158.
296 luca oliva
partitions, namely “assignable parts.” It is noteworthy that space and time
are here defined as “continuous quanta” because they contain all assignable
parts (as undetermined, though), behaving like a common field in which
quanta get differentiated. This field seems to be already predisposed to
being quantified by arbitrary partitions. The difference between discrete
and continuous quanta rests on the possibility of aggregating parts into
wholes. A number is clearly identified with the quantum whose assignable
parts are not only homogeneous and individual, but also increased and
decreased via aggregation.16
This latter says that a number is ultimately an extensive magnitude.
Kant calls extensive those magnitudes which are composed by homoge-
neous parts we can distinguish, because they are assignable.
All magnitudes can be considered in two ways: either extensively or inten-
sively. There are objects in which we do distinguish no multitude of
homogeneous parts; this is intensive magnitude . . . The objects in which
we distinguish a multitude of homogeneous parts have extensive magnitude.
The intensive magnitude is the magnitude of the degree, and the extensive
magnitude is the magnitude of the aggregate. (28:562)
Differently put, “A magnitude [is] extensive when the representation
of the parts makes possible, and therefore necessarily precedes, the
representation of the whole” (A 162/B 203). On the contrary, when the
composition (Zusammensetzung) of parts is a coalition (Koalition) rather
than an aggregate (Aggregation), the magnitude which derives from the
mathematical synthesis of the homogeneous is intensive (B 202). For it
does not allow for any mereological distinction.17 While in order to be an
aggregate of parts a magnitude must be apprehended in an extensive
fashion, namely “through successive synthesis of part to part.” This
is possible only if all intuitions are from the very beginning conceived
in terms of extensive magnitudes (B 202), namely “as aggregates, as
complexes of previously given parts” (A 163/B 204).
16
Schultz says: “1. From several given homogeneous quanta, to generate the concept of one quantum
by their successive connection, i.e. to transform them into one whole. 2. To increase and to diminish any
given quantum by as much as one wants, that is, to infinity.” J. Schultz, Prüfung der kantischen Kritik
der reinen Vernunft (Königsberg: Hartung, 1789), vol. i, 221. It is noteworthy that Rickert shares the
purely logical argument of Leibniz which grounds arithmetical identity on the associativity and
commutativity of addition. The same argument is employed by Kant’s pupil Schultz about
progressing from 7 to 12 by successive additions of 1.
17
Also a continuous quantum can be increased, and that not by means of a mereological addition of
parts rather by means of altering the value of its degree; since an intensive magnitude represents a
degree of intuitions (A 166–176/B 207–218).
Rickert on logical identity
We have seen above how homogeneity and the part–whole relation pertain
to Kant’s notion of numbers in a crucial way. In his The One, the Unity and
the Number One (1924a) Rickert offers an interesting variation of that
notion by diversifying the logical character of the intuitions. Roughly
put, homogeneous parts are intended to represent the logical core of
numbers, while their mereological compositions are developed in mathe-
matical series. The former are ultimately based upon spatial places, the
latter upon temporal series. In what follows, I’ll present the arguments in
support of these conclusions, starting with homogeneity.
Rickert begins with a metaphysical premise. Logic and mathematics share
the same kind of objects, namely the unreal. They both dismiss psychophy-
sical reality, holding that “a number or a line are not real as the paper on
which they are drawn or the mental act by which they are perceived” (Rickert
1924a: 3). Yet their similarity ends here. Even if their areas mainly overlap, it
may be possible to identify parts of them which are not in common. If it is
feasible to do, these parts along with their characteristics could represent
“that which is mathematical” (das Mathematische) on the one hand, and
“that which is truly logical” (das reine Logische) on the other.
What is logical? Rickert’s first concern is to dismiss any psychological
answer in the following way. Anything we may think is an object of thought,
and any object of thought is logically something. There is no such thing as
thinking in general, which may be considered as empty and independent of
any object. We cannot think anything without thinking of something at the
same time. However, whereas thinking is an act of my thought, what is
thought by such an act looks independent of it. Rickert holds that we can
sort out act and object from anything we think, and then we can separately
question the object of thought without involving the act through which we
think of it. Therefore, the act of thinking and the something which is
thought by means of that act can be treated separately.
Rickert accomplishes this goal by carrying out a semantic examination of
objects. According to it, any object of thought gets labeled as ‘something’
and, therefore, gets referred to as a composition of “form and content.” On
the one hand, the word ‘something’ behaves like a rigid designator à la
Kripke so to speak, for it always means that composition. On the other, it can
designate anything specific as well, hence it turns out to be non-rigid. Finally,
he focuses on several definitions of “something.”
(a) In so far as there is a content (whatsoever), there is also a “unifying
form of that content.” What really matters for Rickert is that “form
298 luca oliva
and content” represent irreducible elements which compose any
“something” as its parts or moments. As separated, each of them
was not yet something; for in that case, it would rather become itself
an object of thought, and as such it would be “something” and thus
requiring “form and content” over again. The form itself or the
content itself cannot be thought as more elementary objects because
any object of thought is a composition of them.
(b) In other terms, form and content represent two logical moments
which are sharply different, like “the one and the other” (das Eine und
das Andere). He notices that paradoxically the identity of something is
connoted by the alterity (otherness) of such moments. Since each of
them always remains in relation to the other they form a relation of
relatives (relatio relatorum). Therefore the logical object results in
nothing simple but in a plurality of moments.
The same point is later clarified in Die Logik des Prädikats und das
Problem der Ontologie (Rickert 1930), where Rickert defends the idea
that the identity of something is ultimately the unity of “the one and
the other,” for they stand for the relation (Verbindung) of subject and
predicate, in which alone “the logical meaning is found out in a
proposition which relates a grammatical subject with a grammatical
predicate, or in short: all true and reasonable propositions are linguistic
predications” (Rickert 1930: 30). In this sense, any proposition is the
logical, linguistic synthesis of the one and the other. A conclusion
otherwise reached in Grundprobleme der Philosophie (1934), as Rickert
says: “the empty form of the subject synthesizes objects without being
in turn objectified” for its identity consists in the logical function of
objective unification (Rickert 1934: 115–119).
It follows that the ultimate thing of thought or “the most elementary
object we may think of” is, logically speaking, a manifold, i.e., not one but
two logical places or moments, which are called “pre-objects or pre-
objective elements of any object.” They represent two constitutive parts
of any object of thought, namely of any something in general.
This latter is also addressed as the content which is united by some
form. What matters in this case is that they both are strictly formal,
namely logical. As the logical content is filled out by anything alogical,
the object in general turns particular or empirical. Accordingly, Rickert
distinguishes the formal content (Gehalt) from the material content
(Inhalt or Beschaffenheit) – the former being constitutive therefore neces-
sary, the latter being only contingent. In this sense, it is always the
content that specifies the kind of object, for the material content that
fills the (formal) content is always alogical. It can only be perceived and
never thought.
The form alone or the content alone do represent objects which may be
thought logically. This is correct and must be highlighted. However, it is
misleading to believe that they at the same time represent something
logically more elementary or that they represent that which is logically
ultimate and simplest. On the contrary, that which is theoretically ultimate
and simplest has been already obtained by means of the concept of formed
content or unity of form and content. The ultimate thing which can be
thought as object is never a single thing; such ultimate thing does not
designate a singularity but as soon as it is conceptually thought it designates
a manifold of elements, which as numbered designates a plurality. (Rickert
1924a: 14)
(c) In Rickert’s view, form and content replace subject and predicate. He
explicitly argues for that in Die Logik des Prädikats (1930). In his main
work Der Gegenstand der Erkenntnis (1928), he shows how to justify
this switch. The argument runs as follows. As we say, for instance,
that “the leaf is real” we apply the form of reality (Wirklichkeit) to a
content that is accordingly real (wirklich); however, this is not all. We
also assume that there is a form in general which is empty and which
can be filled with a content; the “leaf” is a real object, that is some-
thing that bears “the immanent, real object that is filled in the
content” by means of the form of reality. It suggests that form and
content of something ultimately mean its subject–predicate relation,
and that both are a priori and formal components of any object that
we can address as something.
(d) Another name for “something in general” is “one and the same.”
With this definition Rickert wants to stress that the form of one is
what endows the content with unity, and provides something with
identity. Such identity connotes a formal property of any object.
Rickert expresses this idea in different formulas like “the content in
general as unified by the form of identity,” or “any content is
identified by the form of unity” (Rickert 1924a: 45).
He also introduces a subtle distinction between “what” (whatsoever)
and “something,” identifying the former with the quality (materiality,
Beschaffenheit) of the latter; the quality of something is determined by its
content. In this sense, form and content are for Rickert replaceable with
identity (quantity) and alterity (quality). And from that he criticizes
Hegel’s dialectical thinking for having based the moment of the anti-thesis
upon negation. In Rickert’s eyes it should be more adequately conceived in
300 luca oliva
terms of hetero-thesis based upon alterity, not just a formal negation but a
completion. The negation follows alterity, which is not the mere negation
of identity but a positive, primitive moment that cannot be simply
deduced from the other. Consequently, he replaces tautology with
heterology.
In conclusion, Rickert’s notion of logical objects describes a triadic
construction made of two moments and their union, namely “the one(1)
and(3) the other(2).” It reproduces the formal synthesis of subject and
predicate, which defines the identity of any type of objects, numbers
included.
What numbers could not be

Rickert defends the following thesis. A number is ultimately a structure
which is not understandable in merely logical terms, although it is math-
ematically elementary. This is due to the fact that its concept includes
alogical components which prevent it from being purely logical.
Consistently, any logical foundation of mathematics is rejected as far as
logic is conceived as restricted to the above unity of S-term and P-term
(form and content).
In order to support his conclusions, Rickert introduces a few arguments
regarding the non-commutativity of subject and predicate, the irreduci-
bility of identity to equality, and the nature of the numerical series. Let’s
call this part “what numbers could not be” and let’s see how it leads to
Rickert’s account of numbers.
(a) “Subject and predicate” are not two numbers yet, and their unity does
not mean the number two. Although their logical elements numeri-
cally designate a plurality, the semantic reference does not mirror the
designator. In fact, counting those elements presupposes rather than
derives numbers; it looks like numbering anything else. However,
let’s suppose the opposite to derive a contradiction like in an indirect
proof. Thus, if “subject and predicate” meant the same as “1+1” and
their “unity” the same as “2,” then the commutative law should rule
over them like it does over any numerical addition. However, while
the two numbers can be switched without any loss, the same cannot
happen with that logical unity.
A clarifying example can be drawn from predicative set-theory. The
unity subject–predicate specifies the former by means of the latter; it can be
read as x possessing the property P, and upon the propositional function
P(x) a set can be built, S = {x : P(x)}. Rickert argues that x and P (respectively
subject and predicate) do not merely fill the logical places of the set, places
which were still remaining after x and P were removed. On the contrary,
they are those places, namely specific building-blocks of any set. As
placeholders in any set-building form, they do not have anything in
common. Consequently, there is no possibility of switching them.
(b) “Subject and predicate” are not equal. They identify something by
means of their unity but do not allow for any equality among them.
Otherwise put, any object is characterized as “one and the same” by
means of the logical relation of its S-term with its P-term. Far from
being equal they are irreducibly different. This is another reason why
we cannot count them as numbers; in fact, there is no homogeneity
between them which may lead to an addition. Rickert holds that
logical identities exclude any duplicate in such a way that we cannot
add anything identical to anything.
He draws upon medieval theories sharing the same argument, like those
of Meister Eckhardt, who states Gleichheit steht in Unterschied (equality
rests on difference), or Thomas Aquinas’ formula aequalitas diversorum est
(equality pertains to differences). Accordingly, he draws the conclusion
that there is no adequate semantic expression for identity. Even the classic
formula “A=A” is misleading because it shows two “As” instead of one (and
the same). As Friedman nicely puts it, “Rickert still identifies formal logic
with the traditional subject-predicate logic, which is indeed confined to
relations of genus and species and thus to the purely symmetrical relations
of identity and difference.”18
(c) Finally, the arrangement of “subject and predicate” does not display
any mathematical series. Their logical unity lacks any temporal
sequence which may lead to a numeration. This is mainly due to
the different meaning which is borne by the mathematical sign for
addition (plus, +) and the logical connective for union (and, •).
Rickert says that the logical unity connects its elements without any
fusion, for the logical elements unified remain separate. It is challen-
ging to express this in current notation.
In my view, Rickert’s aim is just to exclude any possibility of deriving a
series from the disposition of their conjunction: for a series has to do with
homogeneous parts, namely a repetition in time and space where a succes-
sion can literally take place. In this sense it seems quite consistent to stress
the role of mediation played by their union. S-term and P-term are logical
18
M. Friedman, A Parting of the Ways: Carnap, Cassirer, and Heidegger (Chicago: Open Court, 2000).
302 luca oliva
places and their union, the logical conjunction, however intended (and, •),
behaves like a logical medium between heterogeneous parts.
It is noteworthy that switching from “that which is logical” to “that
which is mathematical” has significantly to do with the introduction of
time in the logical relation of places. It suggests that Rickert sharply
differentiates Kant’s notion of spatiotemporal intuition in logical, spatial
placeholders on the one hand, and their alogical, temporal sequences on
the other. Rickert clearly attributes the latter to the subject, “We need to
free ourselves from the very idea of a supposed creation of that which is
‘logical’ by means of the subject, even if such a subject is taken as merely
formal” (Rickert 1924a: 57). Then he reaffirms the main theme of his
philosophy19 as he had already done in Zwei Wege der Erkenntnistheorie
(1909). The thinking subject can only recognize such a logical object as
something which is objectively valid (by itself). Hence, the logical meaning
of everything (numbers included) rests in a sort of super-individual,
independent sphere, like a theoretical value which transcends every cogniz-
ing subject which is ultimately compelled to recognize it as that which
cannot be otherwise. Rickert’s philosophy of values is all about this formal
obligation of validity, which in the case of numbers turns out to be logical.
In a nice little book on the concept-formation of the sciences, titled Zur
Lehre der Definition (1888), Rickert accordingly clarifies the metaphysical
nature of mathematical concepts:
the mathematical concepts are not referred to the sensible, real objects of the
natural sciences, from whose vast multiplicity clear properties [Merkmale]
are first sorted out as essential; but they are valid by virtue of their ideal
being, due to which everything is equally essential, namely the difference
between essential and non-essential just drops. (Rickert 1888: 43)
If logical objects are not yet numbers, then what are numbers?
The numerical series

From what numbers cannot be Rickert derives what they are. The logical
core of numbers is provided by the subject–predicate relation, whose unity
represents the homogeneous part of each numerical whole. What remains
19
Mathematics works with objects which are quantitatively determined. There are three ontological
spheres: logic (that which is valid, or has validity, or simply is without existing), mathematics (ideal
existing), and reality (real existing). The validity of logical objects (objects which only own a general
content, not a particular or special content filling that general content) disregards any quantification
along with all existence. It represents a duty or a norm for the thinking subject who does not have
any other choice than recognizing its validity.
to be discussed are the alogical components of the numbers. They even-
tually coincide with the thought itself, namely the real, psychical process
(act) that takes place in time involving series and quantity. I’ll first survey
Rickert’s argument along with its source (Dedekind), and then I’ll com-
pare both of them with the Neo-Kantians of Marburg.
The alogical components by themselves cannot be numbers, for instance
the acts of thought which count objects in time are not numbers. But they
say something crucial about them. Thinking of the same object several
times does not result in different objects but in different acts of thinking
one and the same object. “The identity of an object can be placed many
times, although to each of these times does not correspond a different
identity,” for there were no two identities but “just two times the same
identity”; more precisely, “the same union of form and content (subject–
predicate relation) is repeated many times” (Rickert 1924a: 56). This
suggests to Rickert that numbers have to do with psychical (empirical)
acts which are repeated in time; by means of them the same object is placed
over and over again. However, in order to accomplish such repetition a
temporal series is needed, for the placement is a psychical reality which
occurs in a time series (Reihe).
In such series we find the same object in different places, something
which gathers together that identity (Ps) and diversity (different places on a
line) required for the mathematical equality. Thus, switching one logical
object with another becomes possible, for the series works like an homo-
geneous medium where one and the same object can be differently placed
somewhere else; and in this sense it can finally be seen as equal. In
comparison, the logical medium was pretty much heterogeneous.
Although that explains equality and commutativity, it does not yet
account for numerical addition. In fact, the series as such (the homoge-
neous medium) just represents a bunch of places filled by one and the same
object. Rickert notices that and introduces quantity as the last alogical
component. He precisely introduces quantity into the formal content of
something in general as a sort of special, qualitative content. In this sense, a
number is something in general (a logical object) which is qualified by a
certain quantity and thus assigned to (placed in) a precise position of any
series whatsoever. In other terms, it is a quantum. The quantification turns
a bunch of places into an ordered series which by virtue of that looks like an
arrow pointing at a progression.
Zijderveld captures Rickert’s idea as follows: “As abstract, pure, ideal
and non-sensual as numerals are, they are nevertheless the substantial,
quantitative objects of mathematics” since “after all numerals do
304 luca oliva
exist.”20 However, I think that Rickert’s reasoning is missing the point.
Introducing quantity in each place of the series is ineffective unless
those places get somehow aggregated (like parts) in a proportional way.
As far as Rickert keeps quantitative sizes as components of numbers, he
could not dismiss part–whole relations. The entire argument asks for a
mereological foundation of numbers in a Kantian sense.
About that, it is interesting to notice how Rickert diverges from Cassirer
after their initial, common endorsement of Dedekind’s theory of num-
bers.21 For Dedekind the whole of arithmetic follows from the act of
counting which bears out “the successive creation of the infinite series of
positive integers in which each individual is defined by the one immedi-
ately preceding.” If a and b represent one and the same rational number,
then a=b as well as b=a. If not, then their difference a−b has either a
positive or negative value. “In the former case a is said to be greater than b,
b less than a,” in symbols as a > b, b < a; as “in the latter case b−a has a
positive value it follows that b > a, a < b” (Dedekind 1901: 4–5). The three
following laws ground Rickert’s notion of numerical series and the entire
mathematical account of the Marburg Neo-Kantians.
i. If a>b, and b>c, then a>c. Whenever a, c are two different (or unequal)
numbers, and b is greater than the one and less than the other, [it geomet-
rically means] b lies between the two numbers a, c. [That is] If p lies to the
right of q, and q to the right of r, then p lies to the right of r, . . . q lies
between the points p and r. ii. If a, c are two different numbers, there are
infinitely many different numbers lying between a, c. [That is] If p, r are two
different points, then there always exist infinitely many points that lie
between p and r. iii. If a is any definite number, then all numbers of the
system R fall into two classes, A1 and A2, each of which contains infinitely
many individuals; . . . A1 comprises all numbers a1 that are <a, . . . A2
comprises all numbers a2 that are >a . . . every number of . . . A1 is less
than every number of . . . A2. [That is] If p is a definite point in L, then all
points in L fall into two classes, P1, P2, each of which contains infinitely
many individuals; . . . P1 contains all the points p1, that lie to the left of p,
and . . . P2 contains all the points p2 that lie to the right of p; the point p itself
may be assigned at pleasure to the first or second class. (Dedekind 1901: 6–7)
Dedekind shows that the “properties of rational numbers recall the
corresponding relations of position of the points of a straight line L”
20
A. Zijderveld, Rickert’s Relevance: The Ontological Nature and Epistemological Functions of Values
(Leiden: Brill, 2006), 156.
21
R. Dedekind, “Continuity and Irrational Numbers,” in Essays on the Theory of Numbers (Chicago:
Open Court, 1901), 1–27.
(Dedekind 1901: 6), where the arithmetic difference assigns each number to
a different place along the series. According to Dedekind’s relational
structure, “numbers are simply places within such a series or progression,”
and their concept is properly defined as relational for it is “entirely
exhausted by the formal properties of a particular kind of relational
structure.”22
This is precisely the notion of numerical series loosely discussed by
Rickert and endorsed by Cassirer. In his Substance and Function (1923)
Cassirer argues that natural numbers are entirely given by their position in
the progression of ordinal numbers. They behave like objects whose system
is ruled by a logic of relations (the concepts of function), objects that are
thus “resolved into a web of relations” (Cassirer 1927: 31–92). Given a
certain law of progression, “to every member there belongs an immediate
successor with which it is connected by an unambiguous transitive and
asymmetrical relation.” Since mathematics “is a system of ideal objects
whose whole content is exhausted in their mutual relations, . . . the
‘essence’ of the numbers is completely expressed in their positions,” and
they solely derive “from purely logical premises” (Cassirer 1923: 38–39).
“When,” says Dedekind in definition, “in the consideration of a simple
infinite system N, arranged by the ‘copying’ (Abbildung) Φ, we totally
abstract from the particular properties of the elements, retain merely their
distinctness, and attend only to the relations in which they are placed to
each other by the ordering ‘copying’ Φ, then these elements are called the
natural numbers or the ordinal numbers or also simply numbers, and the
fundamental element 1 is called the fundamental number of the numerical
series N”. (Cassirer 1923: 38)
Rickert agrees with Cassirer’s premises but not with his conclusions.
The schools they respectively represent, the Baden and Marburg Schools of
Neo-Kantianism, converge at “the relationship between mathematics and
the realm of pure logic.”23 In fact, pure logic is certainly included within
mathematics but Rickert balks at the converse. Mathematical objects like
numbers represent sorts of advanced structures which are not ultimately
reducible in the logical terms of predication or in any of its formal
variations, set-predicative theory included. Like Rickert, Natorp24 argues
that “arithmetic is grounded on a series of relations among elements
(namely the natural numbers ordered by the successor relation) . . . but
he adds that the sequence so defined is not identical to or dependent on
22 23
Friedman, A Parting of the Ways, 29. Friedman, A Parting of the Ways, 29.
24
Natorp 1910.
306 luca oliva
temporal sequence.”25 This is a conclusion that Rickert explicitly rejects
along with Natorp’s logicism. Natorp holds that the fundamental
moments of thought do not concern space and time but the separation
and unification of equals in such a way that temporal succession is
excluded. Rickert challenges this conception arguing that as far as the
objects arranged in serial order are homogeneous their identity is not
derived from that order but it is constantly presupposed in the homoge-
neity which allows for the addition of equals. In contrast to Cassirer’s and
Natorp’s logical idealism, Rickert names his position transcendental empiri-
cism, thus emphasizing the alogical components of the concept number.
In conclusion, Rickert accounts for numbers in terms of quantitative
units placed in a progressive series, by involving logical and alogical
components, along with their spatial and temporal properties. The former
are provided by the subject–predicate relation, the latter are introduced by
a counting subject and Dedekind’s relations of places. The nature of
numbers is ultimately decided by their addition, namely the fusion of
equal (homogeneous) parts which results in the variation of quantity
according to a progression. In this sense, Rickert holds that smaller (less
than, <) and larger (more than, >) numbers are qualitative inequalities of
quantity, namely differentiated identities. This brings Rickert’s account of
numbers closer to Kant’s mereology than to the relationism of the Marburg
Neo-Kantians.
25
J. Heis, “Critical philosophy begins at the very point where logistic leaves off: Cassirer’s response to
Frege and Russell,” Perspectives on Science 18/4 (2010), 383–408; here 393.
Primary sources
The reference system in this book is a combination of both author–date and short-
title. The canonic in-text author–date system is used for primary sources, for which
full references are provided in the list below, whereas references to secondary sources
are provided in the footnotes. Secondary sources are cited in full on their first
occurrence, and they are abbreviated in all subsequent occurrences. The decision to
resort to this somewhat unorthodox method is motivated by the overwhelmingly
vast amount of sources cited by the authors. Using exclusively the author–date
system would have required over forty pages of bibliography at the end of the book.
Using exclusively the footnote system would have led to a number of unnecessary
repetitions for works that are cited frequently. Moreover, considering that not all the
protagonists of the Neo-Kantian movement are very well known, it would have
likely blurred the distinction between references to primary and secondary material.
The present list includes all and only the works by Neo-Kantian philosophers and
other major thinkers discussed in the book. It has the advantage of being compact
and yet giving a comprehensive survey of the key texts of Neo-Kantianism, along
with those of major figures in dialogue with the Neo-Kantian tradition.
Aristotle 1976. De anima. Cambridge University Press.
Austin, John 1832. The Province of Jurisprudence Determined. London: John
Murray.
Bauch, Bruno 1923a. Immanuel Kant. Berlin and Leipzig: de Gruyter.
1923b. Wahrheit, Wert und Wirklichkeit. Leipzig: Meiner.
1923c. “Über den Begriff des Naturgesetzes,” Kant-Studien 19: 303–337.
Beneke, Friedrich Eduard 1832. Kant und die philosophische Aufgabe unserer Zeit:
Eine Jubeldenkschrift auf die Kritik der reinen Vernunft. Berlin: Ernst Siegfried
Mittler.
1833. Die Philosophie in ihrem Verhältnisse zur Erfahrung, zur Spekulation und
zum Leben Dargestellt. Berlin: Ernst Siegfried Mittler.
Bona Meyer, Jürgen 1870. Kants Psychologie. Berlin: Hertz.
Brandom, Robert 1994. Making it Explicit: Reasoning, Representing, and Discursive
Commitment. Cambridge, MA: Harvard University Press.
2001. “Reason, Expression, and the Philosophic Enterprise,” in C. P. Ragland
and S. Heidt (eds.), What Is Philosophy? New Haven and London: Yale
University Press, 74–95.
307
308 Primary sources
Brentano, Franz 1874. Psychologie vom empirischen Standpunkt. Hamburg: Meiner.
Carnap, Rudolf 1937. The Logical Syntax of Language. London and New York:
Kegan Paul.
1950. “Empiricism, Semantics, and Ontology,” Revue Internationale de
Philosophie 4: 20–40.
1999. Der logische Aufbau der Welt. Hamburg: Meiner.
Cassirer, Ernst 1902. Leibniz’ System in seinen wissenschaftlichen Grundlagen.
Marburg: Elwert.
1907. “Kant und die moderne Mathematik (Mit Bezug auf Bertrand Russells
und Louis Couturats Werke über die Prinzipien der Mathematik),” Kant-
Studien 12: 1–49.
1912. “Hermann Cohen und die Erneuerung der Kantischen Philosophie,”
Kant-Studien 17: 252–273.
1923. Substance and Function. Chicago and London: Open Court.
1927. “Erkenntnistheorie nebst den Grenzfragen der Logik und Denkpsychologie,”
Jahrbücher der Philosophie 3: 31–92.
1943. “Hermann Cohen 1842–1918,” Social Research: An International Quarterly
of Political and Social Science 10: 219–232.
1944. An Essay on Man: An Introduction to a Philosophy of Human Culture. New
Haven and London: Yale University Press.
1946. The Myth of the State. New Haven: Yale University Press.
1951. The Philosophy of the Enlightenment. Princeton University Press.
1954. Philosophie der symbolischen Formen. Oxford: Bruno Cassirer.
1956a. Determinism and Indeterminism in Modern Physics. New Haven: Yale
University Press.
1956b. Wesen und Wirkung des Symbolbegriffs. Darmstadt: Wissenschaftliche
Buchgesellschaft.
1963. Rousseau, Kant and Goethe. New York: Harper & Row.
1971. Das Erkenntnisproblem. Hildesheim: Georg Olms.
1975. Freheit und Form. Studien zur deutschen Geistesgeschichte. Darmstadt:
Wissenschaftliche Buchgesellschaft.
1979. Symbol, Myth, and Culture: Essays and Lectures of Ernst Cassirer, 1935–1945.
New Haven: Yale University Press.
1981. Kant’s Life and Thought. New Haven and London: Yale University Press.
1985. The Philosophy of Symbolic Forms, vol. iii. New Haven: Yale University Press.
1993. Erkenntnis, Begriff, Kultur. Hamburg: Meiner.
1994a. Das Erkenntnisproblem in der Philosophie und Wissenschaft der neueren
Zeit, 4 parts. Darmstadt: Wissenschaftliche Buchgesellschaft.
1994b. Substanzbegriff und Funktionsbegriff. Untersuchungen über die Grundfragen
der Erkenntniskritik. Darmstadt: Wissenschaftliche Buchgesellschaft.
1995. Zur Metaphysik der symbolischen Formen. Hamburg: Meiner.
1996. The Philosophy of Symbolic Forms, vol. iv. New Haven: Yale University
Press.
1998. Leibniz’s System in seinen wissenschaftlichen Grundlagen. In Gesammelte
Werke, vol. i. Hamburg: Meiner.
Primary sources 309
2000. Substanzbegriff und Funktionsbegriff. In Gesammelte Werke, vol. vi.
Hamburg: Meiner.
2001. Kants Leben und Lehre. In Gesammelte Werke, vol. viii. Hamburg: Meiner.
2004. Kulturphilosophie. Vorlesungen und Vorträge 1929–1941. Hamburg: Meiner.
2006. An Essay on Man: An Introduction to a Philosophy of Human Culture. In
Gesammelte Werke, vol. xxxiii. Hamburg: Meiner.
2009a. Ausgewählter wissenschaftlicher Briefwechsel. Hamburg: Meiner.
2009b. Schriften zur Philosophie der symbolischen Formen. Hamburg: Meiner.
2010. Philosophie der symbolischen Formen, vol. iii. Hamburg: Meiner.
Cohen, Hermann 1877. Kants Begründung der Ethik. Berlin: Dümmler.
1889. Kants Begründung der Ästhetik. Berlin: Dümmler.
1902. System der Philosophie, part i: Logik der reinen Erkenntnis. Berlin: Bruno
Cassirer.
1904. Ethik des reinen Willens. Berlin: Bruno Cassirer.
1915. Der Begriff der Religion im System der Philosophie. Giessen: Alfred Töpelmann.
1928. “Einleitung mit kritischem Nachtrag zu Langes Geschichte des
Materialismus,” in Schriften zur Philosophie und Zeitgeschichte. Berlin:
Akademie Verlag, 1–29.
1966. Die Religion der Vernunft aus den Quellen des Judentums. Darmstadt: Joseph
Melzer Verlag.
1977. Logik der reinen Erkenntnis. In Werke, vol. v. Hildesheim, Zürich, and
New York: Olms.
1984. DasPrinzip der Infinitesimal-Methode und seine Geschichte. In Werke, vol.
v. Hildesheim, Zürich, and New York: Olms.
1987. Kants Theorie der Erfahrung. In Werke, vol. i.3. Hildesheim, Zürich, and
New York: Olms.
1995. Religion of Reason out of the Sources of Judaism. Simon Kaplan. Atlanta, GA:
Scholars Press.
1996. Der Begriff der Religion im System der Philosophie. In Werke, vol. x.
Hildesheim, Zürich, and New York: Olms.
2001a [1981]. Ethik des reinen Willens. In Werke, vol. vii. Hildesheim, Zürich,
and New York: Olms.
2001b. Kants Begründung der Ethik. In Werke, vol. viii. Hildesheim, Zürich, and
New York: Olms.
Comte, Auguste 1903. A Discourse on the Positive Spirit. London: William Reeves.
Dedekind, Richard 1901. “Continuity and Irrational Numbers,” in Essays on the
Theory of Numbers. Chicago: Open Court, 1–27.
Deleuze, Gilles 1969. Logique du sens. Paris: Les Éditions de Minuit.
1980. Milles plateaux. Paris: Les Éditions de Minuit.
Derrida, Jacques 1972. “La forme et le vouloir-dire. Note sur la phénoménologie
du langage,” in Marges de la Philosophie. Paris: Les Éditions de Minuit, 25–55.
Dilthey, Wilhelm 1960. Weltanschauungslehre: Abhandlungen zur Philosophie der
Philosophie, Gesammelte Schriften, vol. viii. Göttingen: Vandenhoeck &
Ruprecht.
310 Primary sources
Du Bois-Reymond, Emil 1874. “The Limits of Our Knowledge of Nature,” The
Popular Science Monthly 5/2: 17–32.
Durkheim, Emil 1997. The Division of Labor in Society. New York: The Free Press.
Frege, Gottlob 1884. Die Grundlagen der Arithmetik. Breslau: Koebner.
1893. Grundgesetze der Arithmetik, vol. i. Jena: Pohle.
1918. “Der Gedanke. Eine logische Untersuchung,” Beiträge zur Philosophie des
Deutschen Idealismus 1: 57–77. English translation: “The Thought: A Logical
Inquiry,” trans. A. M. Marcelle Quinton, Mind 65 (1956): 289–311.
1977. Begriffsschrift und andere Aufsätze. Hildesheim, Zürich, and New York:
Olms.
1980. Funktion, Begriff, Bedeutung. Göttingen: Vandenhoeck & Ruprecht.
1985. Nachgelassene Schriften. Hamburg: Meiner.
1993. Logische Untersuchungen. Göttingen: Vandenhoeck & Ruprecht.
Fries, Jacob Friedrich 1807. Neue oder anthropologische Kritik der Vernunft.
Heidelberg: Carl Winter.
Goethe, Johann Wolfgang von 1907. Maximen und Reflexionen. In Schriften der
Goethe-Gesellschaft, vol. xxi. Weimar: Verlag der Goethe-Gesellschaft.
Habermas, Jürgen 1976. On the Pragmatics of Social Interaction, trans. B. Fultner.
Cambridge: MIT Press.
1985. The Theory of Communicative Action: Reason and the Rationalization of
Society, vol. i. Boston: Beacon Press.
1987. The Theory of Communicative Action: Lifeworld and System: A Critique of
Functionalist Reason, vol. ii. Boston: Beacon Press.
1993. “Themes in Postmetaphysical Thinking,” in Postmetaphysical Thinking.
Philosophical Essays. Cambridge, MA: MIT Press, 28–53.
1996. Between Facts and Norms: Contributions to a Discourse Theory of Law and
Democracy. Cambridge, MA: MIT Press.
2005. “Faith and Knowledge,” in E. Mendieta (ed.), The Frankfurt School on
Religion: Key Writings by the Major Thinkers. New York: Routledge, 101–115.
2008. “Religion in the Public Sphere,” in Between Naturalism and Religion.
Philosophical Essays. Malden, MA: Polity Press, 114–148.
2009. “What is Meant by ‘Post-Secular Society’? A Discussion on Islam in
Europe,” in Europe: The Faltering Project. Malden, MA: Polity Press, 59–78.
Hammond, William 1901. Review of W. Windelband, Platon, The Philosophical
Review 10/4: 430–436.
Hartmann, Eduard von 1923. Kategorienlehre. Berlin: Dunkers.
Hegel, Georg Wilhelm Friedrich 1951. Wissenschaft der Logik: Zwei Teilen. Leipzig:
Meiner.
1952. Phänomenologie des Geistes. Hamburg: Meiner.
1955. Grundlinien der Philosophie des Rechts. Hamburg: Meiner.
1977. The Phenomenology of Spirit. Oxford University Press.
1991. Enzyklopädie der philosophischen Wissenschaften im Grundrisse (1830).
Hamburg: Meiner.
Heidegger, Martin 1987. Zur Bestimmung der Philosophie. In Gesamtausgabe, vol.
lvi–lvii. Frankfurt am Main: Klostermann.
Primary sources 311
1991. Kant und das Problem der Metaphysik. In Gesamtausgabe, vol. iii. Frankfurt
am Main: Klostermann.
2005. Grundprobleme der Phänomenologie. In Gesamtausgabe, vol. lviii.
Frankfurt am Main: Klostermann.
Helmoltz, Herrman von 1995. “On the Relation of Optics to Painting,” in Science
and Culture: Popular and Philosophical Essays. University of Chicago Press,
279–308.
Hume, David 1985.“Of the Original Contract,” in Essays: Moral, Political and
Literary. Indianapolis: Liberty Fund, 465–487.
Husserl, Edmund 1913. Ideen zu einer reinen Phänomenologie und phänomenologische
Philosophie, book i: Allgemeine Einführung in die reine Phänomenologie. Halle:
Niemeyer.
1968. Logische Untersuchungen, zweiter Band, erster Teil. Tübingen: Niemeyer.
1969. Formal and Transcendental Logic. In Husserliana, vol. xvii. The Hague:
Martinus Nijhoff.
1971. Ideen zu einer reinen Phänomenologie und phänomenologische Philosophie,
Drittes Buch: Die Phänomenologie und die Fundamente der Wissenschaften. In
Husserliana, vol. v. The Hague: Martinus Nijhoff.
1984. Logische Untersuchungen, vol. ii: Untersuchungen zur Phänomenologie und
Theorie der Erkenntnis, part i. In Husserliana, vol. xix.1. The Hague:
Martinus Nijhoff.
1989. Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy,
book ii: Studies in the Phenomenology of Constitution. Dordrecht: Kluwer.
1994. Briefwechsel, vol. v. The Hague: Kluwer.
2014. Ideas for a Pure Phenomenology and Phenomenological Philosophy, book i:
General Introduction to Pure Phenomenology. Indianapolis: Hackett.
Kant, Immanuel 1950. Prolegomena to Any Future Metaphysics. Indianapolis:
Bobbs-Merrill.
1956. Kritik der reinen Vernunft. Hamburg: Meiner.
1978. Prolegomena zu einer jeden künftigen Metaphysik die als Wissenschaft wird
auftreten können. Frankfurt am Main: Suhrkamp.
1981. Grounding for the Metaphysics of Morals [Groundwork]. Indianapolis:
Hackett.
1983. Perpetual Peace and Other Essays. Indianapolis: Hackett.
1996. Practical Philosophy. Cambridge University Press.
1997. Lectures on Ethics. Cambridge University Press.
1998. Critique of Pure Reason. Cambridge University Press.
1999. Religion within the Boundaries of Mere Reason and Other Writings.
Cambridge University Press.
2001. Lectures on Metaphysics. Cambridge University Press.
Kelsen, Hans 1911. Hauptprobleme der Staatsrechtslehre, entwickelt aus der Lehre
vom Rechtsatze. Freiburg and Leipzig: Mohr.
1998. “The Pure Theory of Law, ‘Labandism,’ and Neo-Kantianism: A Letter to
Renato Treves,” in Normativity and Norms: Critical Perspectives on Kelsenian
Themes. Oxford: Clarendon Press, 111–119.
312 Primary sources
Kries, Johannes von 1888. “Über den Begriff der objectiven Möglichkeit und einige
Anwendungen desselben,” Vierteljahresschrift für wissenschaftliche Philosophie
12: 179–240, 287–323, 393–428.
Kuhn, Thomas 1970. The Structure of Scientific Revolutions. University of Chicago
Press.
1997. The Essential Tension: Selected Studies in Scientific Tradition and Change.
University of Chicago Press.
Lange, Friedrich Albert 1887. Geschichte des Materialismus und Kritik seiner
Bedeutung in der Gegenwart. Iserlohn and Leipzig: J. Baedeker.
Laplace, Pierre Simon de 1902. A Philosophical Essay on Probabilities. New York:
John Wiley & Sons; London: Chapman & Hall.
Lask, Emil 1905. Rechtsphilosophie. Heidelberg: Carl Winter.
1911. Die Logik der Philosophie und die Kategorienlehre.
1923. “Rechtsphilosophie,” in Gesammelte Schriften, vol. i. Freiburg and Leipzig:
Mohr.
Leibniz, Gottfried Wilhelm 1996. Neue Abhandlungen über den menschlichen
Verstand. Hamburg: Meiner.
Levy, Heinrich 1927. Die Hegel-Renaissance in der deutschen Philosophie.
Charlottenburg: Pan.
Liebmann, Otto 1880. Zur Analysis der Wirklichkeit. Strasbourg: Trübner.
1912. Kant und die Epigonen: Eine kritische Abhandlung. Berlin: Reuther &
Reichard.
Lotze, Rudolf Hermann 1879. Metaphysik. Leipzig: Hirzel.
1882. Grundzüge der Religionsphilosophie. Leipzig: Hirzel.
1902. Grundzüge der Logik und Encyclopädie der Philosophie. Leipzig: Hirzel.
1989a. Logik Erstes Buch: Vom Denken. Hamburg: Meiner.
1989b. Logik Drittes Buch: Vom Erkennen. Hamburg: Meiner.
MacIntyre, Alasdair 1999. Dependent Rational Animals: Why Human Beings Need
the Virtues. Chicago: Open Court.
Marck, Siegfried 1929. Die Dialektik in der Philosophie der Gegenwart. Freiburg
and Leipzig: Mohr.
Marion, Jean-Luc 1989. Réduction et donation. Paris: PUF.
Maxwell, James Clerk 1859. “On the Stability of Saturn’s Rings,” Monthly Notices
of the Royal Astronomical Society 19: 297–304.
1995. “Drafts of lecture on ‘Molecules’, c. August – September 1873,” in The
Scientific Letters and Papers of James Clerk Maxwell. Cambridge University
Press, 922–933.
Menger, Carl 1963. Problems of Economics and Sociology. University of Illinois Press.
Merleau-Ponty, Maurice 1968. The Visible and the Invisible. Evanston:
Northwestern University Press.
2012. Phenomenology of Perception. London: Routledge.
Mill, John Stuart 1974. A System of Logic Ratiocinative and Inductive: Being a
Connected View of the Principles of Evidence and the Methods of Scientific
Investigation. Toronto and London: University of Toronto Press and
Routledge & Kegan Paul.
Primary sources 313
1994. The Logic of the Moral Sciences. Peru, IL: Open Court.
Mises, Ludwig von 1929. “Soziologie und Geschichte: Epilog zum Methodenstreit
in der Nationalökonomie,” Archiv für Sozialwissenschaft und Sozialpolitik 61/3:
465–512.
Montaigne, Michel de 1958. Essays. London: Penguin.
Nagaoka, Hantarō 1904. “Kinetics of a System of Particles Illustrating the Line and
the Band Spectrum and the Phenomena of Radioactivity,” Philosophical
Magazine 7: 445–455.
Natorp, Paul 1882a. Descartes’ Erkenntnistheorie. Eine Studie zur Vorgeschichte des
Kritizismus. Marburg: Elwert.
1882b. “Galillei als Philosoph,” Philosphische Monatshefte 18: 193–229.
1887. “Ueber objektive und subjektive Begründung der Erkenntnis,” Philosophische
Monatshefte 23: 257–286.
1888. Einleitung in die Psychologie nach kritischer Methode. Freiburg and Leipzig:
Mohr.
1894. Religion innerhalb der Grenzen der Humanität. Freiburg and Leipzig: Mohr.
1901. “Zur Frage der logischen Methode. Mit Beziehung auf Edmund Husserl’s
‘Prolegomena’ zur reinen Logik,” Kant-Studien 6: 270–283.
1904a. Allgemeine Psychologie, in Leitsätzen zu akademischen Vorlesungen. Marburg:
Elwert.
1904b. Sozialpädagogik: Theorie der Willenserziehung auf der Grundlage der
Gemeinschaft. Stuttgart: Frommanns.
1910. Die logischen Grundlagen der exakten Wissenschaften. Leipzig and Berlin:
Teubner.
1912a. Allgemeine Psychologie nach kritischer Methode. Freiburg and Leipzig:
Mohr.
1912b. “Kant und die Marburger Schule,” Kant-Studien 17: 192–221.
1913. “Recht und Sittlichkeit. Ein Beitrag zur kategorialen Begründung der
praktischen Philosophie. Mit besonderer Beziehung auf Hermann Cohens
҅Ethik des reinen Willens’ und Rudolf Stammlers ‘Theorie der
Rechtswissenschaft,’” Kant Studien 18: 1–79.
1917. “Husserl’s Ideen zu einer reinen Phänomenologie,” Logos 7: 224–246.
1918. “Hermann Cohens philosophische Leistung unter dem Gesichtspunkte
des Systems,” Philosophische Vorträge veröffentlicht von der Kant-Gesellschaft
21. Berlin: Reuther & Reichard.
1920. Sozial-Idealismus: Neue Richtlinien sozialer Erziehung. Berlin: Julius
Springer.
1921. Die logischen Grundlagen der exakten Wissenschaften. Leipzig and Berlin:
Teubner.
1925a. Vorlesungen über praktische Philosophie. Erlangen: Verlag der Philosophischen
Akademie.
1925b. Sozialpädagogik. Theorie der Willensbildung auf der Grundlage der
Gemeinschaft. Paderborn: Schöningh.
1981. “On the Objective and Subjective Grounding of Knowledge,” Journal of
the British Society for Phenomenology 12: 245–266.
314 Primary sources
1985. “Leibniz und der Materialismus. Aus dem Nachlaß hrsg. von H.
Holzhey,” Studia Leibnitiana 17: 3–14.
1994. Platons Ideenlehre: Eine Einführung in den Idealismus. Hamburg: Meiner.
2008. Philosophie – ihr Problem und ihre Probleme. Eine Einführung in den
kritischen Idealismus. Göttingen: Ruprecht.
2013. Allgemeine Psychologie nach kritischer Methode. Darmstadt: Wissenschaftliche
Buchgesellschaft.
Plato 1977. Phaedo. Indianapolis: Hackett.
2000. Timaeus. Indianapolis: Hackett.
Popper, Karl 1972. Objective Knowledge: An Evolutionary Approach. Clarendon:
Oxford University Press.
Quine, Willard Van Orman 1969. Ontological Relativity and Other Essays. Columbia
University Press.
Radbruch, Gustav 1914. Grundzüge der Rechtsphilosophie. Leipzig: Quelle and
Meyer.
1950. Rechtsphilosophie; “Legal Philosophy,” in E. W. Patterson (ed.), trans. H.
Wilk, The Legal Philosophies of Lask, Radbruch and Dabin. Cambridge, MA:
Harvard University Press, 43–224.
2003. Rechtsphilosophie (Studienausgabe). Heidelberg: C. F. Müller.
Ranke, Leopold von 1885. Geschichten der romanischen und germanischen Völker
von 1494 bis 1535. Leipzig: Duncker & Humblot.
1890. “Aufsätze zur eigenen Lebensbeschreibung,” in Zur eigenen Lebensgeschichte.
Leipzig: Dunker & Humblot, 32–33.
Rawls, John 1993. Political Liberalism. New York: Columbia University Press.
1999. A Theory of Justice. Cambridge, MA: Belknap Press of Harvard University
Press.
2005. “The Idea of Public Reason Revisited,” in Political Liberalism as
“Lecture ix” (expanded edition). Columbia University Press, 437–490.
2009. A Brief Inquiry into the Meaning of Sin and Faith. Cambridge, MA:
Harvard University Press.
Rickert, Heinrich 1888. Zur Lehre von der Definition. Tübingen: Paul Siebeck.
1896. Die Grenzen der naturwissenschaftlichen Begriffsbildung: Eine logische
Einleitung in die historischen Wissenschaften. Tübingen and Leipzig: Mohr.
1899a. Fichtes Atheismusstreit und die Kantische Philosophie. Berlin: Reuther &
Reichard.
1899b. Kulturwissenschaft und Naturwissenschaft. Tübingen and Leipzig: Mohr.
1900. “Psychophysische Causalität und psychophysischer Parallelismus,” in
Benno Erdmann et al. (eds.), Philosophische Abhandlungen: Christoph Sigwart
zu seinem siebzigsten Geburtstage. Tübingen and Leipzig: Mohr, 59–87.
1902. Die Grenzen der naturwissenschaftlichen Begriffsbildung: Eine logische
Einleitung in die historischen Wissenschaften. Tübingen and Leipzig: Mohr.
1907. “Geschichtsphilosophie,” in W. Windelband (ed.), Die Philosophie im
Beginn des 20. Jahrhunderts: Festschrift für Kuno Fischer. Heidelberg: Carl
Winter’s Universitätsbuchhandlung, 321–342.
Primary sources 315
1909. “Zwei Wege der Erkenntnistheorie. Transcendentalpsychologie und
Transcendentallogik,” Kant-Studien 14: 169–228.
1910. “Vom Begriff der Philosophie,” in Philosophische Aufsätze. Tübingen and
Leipzig: Mohr, 3–36.
1911. “Lebenswerte und Kulturwerte,” Logos 2: 131–166.
1913. Vom System der Werte. Tübingen and Leipzig: Mohr.
1914. “Über logische und ethische Geltung,” Logos 19: 182–221.
1921. System der Philosophie: Erster Teil: Allgemeine Grundlegung der Philosophie.
Tübingen and Leipzig: Mohr.
1924a. Das Eine, die Einheit und die Eins: Bemerkungen zur Logik des Zahlbegriffs.
1924b. Die Probleme der Geschichtsphilosophie: eine Einführung, 2nd edn.
1926. Kulturwissenschaft und Naturwissenschaft. Tübingen and Leipzig: Mohr.
1928. Der Gegenstand der Erkenntnis: Einführung in die Transzendentalphilosophie,
6th edn. Tübingen and Leipzig: Mohr.
1929a. Die Grenzen der naturwissenschaftlichen Begriffsbildung: Eine logische
Einleitung in die historischen Wissenschaften, 5th edn. Tübingen and
Leipzig: Mohr.
1929b. Wilhelm Windelband. Tübingen and Leipzig: Mohr.
1930. Die Logik des Prädikats und das Problem der Ontologie, 2nd edn.
1934. Grundprobleme der Philosophie: Methodologie, Ontologie, Anthropologie.
1939. Unmittelbarkeit und Sinndeutung. Tübingen and Leipzig: Mohr.
1962. Science and History: A Critique of Positivist Epistemology. Princeton,
Toronto, New York, and London: Van Nostrand.
1986. The Limits of Concept Formation in Natural Science: A Logical Introduction
to the Historical Sciences. Cambridge University Press.
1999. Philosophische Aufsätze. Tübingen and Leipzig: Mohr.
Riehl, Alois 1872. Über Begriff und Form der Philosophie: Eine allgemeine
Einführung in das Studium der Philosophie. Berlin: Carl Dunckers.
1903. Zur Einführung in die Philosophie der Gegenwart. Acht Vorträge. Teubner:
Leipzig.
Rosenzweig, Franz 1937. Kleinere Schriften. Berlin: Schocken Verlag.
Rousseau, Jean-Jacques 1984. A Discourse on Inequality. London: Penguin.
Rutherford, Ernest 1911. “The Scattering of α and β Particles of Matter and the
Structure of the Atom,” Philosophical Magazine 21: 669–688.
Schnädelbach, Herbert 1983. Philosophie in Deutschland 1831–1933. Frankfurt am
Main: Suhrkamp.
Schlick, Moritz 1979. “Critical or Empiricist Interpretation of Modern Physics?
Remarks on Ernst Cassirer’s Einstein’s Theory of Relativity,” in Philosophical
Papers, vol. i [1909–1922]. Dordrecht, Boston, and London: Reidel, 327–328.
Simmel, Georg 1890a. “Rembrandt als Erzieher,” Vossische Zeitung (June 1, 1890),
Sunday Supplement 22, cols. 7–10.
316 Primary sources
1890b. Über soziale Differenzierung. Soziologische und psychologische
Untersuchungen. Leipzig: Duncker & Humblot.
1902. “Tendencies in German Life and Though since 1870,” The International
Monthly 5: 93–111 and 166–184.
1911. Philosophische Kultur. Über das Abenteuer, die Geschlechter und die Krise der
Moderne. In Gesammelte Essais. Leipzig: Werner Klinkhardt.
1917. Der Krieg und die geistige Entscheidung. Leipzig: Duncker & Humblot.
1918. Der Konflikt der modernen Kultur. Munich: Duncker & Humblot.
1922. Zur Philosophie der Kunst. Philosophische und kunstphilosophische Aufsätze.
Potsdam: Kiepenheuer.
1953. Rembrandtstudien. Darmstadt: Wissenschaftliche Buchgemeinschaft.
1980. On the History of Philosophy. In Essays on Interpretation in Social Science.
Totowa, NJ: Rowman and Littlefield, 198–204.
1989–2008. Gesamtausgabe in 24 Bände. Frankfurt am Main: Suhrkamp.
1990a. The Philosophy of Money. London: Routledge.
1990b. Vom Wesen der Moderne. Essays zur Philosophie und Ästhetik. Hamburg:
Junius.
1996. Hauptprobleme der Philosophie. In Gesamtausgabe, vol. xiv. Frankfurt am
Main: Suhrkamp.
2005. Rembrandt: An Essay in the Philosophy of Art. New York: Routledge.
2010. The View of Life: Four Metaphysical Essays: University of Chicago Press.
Stadler, August 1912. Kants Teleologie und ihre erkenntnistheoretische Bedeutung.
Berlin: Dümmler.
Stammler, Rudolf 1896a. Die Theorie des Anarchismus. Berlin: Häring.
1896b. Wirtschaft und Recht nach der materialistischen Geschichtsauffassung.
Leipzig: Veit and Comp.
1902. Die Lehre von dem richtigen Rechte. Berlin: Guttentag.
1906. “Wesen des Rechtes und der Rechtswissenschaft,” in Systematische
Rechtswissenschaft, vol. viii, Die Kulture der Gegenwart II. Berlin:
B. G. Teubner, 1–59.
1914. Wirtschaft und Recht nach der materialistischen Geschichtsauffassung. Leipzig:
Veit and Comp.
1922. Lehrbuch der Rechtsphilosophie. Berlin: Walter de Gruyter.
1923a. “Fundamental Tendencies in Modern Jurisprudence,” Michigan Law
Review 21/6: 623–654.
1923b. “Fundamental Tendencies in Modern Jurisprudence (continued),”
Michigan Law Review 21/8: 862–903.
1925. “Legislation and Judicial Decision,” Michigan Law Review 23/4: 362–381.
2000. Die Lehre: The Theory of Justice. Union, NJ: Lawbook Exchange.
Vorländer, Karl 1893. Der Formalismus in der Kantischen Ethik in seiner
Notwendigkeit und Fruchtbarkeit. Marburg: Friedrich.
Weber, Max 1968. Economy and Society: An Outline of Interpretive Sociology, trans.
Guenther Roth and Claus Wittich. New York: Bedminster Press.
1977. Critique of Stammler. New York: Free Press.
Primary sources 317
1978. Economy and Society: An Outline of Interpretive Sociology. Berkeley:
University of California Press.
1982. Gesammelte Aufsätze zur Wissenschaftslehre. Freiburg and Leipzig: Mohr.
1985. Kritische Studien auf dem Gebiet der kulturwissenschaftlichen Logik, in
Gesammelte Aufsätze zur Wissenschaftslehre. Freiburg and Leipzig: Mohr,
215–266.
2012. Collected Methodological Writings. London: Routledge.
Weiszäcker, Karl von 1937. Review of Cassirer’s Determinism and Indeterminism.
Physikalische Zeitschrift 38: 860–861.
Windelband, Wilhelm 1892. Geschichte der Philosophie. Freiburg: Mohr.
1894. Geschichte und Naturwissenschaft: Rede zum Antritt des Rectorats der Kaiser-
Wilhelms-Universität Strassburg. Gehalten am 1. Mai 1894. Strasbourg: Heitz.
1900. Platon. Stuttgart: Frommann.
(ed.) 1907. Die Philosophie im Beginn des 20. Jahrhunderts: Festschrift für Kuno
Fischer. Heidelberg: Carl Winter.
1911. Die Geschichte der neueren Philosophie in ihrem Zusammenhange mit der
Allgemeinenkultur und den besonderen Wissenschaften. Leipzig: Breitkopf &
Härtel.
1915. Präludien: Aufsätze und Reden zur Philosophie und ihrer Geschichte, vol. i.
Tübingen: J. C. B. Mohr.
1956. History of Ancient Philosophy. New York: Dover.
1958a. A History of Philosophy, vol. i: Greek, Roman, Medieval. New York: Harper.
1958b. A History of Philosophy, vol. ii: Renaissance, Enlightenment, and Modern.
New York: Harper.
1980. “History and Natural Science,” History and Theory 19: 165–185.
Zeller, Eduard 1877a. “Über Bedeutung und Aufgabe der Erkenntnistheorie
(1862),” in Vorträge und Abhandlungen II. Leipzig: Fues, 479–496.
1877b. Vorträge und Abhandlungen. Leipzig: Fues.
Index
a priori, ix, 23, 49, 55, 63, 65, 66, 67, 68, 72, 73, 74, Cohen, Hermann, 8, 84, 90, 129, 140, 207, 221
83, 86, 91, 93, 95, 102, 107, 112, 116, 117, 118, Coleman, Jules, 150
122, 127, 130, 131, 132, 135, 136, 154, 168, 194, community, 132, 135, 138, 145, 159, 163, 221, 282
223, 224, 225, 227, 229, 262, 263, 267, 268, Comte, Auguste, 83, 177
273, 275, 276, 277, 279, 280, 281, 285, 286, concept formation, 27, 44, 46, 128, 171, 176, 179,
287, 289, 290, 292, 293, 299 183, 229, 231, 287
a priori cognition, 225, 227 concept of constellation, 184
a priori concepts, 55 concept of function, 230, 265, 266, 269
Adorno, Theodor W., 213 concept of law, 149, 152, 153, 158, 160, 162, 181, 279
analytic philosophy, 59, 60, 61, 62, 63, 64, 65, 66, Concept of Law, The (Hart), 149
67, 68, 69, 70, 71, 72, 75 Conflict of the Faculties, The (Kant), 204
analytic propositions, 287 consciousness, 29, 32, 33, 37, 83, 87, 89, 92, 94, 95,
anti-psychologism, 76 96, 97, 98, 99, 100, 101, 102, 104, 105, 106,
Aristotle, 229, 245, 252, 265, 269 116, 141, 142, 153, 159, 192, 194, 208, 233, 236,
astronomy, 91, 176, 177, 178, 179, 180, 181, 182 241, 242, 243, 244, 245, 246, 247, 248, 249,
Austin, John, 149 250, 251, 252, 253, 254, 255, 256, 257, 258
Austrian philosophy, 64 self-consciousness, 95
autonomy, 52, 53, 54, 56, 92, 124, 131, 132, 133, 138, subjective activity, 52
143 constitution, 128
axiology, 121, 123, 124 continuous quanta, 296
Critique of Practical Reason (Kant), 164, 206
Baudelaire, Charles, 191 Critique of Pure Reason (Kant), 82, 96, 160, 164,
Being and Time (Heidegger), 260 165, 193, 224, 273, 311
Bendemann, Margarete von, 186 culture, vii, 12, 109, 111, 187, 188, 189, 193, 195, 202,
Beneke, Friedrich Eduard, 6, 21, 23 221, 228, 232, 233, 237, 308, 309, 311
Benveniste, Émile, 107
Bergson, Henri, 66, 68, 114, 188, 196 Davos disputation, 2, 6, 68, 92, 106, 235
Berlin Secession, 187, 197 Dedekind, Richard, 266, 304
Berlin Trade Exhibition, 188, 192 Determinism and Indeterminism (Cassirer), 278,
Bloch, Ernst, 188 283, 308, 317
Bolzano, Bernard, 64 Dilthey, Wilhelm, 29, 194
Born, Max, 282 Duhem, Pierre, 266, 272
Brandom, Robert, 81, 121, 237, 238, 307 Dummett, Michael, 61, 73
Brentano, Franz, 64, 67, 73, 87, 98, 245, 308 duty, 143
British empiricism, 64 Dworkin, Ronald, 149, 160
Carnap, Rudolf, 71 Eckhardt, Meister, 301

Cassirer, Ernst, 53, 86, 92, 98, 221, 227, 261 ego, 14, 95, 242, 243, 244, 245, 251, 252, 253, 256,
causality, 15, 25, 134, 137, 140, 153, 157, 174, 278, 259
280, 281, 282, 283 Eliot, T. S., 1
318
Index 319
Einleitung in die Geisteswissenschaften (Dilthey), Grundprobleme der Philosophie (Rickert), 26, 29,
194 315
Einleitung in die Psychologie (Natorp), 241, 313 Gulbenkian Commission, 177
Einstein, Albert, 273
physical theory, 274 Habermas, Jürgen, 13, 14, 85, 148, 207, 213
empirical psychology, 32 Hart, H. L. A., 149
Enlightenment, 57, 92, 190, 204, 213, 223, 235, Hegel, G. W. F., 116, 120, 122, 234
308, 317 “das Wahre ist das Ganze” not: die “Wahrheit
Erkenntnistheorie, see theory of knowledge ist das ‘Ganze’,” 88
Ernst, Paul, 188 Heidegger, Martin, 4, 92, 103, 240, 259
ethical subject, 134, 135, 138, 145, 146 Heisenberg, Werner, 280
ethics, 127, 131, 138, 164, 228 Helmholtz, Hermann von, 85, 113, 183, 262
Ethik des reinen Willens (Cohen), 13, 131, 132, 133, Herz, Marcus, 94
163, 309, 313 Hesse, Mary, 184
Euclid, 286, 287, 291, 292 heterological principle, 26, 36
Euclidean geometry, 8, 271, 276 Hintikka, Jaakko, 285, 290
extensive magnitude, 296 historical sciences, 39, 41, 42, 43, 48, 55, 57, 171,
172, 174, 175, 176, 179
Fichte, Johann Gottlieb, 124 Hohfeld, Wesley, 166
Frank, Philipp, 278 Horkheimer, Max, 213
Frankfurt School, 213, 221 human sciences (Geisteswissenschaften), 8, 29, 39,
freedom, 14, 25, 31, 40, 52, 53, 54, 56, 112, 120, 124, 40, 128, 131, 132, 166, 169
125, 126, 133, 134, 140, 144, 159, 201, 204, 236, Husserl, Edmund, 73, 82, 240, 245
283
Frege, Gottlob, 60, 62, 64, 66, 69, 71, 72, 73, 75, ideal type, 183
77, 78, 79, 81, 135, 266, 310 idealism, 22, 23, 40, 52, 57, 59, 63, 64, 65, 68, 71,
Friedman, Michael, 68 73, 82, 91, 112, 115, 120, 126, 151, 153, 223, 225,
functional concept, 230, 232, 269, 274 226, 228, 229, 232, 272, 276, 306
functional theory of concept, 270 symbolic idealism, 232
Funktion und Begriff (Frege), 79 Ideen I (Husserl), 82, 84, 89, 93, 99, 100, 101, 103,
241, 245, 257
Gabriel, Gottfried, 78, 79 idiographic, 13, 42, 171, 185
Gegenstand der Erkenntnis, Der (Rickert), Ignorabimus controversy, 178
135, 315 Impressionist painting, 187
General Psychology according to the Critical intentionality, 87, 96, 98, 99, 100, 101, 125
Method (Natorp), 240 intuition, 25, 27, 29, 48, 55, 75, 89, 90, 91, 93, 94,
Geny, François, 152 166, 195, 224, 231, 232, 253, 260, 262, 265,
George, Stefan, 188 269, 271, 275, 285, 286, 288, 289, 290, 291,
German Idealism, i, 3, 4, 7, 8, 65, 86, 97, 98, 113, 292, 294, 302
116, 120 invariant theory of experience, 267, 274
Fichte, 124, 133
Hegel, 87 Kant, Immanuel, 67, 93, 94, 104, 118, 143, 154, 157,
Hegel’s speculative idealism, 22 203, 222, 224, 288, 307
God, 19, 23, 129, 191, 192, 203, 205, 206, Kant und die Epigonen (Liebmann), 64, 88, 115,
209, 214 312
Goethe, Johann Wolfgang von, x, xi, 56, 57, 103, Kant und die philosophische Aufgabe unserer Zeit
196, 200, 232, 261, 271, 274, 310 (Beneke), 6, 23, 307
Goodman, Nelson, 105 Kant’s practical philosophy, 56
Grenzen der naturwissenschaftlichen Kants Begründung der Ethik (Cohen), 129
Begriffsbildung: eine logische Einleitung in die Kants Theorie der Erfahrung (Cohen), 85, 86, 97,
historischen Wissenschaften, Die (Rickert), 44 129, 262, 309
Groundwork for the Metaphysics of Morals (Kant), Kelsen, Hans, 13, 148, 161, 167
164 Kitcher, Philip, 20
Grundgesetze der Arithmetik (Frege), 79, 310 Kniel, Gertrude, 187
320 Index
Kries, Johannes von, 183 Natorp, Paul, 48, 90, 141, 144, 221, 240
Kuehn, Manfred, 2, 147, 204 natural attitude, 89, 257
Kuhn, Thomas, 283 natural sciences, 6, 7, 9, 20, 22, 24, 27, 33, 39, 40,
Kulturwissenschaften, 42, 47 41, 42, 43, 44, 47, 48, 62, 66, 70, 71, 73, 82,
83, 85, 130, 131, 161, 163, 171, 172, 173, 174, 175,
Langbehn, Julius, 197 176, 177, 178, 179, 181, 183, 184, 185, 225, 261,
Laplace, Pierre Simon de, 176 264, 273, 278, 302
Lask, Emil, 4, 13, 63, 111, 114, 120, 148, 151, 168, 312, naturalism, 74
314 Nietzsche, Friedrich, 189, 195
Laue, Max von, 282 noema, 99, 245, 255
Lebensphilosophie, 14, 68, 69, 117, 195, 196, noesis, 255
Lebenswelt, 84 nomothetic, 13, 42, 171, 185
Leibniz, Gottfried Wilhelm, 71, 143 non-Euclidean geometries, 271
principle of indiscernibles, 292 normativity, 77, 81, 123, 128, 130, 132, 134, 150, 169
system, 261 legal norm, 155
Liebl, Wilhelm, 197 numbers, 24, 66, 75, 76, 79, 80, 226, 240, 250,
Liebmann, Otto, 63, 76, 88, 115 269, 285, 286, 293, 294, 295, 297, 300, 301,
linguistic turn, 19, 20, 22, 37, 62, 67, 70 302, 303, 304, 305, 306
logic, 8, 22, 39, 41, 48, 54, 62, 64, 66, 69, 70, 71,
72, 74, 76, 77, 79, 80, 88, 104, 112, 116, 120, One, the Unity and the Number One, The
121, 122, 130, 132, 135, 136, 138, 139, 163, 166, (Rickert), 297
194, 227, 228, 229, 230, 231, 232, 233, 234, ontological pluralism, 37, 130
236, 248, 254, 256, 257, 264, 267, 268, 269, ontology, 230
281, 289, 291, 297, 298, 299, 300, 301, 302, ought (Sollen), 77, 118, 123, 128, 129, 130, 132, 136,
305, 311, 312, 313 137, 141, 142, 201
Logical Investigations (Husserl), 90, 100, 106, 241,
251, 269 Parsons, Talcott, 290
logical laws, 30, 76, 77, 79 perception, 1, 2, 5, 9, 53, 79, 94, 95, 96, 98, 102,
logical positivists, 60, 62, 65, 67, 69, 72 103, 104, 107, 119, 137, 154, 172, 194, 199, 246,
Logical Syntax of Language (Carnap), 69, 308 274
Logik der reinen Erkenntnis (Cohen), 262 phenomenological reduction, 83
Logische Aufbau der Welt (Carnap), 71 phenomenology, 82, 84, 87, 91, 101, 107, 241, 247,
Lotze, Rudolf Hermann, 72 251, 253
Lukács, Georg, 4, 188 transcendental phenomenology, 83, 91, 93
Phenomenology of Spirit (Hegel), 22
Main Problems of Philosophy (Rickert), 298 Philosophie der Symbolischen Formen (Cassirer), 70
Marburg School, 5, 8, 11, 14, 39, 48, 51, 63, 71, 85, Philosophie in ihrem Verhältnisse zur Erfahrung,
86, 114, 118, 122, 147, 151, 162, 163, 207, 221, zur Spekulation und zum Leben Dargestellt,
222, 223, 224, 225, 227, 228, 229, 237, 238, Die (Beneke), 24, 307
239, 261, 263, 275 Philosophy of Arithmetic (Husserl), 289, 290
Marx, Karl, 188 philosophy of culture, 12, 111, 112, 114, 117, 118, 120,
mathematics, viii, 15, 285, 290, 291, 293, 295, 297, 123, 124, 125, 127, 223, 224, 228, 229, 234, 235,
302 236, 261, 278
Maxwell, James Clerk, 184, 312 philosophy of law, 12, 13, 147, 148, 149, 150, 151,
McDowell, John, 90 152, 162, 166, 170
McNaughton, David, 20 philosophy of psychology, 241
Menger, Carl, 183 Philosophy of Money (Simmel), 191, 192, 316
Merleau-Ponty, Maurice, 104 Philosophy of Symbolic Forms (Cassirer), 5, 86, 87,
metaphysics, 23, 62, 64, 67, 69, 71, 72, 73, 86, 88, 88, 94, 96, 101, 104, 105, 222, 224, 233, 261,
91, 92, 95, 112, 116, 119, 148, 151, 164, 177, 192, 308
265 Plato, x, 1, 11, 34, 48, 51, 52, 53, 54, 57, 112, 115, 120,
Mises, Ludwig von, 185 130, 141, 209, 210, 225, 283, 310, 314
Platos Ideenlehre (Natorp), 52, 314
Nagaoka, Hantarō, 184 Poincaré, Henri, 266
Index 321
Political Liberalism (Rawls), 211, 314 Schopenhauer and Nietzsche (Simmel), 195
political philosophy, i, 68, 158, 210 science of the whole, 88, 107, 123
power of metaphors, 184 whole whole as a whole, 24
praxis, 5, 43, 144, 146 world as a whole, 28, 31
problem of constitution, 89, 94, 106 science of the whole (Ganzheitswissenschaft), 24
Problem of Ontology (Rickert), 298 self-consciousness, 32, 95, 142, 243
Province of Jurisprudence Determined, The Simmel, Georg, 29, 186–202
(Austin), 149, 307 Sinn und Bedeutung (Frege), 80
psychologism, 74, 77, 97, 194 Sluga, Hans 78
psychology, i, 5, 7, 13, 14, 22, 26, 32, 33, 36, 40, 41, social materialism, 151, 153, 154
42, 66, 69, 71, 72, 74, 77, 83, 95, 97, 99, 105, social sciences, 13, 26, 60, 65, 154, 161, 171, 172, 175,
119, 167, 171, 173, 175, 178, 181, 187, 194, 201, 178, 181, 185
240, 241, 242, 243, 244, 245, 247, 248, 249, social theory, 150, 169
250, 251, 254, 255, 256, 258, 259, 260 sociology, 177, 180
sociology of religion, 214
quantum mechanics, 15, 273, 278, 279, 280, 281, Socrates, 20, 45, 53
282 Sollen, see ought
Quine, Willard Van Orman, 68 Southwest School, 12, 39, 40, 41, 48, 57, 60, 77,
80, 111, 114, 118, 120, 122, 147, 151, 152, 168, 221
space and time, 81, 160, 174, 179, 224, 231, 262,
Radbruch, Gustav, 148 267, 274, 275, 277, 281, 288, 296, 306
rational cosmology, 25 spirit, i, x, 97, 98, 104, 105, 309, 310
Rawls, John, 14, 207, 210, 214 Stammler, Rudolf, 151
reality, 1, 22, 24, 25, 26, 27, 28, 30, 32, 33, 34, 35, 36, subjectivity, 12, 14, 97, 99, 102, 105, 107, 111, 122,
37, 40, 42, 46, 56, 67, 68, 70, 73, 91, 97, 106, 125, 240, 241, 242, 243, 244, 245, 247, 249,
113, 116, 117, 119, 121, 122, 127, 128, 129, 130, 254, 255, 256, 259, 260
131, 135, 136, 137, 138, 139, 140, 141, 142, 145, substance, 53, 92, 99, 143, 195, 198, 229, 230, 265,
146, 172, 173, 175, 179, 180, 181, 182, 190, 194, 266
195, 200, 206, 225, 231, 263, 264, 268, 277, Substance and Function (Cassirer), 15, 94, 229,
279, 297, 299, 302, 303 261, 266, 269, 270, 284, 288, 305, 308
reason, 48, 52, 55, 89, 142, 217 symbolic forms, 12, 85, 86, 88, 94, 96, 101, 102,
Reichenbach, Hans, 277 105, 106, 231, 232, 233, 234, 235, 236, 238, 277,
relational concepts, 173, 174, 175, 176, 179, 180, 182 278
relational theory of concept, 265 symbolic pregnance, 96, 101, 102, 103, 104, 105,
Religion der Vernunft aus den Quellen des 107
Judentums (Cohen), 208, 309 synthetic a priori knowledge, 23
religion of reason, 14, 205, 206, 208, 209, 210, 213, synthetic propositions, 69, 287, 288
215, 217
Religion within the Boundaries of Mere Reason,
(Kant), 203, 205, 206, 311 Theory of Communicative Action, The
Rembrandt, i, viii, 14, 186, 196, 197, 198, 199, 200, (Habermas), 214, 216
201, 316 Theory of Definition, The (Rickert), 302
Rembrandt: An Essay in the Philosophy of Art, Theory of Justice, A (Rawls), 186, 187, 201, 316
(Simmel), 186, 316 theory of knowledge (Erkenntnistheorie), 6, 7, 14,
Rickert, Heinrich, x, 2, 10, 15, 21, 26, 36, 78, 111, 26, 45, 50, 63, 66, 67, 71, 72, 74, 77, 88, 93,
117, 120, 129, 135, 140, 148, 171, 180, 297 111, 113, 116, 117, 120, 121, 136, 151, 164, 175,
Riehl, Alois, 21 208, 222, 224, 228, 247, 250, 258, 262, 266,
Rilke, Rainer Maria, 188 273
Rodin, Auguste, 187 theory of relativity, 70, 273, 274, 276, 277, 278,
Rousseau, Jean-Jacques, 54 282
Rousseau, Kant and Goethe (Cassirer), 52, 54, 308 time-consciousness, 101, 244
Rutherford, Ernest, 184 tragedy of culture, 193
Transcendental Aesthetics, 231, 262, 288
transcendental deduction, 95, 96, 98, 105, 267
schema, 101, 106, 269, 271, 274, 280, 293, 294 transcendental idealism, 92, 119, 122, 134,
Schlick, Moritz, 275 137, 226
322 Index
transcendental method, 14, 86, 164, 165, 167, 225, Vorlesungen über praktische Philosophie (Natorp),
226, 228, 229, 231, 263, 264, 273 135, 141, 313
transcendental philosophy, 14, 63, 74, 77, 82, 85,
112, 116, 118, 125, 126, 160, 163, 222, 233, 263, Warburg, Aby, 86
267, 289 Weber, Max, 181, 182
transcendental subjectivity, 74, 84, 99, 107 Weisse, Christian, 8
truth, ix, 46, 47, 77, 79, 80, 118, 291 Weizsäcker Carl, Friedrich von,
282
validity (Geltung), vii, 13, 35, 42, 46, 74, 77, 112, Weltanschauung, 29, 30, 113, 114
115, 119, 128, 129, 130, 135, 174, 289 philosophical worldview, 31
value (Wert), 35, 42, 43, 47, 77, 112, 125, 140 Windelband, Wilhelm, 41, 77
Vienna Circle, 64, 68, 71, 72, 221, 278 Wirtschaft und Recht (Stammler), 151, 152,
View of Life, The (Simmel) 186, 201–202 155, 316
volition, 22, 51, 133, 135, 138, 141, 143, 149, 159, 168, Wittgenstein, Ludwig, 81
216
Vom Begriff der Philosophie (Rickert), 31, 33, 315 Zeller, Eduard, 6, 88
Vom System der Werte (Rickert), 139 Zurück zu Kant! (Back to Kant!), 8

Staiti, Andrea - Warren, Nicolas de - New Approaches To Neo-Kantianism-Cambridge University Press (2015) PDF

Uploaded by

Copyright:

Available Formats

You might also like

Staiti, Andrea - Warren, Nicolas de - New Approaches To Neo-Kantianism-Cambridge University Press (2015) PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Staiti, Andrea - Warren, Nicolas de - New Approaches To Neo-Kantianism-Cambridge University Press (2015) PDF

Uploaded by

Copyright:

Available Formats

NEW APPROACHES TO NEO-KANTIANISM

After the demise of German Idealism, Neo-Kantianism ﬂourished as

nicolas de warren is Research Professor of Philosophy at the

andrea staiti is Associate Professor of Philosophy at Boston

Cambridge University Press is part of the University of Cambridge.

List of contributors page ix

Introduction: towards a reconsideration of Neo-Kantianism 1

part i neo-kantianism and philosophy 17

part ii ethics and culture 109

part iii the theory of knowledge 219

Primary sources 307

beatrice centi is Professor of History of Philosophy at the University

We would like to thank Andrea Cimino, Leonardo Manfrini, Kevin

In the summer of 1914, T. S. Eliot arrived in Marburg from Harvard

The Neo-Kantians on the meaning

The world-whole and the world’s parts

Neo-Kantian ideas of history

The Southwest School

Neo-Kantianism and analytic philosophy

Analytic philosophy, Austrian philosophy, and Kantianism

The “reﬂective turn” in Neo-Kantianism and analytic philosophy

Neo-Kantianism and science

Lotze, Liebmann, and Frege

Frege and the Southwest German School

Reise um die Welt: Cassirer’s cosmological

Il faut être absolument moderne

Nous sommes embarqués

Die Sachen Selbst

Husserl or the modern necromancer

Spirit or the authentic Proteus

Philosophy as philosophy of culture?

Neo-Kantianism: philosophy as philosophy of culture

Three Hegelian challenges

The validity of norms in Neo-Kantian ethics

Ought and being in Cohen

Ought, validity, and values in Rickert

Structure and function of the ought in Natorp

Neo-Kantianism in the philosophy

Between facts and norms: the dual nature of law

Stammler and the three tasks of legal philosophy

The concept of law on the basis of a concept of society

The legitimacy of law’s coercion

The social ideal and correct law (richtiges Recht)

Cohen’s methodological analogy: law as the mathematics

Conclusion: a conjecture and a hypothesis

Neo-Kantianism and the social sciences: from

The legacy of the Southwest German School

Natural sciences and historical sciences

Astronomy, of all sciences!

Simmel’s Rembrandt and The View of Life

Modern society and the aesthetic point of view

Simmel and Neo-Kantianism

The turn to the philosophy of life (Lebensphilosophie)

Simmel’s Rembrandt and the individual law

The View of Life

The binding of Isaac and the boundaries of reason:

Cohen and the religion of reason

Rawls on comprehensive doctrines